From makc@iinet.net.au Thu Apr 1 00:19:38 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.6 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o315Jbpl093054 for ; Thu, 1 Apr 2010 00:19:38 -0500 X-ASG-Debug-ID: 1270099281-66ce01e70000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from outbound.icp-qv1-irony-out4.iinet.net.au (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 74F40173A46F for ; Wed, 31 Mar 2010 22:21:21 -0700 (PDT) Received: from outbound.icp-qv1-irony-out4.iinet.net.au (outbound.icp-qv1-irony-out4.iinet.net.au [203.59.1.104]) by cuda.sgi.com with ESMTP id 1FHiaeKvigAXMllv for ; Wed, 31 Mar 2010 22:21:21 -0700 (PDT) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AooFAN/Js0vL2Rwk/2dsb2JhbACPRIt1cbRbhQEE X-IronPort-AV: E=Sophos;i="4.51,346,1267372800"; d="scan'208";a="528781361" Received: from unknown (HELO alma.crabbed.net) ([203.217.28.36]) by outbound.icp-qv1-irony-out4.iinet.net.au with ESMTP; 01 Apr 2010 13:21:05 +0800 Received: by alma.crabbed.net (Postfix, from userid 501) id EB525E8945; Thu, 1 Apr 2010 16:21:03 +1100 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <19380.11583.660939.585636@iinet.net.au> Date: Thu, 1 Apr 2010 16:21:03 +1100 From: Max Matveev To: pcp@oss.sgi.com X-ASG-Orig-Subj: Missing commit - pmdaproc changes to deal with Solaris' ping Subject: Missing commit - pmdaproc changes to deal with Solaris' ping X-Mailer: VM 7.17 under 21.4 (patch 19) "Constant Variable" XEmacs Lucid X-Barracuda-Connect: outbound.icp-qv1-irony-out4.iinet.net.au[203.59.1.104] X-Barracuda-Start-Time: 1270099284 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0757 1.0000 -1.5399 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.54 X-Barracuda-Spam-Status: No, SCORE=-1.54 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.26394 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Nathan, Martin, It looks like the change which I've sent on Australia Day never made it into dev or master: Small fix to teach pmdaproc about Solaris' ping. The following changes since commit 4318547118e4ab304bf4c2acf6baf7e6c116fb00: Nathan Scott (1): Bump development branch version to 3.1.0 for planned release. are available in the git repository at: git://oss.sgi.com/makc/pcp pmdaproc Max Matveev (1): Teach pmdaproc about Solaris' ping. src/pmcd/pmdaproc.sh | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) Do you mind if I push it? I'll also push in my change to reserve domain name for IPMI pmda. max From mailman-bounces@oss.sgi.com Thu Apr 1 05:04:12 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.8 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00, FH_DATE_PAST_20XX,J_CHICKENPOX_25 autolearn=no version=3.3.0-rupdated Received: from oss.sgi.com (localhost [127.0.0.1]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o31A0hrd112167 for ; Thu, 1 Apr 2010 05:00:43 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Subject: oss.sgi.com mailing list memberships reminder From: mailman-owner@oss.sgi.com To: pcp@oss.sgi.com X-No-Archive: yes Message-ID: Date: Thu, 01 Apr 2010 05:00:27 -0500 Precedence: bulk X-BeenThere: mailman@oss.sgi.com X-Mailman-Version: 2.1.11 List-Id: Mailman mail X-List-Administrivia: yes Sender: mailman-bounces@oss.sgi.com Errors-To: mailman-bounces@oss.sgi.com This is a reminder, sent out once a month, about your oss.sgi.com mailing list memberships. It includes your subscription info and how to use it to change it or unsubscribe from a list. You can visit the URLs to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on. In addition to the URL interfaces, you can also use email to make such changes. For more info, send a message to the '-request' address of the list (for example, mailman-request@oss.sgi.com) containing just the word 'help' in the message body, and an email message will be sent to you with instructions. If you have questions, problems, comments, etc, send them to mailman-owner@oss.sgi.com. Thanks! Passwords for pcp@oss.sgi.com: List Password // URL ---- -------- pcp-announce@oss.sgi.com xE+IBctR http://oss.sgi.com/mailman/options/pcp-announce/pcp%40oss.sgi.com From makc@iinet.net.au Tue Apr 6 19:24:25 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.6 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o370OOa1146428 for ; Tue, 6 Apr 2010 19:24:25 -0500 X-ASG-Debug-ID: 1270599972-3f5202ac0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from outbound.icp-qv1-irony-out2.iinet.net.au (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 32360FA661A for ; Tue, 6 Apr 2010 17:26:13 -0700 (PDT) Received: from outbound.icp-qv1-irony-out2.iinet.net.au (outbound.icp-qv1-irony-out2.iinet.net.au [203.59.1.107]) by cuda.sgi.com with ESMTP id uihG99lseA2SsQMS for ; Tue, 06 Apr 2010 17:26:13 -0700 (PDT) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AvgFADNuu0vL2Rwk/2dsb2JhbACPU4tocbpLhQkE X-IronPort-AV: E=Sophos;i="4.51,375,1267372800"; d="scan'208";a="632351310" Received: from unknown (HELO alma.crabbed.net) ([203.217.28.36]) by outbound.icp-qv1-irony-out2.iinet.net.au with ESMTP; 07 Apr 2010 08:26:10 +0800 Received: by alma.crabbed.net (Postfix, from userid 501) id 060BAEB94C; Wed, 7 Apr 2010 10:26:10 +1000 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <19387.53537.727368.123975@iinet.net.au> Date: Wed, 7 Apr 2010 10:26:09 +1000 From: Max Matveev To: pcp@oss.sgi.com X-ASG-Orig-Subj: pmlogger_check is fooled by localhost Subject: pmlogger_check is fooled by localhost X-Mailer: VM 7.17 under 21.4 (patch 19) "Constant Variable" XEmacs Lucid X-Barracuda-Connect: outbound.icp-qv1-irony-out2.iinet.net.au[203.59.1.107] X-Barracuda-Start-Time: 1270599975 X-Barracuda-Bayes: INNOCENT GLOBAL 0.4404 1.0000 0.0000 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: 0.00 X-Barracuda-Spam-Status: No, SCORE=0.00 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.26885 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean I've got caught in pmlogger stampede after using the following incantation in the control file: # special local logger localhost n n PCP_LOG_DIR/pmlogger/mumble -c /var/lib/pcp/config/mumble.conf pmlogger was started with -h localhost argument which was helpfully replaced in the map file with the actual host name, that host name did not match what pmlogger_check and pmlogger_daily were expecting, pmlogger_check went troppo and started starting new logger every time it was invoked, pmlogger_daily tried to merge archives which were still written by the horde of pmlogger - it wasn't thrilled by that. The reason I've used 'localhost' instead of LOCALHOSTNAME was because on that machine's name is not resolvable: it's not in DNS and it's not in the /etc/hosts (DHCP addresses and generic hostnames). I've added the following in pmlogger_check --- /usr/libexec/pcp/bin/pmlogger_check 2010-02-04 10:58:38.000000000 +1100 +++ ./pmlogger_check 2010-04-07 10:23:52.587928390 +1000 @@ -288,6 +288,9 @@ | while read host primary socks dir args do line=`expr $line + 1` + if [ "$host" == "localhost" ]; then + host=$LOCALHOSTNAME + fi $VERY_VERBOSE && echo "[control:$line] host=\"$host\" primary=\"$primary\" socks=\"$socks\" dir=\"$dir\" args=\"$args\"" case "$host" in but I wonder if it would be better to remove automatic replacement which pmlogger does. max From kenj@internode.on.net Tue Apr 6 20:25:35 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.1 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o371PY0f149103 for ; Tue, 6 Apr 2010 20:25:35 -0500 X-ASG-Debug-ID: 1270603643-1bab004d0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id E5787FB1AD2 for ; Tue, 6 Apr 2010 18:27:23 -0700 (PDT) Received: from mail.internode.on.net (bld-mail17.adl2.internode.on.net [150.101.137.102]) by cuda.sgi.com with ESMTP id xODzR05ptvnWXMnl for ; Tue, 06 Apr 2010 18:27:23 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.40.21]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 19480260-1927428 for multiple; Wed, 07 Apr 2010 10:57:21 +1030 (CDT) X-ASG-Orig-Subj: Re: [pcp] pmlogger_check is fooled by localhost Subject: Re: [pcp] pmlogger_check is fooled by localhost From: Ken McDonell Reply-To: kenj@internode.on.net To: Max Matveev Cc: pcp@oss.sgi.com In-Reply-To: <19387.53537.727368.123975@iinet.net.au> References: <19387.53537.727368.123975@iinet.net.au> Content-Type: text/plain; charset="UTF-8" Date: Wed, 07 Apr 2010 11:27:01 +1000 Message-ID: <1270603621.15121.79.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail17.adl2.internode.on.net[150.101.137.102] X-Barracuda-Start-Time: 1270603644 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0615 1.0000 -1.6277 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.63 X-Barracuda-Spam-Status: No, SCORE=-1.63 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.26889 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean The hostname is branded in the archive label and is used later on to match archives first to merge and then for config files used by tools like pmie and pmchart and ... So not having a unique and real hostname could present something of a problem for merging archives and using the archives for cross-host analysis. Without looking at the code, the patch you're suggesting "doesn't feel right" ... whatever the control scripts do should be compatible with pmlogger. After looking at the code, the status quo would appear to work fine provided what gethostname(3) returns matches what hostname(1) returns up to the first `.'. So in you specific case, what does gethostname(3) and hostname(1) return? And for extra points, what does uname -n report? pmlogger's automatic replacement is the right thing to do in my opinion, so generic config files can be deployed in environments where hosts are not confused about their names. On Wed, 2010-04-07 at 10:26 +1000, Max Matveev wrote: > I've got caught in pmlogger stampede after using the following > incantation in the control file: > > # special local logger > localhost n n PCP_LOG_DIR/pmlogger/mumble -c /var/lib/pcp/config/mumble.conf > > pmlogger was started with -h localhost argument which was helpfully > replaced in the map file with the actual host name, that host name did > not match what pmlogger_check and pmlogger_daily were expecting, > pmlogger_check went troppo and started starting new logger every time > it was invoked, pmlogger_daily tried to merge archives which were > still written by the horde of pmlogger - it wasn't thrilled by that. > > The reason I've used 'localhost' instead of LOCALHOSTNAME was because > on that machine's name is not resolvable: it's not in DNS and it's not > in the /etc/hosts (DHCP addresses and generic hostnames). > > I've added the following in pmlogger_check > > --- /usr/libexec/pcp/bin/pmlogger_check 2010-02-04 10:58:38.000000000 +1100 > +++ ./pmlogger_check 2010-04-07 10:23:52.587928390 +1000 > @@ -288,6 +288,9 @@ > | while read host primary socks dir args > do > line=`expr $line + 1` > + if [ "$host" == "localhost" ]; then > + host=$LOCALHOSTNAME > + fi > $VERY_VERBOSE && echo "[control:$line] host=\"$host\" > primary=\"$primary\" socks=\"$socks\" dir=\"$dir\" > args=\"$args\"" > case "$host" > in > > but I wonder if it would be better to remove automatic replacement > which pmlogger does. > > max > > _______________________________________________ > pcp mailing list > pcp@oss.sgi.com > http://oss.sgi.com/mailman/listinfo/pcp From makc@iinet.net.au Tue Apr 6 22:40:50 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.5 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX,J_CHICKENPOX_25,J_CHICKENPOX_53 autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o373emcr154152 for ; Tue, 6 Apr 2010 22:40:49 -0500 X-ASG-Debug-ID: 1270611757-2edf02ec0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from outbound.icp-qv1-irony-out2.iinet.net.au (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 11B21FB202F for ; Tue, 6 Apr 2010 20:42:37 -0700 (PDT) Received: from outbound.icp-qv1-irony-out2.iinet.net.au (outbound.icp-qv1-irony-out2.iinet.net.au [203.59.1.107]) by cuda.sgi.com with ESMTP id bbv8d9S6hJ4cM8dC for ; Tue, 06 Apr 2010 20:42:37 -0700 (PDT) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AvsEAKqbu0vL2Rwk/2dsb2JhbACbPHG6dIUJBA X-IronPort-AV: E=Sophos;i="4.51,376,1267372800"; d="scan'208";a="632455979" Received: from unknown (HELO alma.crabbed.net) ([203.217.28.36]) by outbound.icp-qv1-irony-out2.iinet.net.au with ESMTP; 07 Apr 2010 11:42:34 +0800 Received: by alma.crabbed.net (Postfix, from userid 501) id 5A1A1EBA10; Wed, 7 Apr 2010 13:42:34 +1000 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <19387.65322.68301.468516@iinet.net.au> Date: Wed, 7 Apr 2010 13:42:34 +1000 From: Max Matveev To: kenj@internode.on.net Cc: pcp@oss.sgi.com X-ASG-Orig-Subj: Re: [pcp] pmlogger_check is fooled by localhost Subject: Re: [pcp] pmlogger_check is fooled by localhost In-Reply-To: <1270603621.15121.79.camel@bozo.localdomain> References: <19387.53537.727368.123975@iinet.net.au> <1270603621.15121.79.camel@bozo.localdomain> X-Mailer: VM 7.17 under 21.4 (patch 19) "Constant Variable" XEmacs Lucid X-Barracuda-Connect: outbound.icp-qv1-irony-out2.iinet.net.au[203.59.1.107] X-Barracuda-Start-Time: 1270611759 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0950 1.0000 -1.4228 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.42 X-Barracuda-Spam-Status: No, SCORE=-1.42 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.26898 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Wed, 07 Apr 2010 11:27:01 +1000, Ken McDonell wrote: kenj> The hostname is branded in the archive label and is used later on to kenj> match archives first to merge and then for config files used by tools kenj> like pmie and pmchart and ... I understand that... kenj> So not having a unique and real hostname could present something of a kenj> problem for merging archives and using the archives for cross-host kenj> analysis. I can get unique hostnames in my setup, I just cannot guarantee that I'll be able to map those names to IP addresses which is required for -h processing (see below). kenj> Without looking at the code, the patch you're suggesting "doesn't feel kenj> right" ... whatever the control scripts do should be compatible with kenj> pmlogger. That's the point - pmlogger_check's processing of the localhost in the control file does not match pmlogger' use of localhost, e.g.: sonya:~$ cat lhost.ctl $version=1.1 # === LOGGER CONTROL SPECIFICATIONS === # #Host P? S? directory args # local primary logger localhost n n /export/home/makc/pmlogger -c /export/home/makc/conf sonya:~$ /usr/libexec/pcp/bin/pmlogger_check -VV -c lhost.ctl [control:1] host="$version=1.1" primary="" socks="" dir="" args="" [control:2] host="" primary="" socks="" dir="" args="" [control:3] host="#" primary="===" socks="LOGGER" dir="CONTROL" args="SPECIFICATIONS ===" [control:4] host="#" primary="" socks="" dir="" args="" [control:5] host="#Host" primary="P?" socks="S?" dir="directory" args="args" [control:6] host="" primary="" socks="" dir="" args="" [control:7] host="#" primary="local" socks="primary" dir="logger" args="" [control:8] host="localhost" primary="n" socks="n" dir="/export/home/makc/pmlogger" args="-c /export/home/makc/conf" Check pmlogger -h localhost ... in /export/home/makc/pmlogger ... ... try /var/tmp/pmlogger/7146: match=0 different host, skip ... try /var/tmp/pmlogger/8546: match=0 different host, skip ... try /var/tmp/pmlogger/9999: match=0 different host, skip Restarting pmlogger for host "localhost" ... [process 14013] done Latest folio created for 20100407.13.34 [control:9] host="" primary="" socks="" dir="" args="" sonya:~$ cat /var/tmp/pmlogger/14013 4332 sonya /export/home/makc/pmlogger/20100407.13.34 sonya:~$ /usr/libexec/pcp/bin/pmlogger_check -c lhost.ctl sonya:~$ /usr/libexec/pcp/bin/pmlogger_check -V -c lhost.ctl Duplicate archive basename ... rename 20100407.13.34.* files to 20100407.13.34-01.* Restarting pmlogger for host "localhost" ... [process 14160] done Latest folio created for 20100407.13.34 sonya:~$ cat /var/tmp/pmlogger/14160 4334 sonya /export/home/makc/pmlogger/20100407.13.34 sonya:~$ ps -ef | grep pmlogger makc 14184 10822 0 13:36:35 pts/3 0:00 grep pmlogger makc 14085 1 0 13:34:33 pts/3 0:00 pmlogger -h localhost -c /export/home/makc/conf 20100407.13.34 makc 14160 1 0 13:34:39 pts/3 0:00 pmlogger -h localhost -c /export/home/makc/conf 20100407.13.34 root 7146 1 0 Mar 30 ? 0:01 /usr/libexec/pcp/bin/pmlogger -c conf test.1 makc 14013 1 0 13:34:09 pts/3 0:00 pmlogger -h localhost -c /export/home/makc/conf 20100407.13.34 kenj> After looking at the code, the status quo would appear to work fine kenj> provided what gethostname(3) returns matches what hostname(1) returns up kenj> to the first `.'. So in you specific case, what does gethostname(3) and kenj> hostname(1) return? Both return 'sonya' kenj> And for extra points, what does uname -n report? sonya as well. And if I use 'sonya' as hostname, I'm getting sonya:~$ /usr/libexec/pcp/bin/pmlogger -h sonya -c conf no.sonya sonya:~$ cat pmlogger.log Log for pmlogger on sonya started Wed Apr 7 13:39:19 2010 pmlogger: Cannot connect to PMCD on host "sonya": No route to host Log finished Wed Apr 7 13:39:19 2010 This is because sonya is not in DNS or /etc/hosts. max From kenj@internode.on.net Wed Apr 7 07:07:31 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.2 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, J_CHICKENPOX_25,J_CHICKENPOX_53 autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o37C7V9Z186718 for ; Wed, 7 Apr 2010 07:07:31 -0500 X-ASG-Debug-ID: 1270642159-598c03420000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id DCC001D3E9E2 for ; Wed, 7 Apr 2010 05:09:20 -0700 (PDT) Received: from mail.internode.on.net (bld-mail14.adl6.internode.on.net [150.101.137.99]) by cuda.sgi.com with ESMTP id CULgfkCijb809KaU for ; Wed, 07 Apr 2010 05:09:20 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.40.21]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 19962527-1927428 for multiple; Wed, 07 Apr 2010 21:39:18 +1030 (CDT) X-ASG-Orig-Subj: Re: [pcp] pmlogger_check is fooled by localhost Subject: Re: [pcp] pmlogger_check is fooled by localhost From: Ken McDonell Reply-To: kenj@internode.on.net To: Max Matveev Cc: pcp@oss.sgi.com In-Reply-To: <19387.65322.68301.468516@iinet.net.au> References: <19387.53537.727368.123975@iinet.net.au> <1270603621.15121.79.camel@bozo.localdomain> <19387.65322.68301.468516@iinet.net.au> Content-Type: text/plain; charset="UTF-8" Date: Wed, 07 Apr 2010 22:08:57 +1000 Message-ID: <1270642137.15121.125.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail14.adl6.internode.on.net[150.101.137.99] X-Barracuda-Start-Time: 1270642161 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0138 1.0000 -1.9313 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.93 X-Barracuda-Spam-Status: No, SCORE=-1.93 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.26927 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Thanks for the detailed explanation Max. I was going down the path of pmlogger conditionally doing the remapping from localhost to whatever gethostname(3) returns, but that gets ugly as sin when one considers what pmlogger_check would need to do to second guess pmlogger's behaviour. So I change my vote ... I think you're original patch suggestion for pmlogger_check and sticking with localhost in the control file is the best line of attack. Could we please have a comment added in pmlogger_check to explain the rationale? Cheers, Ken. On Wed, 2010-04-07 at 13:42 +1000, Max Matveev wrote: > On Wed, 07 Apr 2010 11:27:01 +1000, Ken McDonell wrote: > > kenj> The hostname is branded in the archive label and is used later on to > kenj> match archives first to merge and then for config files used by tools > kenj> like pmie and pmchart and ... > I understand that... > > kenj> So not having a unique and real hostname could present something of a > kenj> problem for merging archives and using the archives for cross-host > kenj> analysis. > I can get unique hostnames in my setup, I just cannot guarantee that > I'll be able to map those names to IP addresses which is required for > -h processing (see below). > > kenj> Without looking at the code, the patch you're suggesting "doesn't feel > kenj> right" ... whatever the control scripts do should be compatible with > kenj> pmlogger. > That's the point - pmlogger_check's processing of the localhost in the > control file does not match pmlogger' use of localhost, e.g.: > > sonya:~$ cat lhost.ctl > $version=1.1 > > # === LOGGER CONTROL SPECIFICATIONS === > # > #Host P? S? directory args > > # local primary logger > localhost n n /export/home/makc/pmlogger -c > /export/home/makc/conf > sonya:~$ /usr/libexec/pcp/bin/pmlogger_check -VV -c lhost.ctl > [control:1] host="$version=1.1" primary="" socks="" dir="" args="" > [control:2] host="" primary="" socks="" dir="" args="" > [control:3] host="#" primary="===" socks="LOGGER" dir="CONTROL" > args="SPECIFICATIONS ===" > [control:4] host="#" primary="" socks="" dir="" args="" > [control:5] host="#Host" primary="P?" socks="S?" dir="directory" > args="args" > [control:6] host="" primary="" socks="" dir="" args="" > [control:7] host="#" primary="local" socks="primary" dir="logger" > args="" > [control:8] host="localhost" primary="n" socks="n" > dir="/export/home/makc/pmlogger" args="-c /export/home/makc/conf" > Check pmlogger -h localhost ... in /export/home/makc/pmlogger ... > ... try /var/tmp/pmlogger/7146: match=0 different host, skip > ... try /var/tmp/pmlogger/8546: match=0 different host, skip > ... try /var/tmp/pmlogger/9999: match=0 different host, skip > Restarting pmlogger for host "localhost" ... [process 14013] done > Latest folio created for 20100407.13.34 > [control:9] host="" primary="" socks="" dir="" args="" > sonya:~$ cat /var/tmp/pmlogger/14013 > 4332 > sonya > /export/home/makc/pmlogger/20100407.13.34 > sonya:~$ /usr/libexec/pcp/bin/pmlogger_check -c lhost.ctl > sonya:~$ /usr/libexec/pcp/bin/pmlogger_check -V -c lhost.ctl > Duplicate archive basename ... rename 20100407.13.34.* files to > 20100407.13.34-01.* > Restarting pmlogger for host "localhost" ... [process 14160] done > Latest folio created for 20100407.13.34 > sonya:~$ cat /var/tmp/pmlogger/14160 > 4334 > sonya > /export/home/makc/pmlogger/20100407.13.34 > sonya:~$ ps -ef | grep pmlogger > makc 14184 10822 0 13:36:35 pts/3 0:00 grep pmlogger > makc 14085 1 0 13:34:33 pts/3 0:00 pmlogger -h localhost -c /export/home/makc/conf 20100407.13.34 > makc 14160 1 0 13:34:39 pts/3 0:00 pmlogger -h localhost -c /export/home/makc/conf 20100407.13.34 > root 7146 1 0 Mar 30 ? 0:01 /usr/libexec/pcp/bin/pmlogger -c conf test.1 > makc 14013 1 0 13:34:09 pts/3 0:00 pmlogger -h localhost -c /export/home/makc/conf 20100407.13.34 > > kenj> After looking at the code, the status quo would appear to work fine > kenj> provided what gethostname(3) returns matches what hostname(1) returns up > kenj> to the first `.'. So in you specific case, what does gethostname(3) and > kenj> hostname(1) return? > Both return 'sonya' > > kenj> And for extra points, what does uname -n report? > sonya as well. > > And if I use 'sonya' as hostname, I'm getting > > sonya:~$ /usr/libexec/pcp/bin/pmlogger -h sonya -c conf no.sonya > sonya:~$ cat pmlogger.log > Log for pmlogger on sonya started Wed Apr 7 13:39:19 2010 > > pmlogger: Cannot connect to PMCD on host "sonya": No route to host > > Log finished Wed Apr 7 13:39:19 2010 > > This is because sonya is not in DNS or /etc/hosts. > > max From makc@iinet.net.au Wed Apr 7 23:51:49 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.6 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o384pm6Q251599 for ; Wed, 7 Apr 2010 23:51:49 -0500 X-ASG-Debug-ID: 1270702415-046702670000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from outbound.icp-qv1-irony-out6.iinet.net.au (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 81D7BFBF913 for ; Wed, 7 Apr 2010 21:53:36 -0700 (PDT) Received: from outbound.icp-qv1-irony-out6.iinet.net.au (outbound.icp-qv1-irony-out6.iinet.net.au [203.59.1.109]) by cuda.sgi.com with ESMTP id KlMeuXvIBw6qpwPD for ; Wed, 07 Apr 2010 21:53:36 -0700 (PDT) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApQFAEf+vEvL2Rwk/2dsb2JhbACPYItLcbhEhQkE X-IronPort-AV: E=Sophos;i="4.52,168,1270396800"; d="scan'208";a="91147435" Received: from unknown (HELO alma.crabbed.net) ([203.217.28.36]) by outbound.icp-qv1-irony-out6.iinet.net.au with ESMTP; 08 Apr 2010 12:53:34 +0800 Received: by alma.crabbed.net (Postfix, from userid 501) id 6BF70EC1AB; Thu, 8 Apr 2010 14:53:31 +1000 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <19389.24907.135424.300701@iinet.net.au> Date: Thu, 8 Apr 2010 14:53:31 +1000 From: Max Matveev To: kenj@internode.on.net Cc: pcp@oss.sgi.com X-ASG-Orig-Subj: Re: [pcp] pmlogger_check is fooled by localhost Subject: Re: [pcp] pmlogger_check is fooled by localhost In-Reply-To: <1270642137.15121.125.camel@bozo.localdomain> References: <19387.53537.727368.123975@iinet.net.au> <1270603621.15121.79.camel@bozo.localdomain> <19387.65322.68301.468516@iinet.net.au> <1270642137.15121.125.camel@bozo.localdomain> X-Mailer: VM 7.17 under 21.4 (patch 19) "Constant Variable" XEmacs Lucid X-Barracuda-Connect: outbound.icp-qv1-irony-out6.iinet.net.au[203.59.1.109] X-Barracuda-Start-Time: 1270702418 X-Barracuda-Bayes: INNOCENT GLOBAL 0.3808 1.0000 -0.0526 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -0.05 X-Barracuda-Spam-Status: No, SCORE=-0.05 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.26979 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Wed, 07 Apr 2010 22:08:57 +1000, Ken McDonell wrote: kenj> Thanks for the detailed explanation Max. kenj> I was going down the path of pmlogger conditionally doing the remapping kenj> from localhost to whatever gethostname(3) returns, but that gets ugly as kenj> sin when one considers what pmlogger_check would need to do to second kenj> guess pmlogger's behaviour. kenj> So I change my vote ... I think you're original patch suggestion for kenj> pmlogger_check and sticking with localhost in the control file is the kenj> best line of attack. Could we please have a comment added in kenj> pmlogger_check to explain the rationale? Um, maybe not - I've changed my mind (again): pcp really like to be able to resolve the hostname - I've tried running 'pmlogger -h localhost' on sonya and first thing it tripped is "No route to host 'sonya'" from pmNewContext which means that simple use of localhost is not going to cut it. And even if I fix pmlogger to not do that then pmcd gets in a way because it also likes to be able to resolve the name returned by gethostname(3). I've got a patch (git://oss.sgi.com/makc/pcp pmlogger) which with leaves localhost alone but bakes the "actual" hostname into the archives but I'm not sure if I'll even need it because it does not really solve my underlaying problem which is not having the /etc/hosts entry for host I'm running on. What I can of worms... max From kenj@internode.on.net Thu Apr 8 00:34:01 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.1 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o385Y0a2256513 for ; Thu, 8 Apr 2010 00:34:00 -0500 X-ASG-Debug-ID: 1270704949-3437014e0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 5B87BFBF972 for ; Wed, 7 Apr 2010 22:35:50 -0700 (PDT) Received: from mail.internode.on.net (bld-mail19.adl2.internode.on.net [150.101.137.104]) by cuda.sgi.com with ESMTP id VcfSFQB0Y3eB6C2o for ; Wed, 07 Apr 2010 22:35:50 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.40.21]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 19717638-1927428 for multiple; Thu, 08 Apr 2010 15:05:48 +0930 (CST) X-ASG-Orig-Subj: Re: [pcp] pmlogger_check is fooled by localhost Subject: Re: [pcp] pmlogger_check is fooled by localhost From: Ken McDonell Reply-To: kenj@internode.on.net To: Max Matveev Cc: pcp@oss.sgi.com In-Reply-To: <19389.24907.135424.300701@iinet.net.au> References: <19387.53537.727368.123975@iinet.net.au> <1270603621.15121.79.camel@bozo.localdomain> <19387.65322.68301.468516@iinet.net.au> <1270642137.15121.125.camel@bozo.localdomain> <19389.24907.135424.300701@iinet.net.au> Content-Type: text/plain; charset="UTF-8" Date: Thu, 08 Apr 2010 15:35:27 +1000 Message-ID: <1270704927.8807.7.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail19.adl2.internode.on.net[150.101.137.104] X-Barracuda-Start-Time: 1270704951 X-Barracuda-Bayes: INNOCENT GLOBAL 0.3816 1.0000 -0.0502 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -0.05 X-Barracuda-Spam-Status: No, SCORE=-0.05 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.26983 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Hmmm ... since we need to open a socket to talk to pmcd, and these sockets are in the AF_INET domain, nothing much will work unless hostname to IP addr via gethostname() works, which means you're stuck with localhost -> 127.0.0.1 as your only option. This means stopping pmlogger doing the localhost -> hostname() mapping, which probably means another command line argument to pmlogger to not do the mapping, while maintaining compatibility by doing the mapping in the default case. On Thu, 2010-04-08 at 14:53 +1000, Max Matveev wrote: > On Wed, 07 Apr 2010 22:08:57 +1000, Ken McDonell wrote: > > kenj> Thanks for the detailed explanation Max. > kenj> I was going down the path of pmlogger conditionally doing the remapping > kenj> from localhost to whatever gethostname(3) returns, but that gets ugly as > kenj> sin when one considers what pmlogger_check would need to do to second > kenj> guess pmlogger's behaviour. > > kenj> So I change my vote ... I think you're original patch suggestion for > kenj> pmlogger_check and sticking with localhost in the control file is the > kenj> best line of attack. Could we please have a comment added in > kenj> pmlogger_check to explain the rationale? > Um, maybe not - I've changed my mind (again): pcp really like to be able to > resolve the hostname - I've tried running 'pmlogger -h localhost' on > sonya and first thing it tripped is "No route to host 'sonya'" from > pmNewContext which means that simple use of localhost is not going to > cut it. And even if I fix pmlogger to not do that then pmcd gets in a > way because it also likes to be able to resolve the name returned by > gethostname(3). > > I've got a patch (git://oss.sgi.com/makc/pcp pmlogger) which with > leaves localhost alone but bakes the "actual" hostname into the > archives but I'm not sure if I'll even need it because it does not > really solve my underlaying problem which is not having the /etc/hosts > entry for host I'm running on. > > What I can of worms... > > max From makc@iinet.net.au Fri Apr 9 01:52:50 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.4 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX,J_CHICKENPOX_42,J_CHICKENPOX_43,J_CHICKENPOX_64 autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o396qmUE081380 for ; Fri, 9 Apr 2010 01:52:50 -0500 X-ASG-Debug-ID: 1270796077-262d02510000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from outbound.icp-qv1-irony-out4.iinet.net.au (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 18B7EFCDD49 for ; Thu, 8 Apr 2010 23:54:38 -0700 (PDT) Received: from outbound.icp-qv1-irony-out4.iinet.net.au (outbound.icp-qv1-irony-out4.iinet.net.au [203.59.1.104]) by cuda.sgi.com with ESMTP id pWWZczkOLF7ndolh for ; Thu, 08 Apr 2010 23:54:38 -0700 (PDT) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AqMGAJtsvkvL2Rwk/2dsb2JhbACPbYtPcbh9hQkE X-IronPort-AV: E=Sophos;i="4.52,176,1270396800"; d="scan'208";a="534044698" Received: from unknown (HELO alma.crabbed.net) ([203.217.28.36]) by outbound.icp-qv1-irony-out4.iinet.net.au with ESMTP; 09 Apr 2010 14:54:20 +0800 Received: by alma.crabbed.net (Postfix, from userid 501) id 34C0AECB98; Fri, 9 Apr 2010 16:54:20 +1000 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <19390.53020.92570.731051@iinet.net.au> Date: Fri, 9 Apr 2010 16:54:20 +1000 From: Max Matveev To: kenj@internode.on.net Cc: pcp@oss.sgi.com X-ASG-Orig-Subj: Re: [pcp] pmlogger_check is fooled by localhost Subject: Re: [pcp] pmlogger_check is fooled by localhost In-Reply-To: <1270704927.8807.7.camel@bozo.localdomain> References: <19387.53537.727368.123975@iinet.net.au> <1270603621.15121.79.camel@bozo.localdomain> <19387.65322.68301.468516@iinet.net.au> <1270642137.15121.125.camel@bozo.localdomain> <19389.24907.135424.300701@iinet.net.au> <1270704927.8807.7.camel@bozo.localdomain> X-Mailer: VM 7.17 under 21.4 (patch 19) "Constant Variable" XEmacs Lucid X-Barracuda-Connect: outbound.icp-qv1-irony-out4.iinet.net.au[203.59.1.104] X-Barracuda-Start-Time: 1270796080 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0038 1.0000 -1.9963 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.00 X-Barracuda-Spam-Status: No, SCORE=-2.00 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27066 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Thu, 08 Apr 2010 15:35:27 +1000, Ken McDonell wrote: kenj> This means stopping pmlogger doing the localhost -> hostname() mapping, kenj> which probably means another command line argument to pmlogger to not do kenj> the mapping, while maintaining compatibility by doing the mapping in the kenj> default case. Do you mean like this (with a matching change to pmlogger.1): diff --git a/src/pmlogger/pmlogger.c b/src/pmlogger/pmlogger.c index 25310d3..d60a223 100644 --- a/src/pmlogger/pmlogger.c +++ b/src/pmlogger/pmlogger.c @@ -483,6 +483,7 @@ main(int argc, char **argv) fd_set readyfds; char *p; char *runtime = NULL; + int dont_use_localhost = 1; __pmSetProgname(argv[0]); @@ -492,7 +493,7 @@ main(int argc, char **argv) * corresponding changes are made to pmnewlog when pmlogger * options are passed through from the control file */ - while ((c = getopt(argc, argv, "c:D:h:l:Ln:Prs:T:t:uv:V:x:?")) != EOF) { + while ((c = getopt(argc, argv, "c:D:H:h:l:Ln:Prs:T:t:uv:V:x:?")) != EOF) { switch (c) { case 'c': /* config file */ @@ -527,6 +528,9 @@ main(int argc, char **argv) pmDebug |= sts; break; + case 'H': + dont_use_localhost = 0; + /* fallthru */ case 'h': /* hostname for PMCD to contact */ pmcd_host = optarg; break; @@ -628,6 +632,7 @@ main(int argc, char **argv) Options:\n\ -c configfile file to load configuration from\n\ -h host metrics source is PMCD on host\n\ + -H host metrics source is PMCD on host(do not subsitute the host name)\n -l logfile redirect diagnostics and trace output\n\ -L linger, even if not primary logger instance and nothing to log\n -n pmnsfile use an alternative PMNS\n\ @@ -658,7 +663,7 @@ Options:\n\ if (pmcd_host == NULL || strcmp(pmcd_host, "localhost") == 0) { (void)gethostname(archive_host, sizeof(archive_host)); archive_host[sizeof(archive_host)-1] = '\0'; - if (pmcd_host == NULL) + if ((pmcd_host == NULL) || dont_use_localhost) pmcd_host = archive_host; } else { strncmp(archive_host, pmcd_host, sizeof(archive_host)); max From kenj@internode.on.net Fri Apr 9 16:33:16 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.2 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, J_CHICKENPOX_42,J_CHICKENPOX_43,J_CHICKENPOX_64 autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o39LXFeB140537 for ; Fri, 9 Apr 2010 16:33:15 -0500 X-ASG-Debug-ID: 1270848905-29f4030d0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 5B9AB1BCFE1A for ; Fri, 9 Apr 2010 14:35:05 -0700 (PDT) Received: from mail.internode.on.net (bld-mail17.adl2.internode.on.net [150.101.137.102]) by cuda.sgi.com with ESMTP id GP1ABP1QhpmOjqUX for ; Fri, 09 Apr 2010 14:35:05 -0700 (PDT) Received: from [124.176.220.88] (unverified [124.176.220.88]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 19894381-1927428 for multiple; Sat, 10 Apr 2010 07:05:03 +0930 (CST) X-ASG-Orig-Subj: Re: [pcp] pmlogger_check is fooled by localhost Subject: Re: [pcp] pmlogger_check is fooled by localhost From: Ken McDonell Reply-To: kenj@internode.on.net To: Max Matveev Cc: pcp@oss.sgi.com In-Reply-To: <19390.53020.92570.731051@iinet.net.au> References: <19387.53537.727368.123975@iinet.net.au> <1270603621.15121.79.camel@bozo.localdomain> <19387.65322.68301.468516@iinet.net.au> <1270642137.15121.125.camel@bozo.localdomain> <19389.24907.135424.300701@iinet.net.au> <1270704927.8807.7.camel@bozo.localdomain> <19390.53020.92570.731051@iinet.net.au> Content-Type: multipart/mixed; boundary="=-FcqcpWTn31I98ns0cu4s" Date: Sat, 10 Apr 2010 07:15:50 +1000 Message-ID: <1270847750.18793.8.camel@bozo-laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 X-Barracuda-Connect: bld-mail17.adl2.internode.on.net[150.101.137.102] X-Barracuda-Start-Time: 1270848907 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0008 1.0000 -2.0156 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27091 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean --=-FcqcpWTn31I98ns0cu4s Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Max, Yep, pretty close to my patch (no man page from me either) ... I just went for the -H as a flag, rather than -H hostname (which is the same as -h hostname, unless hostname happens to be localhost). But more importantly your source and mine don't match without the patch ... I don't have archive_host[] in my source, it uses local[] there and there is no second test for pmcd_host being NULL, no strncmp() in an else branch ... ??? Attached is my patch. On Fri, 2010-04-09 at 16:54 +1000, Max Matveev wrote: > On Thu, 08 Apr 2010 15:35:27 +1000, Ken McDonell wrote: > > kenj> This means stopping pmlogger doing the localhost -> hostname() mapping, > kenj> which probably means another command line argument to pmlogger to not do > kenj> the mapping, while maintaining compatibility by doing the mapping in the > kenj> default case. > Do you mean like this (with a matching change to pmlogger.1): > > diff --git a/src/pmlogger/pmlogger.c b/src/pmlogger/pmlogger.c > index 25310d3..d60a223 100644 > --- a/src/pmlogger/pmlogger.c > +++ b/src/pmlogger/pmlogger.c > @@ -483,6 +483,7 @@ main(int argc, char **argv) > fd_set readyfds; > char *p; > char *runtime = NULL; > + int dont_use_localhost = 1; > > __pmSetProgname(argv[0]); > > @@ -492,7 +493,7 @@ main(int argc, char **argv) > * corresponding changes are made to pmnewlog when > pmlogger > * options are passed through from the control file > */ > - while ((c = getopt(argc, argv, "c:D:h:l:Ln:Prs:T:t:uv:V:x:?")) != EOF) { > + while ((c = getopt(argc, argv, "c:D:H:h:l:Ln:Prs:T:t:uv:V:x:?")) != EOF) { > switch (c) { > > case 'c': /* config file */ > @@ -527,6 +528,9 @@ main(int argc, char **argv) > pmDebug |= sts; > break; > > + case 'H': > + dont_use_localhost = 0; > + /* fallthru */ > case 'h': /* hostname for PMCD to contact */ > pmcd_host = optarg; > break; > @@ -628,6 +632,7 @@ main(int argc, char **argv) > Options:\n\ > -c configfile file to load configuration from\n\ > -h host metrics source is PMCD on host\n\ > + -H host metrics source is PMCD on host(do not subsitute the host name)\n > -l logfile redirect diagnostics and trace output\n\ > -L linger, even if not primary logger instance and > nothing to log\n > -n pmnsfile use an alternative PMNS\n\ > @@ -658,7 +663,7 @@ Options:\n\ > if (pmcd_host == NULL || strcmp(pmcd_host, "localhost") == 0) { > (void)gethostname(archive_host, sizeof(archive_host)); > archive_host[sizeof(archive_host)-1] = '\0'; > - if (pmcd_host == NULL) > + if ((pmcd_host == NULL) || dont_use_localhost) > pmcd_host = archive_host; > } else { > strncmp(archive_host, pmcd_host, sizeof(archive_host)); > > max --=-FcqcpWTn31I98ns0cu4s Content-Disposition: attachment; filename="patch.localhost" Content-Type: text/x-patch; name="patch.localhost"; charset="UTF-8" Content-Transfer-Encoding: 7bit diff --git a/src/pmlogger/pmlogger.c b/src/pmlogger/pmlogger.c index 3daa339..e420109 100644 --- a/src/pmlogger/pmlogger.c +++ b/src/pmlogger/pmlogger.c @@ -38,6 +38,7 @@ int parse_done = 0; int primary = 0; /* Non-zero for primary pmlogger */ char *archBase; /* base name for log files */ char *pmcd_host = NULL; +int map_localhost = 1; /* translate localhost */ struct timeval epoch; int archive_version = PM_LOG_VERS02; /* Type of archive to create */ int linger = 0; /* linger with no tasks/events */ @@ -493,7 +494,7 @@ main(int argc, char **argv) * corresponding changes are made to pmnewlog when pmlogger * options are passed through from the control file */ - while ((c = getopt(argc, argv, "c:D:h:l:Ln:Prs:T:t:uv:V:x:?")) != EOF) { + while ((c = getopt(argc, argv, "c:D:h:Hl:Ln:Prs:T:t:uv:V:x:?")) != EOF) { switch (c) { case 'c': /* config file */ @@ -532,6 +533,10 @@ main(int argc, char **argv) pmcd_host = optarg; break; + case 'H': /* do not translate localhost */ + map_localhost = 0; + break; + case 'l': /* log file name */ logfile = optarg; break; @@ -629,6 +634,7 @@ main(int argc, char **argv) Options:\n\ -c configfile file to load configuration from\n\ -h host metrics source is PMCD on host\n\ + -H do not translate localhost to host's name\n\ -l logfile redirect diagnostics and trace output\n\ -L linger, even if not primary logger instance and nothing to log\n\ -n pmnsfile use an alternative PMNS\n\ @@ -656,7 +662,9 @@ Options:\n\ /* base name for archive is here ... */ archBase = argv[optind]; - if (pmcd_host == NULL || strcmp(pmcd_host, "localhost") == 0) { + if (pmcd_host == NULL) + pmcd_host = "localhost"; + if (map_localhost && strcmp(pmcd_host, "localhost") == 0) { (void)gethostname(local, MAXHOSTNAMELEN); local[MAXHOSTNAMELEN-1] = '\0'; pmcd_host = local; --=-FcqcpWTn31I98ns0cu4s-- From kenj@internode.on.net Sat Apr 10 17:19:27 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.6 required=5.0 tests=BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3AMJQhs229703 for ; Sat, 10 Apr 2010 17:19:27 -0500 X-ASG-Debug-ID: 1270938047-6996022a0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 3A7331AD885C for ; Sat, 10 Apr 2010 15:20:48 -0700 (PDT) Received: from mail.internode.on.net (bld-mail14.adl6.internode.on.net [150.101.137.99]) by cuda.sgi.com with ESMTP id FVFuBKkk4RMQugih for ; Sat, 10 Apr 2010 15:20:48 -0700 (PDT) Received: from [124.181.152.16] (unverified [124.181.152.16]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20376839-1927428 for ; Sun, 11 Apr 2010 07:50:45 +0930 (CST) X-ASG-Orig-Subj: pcp updates - PM_CONTEXT_LOCAL generalization Subject: pcp updates - PM_CONTEXT_LOCAL generalization From: Ken McDonell Reply-To: kenj@internode.on.net To: pcp@oss.sgi.com Content-Type: text/plain; charset="UTF-8" Date: Sun, 11 Apr 2010 08:20:51 +1000 Message-ID: <1270938051.30492.12.camel@bozo-laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail14.adl6.internode.on.net[150.101.137.99] X-Barracuda-Start-Time: 1270938079 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.52 X-Barracuda-Spam-Status: No, SCORE=-1.52 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=BSF_RULE7568M X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27177 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.50 BSF_RULE7568M Custom Rule 7568M X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Suggest the sgi folk pull these and make the associated changes to the cluster PMDA so it can use the IB PMDA again. All you should need there would be (before the pmNewContext() call to create the PM_CONTEXT_LOCAL connection) ... sts = __pmLocalPMDA(PM_LOCAL_CLEAR, 0, NULL, NULL); sts = __pmLocalPMDA(PM_LOCAL_ADD, 91, "ib/pmda_ib", "ib_init"); (and the first of these is optional). If that goes well, then these should be ready for merging into the main tree, modulo consideration of any review feedback. Changes committed to git://oss.sgi.com/kenj/pcp.git dev Makepkgs | 34 +++++- man/man1/pminfo.1 | 81 +++++++++++++- man/man1/pmprobe.1 | 56 +++++++++ man/man1/pmstat.1 | 2 man/man1/pmval.1 | 70 ++++++++++++ man/man3/pmlocalpmda.3 | 139 ++++++++++++++++++++++++ src/include/impl.h | 9 + src/libpcp/src/GNUmakefile | 14 -- src/libpcp/src/connect.c | 226 +++++++++++++++++++++++++++++++++------- src/libpcp/src/p_profile.c | 42 ++++++- src/pminfo/GNUmakefile | 3 src/pminfo/do_local_spec.c | 107 ++++++++++++++++++ src/pminfo/pminfo.c | 36 ++---- src/pmlogextract/pmlogextract.c | 20 +-- src/pmprobe/.gitignore | 1 src/pmprobe/GNUmakefile | 12 +- src/pmprobe/pmprobe.c | 36 ++---- src/pmstat/pmstat.c | 4 src/pmval/.gitignore | 1 src/pmval/GNUmakefile | 11 + src/pmval/pmval.c | 10 + 21 files changed, 788 insertions(+), 126 deletions(-) commit b7d1ee7da124ef434272ffd47ea0fcafaeca9710 Merge: 94a28e4 a1b2d60 Author: Ken McDonell Date: Sun Apr 11 08:10:41 2010 +1000 Merge branch 'localpmda' into dev commit a1b2d605b25c84e0a6bf402059551a13f61fcd29 Author: Ken McDonell Date: Sun Apr 11 08:08:31 2010 +1000 Generalized DSO PMDA support for PM_CONTEXT_LOCAL [second commit, this time for all the modified files, duh!] This batch of changes adds to the existing hardcoded table (and funky environment variable rules) that control which DSO PMDAs may be used with PM_CONTEXT_LOCAL. Specifically, __pmLocalPMDA() is added to libpcp to allow updates to the table, i.e. add a new entry, delete an existing entry, clear all entries. The monitoring tools that understand -L from the command line (pminfo and pmprobe) or support the equivalent @:metric syntax from pmParseMetricSpec() (pmval) have been given a new command line option, -K to get access to __pmLocalPMDA(). pmstat groks -L, but the metrics it uses are hardcoded to the OS PMDA, so there was no point in addding -K for pmstat. In the process the #define PM_USE_CONTEXT_LOCAL (which was always set) was removed from the source allowing some dead code to be removed. commit a0eb58cf0b14f85f33cb70c4318ee12a07ad089c Author: Ken McDonell Date: Sun Apr 11 07:59:08 2010 +1000 Generalized DSO PMDA support for PM_CONTEXT_LOCAL This batch of changes adds to the existing hardcoded table (and funky environment variable rules) that control which DSO PMDAs may be used with PM_CONTEXT_LOCAL. Specifically, __pmLocalPMDA() is added to libpcp to allow updates to the table, i.e. add a new entry, delete an existing entry, clear all entries. The monitoring tools that understand -L from the command line (pminfo and pmprobe) or support the equivalent @:metric syntax from pmParseMetricSpec() (pmval) have been given a new command line option, -K to get access to __pmLocalPMDA(). pmstat groks -L, but the metrics it uses are hardcoded to the OS PMDA, so there was no point in addding -K for pmstat. In the process the #define PM_USE_CONTEXT_LOCAL (which was always set) was removed from the source allowing some dead code to be removed. Some minor clean up of the man pages also was done, e.g. documenting the -c option for pminfo (hang over from the derived metrics changes). commit 0b7e3d32354f43f33d1267ca4b9c3ba00db7c3d3 Author: Ken McDonell Date: Sun Apr 11 07:50:51 2010 +1000 Grammatical typo. commit dd38b5b3d11ca1a3c1dbaae20535bf8e8c1d62ee Author: Ken McDonell Date: Sun Apr 11 07:47:10 2010 +1000 Makepkgs - Ubuntu toolchain problems persist Comment ahead of new checks says it all ... Sanity checks ... this is sick, but I'm really tired of QA failing because of bad binaries being built from the Ubuntu toolchain. commit e3273a1eabb0225db33b02d6557a6e000358f9f0 Author: Ken McDonell Date: Sun Apr 11 06:58:57 2010 +1000 pmlogextract - clean up diagnostics Report pmid in a more human friendly way, correct an endian issue in the reported times for log volume changes. commit 94a28e47f402981fc13de6601a98e94decfa6653 Author: Ken McDonell Date: Sat Mar 27 07:45:45 2010 +1100 p_profile.c better cleanup after failure First part of a cleanup and DOS defence. Still to come is some "reasonable" size limits when and if we can agree on "reasonable". From makc@iinet.net.au Sat Apr 10 20:05:25 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.5 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3B15OKA239656 for ; Sat, 10 Apr 2010 20:05:25 -0500 X-ASG-Debug-ID: 1270948006-521b03ba0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from outbound.icp-qv1-irony-out5.iinet.net.au (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 322AEFD6C1A for ; Sat, 10 Apr 2010 18:06:46 -0700 (PDT) Received: from outbound.icp-qv1-irony-out5.iinet.net.au (outbound.icp-qv1-irony-out5.iinet.net.au [203.59.1.105]) by cuda.sgi.com with ESMTP id Tg4TeAmsxB57diCR for ; Sat, 10 Apr 2010 18:06:46 -0700 (PDT) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: Aq0GABq9wEvL2Rwk/2dsb2JhbACPdYtTcbc/hQwE X-IronPort-AV: E=Sophos;i="4.52,183,1270396800"; d="scan'208";a="128726300" Received: from unknown (HELO alma.crabbed.net) ([203.217.28.36]) by outbound.icp-qv1-irony-out5.iinet.net.au with ESMTP; 11 Apr 2010 09:06:44 +0800 Received: by alma.crabbed.net (Postfix, from userid 501) id CAE21ED44E; Sun, 11 Apr 2010 11:06:43 +1000 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <19393.8355.508754.477098@iinet.net.au> Date: Sun, 11 Apr 2010 11:06:43 +1000 From: Max Matveev To: kenj@internode.on.net Cc: pcp@oss.sgi.com X-ASG-Orig-Subj: Re: [pcp] pmlogger_check is fooled by localhost Subject: Re: [pcp] pmlogger_check is fooled by localhost In-Reply-To: <1270847750.18793.8.camel@bozo-laptop> References: <19387.53537.727368.123975@iinet.net.au> <1270603621.15121.79.camel@bozo.localdomain> <19387.65322.68301.468516@iinet.net.au> <1270642137.15121.125.camel@bozo.localdomain> <19389.24907.135424.300701@iinet.net.au> <1270704927.8807.7.camel@bozo.localdomain> <19390.53020.92570.731051@iinet.net.au> <1270847750.18793.8.camel@bozo-laptop> X-Mailer: VM 7.17 under 21.4 (patch 19) "Constant Variable" XEmacs Lucid X-Barracuda-Connect: outbound.icp-qv1-irony-out5.iinet.net.au[203.59.1.105] X-Barracuda-Start-Time: 1270948037 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0385 1.0000 -1.7725 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.77 X-Barracuda-Spam-Status: No, SCORE=-1.77 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27185 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Sat, 10 Apr 2010 07:15:50 +1000, Ken McDonell wrote: kenj> Yep, pretty close to my patch (no man page from me either) ... I just kenj> went for the -H as a flag, rather than -H hostname (which is the same as kenj> -h hostname, unless hostname happens to be localhost). I can live with H being a flag, but I really like to remove gethostbyname from GetPort - it's just another place where pmlogger is trying to second-guess user's input and yet another place when it can go wrong. kenj> But more importantly your source and mine don't match without the kenj> patch ... I don't have archive_host[] in my source, it uses local[] kenj> there and there is no second test for pmcd_host being NULL, no strncmp() kenj> in an else branch ... ??? That's because patch was on top of the earlier change I've made which is available from http://oss.sgi.com/cgi-bin/gitweb.cgi?p=makc/pcp;a=shortlog;h=refs/heads/pmlogger It includes using different names for __pmLogCreate() and for the pmNewConext and not trying to resolve the name when creating a map file. And strncmp is a brain-fart - I'm surpised that the stuff works. max From gnb@evostor.com Sun Apr 11 07:12:33 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: ** X-Spam-Status: No, score=2.5 required=5.0 tests=BAYES_00,FH_DATE_PAST_20XX, URIBL_BLACK autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3BCCXLR019199 for ; Sun, 11 Apr 2010 07:12:33 -0500 X-ASG-Debug-ID: 1270988035-629900a80000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from relay.ihostexchange.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 6FE4D1DD0B20 for ; Sun, 11 Apr 2010 05:13:56 -0700 (PDT) Received: from relay.ihostexchange.net (relay.ihostexchange.net [66.46.182.55]) by cuda.sgi.com with ESMTP id BtEx9rvibSFRpzQo for ; Sun, 11 Apr 2010 05:13:56 -0700 (PDT) Received: from mailhost.internal (203.206.165.193) by smtp.ihostexchange.net (66.46.182.50) with Microsoft SMTP Server (TLS) id 8.1.393.1; Sun, 11 Apr 2010 08:13:39 -0400 Received: from [172.29.254.198] (unknown [172.29.254.198]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mailhost.internal (Postfix) with ESMTPSA id 7045C30A50 for ; Sun, 11 Apr 2010 22:13:41 +1000 (EST) Message-ID: <4BC1BD0D.3040803@evostor.com> Date: Sun, 11 Apr 2010 22:14:05 +1000 From: Greg Banks User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: PCP Mailing List X-ASG-Orig-Subj: Format .TS in manpages correctly on Solaris Subject: Format .TS in manpages correctly on Solaris Content-Type: text/plain; charset="ISO-8859-1"; format=flowed Content-Transfer-Encoding: 7bit X-Barracuda-Connect: relay.ihostexchange.net[66.46.182.55] X-Barracuda-Start-Time: 1270988066 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0692 1.0000 -1.5802 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.58 X-Barracuda-Spam-Status: No, SCORE=-1.58 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27216 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean G'day, commit 326c5a61d378625310c1c72750df7e10e1a081cd Author: Greg Banks Date: Sun Apr 11 21:52:16 2010 +1000 Format .TS in manpages correctly on Solaris Solaris needs the right obscure magic in the first line of each manpage in order for the .TS/.TE table macros to work. The PCP manpages are formatted with a different piece of obscure magic designed to work on Irix. This change automatically detects the old fashioned .\" t magic in a system manpage at configure time, and if present tweaks the PCP manpages at install time to contain it. diff --git a/.gitignore b/.gitignore index 1581608..1ac435c 100644 --- a/.gitignore +++ b/.gitignore @@ -19,3 +19,4 @@ libs_files libs_files.rpm gram*.tab.c perl-pcp-*.list +*.[1-9].tmp diff --git a/configure.in b/configure.in index 6b73ccc..1547bef 100644 --- a/configure.in +++ b/configure.in @@ -1419,6 +1419,8 @@ AC_SUBST(pcp_x11_appdefs_dir) dnl man pages (source) have_gzipped_manpages=false have_bzip2ed_manpages=false +need_old_tbl_header=false +man_header= pcp_man_dir=`eval echo $mandir` pcp_man_dir=`eval echo $pcp_man_dir` if test "`echo $pcp_man_dir | sed 's;/.*\$;;'`" = NONE @@ -1443,29 +1445,35 @@ then then have_gzipped_manpages=true pcp_man_dir=$d + man_header=`$ZIP -d < $d/man1/man.1.gz | head -1` break elif test -f $d/man1/man.1.bz2 then have_bzip2ed_manpages=true pcp_man_dir=$d + man_header=`$BZIP2 -d < $d/man1/man.1.bz2 | head -1` break elif test -f $d/man1/man.1 then pcp_man_dir=$d + man_header=`head -1 $d/man1/man.1` break elif test -f $d/sman1/man.1.gz then have_gzipped_manpages=true pcp_man_dir=$d + man_header=`$ZIP -d < $d/sman1/man.1.gz | head -1` break elif test -f $d/sman1/man.1.bz2 then have_bzip2ed_manpages=true pcp_man_dir=$d + man_header=`$BZIP2 -d < $d/sman1/man.1.bz2 | head -1` break elif test -f $d/sman1/man.1 then pcp_man_dir=$d + man_header=`head -1 $d/sman1/man.1` break elif test -f $d/cat1/man.1 then @@ -1484,9 +1492,13 @@ then echo "FATAL ERROR: could not find pcp_man_dir directory." exit 1 fi +if test x"$man_header" = "x'\\\" t" -o x"$man_header" = "x'\\\" te" ; then + need_old_tbl_header=true +fi AC_SUBST(pcp_man_dir) AC_SUBST(have_gzipped_manpages) AC_SUBST(have_bzip2ed_manpages) +AC_SUBST(need_old_tbl_header) dnl public binaries pcp_bin_dir=`eval echo $bindir` diff --git a/src/include/builddefs.in b/src/include/builddefs.in index d2675e2..90faff4 100644 --- a/src/include/builddefs.in +++ b/src/include/builddefs.in @@ -172,6 +172,7 @@ MD5SUM = @md5sum@ HAVE_GZIPPED_MANPAGES = @have_gzipped_manpages@ HAVE_BZIP2ED_MANPAGES = @have_bzip2ed_manpages@ +NEED_OLD_TBL_HEADER = @need_old_tbl_header@ CPP_SIMPLE = @cpp_simple@ CPP_SIMPLE_ARGS = @cpp_simple_args@ RDYNAMIC_FLAG = @rdynamic_flag@ @@ -241,15 +242,19 @@ INSTALL_MAN = \ [ -z "$$m" -o "$$m" = "\\" ] && continue; \ t=$(MAN_DEST)/$$m.$(MAN_SECTION); \ if $$first; then \ + _tfx= ;\ + if $(NEED_OLD_TBL_HEADER) ; then \ + $(SED) -e "1s/^'\\\\\"! tbl.*/'\\\\\" t/" $$d > $$d.tmp; _tfx=.tmp; \ + fi; \ if $(HAVE_GZIPPED_MANPAGES) ; then \ - $(ZIP) -c $$d > $$d.gz; _sfx=.gz; \ + $(ZIP) -c $$d$$_tfx > $$d.gz; _tfx=.gz; _sfx=.gz; \ fi; \ if $(HAVE_BZIP2ED_MANPAGES) ; then \ - $(BZIP2) -c $$d > $$d.bz2; _sfx=.bz2; \ + $(BZIP2) -c $$d$$_tfx > $$d.bz2; _tfx=.bz2; _sfx=.bz2; \ fi; \ u=$$m.$(MAN_SECTION)$$_sfx; \ - echo $(INSTALL) -m 644 $${d}$$_sfx $${t}$$_sfx; \ - $(INSTALL) -m 644 $${d}$$_sfx $${t}$$_sfx; \ + echo $(INSTALL) -m 644 $${d}$$_tfx $${t}$$_sfx; \ + $(INSTALL) -m 644 $${d}$$_tfx $${t}$$_sfx; \ else \ if test ! -z $(CASE_INSENSITIVE_FS) -a $(CASE_INSENSITIVE_FS); then \ if test "`echo $$u | tr 'a-z' 'A-Z'`" != "`basename $${t}$$_sfx | tr 'a-z' 'A-Z'`"; then \ -- Greg. From kenj@internode.on.net Sun Apr 11 17:04:41 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: * X-Spam-Status: No, score=1.1 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, URIBL_BLACK autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3BM4eT4052956 for ; Sun, 11 Apr 2010 17:04:41 -0500 X-ASG-Debug-ID: 1271023561-303103210000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 51F442B5341 for ; Sun, 11 Apr 2010 15:06:01 -0700 (PDT) Received: from mail.internode.on.net (bld-mail13.adl6.internode.on.net [150.101.137.98]) by cuda.sgi.com with ESMTP id RGHSJT1vfN62Etey for ; Sun, 11 Apr 2010 15:06:01 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20328447-1927428 for multiple; Mon, 12 Apr 2010 07:36:00 +0930 (CST) X-ASG-Orig-Subj: Re: [pcp] Format .TS in manpages correctly on Solaris Subject: Re: [pcp] Format .TS in manpages correctly on Solaris From: Ken McDonell Reply-To: kenj@internode.on.net To: Greg Banks Cc: PCP Mailing List In-Reply-To: <4BC1BD0D.3040803@evostor.com> References: <4BC1BD0D.3040803@evostor.com> Content-Type: text/plain; charset="UTF-8" Date: Mon, 12 Apr 2010 08:05:35 +1000 Message-ID: <1271023535.24244.6.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail13.adl6.internode.on.net[150.101.137.98] X-Barracuda-Start-Time: 1271023594 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0703 1.0000 -1.5731 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.57 X-Barracuda-Spam-Status: No, SCORE=-1.57 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27251 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Sigh ... why do people in our business (in this case Sun or BSD, but SGI was no better) mistakenly think they are cleverer than the previous person, and make changes for the sake of being different not better? I did consider doing the troff pre-processing in the build (or even in the source tree, maintaining both the original and the post-processed versions in the repository) ... but there are 9 files and that is probably just beyond the point of pain for doing it by hand and avoiding edits going in the wrong file. So, Greg, would you like me to push this forward, or will you piggy-back on Max's git tree at oss.sgi.com? On Sun, 2010-04-11 at 22:14 +1000, Greg Banks wrote: > G'day, > > commit 326c5a61d378625310c1c72750df7e10e1a081cd > Author: Greg Banks > Date: Sun Apr 11 21:52:16 2010 +1000 > > Format .TS in manpages correctly on Solaris > > Solaris needs the right obscure magic in the first line of > each manpage in order for the .TS/.TE table macros to work. > The PCP manpages are formatted with a different piece of > obscure magic designed to work on Irix. > > This change automatically detects the old fashioned .\" t > magic in a system manpage at configure time, and if present > tweaks the PCP manpages at install time to contain it. > > diff --git a/.gitignore b/.gitignore > index 1581608..1ac435c 100644 > --- a/.gitignore > +++ b/.gitignore > @@ -19,3 +19,4 @@ libs_files > libs_files.rpm > gram*.tab.c > perl-pcp-*.list > +*.[1-9].tmp > diff --git a/configure.in b/configure.in > index 6b73ccc..1547bef 100644 > --- a/configure.in > +++ b/configure.in > @@ -1419,6 +1419,8 @@ AC_SUBST(pcp_x11_appdefs_dir) > dnl man pages (source) > have_gzipped_manpages=false > have_bzip2ed_manpages=false > +need_old_tbl_header=false > +man_header= > pcp_man_dir=`eval echo $mandir` > pcp_man_dir=`eval echo $pcp_man_dir` > if test "`echo $pcp_man_dir | sed 's;/.*\$;;'`" = NONE > @@ -1443,29 +1445,35 @@ then > then > have_gzipped_manpages=true > pcp_man_dir=$d > + man_header=`$ZIP -d < $d/man1/man.1.gz | head -1` > break > elif test -f $d/man1/man.1.bz2 > then > have_bzip2ed_manpages=true > pcp_man_dir=$d > + man_header=`$BZIP2 -d < $d/man1/man.1.bz2 | head -1` > break > elif test -f $d/man1/man.1 > then > pcp_man_dir=$d > + man_header=`head -1 $d/man1/man.1` > break > elif test -f $d/sman1/man.1.gz > then > have_gzipped_manpages=true > pcp_man_dir=$d > + man_header=`$ZIP -d < $d/sman1/man.1.gz | head -1` > break > elif test -f $d/sman1/man.1.bz2 > then > have_bzip2ed_manpages=true > pcp_man_dir=$d > + man_header=`$BZIP2 -d < $d/sman1/man.1.bz2 | head -1` > break > elif test -f $d/sman1/man.1 > then > pcp_man_dir=$d > + man_header=`head -1 $d/sman1/man.1` > break > elif test -f $d/cat1/man.1 > then > @@ -1484,9 +1492,13 @@ then > echo "FATAL ERROR: could not find pcp_man_dir directory." > exit 1 > fi > +if test x"$man_header" = "x'\\\" t" -o x"$man_header" = "x'\\\" te" ; then > + need_old_tbl_header=true > +fi > AC_SUBST(pcp_man_dir) > AC_SUBST(have_gzipped_manpages) > AC_SUBST(have_bzip2ed_manpages) > +AC_SUBST(need_old_tbl_header) > > dnl public binaries > pcp_bin_dir=`eval echo $bindir` > diff --git a/src/include/builddefs.in b/src/include/builddefs.in > index d2675e2..90faff4 100644 > --- a/src/include/builddefs.in > +++ b/src/include/builddefs.in > @@ -172,6 +172,7 @@ MD5SUM = @md5sum@ > > HAVE_GZIPPED_MANPAGES = @have_gzipped_manpages@ > HAVE_BZIP2ED_MANPAGES = @have_bzip2ed_manpages@ > +NEED_OLD_TBL_HEADER = @need_old_tbl_header@ > CPP_SIMPLE = @cpp_simple@ > CPP_SIMPLE_ARGS = @cpp_simple_args@ > RDYNAMIC_FLAG = @rdynamic_flag@ > @@ -241,15 +242,19 @@ INSTALL_MAN = \ > [ -z "$$m" -o "$$m" = "\\" ] && continue; \ > t=$(MAN_DEST)/$$m.$(MAN_SECTION); \ > if $$first; then \ > + _tfx= ;\ > + if $(NEED_OLD_TBL_HEADER) ; then \ > + $(SED) -e "1s/^'\\\\\"! tbl.*/'\\\\\" t/" $$d > $$d.tmp; > _tfx=.tmp; \ > + fi; \ > if $(HAVE_GZIPPED_MANPAGES) ; then \ > - $(ZIP) -c $$d > $$d.gz; _sfx=.gz; \ > + $(ZIP) -c $$d$$_tfx > $$d.gz; _tfx=.gz; _sfx=.gz; \ > fi; \ > if $(HAVE_BZIP2ED_MANPAGES) ; then \ > - $(BZIP2) -c $$d > $$d.bz2; _sfx=.bz2; \ > + $(BZIP2) -c $$d$$_tfx > $$d.bz2; _tfx=.bz2; _sfx=.bz2; \ > fi; \ > u=$$m.$(MAN_SECTION)$$_sfx; \ > - echo $(INSTALL) -m 644 $${d}$$_sfx $${t}$$_sfx; \ > - $(INSTALL) -m 644 $${d}$$_sfx $${t}$$_sfx; \ > + echo $(INSTALL) -m 644 $${d}$$_tfx $${t}$$_sfx; \ > + $(INSTALL) -m 644 $${d}$$_tfx $${t}$$_sfx; \ > else \ > if test ! -z $(CASE_INSENSITIVE_FS) -a $(CASE_INSENSITIVE_FS); > then \ > if test "`echo $$u | tr 'a-z' 'A-Z'`" != "`basename > $${t}$$_sfx | tr 'a-z' 'A-Z'`"; then \ > > > > From nathans@aconex.com Sun Apr 11 18:56:10 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.7 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX,J_CHICKENPOX_21,J_CHICKENPOX_43,J_CHICKENPOX_64, J_CHICKENPOX_71,J_CHICKENPOX_73,LOCAL_GNU_PATCH autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3BNu945061098 for ; Sun, 11 Apr 2010 18:56:10 -0500 X-ASG-Debug-ID: 1271030248-489f02740000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 60751125AC53 for ; Sun, 11 Apr 2010 16:57:29 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id BtgKu6566TNH7wW6 for ; Sun, 11 Apr 2010 16:57:29 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id CAA258B9F4E; Mon, 12 Apr 2010 09:57:21 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id 1ap5JbOtvoCLW3UZ; Mon, 12 Apr 2010 09:57:21 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 7F57DA5011D; Mon, 12 Apr 2010 09:54:21 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 8113F48859E; Mon, 12 Apr 2010 09:57:21 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id QHDF9OAnNFT2; Mon, 12 Apr 2010 09:57:16 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id BD0194885A9; Mon, 12 Apr 2010 09:57:16 +1000 (EST) Date: Mon, 12 Apr 2010 09:57:15 +1000 (EST) From: nathans@aconex.com Sender: nscott@aconex.com To: kenj@internode.on.net Cc: pcp@oss.sgi.com Message-ID: <523449220.539791271030235812.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1403858388.539631271030117913.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: Re: [pcp] pcp updates - PM_CONTEXT_LOCAL generalization Subject: Re: [pcp] pcp updates - PM_CONTEXT_LOCAL generalization MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_39417_1931723968.1271030235805" X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271030282 X-Barracuda-Bayes: INNOCENT GLOBAL 0.2665 1.0000 -0.4975 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: 0.00 X-Barracuda-Spam-Status: No, SCORE=0.00 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=BSF_RULE7568M, NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27259 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.00 NO_REAL_NAME From: does not include a real name 0.50 BSF_RULE7568M Custom Rule 7568M X-Virus-Status: Clean ------=_Part_39417_1931723968.1271030235805 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit ----- "Ken McDonell" wrote: > ... > If that goes well, then these should be ready for merging into the > main tree, modulo consideration of any review feedback. > Looking good. Couple of minor things... - the error messages in __pmLocalPMDA are using an old name for the routine (__pmAddLocalPMDA) - seems like do_pmda_spec() should be a libpcp API, ala the other parsing routines, since three tools (so far) use it & it is kinda like one of the "standard" command line options, almost. - at this stage, we could probably split libpcp/src/connect.c into a pmcd version and a local version, ala fetch.c & fetchlocal.c. Here's a patch that addresses these. Its untested (ran out of time on the train this morning, have to look at other stuff now). The parseError() code was pulled from the time window parser, might be time to make that shared too (in libpcp). cheers. -- Nathan ------=_Part_39417_1931723968.1271030235805 Content-Type: text/x-patch; name=localcontext.patch Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename=localcontext.patch diff --git a/src/include/impl.h b/src/include/impl.h index 77044f4..8367c89 100644 --- a/src/include/impl.h +++ b/src/include/impl.h @@ -1202,6 +1202,7 @@ extern void __dmpostfetch(__pmContext *, pmResult **); #define PM_LOCAL_DEL 2 #define PM_LOCAL_CLEAR 3 extern int __pmLocalPMDA(int, int, const char *, const char *); +extern int __pmParseLocalPMDA(const char *, char **); #ifdef __cplusplus } diff --git a/src/libpcp/src/GNUmakefile b/src/libpcp/src/GNUmakefile index 4a867fa..9760aaf 100644 --- a/src/libpcp/src/GNUmakefile +++ b/src/libpcp/src/GNUmakefile @@ -29,7 +29,7 @@ CFILES = connect.c context.c desc.c err.c fetch.c freeresult.c \ checksum.c rtime.c tv.c spec.c fetchlocal.c optfetch.c AF.c \ stuffvalue.c endian.c config.c access.c authcookie.c auxconnect.c \ p_lcontrol.c p_lrequest.c p_lstatus.c logconnect.c logcontrol.c \ - derive.c derive_fetch.c + connectlocal.c derive.c derive_fetch.c HFILES = derive.h diff --git a/src/libpcp/src/connect.c b/src/libpcp/src/connect.c index 47c65f7..206c940 100644 --- a/src/libpcp/src/connect.c +++ b/src/libpcp/src/connect.c @@ -19,13 +19,6 @@ #include "pmapi.h" #include "impl.h" #include "pmda.h" -#include "dsotbl.h" - -static __pmDSO *dsotab = dsotab_i; - -#if defined(HAVE_DLFCN_H) -#include -#endif /* MY_BUFLEN needs to big enough to hold "hostname port" */ #define MY_BUFLEN (MAXHOSTNAMELEN+10) @@ -382,359 +375,3 @@ __pmConnectPMCD(pmHostSpec *hosts, int nhosts) #endif return fd; } - -/* - * As of PCP version 2.1, we're no longer searching for DSO's; - * pmcd's config file should have full paths to each of 'em. - */ -const char * -__pmFindPMDA(const char *name) -{ - return (access(name, F_OK) == 0) ? name : NULL; -} - -__pmDSO * -__pmLookupDSO(int domain) -{ - int i; - for (i = 0; i < numdso; i++) { - if (dsotab[i].domain == domain && dsotab[i].handle != NULL) - return &dsotab[i]; - } - return NULL; -} - -int -__pmConnectLocal(void) -{ - int i; - __pmDSO *dp; - char pathbuf[MAXPATHLEN]; - const char *path; -#if defined(HAVE_DLOPEN) - unsigned int challenge; - void (*initp)(pmdaInterface *); -#endif - - for (i = 0; i < numdso; i++) { - dp = &dsotab[i]; - if (dp->domain == -1 || dp->handle != NULL) - continue; - if (dp->domain == SAMPLE_DSO) { - /* - * only attach sample pmda dso if env var PCP_LITE_SAMPLE - * or PMDA_LOCAL_SAMPLE is set - */ - if (getenv("PCP_LITE_SAMPLE") == NULL && - getenv("PMDA_LOCAL_SAMPLE") == NULL) { - /* no sample pmda */ - dp->domain = -1; - continue; - } - } -#if defined(PROC_DSO) - /* - * For Linux (and perhaps anything other than IRIX), the proc - * PMDA is part of the OS PMDA, so this one cannot be optional - * ... the makefile will ensure dsotbl.h is set up correctly - * and PROC_DSO will or will not be defined as required - */ - if (dp->domain == PROC_DSO) { - /* - * only attach proc pmda dso if env var PMDA_LOCAL_PROC - * is set - */ - if (getenv("PMDA_LOCAL_PROC") == NULL) { - /* no proc pmda */ - dp->domain = -1; - continue; - } - } -#endif - - /* - * __pmLocalPMDA() means the path to the DSO may be something - * other than relative to $PCP_PMDAS_DIR ... need to try both - * options and also with and without DSO_SUFFIX (so, dll, etc) - */ - snprintf(pathbuf, sizeof(pathbuf), "%s%c%s", - pmGetConfig("PCP_PMDAS_DIR"), __pmPathSeparator(), dp->name); - if ((path = __pmFindPMDA(pathbuf)) == NULL) { - snprintf(pathbuf, sizeof(pathbuf), "%s%c%s.%s", - pmGetConfig("PCP_PMDAS_DIR"), __pmPathSeparator(), dp->name, DSO_SUFFIX); - if ((path = __pmFindPMDA(pathbuf)) == NULL) { - if ((path = __pmFindPMDA(dp->name)) == NULL) { - snprintf(pathbuf, sizeof(pathbuf), "%s.%s", dp->name, DSO_SUFFIX); - if ((path = __pmFindPMDA(pathbuf)) == NULL) { - pmprintf("__pmConnectLocal: Warning: cannot find DSO at \"%s\" or \"%s\"\n", - pathbuf, dp->name); - pmflush(); - dp->domain = -1; - dp->handle = NULL; - continue; - } - } - } - } -#if defined(HAVE_DLOPEN) - dp->handle = dlopen(path, RTLD_NOW); - if (dp->handle == NULL) { - pmprintf("__pmConnectLocal: Warning: error attaching DSO " - "\"%s\"\n%s\n\n", path, dlerror()); - pmflush(); - dp->domain = -1; - } -#else /* ! HAVE_DLOPEN */ - dp->handle = NULL; - pmprintf("__pmConnectLocal: Warning: error attaching DSO \"%s\"\n", - path); - pmprintf("No dynamic DSO/DLL support on this platform\n\n"); - pmflush(); - dp->domain = -1; -#endif - - if (dp->handle == NULL) - continue; - -#if defined(HAVE_DLOPEN) - /* - * rest of this only makes sense if the dlopen() worked - */ - if (dp->init == NULL) - initp = NULL; - else - initp = (void (*)(pmdaInterface *))dlsym(dp->handle, dp->init); - if (initp == NULL) { - pmprintf("__pmConnectLocal: Warning: couldn't find init function " - "\"%s\" in DSO \"%s\"\n", dp->init, path); - pmflush(); - dlclose(dp->handle); - dp->domain = -1; - continue; - } - - /* - * Pass in the expected domain id. - * The PMDA initialization routine can (a) ignore it, (b) check it - * is the expected value, or (c) self-adapt. - */ - dp->dispatch.domain = dp->domain; - - /* - * the PMDA interface / PMAPI version discovery as a "challenge" ... - * for pmda_interface it is all the bits being set, - * for pmapi_version it is the complement of the one you are using now - */ - challenge = 0xff; - dp->dispatch.comm.pmda_interface = challenge; - dp->dispatch.comm.pmapi_version = ~PMAPI_VERSION; - - dp->dispatch.comm.flags = 0; - dp->dispatch.status = 0; - - (*initp)(&dp->dispatch); - - if (dp->dispatch.status != 0) { - /* initialization failed for some reason */ - pmprintf("__pmConnectLocal: Warning: initialization " - "routine \"%s\" failed in DSO \"%s\": %s\n", - dp->init, path, pmErrStr(dp->dispatch.status)); - pmflush(); - dlclose(dp->handle); - dp->domain = -1; - } - else { - if (dp->dispatch.comm.pmda_interface == challenge) { - /* - * DSO did not change pmda_interface, assume PMAPI version 1 - * from PCP 1.x and PMDA_INTERFACE_1 - */ - dp->dispatch.comm.pmda_interface = PMDA_INTERFACE_1; - dp->dispatch.comm.pmapi_version = PMAPI_VERSION_1; - } - else { - /* - * gets a bit tricky ... - * interface_version (8-bits) used to be version (4-bits), - * so it is possible that only the bottom 4 bits were - * changed and in this case the PMAPI version is 1 for - * PCP 1.x - */ - if ((dp->dispatch.comm.pmda_interface & 0xf0) == (challenge & 0xf0)) { - dp->dispatch.comm.pmda_interface &= 0x0f; - dp->dispatch.comm.pmapi_version = PMAPI_VERSION_1; - } - } - - if (dp->dispatch.comm.pmda_interface < PMDA_INTERFACE_1 || - dp->dispatch.comm.pmda_interface > PMDA_INTERFACE_LATEST) { - pmprintf("__pmConnectLocal: Error: Unknown PMDA interface " - "version %d in \"%s\" DSO\n", - dp->dispatch.comm.pmda_interface, path); - pmflush(); - dlclose(dp->handle); - dp->domain = -1; - } - - if (dp->dispatch.comm.pmapi_version != PMAPI_VERSION_1 && - dp->dispatch.comm.pmapi_version != PMAPI_VERSION_2) { - pmprintf("__pmConnectLocal: Error: Unknown PMAPI version %d " - "in \"%s\" DSO\n", - dp->dispatch.comm.pmapi_version, path); - pmflush(); - dlclose(dp->handle); - dp->domain = -1; - } - } -#endif /* HAVE_DLOPEN */ - } - - return 0; -} - -int -__pmLocalPMDA(int op, int domain, const char *name, const char *init) -{ - int sts = 0; - int i; - -#ifdef PCP_DEBUG - if (pmDebug & DBG_TRACE_CONTEXT) { - fprintf(stderr, "__pmLocalPMDA(op="); - if (op == PM_LOCAL_ADD) fprintf(stderr, "ADD"); - else if (op == PM_LOCAL_DEL) fprintf(stderr, "DEL"); - else if (op == PM_LOCAL_CLEAR) fprintf(stderr, "CLEAR"); - else fprintf(stderr, "%d ???", op); - fprintf(stderr, ", domain=%d, name=%s, init=%s)\n", domain, name, init); - } -#endif - - if (dsotab == dsotab_i) { - /* - * first call, so promote dsotab[] to be a malloc'd copy of - * dsotab_i[] so we can realloc from here on ... - */ - if ((dsotab = (__pmDSO *)malloc(numdso*sizeof(__pmDSO))) == NULL) { - sts = -errno; - __pmNoMem("__pmAddLocalPMDA malloc", numdso*sizeof(__pmDSO), PM_RECOV_ERR); - dsotab = dsotab_i; - return sts; - } - memcpy((void *)dsotab, (void *)dsotab_i, numdso*sizeof(__pmDSO)); - /* - * need to strdup name and init so PM_LOCAL_CLEAR works for - * all entries - */ - for (i = 0; i < numdso; i++) { - if ((dsotab[i].name = strdup(dsotab_i[i].name)) == NULL) { - sts = -errno; - __pmNoMem("__pmAddLocalPMDA init name", strlen(dsotab_i[i].name)+1, PM_RECOV_ERR); - i--; - while (i >= 0) { - free(dsotab[i].name); - free(dsotab[i].init); - i--; - } - free(dsotab); - dsotab = dsotab_i; - return sts; - } - if ((dsotab[i].init = strdup(dsotab_i[i].init)) == NULL) { - sts = -errno; - __pmNoMem("__pmAddLocalPMDA init", strlen(dsotab_i[i].init)+1, PM_RECOV_ERR); - free(dsotab[i].name); - i--; - while (i >= 0) { - free(dsotab[i].name); - free(dsotab[i].init); - i--; - } - free(dsotab); - dsotab = dsotab_i; - return sts; - } - } - } - - switch (op) { - case PM_LOCAL_ADD: - if ((dsotab = (__pmDSO *)realloc(dsotab, (numdso+1)*sizeof(__pmDSO))) == NULL) { - __pmNoMem("__pmAddLocalPMDA realloc", (numdso+1)*sizeof(__pmDSO), PM_FATAL_ERR); - /*NOTREACHED*/ - } - dsotab[numdso].domain = domain; - if (name == NULL) { - /* odd, will fail later at dlopen */ - dsotab[numdso].name = NULL; - } - else { - if ((dsotab[numdso].name = strdup(name)) == NULL) { - sts = -errno; - __pmNoMem("__pmAddLocalPMDA name", strlen(name)+1, PM_RECOV_ERR); - return sts; - } - } - if (init == NULL) { - /* odd, will fail later at initialization call */ - dsotab[numdso].init = NULL; - } - else { - if ((dsotab[numdso].init = strdup(init)) == NULL) { - sts = -errno; - __pmNoMem("__pmAddLocalPMDA init", strlen(init)+1, PM_RECOV_ERR); - return sts; - } - } - dsotab[numdso].handle = NULL; - numdso++; - break; - - case PM_LOCAL_DEL: - sts = PM_ERR_INDOM; - for (i = 0; i < numdso; i++) { - if ((domain != -1 && dsotab[i].domain == domain) || - (name != NULL && strcmp(dsotab[i].name, name) == 0)) { - if (dsotab[i].handle) { - dlclose(dsotab[i].handle); - dsotab[i].handle = NULL; - } - dsotab[i].domain = -1; - sts = 0; - } - } - break; - - case PM_LOCAL_CLEAR: - for (i = 0; i < numdso; i++) { - free(dsotab[i].name); - free(dsotab[i].init); - if (dsotab[i].handle) - dlclose(dsotab[i].handle); - } - free(dsotab); - dsotab = NULL; - numdso = 0; - break; - - default: - sts = PM_ERR_CONV; - break; - } - -#ifdef PCP_DEBUG - if (pmDebug & DBG_TRACE_CONTEXT) { - if (sts != 0) - fprintf(stderr, "__pmLocalPMDA -> %s\n", pmErrStr(sts)); - fprintf(stderr, "Local Context PMDA Table"); - if (numdso == 0) - fprintf(stderr, " ... empty"); - fputc('\n', stderr); - for (i = 0; i < numdso; i++) { - fprintf(stderr, "%p [%d] domain=%d name=%s init=%s handle=%p\n", - &dsotab[i], i, dsotab[i].domain, dsotab[i].name, dsotab[i].init, dsotab[i].handle); - } - } -#endif - - return sts; -} diff --git a/src/libpcp/src/connectlocal.c b/src/libpcp/src/connectlocal.c new file mode 100644 index 0000000..4baacfa --- /dev/null +++ b/src/libpcp/src/connectlocal.c @@ -0,0 +1,496 @@ +/* + * Copyright (c) 1995-2002,2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2010 Ken McDonell. All Rights Reserved. + * + * This library is free software; you can redistribute it and/or modify it + * under the terms of the GNU Lesser General Public License as published + * by the Free Software Foundation; either version 2.1 of the License, or + * (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public + * License for more details. + */ + +#include "pmapi.h" +#include "impl.h" +#include "pmda.h" +#include "dsotbl.h" +#include + +static __pmDSO *dsotab = dsotab_i; + +#if defined(HAVE_DLFCN_H) +#include +#endif + +/* + * As of PCP version 2.1, we're no longer searching for DSO's; + * pmcd's config file should have full paths to each of 'em. + */ +const char * +__pmFindPMDA(const char *name) +{ + return (access(name, F_OK) == 0) ? name : NULL; +} + +__pmDSO * +__pmLookupDSO(int domain) +{ + int i; + for (i = 0; i < numdso; i++) { + if (dsotab[i].domain == domain && dsotab[i].handle != NULL) + return &dsotab[i]; + } + return NULL; +} + +int +__pmConnectLocal(void) +{ + int i; + __pmDSO *dp; + char pathbuf[MAXPATHLEN]; + const char *path; +#if defined(HAVE_DLOPEN) + unsigned int challenge; + void (*initp)(pmdaInterface *); +#endif + + for (i = 0; i < numdso; i++) { + dp = &dsotab[i]; + if (dp->domain == -1 || dp->handle != NULL) + continue; + if (dp->domain == SAMPLE_DSO) { + /* + * only attach sample pmda dso if env var PCP_LITE_SAMPLE + * or PMDA_LOCAL_SAMPLE is set + */ + if (getenv("PCP_LITE_SAMPLE") == NULL && + getenv("PMDA_LOCAL_SAMPLE") == NULL) { + /* no sample pmda */ + dp->domain = -1; + continue; + } + } +#if defined(PROC_DSO) + /* + * For Linux (and perhaps anything other than IRIX), the proc + * PMDA is part of the OS PMDA, so this one cannot be optional + * ... the makefile will ensure dsotbl.h is set up correctly + * and PROC_DSO will or will not be defined as required + */ + if (dp->domain == PROC_DSO) { + /* + * only attach proc pmda dso if env var PMDA_LOCAL_PROC + * is set + */ + if (getenv("PMDA_LOCAL_PROC") == NULL) { + /* no proc pmda */ + dp->domain = -1; + continue; + } + } +#endif + + /* + * __pmLocalPMDA() means the path to the DSO may be something + * other than relative to $PCP_PMDAS_DIR ... need to try both + * options and also with and without DSO_SUFFIX (so, dll, etc) + */ + snprintf(pathbuf, sizeof(pathbuf), "%s%c%s", + pmGetConfig("PCP_PMDAS_DIR"), __pmPathSeparator(), dp->name); + if ((path = __pmFindPMDA(pathbuf)) == NULL) { + snprintf(pathbuf, sizeof(pathbuf), "%s%c%s.%s", + pmGetConfig("PCP_PMDAS_DIR"), __pmPathSeparator(), dp->name, DSO_SUFFIX); + if ((path = __pmFindPMDA(pathbuf)) == NULL) { + if ((path = __pmFindPMDA(dp->name)) == NULL) { + snprintf(pathbuf, sizeof(pathbuf), "%s.%s", dp->name, DSO_SUFFIX); + if ((path = __pmFindPMDA(pathbuf)) == NULL) { + pmprintf("__pmConnectLocal: Warning: cannot find DSO at \"%s\" or \"%s\"\n", + pathbuf, dp->name); + pmflush(); + dp->domain = -1; + dp->handle = NULL; + continue; + } + } + } + } +#if defined(HAVE_DLOPEN) + dp->handle = dlopen(path, RTLD_NOW); + if (dp->handle == NULL) { + pmprintf("__pmConnectLocal: Warning: error attaching DSO " + "\"%s\"\n%s\n\n", path, dlerror()); + pmflush(); + dp->domain = -1; + } +#else /* ! HAVE_DLOPEN */ + dp->handle = NULL; + pmprintf("__pmConnectLocal: Warning: error attaching DSO \"%s\"\n", + path); + pmprintf("No dynamic DSO/DLL support on this platform\n\n"); + pmflush(); + dp->domain = -1; +#endif + + if (dp->handle == NULL) + continue; + +#if defined(HAVE_DLOPEN) + /* + * rest of this only makes sense if the dlopen() worked + */ + if (dp->init == NULL) + initp = NULL; + else + initp = (void (*)(pmdaInterface *))dlsym(dp->handle, dp->init); + if (initp == NULL) { + pmprintf("__pmConnectLocal: Warning: couldn't find init function " + "\"%s\" in DSO \"%s\"\n", dp->init, path); + pmflush(); + dlclose(dp->handle); + dp->domain = -1; + continue; + } + + /* + * Pass in the expected domain id. + * The PMDA initialization routine can (a) ignore it, (b) check it + * is the expected value, or (c) self-adapt. + */ + dp->dispatch.domain = dp->domain; + + /* + * the PMDA interface / PMAPI version discovery as a "challenge" ... + * for pmda_interface it is all the bits being set, + * for pmapi_version it is the complement of the one you are using now + */ + challenge = 0xff; + dp->dispatch.comm.pmda_interface = challenge; + dp->dispatch.comm.pmapi_version = ~PMAPI_VERSION; + + dp->dispatch.comm.flags = 0; + dp->dispatch.status = 0; + + (*initp)(&dp->dispatch); + + if (dp->dispatch.status != 0) { + /* initialization failed for some reason */ + pmprintf("__pmConnectLocal: Warning: initialization " + "routine \"%s\" failed in DSO \"%s\": %s\n", + dp->init, path, pmErrStr(dp->dispatch.status)); + pmflush(); + dlclose(dp->handle); + dp->domain = -1; + } + else { + if (dp->dispatch.comm.pmda_interface == challenge) { + /* + * DSO did not change pmda_interface, assume PMAPI version 1 + * from PCP 1.x and PMDA_INTERFACE_1 + */ + dp->dispatch.comm.pmda_interface = PMDA_INTERFACE_1; + dp->dispatch.comm.pmapi_version = PMAPI_VERSION_1; + } + else { + /* + * gets a bit tricky ... + * interface_version (8-bits) used to be version (4-bits), + * so it is possible that only the bottom 4 bits were + * changed and in this case the PMAPI version is 1 for + * PCP 1.x + */ + if ((dp->dispatch.comm.pmda_interface & 0xf0) == (challenge & 0xf0)) { + dp->dispatch.comm.pmda_interface &= 0x0f; + dp->dispatch.comm.pmapi_version = PMAPI_VERSION_1; + } + } + + if (dp->dispatch.comm.pmda_interface < PMDA_INTERFACE_1 || + dp->dispatch.comm.pmda_interface > PMDA_INTERFACE_LATEST) { + pmprintf("__pmConnectLocal: Error: Unknown PMDA interface " + "version %d in \"%s\" DSO\n", + dp->dispatch.comm.pmda_interface, path); + pmflush(); + dlclose(dp->handle); + dp->domain = -1; + } + + if (dp->dispatch.comm.pmapi_version != PMAPI_VERSION_1 && + dp->dispatch.comm.pmapi_version != PMAPI_VERSION_2) { + pmprintf("__pmConnectLocal: Error: Unknown PMAPI version %d " + "in \"%s\" DSO\n", + dp->dispatch.comm.pmapi_version, path); + pmflush(); + dlclose(dp->handle); + dp->domain = -1; + } + } +#endif /* HAVE_DLOPEN */ + } + + return 0; +} + +int +__pmLocalPMDA(int op, int domain, const char *name, const char *init) +{ + int sts = 0; + int i; + +#ifdef PCP_DEBUG + if (pmDebug & DBG_TRACE_CONTEXT) { + fprintf(stderr, "__pmLocalPMDA(op="); + if (op == PM_LOCAL_ADD) fprintf(stderr, "ADD"); + else if (op == PM_LOCAL_DEL) fprintf(stderr, "DEL"); + else if (op == PM_LOCAL_CLEAR) fprintf(stderr, "CLEAR"); + else fprintf(stderr, "%d ???", op); + fprintf(stderr, ", domain=%d, name=%s, init=%s)\n", domain, name, init); + } +#endif + + if (dsotab == dsotab_i) { + /* + * first call, so promote dsotab[] to be a malloc'd copy of + * dsotab_i[] so we can realloc from here on ... + */ + if ((dsotab = (__pmDSO *)malloc(numdso*sizeof(__pmDSO))) == NULL) { + sts = -errno; + __pmNoMem("__pmLocalPMDA malloc", numdso*sizeof(__pmDSO), PM_RECOV_ERR); + dsotab = dsotab_i; + return sts; + } + memcpy((void *)dsotab, (void *)dsotab_i, numdso*sizeof(__pmDSO)); + /* + * need to strdup name and init so PM_LOCAL_CLEAR works for + * all entries + */ + for (i = 0; i < numdso; i++) { + if ((dsotab[i].name = strdup(dsotab_i[i].name)) == NULL) { + sts = -errno; + __pmNoMem("__pmLocalPMDA init name", strlen(dsotab_i[i].name)+1, PM_RECOV_ERR); + i--; + while (i >= 0) { + free(dsotab[i].name); + free(dsotab[i].init); + i--; + } + free(dsotab); + dsotab = dsotab_i; + return sts; + } + if ((dsotab[i].init = strdup(dsotab_i[i].init)) == NULL) { + sts = -errno; + __pmNoMem("__pmLocalPMDA init", strlen(dsotab_i[i].init)+1, PM_RECOV_ERR); + free(dsotab[i].name); + i--; + while (i >= 0) { + free(dsotab[i].name); + free(dsotab[i].init); + i--; + } + free(dsotab); + dsotab = dsotab_i; + return sts; + } + } + } + + switch (op) { + case PM_LOCAL_ADD: + if ((dsotab = (__pmDSO *)realloc(dsotab, (numdso+1)*sizeof(__pmDSO))) == NULL) { + __pmNoMem("__pmLocalPMDA realloc", (numdso+1)*sizeof(__pmDSO), PM_FATAL_ERR); + /*NOTREACHED*/ + } + dsotab[numdso].domain = domain; + if (name == NULL) { + /* odd, will fail later at dlopen */ + dsotab[numdso].name = NULL; + } + else { + if ((dsotab[numdso].name = strdup(name)) == NULL) { + sts = -errno; + __pmNoMem("__pmLocalPMDA name", strlen(name)+1, PM_RECOV_ERR); + return sts; + } + } + if (init == NULL) { + /* odd, will fail later at initialization call */ + dsotab[numdso].init = NULL; + } + else { + if ((dsotab[numdso].init = strdup(init)) == NULL) { + sts = -errno; + __pmNoMem("__pmLocalPMDA init", strlen(init)+1, PM_RECOV_ERR); + return sts; + } + } + dsotab[numdso].handle = NULL; + numdso++; + break; + + case PM_LOCAL_DEL: + sts = PM_ERR_INDOM; + for (i = 0; i < numdso; i++) { + if ((domain != -1 && dsotab[i].domain == domain) || + (name != NULL && strcmp(dsotab[i].name, name) == 0)) { + if (dsotab[i].handle) { + dlclose(dsotab[i].handle); + dsotab[i].handle = NULL; + } + dsotab[i].domain = -1; + sts = 0; + } + } + break; + + case PM_LOCAL_CLEAR: + for (i = 0; i < numdso; i++) { + free(dsotab[i].name); + free(dsotab[i].init); + if (dsotab[i].handle) + dlclose(dsotab[i].handle); + } + free(dsotab); + dsotab = NULL; + numdso = 0; + break; + + default: + sts = PM_ERR_CONV; + break; + } + +#ifdef PCP_DEBUG + if (pmDebug & DBG_TRACE_CONTEXT) { + if (sts != 0) + fprintf(stderr, "__pmLocalPMDA -> %s\n", pmErrStr(sts)); + fprintf(stderr, "Local Context PMDA Table"); + if (numdso == 0) + fprintf(stderr, " ... empty"); + fputc('\n', stderr); + for (i = 0; i < numdso; i++) { + fprintf(stderr, "%p [%d] domain=%d name=%s init=%s handle=%p\n", + &dsotab[i], i, dsotab[i].domain, dsotab[i].name, dsotab[i].init, dsotab[i].handle); + } + } +#endif + + return sts; +} + +/* Construct error message buffer for syntactic error */ +static void +parseError(const char *spec, const char *point, char *msg, char **rslt) +{ + int need = 2 * (int)strlen(spec) + (int)strlen(msg) + 8; + const char *p; + char *q; + + if ((*rslt = malloc(need)) == NULL) + __pmNoMem("__pmParseLocalPMDA", need, PM_FATAL_ERR); + q = *rslt; + + for (p = spec; *p != '\0'; p++) + *q++ = *p; + *q++ = '\n'; + for (p = spec; p != point; p++) + *q++ = isgraph((int)*p) ? ' ' : *p; + sprintf(q, "^ -- "); + q += 5; + for (p = msg; *p != '\0'; p++) + *q++ = *p; + *q++ = '\n'; + *q = '\0'; +} + +/* + * Parse a command line string that encodes arguments to __pmLocalPMDA(), + * then call __pmLocalPMDA(). + * + * The syntax for the string is 1 to 4 fields separated by colons: + * - op ("add" for add, "del" for delete, "clear" for clear) + * - domain (PMDA's PMD) + * - path (path to DSO PMDA) + * - init (name of DSO's initialization routine) + */ +int +__pmParseLocalPMDA(const char *spec, char **errmsg) +{ + int op; + int domain = -1; + char *name = NULL; + char *init = NULL; + int sts; + char *arg; + char *ap; + + if ((arg = strdup(spec)) == NULL) { + sts = -errno; + __pmNoMem("__pmParseLocalPMDA dup spec", strlen(spec)+1, PM_RECOV_ERR); + return 1; + } + if (strncmp(arg, "add", 3) == 0) { + op = PM_LOCAL_ADD; + ap = &arg[3]; + } + else if (strncmp(arg, "del", 3) == 0) { + op = PM_LOCAL_DEL; + ap = &arg[3]; + } + else if (strncmp(arg, "clear", 5) == 0) { + op = PM_LOCAL_CLEAR; + ap = &arg[5]; + } + else { + parseError(spec, spec, "bad op in spec", errmsg); + return -1; + } + if (op == PM_LOCAL_CLEAR && *ap == '\0') + goto doit; + + if (*ap != ',') { + parseError(spec, ap, "bad spec", errmsg); + return -1; + } + arg = ++ap; + if (*ap != ',' && *ap != '\0') { + domain = (int)strtol(arg, &ap, 10); + if ((*ap != ',' && *ap != '\0') || domain < 0 || domain > 510) { + parseError(spec, ap, "bad domain in spec", errmsg); + return -1; + } + } + if (*ap == ',') { + ap++; + if (*ap == ',') { + /* no name, could have init (not useful but possible!) */ + ap++; + if (*ap != '\0') + init = ap; + } + else if (*ap != '\0') { + /* have name and possibly init */ + name = ap; + while (*ap != ',' && *ap != '\0') + ap++; + if (*ap == ',') { + *ap++ = '\0'; + if (*ap != '\0') + init = ap; + } + } + } + +doit: + sts = __pmLocalPMDA(op, domain, name, init); + if (sts < 0) { + char buffer[256]; + snprintf(buffer, sizeof(buffer), "__pmLocalPMDA: %s", pmErrStr(sts)); + *errmsg = strdup(buffer); + return -1; + } + return 0; +} diff --git a/src/pminfo/GNUmakefile b/src/pminfo/GNUmakefile index e0aeff4..89ea603 100644 --- a/src/pminfo/GNUmakefile +++ b/src/pminfo/GNUmakefile @@ -19,7 +19,7 @@ TOPDIR = ../.. include $(TOPDIR)/src/include/builddefs -CFILES = pminfo.c do_local_spec.c +CFILES = pminfo.c LLDLIBS = -lpcp CMDTARGET = pminfo$(EXECSUFFIX) diff --git a/src/pminfo/do_local_spec.c b/src/pminfo/do_local_spec.c deleted file mode 100644 index 2767282..0000000 --- a/src/pminfo/do_local_spec.c +++ /dev/null @@ -1,107 +0,0 @@ -/* - * Copyright (c) 2010 Ken McDonell. All Rights Reserved. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License as published by the - * Free Software Foundation; either version 2 of the License, or (at your - * option) any later version. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY - * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License - * for more details. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write to the Free Software Foundation, Inc., - * 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - */ - -/* - * Parse a command line string that encodes arguments to __pmLocalPMDA(), - * then call __pmLocalPMDA(). - * - * The syntax for the string is 1 to 4 fields separated by colons: - * - op ("add" for add, "del" for delete, "clear" for clear) - * - domain (PMDA's PMD) - * - path (path to DSO PMDA) - * - init (name of DSO's initialization routine) - */ - -#include "pmapi.h" -#include "impl.h" - -int -do_local_spec(const char *spec) -{ - int op; - int domain = -1; - char *name = NULL; - char *init = NULL; - int sts; - char *arg; - char *ap; - - if ((arg = strdup(spec)) == NULL) { - fprintf(stderr, "%s: cannot dup spec \"%s\": %s\n", pmProgname, spec, pmErrStr(-errno)); - return 1; - } - if (strncmp(arg, "add", 3) == 0) { - op = PM_LOCAL_ADD; - ap = &arg[3]; - } - else if (strncmp(arg, "del", 3) == 0) { - op = PM_LOCAL_DEL; - ap = &arg[3]; - } - else if (strncmp(arg, "clear", 5) == 0) { - op = PM_LOCAL_CLEAR; - ap = &arg[5]; - } - else { - fprintf(stderr, "%s: bad op in spec \"%s\"\n", pmProgname, spec); - return 1; - } - if (op == PM_LOCAL_CLEAR && *ap == '\0') - goto doit; - - if (*ap != ',') { - fprintf(stderr, "%s: bad spec \"%s\"\n", pmProgname, spec); - return 1; - } - arg = ++ap; - if (*ap != ',' && *ap != '\0') { - domain = (int)strtol(arg, &ap, 10); - if ((*ap != ',' && *ap != '\0') || domain < 0 || domain > 510) { - fprintf(stderr, "%s: bad domain (%d) in spec \"%s\"\n", pmProgname, domain, spec); - return 1; - } - } - if (*ap == ',') { - ap++; - if (*ap == ',') { - // no name, could have init (not useful but possible!) - ap++; - if (*ap != '\0') - init = ap; - } - else if (*ap != '\0') { - // have name and possibly init - name = ap; - while (*ap != ',' && *ap != '\0') - ap++; - if (*ap == ',') { - *ap++ = '\0'; - if (*ap != '\0') - init = ap; - } - } - } - -doit: - sts = __pmLocalPMDA(op, domain, name, init); - if (sts < 0) { - fprintf(stderr, "%s: __pmAddLocalPMDA failed: %s\n", pmProgname, pmErrStr(sts)); - return 1; - } - return 0; -} diff --git a/src/pminfo/pminfo.c b/src/pminfo/pminfo.c index 310ebfb..4409d40 100644 --- a/src/pminfo/pminfo.c +++ b/src/pminfo/pminfo.c @@ -45,8 +45,6 @@ static int zflag; /* for -z */ static char *tz; /* for -Z timezone */ static struct timeval start; /* start of time window */ -extern int do_local_spec(const char *); - /* * stolen from pmprobe.c ... cache all of the most recently requested * pmInDom ... @@ -435,7 +433,10 @@ ParseOptions(int argc, char *argv[]) break; case 'K': /* update local PMDA table */ - errflag += do_local_spec(optarg); + if (__pmParseLocalPMDA(optarg, &endnum) < 0) { + fprintf(stderr, "%s: __pmParseLocalPMDA failed\n%s\n", pmProgname, endnum); + errflag++; + } break; case 'L': /* local PMDA connection, no PMCD */ diff --git a/src/pmprobe/GNUmakefile b/src/pmprobe/GNUmakefile index a28d0b3..5d1e383 100644 --- a/src/pmprobe/GNUmakefile +++ b/src/pmprobe/GNUmakefile @@ -19,10 +19,9 @@ TOPDIR = ../.. include $(TOPDIR)/src/include/builddefs -CFILES = pmprobe.c do_local_spec.c +CFILES = pmprobe.c LLDLIBS = -lpcp CMDTARGET = pmprobe$(EXECSUFFIX) -LDIRT = do_local_spec.c default: $(CMDTARGET) @@ -35,10 +34,3 @@ default_pcp: default install_pcp: install -do_local_spec.c: ../pminfo/do_local_spec.c - rm -f do_local_spec.c - @echo "/********************************************************" >do_local_spec.c - @echo " * copied from ../pminfo/do_local_spec.c -- DO NOT EDIT *" >>do_local_spec.c - @echo " ********************************************************/" >>do_local_spec.c - cat ../pminfo/do_local_spec.c >>do_local_spec.c - chmod 444 do_local_spec.c diff --git a/src/pmprobe/pmprobe.c b/src/pmprobe/pmprobe.c index 39ba030..1c19433 100644 --- a/src/pmprobe/pmprobe.c +++ b/src/pmprobe/pmprobe.c @@ -159,7 +159,10 @@ main(int argc, char **argv) break; case 'K': /* update local PMDA table */ - errflag += do_local_spec(optarg); + if (__pmParseLocalPMDA(optarg, &msg) < 0) { + fprintf(stderr, "%s: __pmParseLocalPMDA failed\n%s\n", pmProgname, msg); + errflag++; + } break; case 'L': /* local PMDA connection, no PMCD */ diff --git a/src/pmval/GNUmakefile b/src/pmval/GNUmakefile index 7babacc..0b26c32 100644 --- a/src/pmval/GNUmakefile +++ b/src/pmval/GNUmakefile @@ -20,10 +20,9 @@ TOPDIR = ../.. include $(TOPDIR)/src/include/builddefs -CFILES = pmval.c do_local_spec.c +CFILES = pmval.c LLDFLAGS = -L$(TOPDIR)/src/libpcp_gui/src LLDLIBS = -lpcp_gui -lpcp -lm -LDIRT = do_local_spec.c CMDTARGET = pmval$(EXECSUFFIX) @@ -38,10 +37,3 @@ default_pcp: default install_pcp: install -do_local_spec.c: ../pminfo/do_local_spec.c - rm -f do_local_spec.c - @echo "/********************************************************" >do_local_spec.c - @echo " * copied from ../pminfo/do_local_spec.c -- DO NOT EDIT *" >>do_local_spec.c - @echo " ********************************************************/" >>do_local_spec.c - cat ../pminfo/do_local_spec.c >>do_local_spec.c - chmod 444 do_local_spec.c diff --git a/src/pmval/pmval.c b/src/pmval/pmval.c index 7d495b7..fd2aeac 100644 --- a/src/pmval/pmval.c +++ b/src/pmval/pmval.c @@ -772,8 +772,6 @@ getinstance(char *p) return start; } -extern int do_local_spec(const char *); - /* extract command line arguments - exits on error */ static void getargs(int argc, /* in - command line argument count */ @@ -886,7 +884,10 @@ getargs(int argc, /* in - command line argument count */ break; case 'K': /* update local PMDA table */ - errflag += do_local_spec(optarg); + if (__pmParseLocalPMDA(optarg, &endnum) < 0) { + fprintf(stderr, "%s: __pmParseLocalPMDA failed\n%s\n", pmProgname, endnum); + errflag++; + } break; case 'n': /* alternative name space file */ ------=_Part_39417_1931723968.1271030235805-- From kenj@internode.on.net Sun Apr 11 18:57:40 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.1 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3BNveeZ061181 for ; Sun, 11 Apr 2010 18:57:40 -0500 X-ASG-Debug-ID: 1271030343-3e6802d80000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id EC27F1258B89 for ; Sun, 11 Apr 2010 16:59:03 -0700 (PDT) Received: from mail.internode.on.net (bld-mail18.adl2.internode.on.net [150.101.137.103]) by cuda.sgi.com with ESMTP id bv69FvwglJXyJIvn for ; Sun, 11 Apr 2010 16:59:03 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20209541-1927428 for multiple; Mon, 12 Apr 2010 09:29:02 +0930 (CST) X-ASG-Orig-Subj: Re: [pcp] pmlogger_check is fooled by localhost Subject: Re: [pcp] pmlogger_check is fooled by localhost From: Ken McDonell Reply-To: kenj@internode.on.net To: Max Matveev Cc: pcp@oss.sgi.com In-Reply-To: <19393.8355.508754.477098@iinet.net.au> References: <19387.53537.727368.123975@iinet.net.au> <1270603621.15121.79.camel@bozo.localdomain> <19387.65322.68301.468516@iinet.net.au> <1270642137.15121.125.camel@bozo.localdomain> <19389.24907.135424.300701@iinet.net.au> <1270704927.8807.7.camel@bozo.localdomain> <19390.53020.92570.731051@iinet.net.au> <1270847750.18793.8.camel@bozo-laptop> <19393.8355.508754.477098@iinet.net.au> Content-Type: text/plain; charset="UTF-8" Date: Mon, 12 Apr 2010 09:58:38 +1000 Message-ID: <1271030318.24244.54.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail18.adl2.internode.on.net[150.101.137.103] X-Barracuda-Start-Time: 1271030373 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0209 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27259 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean I'm afraid it is more than just GetPort() ... we try and get the fqdn of the pmcd host in several places ... * GetPort() - to be written to the port map file (/tmp/pmlogger/* in unix-land) * sendstatus() - for response to a pmlc "status" request * do_preamble() - for the "fake" initial pmResult that contains the metrics pmcd.pmlogger.host, pmcd.pmlogger.port and pmcd.pmlogger.archive from pmlogger (not pmcd) So, one option could be for -H to disable all of the gethostname() and gethostbyname() calls and use the -h from the command line (or "localhost" by default) without translation. This would still leave the messy question of what to do with -P? You cannot have -h with -P, and -P needs to be able to talk to pmcd on the local host, so should -P imply "localhost" with -H, or the result of gethostname() without -H? But before exploring that, in your dns-free environment, what does gethostbyname("localhost") return? And I assume pcp -h localhost works as expected, correct? On Sun, 2010-04-11 at 11:06 +1000, Max Matveev wrote: > On Sat, 10 Apr 2010 07:15:50 +1000, Ken McDonell wrote: > > kenj> Yep, pretty close to my patch (no man page from me either) ... I just > kenj> went for the -H as a flag, rather than -H hostname (which is the same as > kenj> -h hostname, unless hostname happens to be localhost). > I can live with H being a flag, but I really like to remove > gethostbyname from GetPort - it's just another place where pmlogger is > trying to second-guess user's input and yet another place when it can > go wrong. > > kenj> But more importantly your source and mine don't match without the > kenj> patch ... I don't have archive_host[] in my source, it uses local[] > kenj> there and there is no second test for pmcd_host being NULL, no strncmp() > kenj> in an else branch ... ??? > That's because patch was on top of the earlier change I've made which is > available from > > http://oss.sgi.com/cgi-bin/gitweb.cgi?p=makc/pcp;a=shortlog;h=refs/heads/pmlogger > > It includes using different names for __pmLogCreate() and for the > pmNewConext and not trying to resolve the name when creating a map > file. > > And strncmp is a brain-fart - I'm surpised that the stuff works. > > max From nathans@aconex.com Sun Apr 11 20:56:03 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3C1u3VZ068885 for ; Sun, 11 Apr 2010 20:56:03 -0500 X-ASG-Debug-ID: 1271037444-489303b30000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 54A9F125BDB7 for ; Sun, 11 Apr 2010 18:57:25 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id oNXWCdCUb9G1nHEE for ; Sun, 11 Apr 2010 18:57:25 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id BD5A08BB000; Mon, 12 Apr 2010 11:56:55 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id rNVmvMrP3xI4zejY; Mon, 12 Apr 2010 11:56:55 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 7C01EA5011D; Mon, 12 Apr 2010 11:53:55 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 83C7C4885A9; Mon, 12 Apr 2010 11:56:55 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id DNg43mIvw7-p; Mon, 12 Apr 2010 11:56:50 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id CDE7948859E; Mon, 12 Apr 2010 11:56:50 +1000 (EST) Date: Mon, 12 Apr 2010 11:56:50 +1000 (EST) From: nathans@aconex.com Sender: nscott@aconex.com To: kenj Cc: pcp Message-ID: <1988890790.547131271037410710.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1688696220.546661271036941381.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: Local context vs dynamic namespace Subject: Local context vs dynamic namespace MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271037476 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0005 1.0000 -2.0180 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27266 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.00 NO_REAL_NAME From: does not include a real name X-Virus-Status: Clean Hi Ken, We're a long way down the path of converting all of our production environment over to using MMV and the dynamic namespace extensions. All going swimmingly so far. Just this morning we came across an issue with the bash metric name auto-completion. This uses pminfo in local context mode (-L) to provide a list of metric names which the shell can use for name completion with several pcp commands. In our namespace we have MMV:*:* (a top level mmv namespace entry) and ACONEX:*:* as well (likewise, at the top level). In local mode pminfo does not traverse, just reports "mmv" and "aconex". We can change the bash code to not use -L, but that seems like its just avoiding the real problem - should local context mode be able to descend into these parts of the namespace? I woulda thought it should, just like regular pmcd mode...? cheers. -- Nathan From gnb@evostor.com Sun Apr 11 21:21:19 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: * X-Spam-Status: No, score=1.6 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3C2LJXx070308 for ; Sun, 11 Apr 2010 21:21:19 -0500 X-ASG-Debug-ID: 1271038962-300602960000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from relay.ihostexchange.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id CB52D2B5711 for ; Sun, 11 Apr 2010 19:22:42 -0700 (PDT) Received: from relay.ihostexchange.net (relay.ihostexchange.net [66.46.182.55]) by cuda.sgi.com with ESMTP id tLsU1Xx6XlY8Eptb for ; Sun, 11 Apr 2010 19:22:42 -0700 (PDT) Received: from mailhost.internal (203.206.165.193) by smtp.ihostexchange.net (66.46.182.50) with Microsoft SMTP Server (TLS) id 8.1.393.1; Sun, 11 Apr 2010 22:22:25 -0400 Received: from [172.29.254.198] (unknown [172.29.254.198]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mailhost.internal (Postfix) with ESMTPSA id E4FBC308E6; Mon, 12 Apr 2010 12:22:25 +1000 (EST) Message-ID: <4BC283FC.3030604@evostor.com> Date: Mon, 12 Apr 2010 12:22:52 +1000 From: Greg Banks User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: "kenj@internode.on.net" CC: PCP Mailing List X-ASG-Orig-Subj: Re: [pcp] Format .TS in manpages correctly on Solaris Subject: Re: [pcp] Format .TS in manpages correctly on Solaris References: <4BC1BD0D.3040803@evostor.com> <1271023535.24244.6.camel@bozo.localdomain> In-Reply-To: <1271023535.24244.6.camel@bozo.localdomain> Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: 7bit X-Barracuda-Connect: relay.ihostexchange.net[66.46.182.55] X-Barracuda-Start-Time: 1271038992 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0001 1.0000 -2.0205 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27267 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Ken McDonell wrote: > So, Greg, would you like me to push this forward, or will you piggy-back > on Max's git tree at oss.sgi.com? > Could you please push it? I have to confess I have no idea of the shape of the current PCP development workflow and what tree is definitive and so forth, so the option where you deal with it is very attractive right now. -- Greg. From kenj@internode.on.net Sun Apr 11 21:30:11 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.2 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3C2UAxu070837 for ; Sun, 11 Apr 2010 21:30:11 -0500 X-ASG-Debug-ID: 1271039493-202300420000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 81920FDED9D for ; Sun, 11 Apr 2010 19:31:33 -0700 (PDT) Received: from mail.internode.on.net (bld-mail14.adl6.internode.on.net [150.101.137.99]) by cuda.sgi.com with ESMTP id BGyVaEaxc2Ajn1Wb for ; Sun, 11 Apr 2010 19:31:33 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20479117-1927428 for multiple; Mon, 12 Apr 2010 12:01:32 +0930 (CST) X-ASG-Orig-Subj: Re: [pcp] pcp updates - PM_CONTEXT_LOCAL generalization Subject: Re: [pcp] pcp updates - PM_CONTEXT_LOCAL generalization From: Ken McDonell Reply-To: kenj@internode.on.net To: nathans@aconex.com Cc: pcp@oss.sgi.com In-Reply-To: <523449220.539791271030235812.JavaMail.root@mail-au.aconex.com> References: <523449220.539791271030235812.JavaMail.root@mail-au.aconex.com> Content-Type: text/plain; charset="UTF-8" Date: Mon, 12 Apr 2010 12:31:07 +1000 Message-ID: <1271039467.24244.88.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail14.adl6.internode.on.net[150.101.137.99] X-Barracuda-Start-Time: 1271039524 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0767 1.0000 -1.5342 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.53 X-Barracuda-Spam-Status: No, SCORE=-1.53 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27267 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Mon, 2010-04-12 at 09:57 +1000, nathans@aconex.com wrote: > ----- "Ken McDonell" wrote: > > ... > > If that goes well, then these should be ready for merging into the > > main tree, modulo consideration of any review feedback. > > > > Looking good. Couple of minor things... > - the error messages in __pmLocalPMDA are using an old name for the > routine (__pmAddLocalPMDA) Thanks for catching that. > - seems like do_pmda_spec() should be a libpcp API, ala the other > parsing routines, since three tools (so far) use it & it is kinda > like one of the "standard" command line options, almost. I toyed with this ... I decided not to because libpcp seems to have become a dumping ground ... PCP started life with a "mean and lean" charter and once upon a time there were a only handful of routines in libpcp. Today we're exposing 320+ external functions in libpcp, and of these 50 are NOT called anywhere in the main pcp source tree (I know there are proprietary bits still in hiding and the usage from pmchart and other client tools outside the main pcp source tree, but still!). A further 55 external routines are only called from one source file. Now I'm not saying __pmParseLocalPMDA() is not a candidate for libpcp, I'm just not sure ... although it seems to have a stronger case than lots of other routines that are included in the library ... 8^(> > - at this stage, we could probably split libpcp/src/connect.c into > a pmcd version and a local version, ala fetch.c & fetchlocal.c. This is more of a religious issue ... I personally don't think code quality and engineer productivity has much to do with source line count, but then again I'm just an old fart who hates gratuitious comments in the source. > Here's a patch that addresses these. Its untested (ran out of time > on the train this morning, have to look at other stuff now). The > parseError() code was pulled from the time window parser, might be > time to make that shared too (in libpcp). __pmParseLocalPMDA() eats the whole buffer, so probably does not need the second argument ... the routine could return NULL for success, else a pointer to an error message. Do you want me to refine my changes or are you going to finish your patch? I think the end result will be much the same, so I'm easy ... just let me know. From nathans@aconex.com Sun Apr 11 22:00:53 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3C30rrH072681 for ; Sun, 11 Apr 2010 22:00:53 -0500 X-ASG-Debug-ID: 1271041332-255c02cc0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 5008A125C37A for ; Sun, 11 Apr 2010 20:02:13 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id loXrt7TwrYqhLCDJ for ; Sun, 11 Apr 2010 20:02:13 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 517A68BB5F6; Mon, 12 Apr 2010 13:02:11 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id WfUsBQmzFoXk7qjK; Mon, 12 Apr 2010 13:02:11 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 3B53EA5011D; Mon, 12 Apr 2010 12:59:11 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 45A9C4885A9; Mon, 12 Apr 2010 13:02:11 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id HOcrzPvVOKgb; Mon, 12 Apr 2010 13:02:06 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 896E24885AC; Mon, 12 Apr 2010 13:02:06 +1000 (EST) Date: Mon, 12 Apr 2010 13:02:06 +1000 (EST) From: Nathan Scott To: kenj@internode.on.net Cc: pcp@oss.sgi.com Message-ID: <1458453143.550201271041326429.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1271039467.24244.88.camel@bozo.localdomain> X-ASG-Orig-Subj: Re: [pcp] pcp updates - PM_CONTEXT_LOCAL generalization Subject: Re: [pcp] pcp updates - PM_CONTEXT_LOCAL generalization MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271041367 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0736 1.0000 -1.5533 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.55 X-Barracuda-Spam-Status: No, SCORE=-1.55 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27270 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean ----- "Ken McDonell" wrote: > On Mon, 2010-04-12 at 09:57 +1000, nathans@aconex.com wrote: > > ----- "Ken McDonell" wrote: > > > ... > > Now I'm not saying __pmParseLocalPMDA() is not a candidate for > libpcp, > I'm just not sure ... although it seems to have a stronger case than > lots of other routines that are included in the library ... 8^(> > Yeah. I think we should. The "make copies of the code, recompiled for each binary" thing was more of a downside than one more libpcp routine, IMO. pmdumptext and pmchart can use this treatment too, and to get the "same" code over there will be really painful without having it in libpcp. > > - at this stage, we could probably split libpcp/src/connect.c into > > a pmcd version and a local version, ala fetch.c & fetchlocal.c. > > This is more of a religious issue ... I personally don't think code > quality and engineer productivity has much to do with source line > count, > but then again I'm just an old fart who hates gratuitious comments in > the source. It was more about having "like" stuff together, and that file is pretty much two very different chunks of independent code. Was an easy split, I think we should still do it. > Do you want me to refine my changes or are you going to finish your > patch? I think the end result will be much the same, so I'm easy ... > just let me know. I'd prefer if you could pick it up and run with it, then I can get back to hacking pmchart on the train. :) cheers. -- Nathan From kenj@internode.on.net Sun Apr 11 22:35:38 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.2 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3C3Zco2074900 for ; Sun, 11 Apr 2010 22:35:38 -0500 X-ASG-Debug-ID: 1271043420-44b0027c0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id C9A4C2B58CE for ; Sun, 11 Apr 2010 20:37:01 -0700 (PDT) Received: from mail.internode.on.net (bld-mail19.adl2.internode.on.net [150.101.137.104]) by cuda.sgi.com with ESMTP id BzF1FxUum0MNSUlT for ; Sun, 11 Apr 2010 20:37:01 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20136050-1927428 for multiple; Mon, 12 Apr 2010 13:06:59 +0930 (CST) X-ASG-Orig-Subj: Re: [pcp] pcp updates - PM_CONTEXT_LOCAL generalization Subject: Re: [pcp] pcp updates - PM_CONTEXT_LOCAL generalization From: Ken McDonell Reply-To: kenj@internode.on.net To: Nathan Scott Cc: pcp@oss.sgi.com In-Reply-To: <1458453143.550201271041326429.JavaMail.root@mail-au.aconex.com> References: <1458453143.550201271041326429.JavaMail.root@mail-au.aconex.com> Content-Type: text/plain; charset="UTF-8" Date: Mon, 12 Apr 2010 13:36:35 +1000 Message-ID: <1271043395.24244.89.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail19.adl2.internode.on.net[150.101.137.104] X-Barracuda-Start-Time: 1271043451 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0020 1.0000 -2.0077 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.01 X-Barracuda-Spam-Status: No, SCORE=-2.01 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27272 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Mon, 2010-04-12 at 13:02 +1000, Nathan Scott wrote: > ... Nod. > I'd prefer if you could pick it up and run with it, then I can get back > to hacking pmchart on the train. :) OK, I'll get started on it ... From kenj@internode.on.net Sun Apr 11 22:52:00 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.2 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3C3pxSG075855 for ; Sun, 11 Apr 2010 22:52:00 -0500 X-ASG-Debug-ID: 1271044399-2d8200f80000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id EECB02B2A6A for ; Sun, 11 Apr 2010 20:53:20 -0700 (PDT) Received: from mail.internode.on.net (bld-mail13.adl6.internode.on.net [150.101.137.98]) by cuda.sgi.com with ESMTP id XTFgYpmowSGqceZM for ; Sun, 11 Apr 2010 20:53:20 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20366810-1927428 for multiple; Mon, 12 Apr 2010 13:23:19 +0930 (CST) X-ASG-Orig-Subj: Re: Local context vs dynamic namespace Subject: Re: Local context vs dynamic namespace From: Ken McDonell Reply-To: kenj@internode.on.net To: nathans@aconex.com Cc: pcp In-Reply-To: <1988890790.547131271037410710.JavaMail.root@mail-au.aconex.com> References: <1988890790.547131271037410710.JavaMail.root@mail-au.aconex.com> Content-Type: text/plain; charset="UTF-8" Date: Mon, 12 Apr 2010 13:52:54 +1000 Message-ID: <1271044374.24244.105.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail13.adl6.internode.on.net[150.101.137.98] X-Barracuda-Start-Time: 1271044433 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27273 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean This is pretty much outside the scope of what I was aiming for with the dynamic metrics, although it is not unreasonable. It will mean even more code being copied and changed from pmcd to libpcp as the current code in the library has no support for making the needed DSO calls. So it is not a simple change. The mmv pmda will sneak in on the back of the default dso table for PM_CONTEXT_LOCAL, but how will you get the aconex pmda into the dso table (pminfo -L -K 123,aconex/pmda_aconex.so,aconex_init ... in the bash stuff)? On Mon, 2010-04-12 at 11:56 +1000, nathans@aconex.com wrote: > Hi Ken, > > We're a long way down the path of converting all of our production > environment over to using MMV and the dynamic namespace extensions. > All going swimmingly so far. Just this morning we came across an > issue with the bash metric name auto-completion. This uses pminfo > in local context mode (-L) to provide a list of metric names which > the shell can use for name completion with several pcp commands. > > In our namespace we have MMV:*:* (a top level mmv namespace entry) > and ACONEX:*:* as well (likewise, at the top level). In local mode > pminfo does not traverse, just reports "mmv" and "aconex". > > We can change the bash code to not use -L, but that seems like its > just avoiding the real problem - should local context mode be able > to descend into these parts of the namespace? I woulda thought it > should, just like regular pmcd mode...? > > cheers. > From nathans@aconex.com Mon Apr 12 00:30:15 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3C5UF7l083510 for ; Mon, 12 Apr 2010 00:30:15 -0500 X-ASG-Debug-ID: 1271050297-2d53024b0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 423F62A900C for ; Sun, 11 Apr 2010 22:31:38 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id O2akMjQBdkJgd7F6 for ; Sun, 11 Apr 2010 22:31:38 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id CC23E440F86; Mon, 12 Apr 2010 15:31:35 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id EAocrCePPyOGJbir; Mon, 12 Apr 2010 15:31:35 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 9788CA5011D; Mon, 12 Apr 2010 15:28:35 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id B5D6F4885B0; Mon, 12 Apr 2010 15:31:35 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id a4zltiyhxWfo; Mon, 12 Apr 2010 15:31:31 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 09FB59D0001; Mon, 12 Apr 2010 15:31:31 +1000 (EST) Date: Mon, 12 Apr 2010 15:31:29 +1000 (EST) From: nathans@aconex.com Sender: nscott@aconex.com To: kenj@internode.on.net Cc: pcp Message-ID: <805917791.558821271050288776.JavaMail.root@mail-au.aconex.com> In-Reply-To: <121784609.558421271050008299.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: Re: Local context vs dynamic namespace Subject: Re: Local context vs dynamic namespace MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271050329 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0050 1.0000 -1.9881 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.99 X-Barracuda-Spam-Status: No, SCORE=-1.99 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27280 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.00 NO_REAL_NAME From: does not include a real name X-Virus-Status: Clean ----- "Ken McDonell" wrote: > This is pretty much outside the scope of what I was aiming for with > the dynamic metrics, although it is not unreasonable. > > It will mean even more code being copied and changed from pmcd to > libpcp > as the current code in the library has no support for making the > needed DSO calls. > > So it is not a simple change. > > The mmv pmda will sneak in on the back of the default dso table for > PM_CONTEXT_LOCAL, but how will you get the aconex pmda into the dso > table (pminfo -L -K 123,aconex/pmda_aconex.so,aconex_init ... in the > bash stuff)? > Good point. I'll go with dropping -L in the bash completion, I think. Guess we could do something like the /etc/ld.so.conf.d/ approach with local context pmdas, to specify all that should be open when clients open local contexts, but probably overkill at this stage (but perhaps thats worth thinking about, instead of putting -K in all the tools?). Hmmm, probably not a good time to be suggesting that, now that all the work is done ... :) cheers. -- Nathan From kenj@internode.on.net Mon Apr 12 23:32:38 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.2 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3D4WcZa190329 for ; Mon, 12 Apr 2010 23:32:38 -0500 X-ASG-Debug-ID: 1271133270-47d1029a0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 707001D410C3 for ; Mon, 12 Apr 2010 21:34:30 -0700 (PDT) Received: from mail.internode.on.net (bld-mail19.adl2.internode.on.net [150.101.137.104]) by cuda.sgi.com with ESMTP id bBhUtlvOFp6EnBSX for ; Mon, 12 Apr 2010 21:34:30 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20279195-1927428 for ; Tue, 13 Apr 2010 14:04:29 +0930 (CST) X-ASG-Orig-Subj: pcp updates - Generalized DSO PMDA support for PM_CONTEXT_LOCAL (part 2) Subject: pcp updates - Generalized DSO PMDA support for PM_CONTEXT_LOCAL (part 2) From: Ken McDonell Reply-To: kenj@internode.on.net To: pcp@oss.sgi.com Content-Type: text/plain; charset="UTF-8" Date: Tue, 13 Apr 2010 14:34:04 +1000 Message-ID: <1271133244.24244.160.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail19.adl2.internode.on.net[150.101.137.104] X-Barracuda-Start-Time: 1271133272 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0036 1.0000 -1.9976 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.00 X-Barracuda-Spam-Status: No, SCORE=-2.00 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27356 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Second round of changes incorporating Nathan's review comments. Changes committed to git://oss.sgi.com/kenj/pcp.git dev man/man1/pminfo.1 | 32 -- man/man1/pmprobe.1 | 32 -- man/man1/pmval.1 | 33 -- man/man3/GNUmakefile | 3 man/man3/pmlocalpmda.3 | 5 man/man3/pmspeclocalpmda.3 | 118 ++++++++++ src/include/impl.h | 1 src/libpcp/src/GNUmakefile | 2 src/libpcp/src/connect.c | 363 -------------------------------- src/libpcp/src/connectlocal.c | 474 ++++++++++++++++++++++++++++++++++++++++++ src/pminfo/GNUmakefile | 2 src/pminfo/do_local_spec.c | 107 --------- src/pminfo/pminfo.c | 8 src/pmprobe/GNUmakefile | 10 src/pmprobe/pmprobe.c | 5 src/pmval/GNUmakefile | 10 src/pmval/pmval.c | 8 17 files changed, 627 insertions(+), 586 deletions(-) commit 05ded3f96c18ae7cddec542e4dbf798b042ac9fa Author: Ken McDonell Date: Tue Apr 13 14:04:24 2010 +1000 Generalized DSO PMDA support for PM_CONTEXT_LOCAL (part 2) Changes after Nathan's review. Mostly this involves renaming the command-line -K "spec" parsing routine to __pmSpecLocalPMDA() and promoting this to libpcp. From kenj@internode.on.net Tue Apr 13 04:34:35 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: * X-Spam-Status: No, score=1.2 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, URIBL_BLACK autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3D9YYJH213128 for ; Tue, 13 Apr 2010 04:34:35 -0500 X-ASG-Debug-ID: 1271151386-40ad02030000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 105F51DD259D for ; Tue, 13 Apr 2010 02:36:27 -0700 (PDT) Received: from mail.internode.on.net (bld-mail16.adl2.internode.on.net [150.101.137.101]) by cuda.sgi.com with ESMTP id jfJWGG59QRI82j6B for ; Tue, 13 Apr 2010 02:36:27 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20426768-1927428 for ; Tue, 13 Apr 2010 19:06:25 +0930 (CST) X-ASG-Orig-Subj: pcp updates Subject: pcp updates From: Ken McDonell Reply-To: kenj@internode.on.net To: pcp@oss.sgi.com Content-Type: text/plain; charset="UTF-8" Date: Tue, 13 Apr 2010 19:36:00 +1000 Message-ID: <1271151360.24244.172.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail16.adl2.internode.on.net[150.101.137.101] X-Barracuda-Start-Time: 1271151389 X-Barracuda-Bayes: INNOCENT GLOBAL 0.2217 1.0000 -0.7153 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -0.72 X-Barracuda-Spam-Status: No, SCORE=-0.72 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27374 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Nathan/Martin, Can you please pull these into the mainline pcp tree at your convenience? Ta. Changes committed to git://oss.sgi.com/kenj/pcp.git dev configure.in | 14 ++++++++++++++ man/.gitignore | 1 + src/include/builddefs.in | 13 +++++++++---- 3 files changed, 24 insertions(+), 4 deletions(-) commit 10e26829ca631fcd1a03f644ebad1d9e55fb2e9d Author: Ken McDonell Date: Tue Apr 13 19:33:28 2010 +1000 Solaris man page build changes Patch from Greg Banks to allow PCP man pages with embedded tbl(1) fragments to be correctly formatted on OpenSolaris. From kenj@internode.on.net Tue Apr 13 04:59:53 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.3 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, J_CHICKENPOX_35,J_CHICKENPOX_36 autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3D9xqxb215038 for ; Tue, 13 Apr 2010 04:59:53 -0500 X-ASG-Debug-ID: 1271152905-5c4700f00000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 6C8201DD263A for ; Tue, 13 Apr 2010 03:01:46 -0700 (PDT) Received: from mail.internode.on.net (bld-mail17.adl2.internode.on.net [150.101.137.102]) by cuda.sgi.com with ESMTP id 8fkesk3s66i08fWR for ; Tue, 13 Apr 2010 03:01:46 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20261876-1927428 for multiple; Tue, 13 Apr 2010 19:31:44 +0930 (CST) X-ASG-Orig-Subj: Re: [pcp] Local context vs dynamic namespace Subject: Re: [pcp] Local context vs dynamic namespace From: Ken McDonell Reply-To: kenj@internode.on.net To: nathans@aconex.com Cc: pcp In-Reply-To: <1271044374.24244.105.camel@bozo.localdomain> References: <1988890790.547131271037410710.JavaMail.root@mail-au.aconex.com> <1271044374.24244.105.camel@bozo.localdomain> Content-Type: text/plain; charset="UTF-8" Date: Tue, 13 Apr 2010 20:01:19 +1000 Message-ID: <1271152879.24244.182.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail17.adl2.internode.on.net[150.101.137.102] X-Barracuda-Start-Time: 1271152907 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27375 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean OK, this looks like getting pretty ugly. Consider the following 4 outputs from pminfo ... kenj@bozo:~/src/pcp$ pminfo -n /var/lib/pcp/pmns/root -L -m mmv mmv PMID: 70.*.* kenj@bozo:~/src/pcp$ pminfo -n /var/lib/pcp/pmns/root -m mmv mmv PMID: 70.*.* kenj@bozo:~/src/pcp$ pminfo -L -m mmv mmv PMID: 70.*.* kenj@bozo:~/src/pcp$ pminfo -m mmv mmv.test.strings PMID: 70.1.6 mmv.test.string PMID: 70.1.5 mmv.test.interval PMID: 70.1.4 mmv.test.indom PMID: 70.1.3 mmv.test.discrete PMID: 70.1.2 mmv.test.counter PMID: 70.1.1 mmv.debug PMID: 70.0.1 mmv.reload PMID: 70.0.0 According to my original scope, these are all correct. Nathan would like the third one to produce the same output as the last one. If this was to happen, what is the "correct" output for the first and second cases? I'd like to see some justification, not just votes please. Of course pminfo is a worst case scenario because it uses unusual logic to try very hard to resolve PMNS-only requests without reference to pmcd (or pmdas). On Mon, 2010-04-12 at 13:52 +1000, Ken McDonell wrote: > This is pretty much outside the scope of what I was aiming for with the > dynamic metrics, although it is not unreasonable. > > It will mean even more code being copied and changed from pmcd to libpcp > as the current code in the library has no support for making the needed > DSO calls. > > So it is not a simple change. > > The mmv pmda will sneak in on the back of the default dso table for > PM_CONTEXT_LOCAL, but how will you get the aconex pmda into the dso > table (pminfo -L -K 123,aconex/pmda_aconex.so,aconex_init ... in the > bash stuff)? > > On Mon, 2010-04-12 at 11:56 +1000, nathans@aconex.com wrote: > > Hi Ken, > > > > We're a long way down the path of converting all of our production > > environment over to using MMV and the dynamic namespace extensions. > > All going swimmingly so far. Just this morning we came across an > > issue with the bash metric name auto-completion. This uses pminfo > > in local context mode (-L) to provide a list of metric names which > > the shell can use for name completion with several pcp commands. > > > > In our namespace we have MMV:*:* (a top level mmv namespace entry) > > and ACONEX:*:* as well (likewise, at the top level). In local mode > > pminfo does not traverse, just reports "mmv" and "aconex". > > > > We can change the bash code to not use -L, but that seems like its > > just avoiding the real problem - should local context mode be able > > to descend into these parts of the namespace? I woulda thought it > > should, just like regular pmcd mode...? > > > > cheers. > > > > > _______________________________________________ > pcp mailing list > pcp@oss.sgi.com > http://oss.sgi.com/mailman/listinfo/pcp From nathans@aconex.com Tue Apr 13 15:22:17 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.7 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, J_CHICKENPOX_35,J_CHICKENPOX_36 autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3DKMH7G261497 for ; Tue, 13 Apr 2010 15:22:17 -0500 X-ASG-Debug-ID: 1271190249-02ab01270000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id CC3C32BDB58 for ; Tue, 13 Apr 2010 13:24:10 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id W24OhVc5xpcPzkLe for ; Tue, 13 Apr 2010 13:24:10 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 26F348C4EB7; Wed, 14 Apr 2010 06:24:08 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id vkQcYT3XZOgk7TNE; Wed, 14 Apr 2010 06:24:08 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 0BEF9A5028C; Wed, 14 Apr 2010 06:21:07 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 083DE4884FC; Wed, 14 Apr 2010 06:24:08 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 82M7pZhSCN7G; Wed, 14 Apr 2010 06:24:03 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 511EA48828C; Wed, 14 Apr 2010 06:24:03 +1000 (EST) Date: Wed, 14 Apr 2010 06:24:03 +1000 (EST) From: Nathan Scott To: kenj@internode.on.net Cc: pcp Message-ID: <1927427945.612971271190242960.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1271152879.24244.182.camel@bozo.localdomain> X-ASG-Orig-Subj: Re: [pcp] Local context vs dynamic namespace Subject: Re: [pcp] Local context vs dynamic namespace MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [220.237.111.48] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - SAF3 (Mac)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271190251 X-Barracuda-Bayes: INNOCENT GLOBAL 0.3206 1.0000 -0.2634 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -0.26 X-Barracuda-Spam-Status: No, SCORE=-0.26 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27411 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean ----- "Ken McDonell" wrote: > OK, this looks like getting pretty ugly. > > Consider the following 4 outputs from pminfo ... > > kenj@bozo:~/src/pcp$ pminfo -n /var/lib/pcp/pmns/root -L -m mmv > mmv PMID: 70.*.* > > kenj@bozo:~/src/pcp$ pminfo -n /var/lib/pcp/pmns/root -m mmv > mmv PMID: 70.*.* > > kenj@bozo:~/src/pcp$ pminfo -L -m mmv > mmv PMID: 70.*.* > > kenj@bozo:~/src/pcp$ pminfo -m mmv > mmv.test.strings PMID: 70.1.6 > mmv.test.string PMID: 70.1.5 > mmv.test.interval PMID: 70.1.4 > mmv.test.indom PMID: 70.1.3 > mmv.test.discrete PMID: 70.1.2 > mmv.test.counter PMID: 70.1.1 > mmv.debug PMID: 70.0.1 > mmv.reload PMID: 70.0.0 > > According to my original scope, these are all correct. > > Nathan would like the third one to produce the same output as the > last > one. > > If this was to happen, what is the "correct" output for the first and > second cases? > > I'd like to see some justification, not just votes please. I think first two are OK as is. I'm wondering also if a pminfo flag is needed for both cases 3 and 4, so that by default they do the same thing (case 4), but with the new flag report on dynamic nodes (AIUI, there's no way currently to tell which remote pmcd nodes are dynamic?) The justification is the local context should not behave differently to using host context for no real reason, it'll result in bugs & some surprises for users (like I got!). cheers. -- Nathan From kenj@internode.on.net Tue Apr 13 15:45:42 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.2 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3DKjgmh000880 for ; Tue, 13 Apr 2010 15:45:42 -0500 X-ASG-Debug-ID: 1271191655-4803033f0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 4A02D2BDE7E for ; Tue, 13 Apr 2010 13:47:35 -0700 (PDT) Received: from mail.internode.on.net (bld-mail16.adl2.internode.on.net [150.101.137.101]) by cuda.sgi.com with ESMTP id qogHsT4GOALgtClz for ; Tue, 13 Apr 2010 13:47:35 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20478472-1927428 for multiple; Wed, 14 Apr 2010 06:17:34 +0930 (CST) X-ASG-Orig-Subj: Re: Local context vs dynamic namespace Subject: Re: Local context vs dynamic namespace From: Ken McDonell Reply-To: kenj@internode.on.net To: nathans@aconex.com Cc: pcp In-Reply-To: <805917791.558821271050288776.JavaMail.root@mail-au.aconex.com> References: <805917791.558821271050288776.JavaMail.root@mail-au.aconex.com> Content-Type: text/plain; charset="UTF-8" Date: Wed, 14 Apr 2010 06:47:08 +1000 Message-ID: <1271191628.24244.205.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail16.adl2.internode.on.net[150.101.137.101] X-Barracuda-Start-Time: 1271191657 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0004 1.0000 -2.0185 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27411 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Nathan, I know you don't particularly like the environment variable approach, but following the $PCP_DERIVED_CONFIG example, we could introduce $PCP_LOCAL_CONFIG that might specify the name of a file (perhaps stashed in $HOME/.pcp at the user's discretion) that contained local PMDA specifications (in the syntax for __pmSpecLocalPMDA that was recently created). So in your case, the file might contain clear add,60,linux/pmda_linux,linux_init add,70,mmv/pmda_mmv,mmv_init add,123,aconex/pmda_aconex,aconex_init or indeed, simply take the default and add the aconex PMDA, as in add,123,aconex/pmda_aconex,aconex_init In some ways this is less ugly than the -K stuff for cases where the set of PMDAs you'd like to be available for PM_CONTEXT_LOCAL is non-standard and fixed. Thoughts? Of course all of this is moot until the issues with PM_CONTEXT_LOCAL and dynamic pmns entries is resolved, because until then the mmv metrics will remain hidden for clients using PM_CONTEXT_LOCAL contexts. On Mon, 2010-04-12 at 15:31 +1000, nathans@aconex.com wrote: > ----- "Ken McDonell" wrote: > > > This is pretty much outside the scope of what I was aiming for with > > the dynamic metrics, although it is not unreasonable. > > > > It will mean even more code being copied and changed from pmcd to > > libpcp > > as the current code in the library has no support for making the > > needed DSO calls. > > > > So it is not a simple change. > > > > The mmv pmda will sneak in on the back of the default dso table for > > PM_CONTEXT_LOCAL, but how will you get the aconex pmda into the dso > > table (pminfo -L -K 123,aconex/pmda_aconex.so,aconex_init ... in the > > bash stuff)? > > > > Good point. I'll go with dropping -L in the bash completion, I think. > Guess we could do something like the /etc/ld.so.conf.d/ approach with > local context pmdas, to specify all that should be open when clients > open local contexts, but probably overkill at this stage (but perhaps > thats worth thinking about, instead of putting -K in all the tools?). > > Hmmm, probably not a good time to be suggesting that, now that all the > work is done ... :) > > cheers. > From nathans@aconex.com Tue Apr 13 17:00:36 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.6 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3DM0amL005927 for ; Tue, 13 Apr 2010 17:00:36 -0500 X-ASG-Debug-ID: 1271196149-3886030d0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 86F8C12757A5 for ; Tue, 13 Apr 2010 15:02:30 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id 6e23wNPvTWYLjHXO for ; Tue, 13 Apr 2010 15:02:30 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 3C1488C50B3; Wed, 14 Apr 2010 08:02:28 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id fboPvGJEwHpJZ6ak; Wed, 14 Apr 2010 08:02:28 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id E88DFA5028C; Wed, 14 Apr 2010 07:59:26 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id F0E934884FC; Wed, 14 Apr 2010 08:02:27 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id pyIoAyBTFrNe; Wed, 14 Apr 2010 08:02:23 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 46471488231; Wed, 14 Apr 2010 08:02:23 +1000 (EST) Date: Wed, 14 Apr 2010 08:02:22 +1000 (EST) From: Nathan Scott To: kenj@internode.on.net Cc: pcp Message-ID: <378220262.614601271196142712.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1271191628.24244.205.camel@bozo.localdomain> X-ASG-Orig-Subj: Re: Local context vs dynamic namespace Subject: Re: Local context vs dynamic namespace MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [220.237.111.48] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - SAF3 (Mac)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271196151 X-Barracuda-Bayes: INNOCENT GLOBAL 0.2388 1.0000 -0.6300 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -0.63 X-Barracuda-Spam-Status: No, SCORE=-0.63 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27416 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean ----- "Ken McDonell" wrote: > Nathan, > > I know you don't particularly like the environment variable approach, > but following the $PCP_DERIVED_CONFIG example, we could introduce > $PCP_LOCAL_CONFIG that might specify the name of a file (perhaps > stashed > in $HOME/.pcp at the user's discretion) that contained local PMDA > specifications (in the syntax for __pmSpecLocalPMDA that was recently > created). > > So in your case, the file might contain > > clear > add,60,linux/pmda_linux,linux_init > add,70,mmv/pmda_mmv,mmv_init > add,123,aconex/pmda_aconex,aconex_init > > or indeed, simply take the default and add the aconex PMDA, as in > > add,123,aconex/pmda_aconex,aconex_init > > In some ways this is less ugly than the -K stuff for cases where the > set > of PMDAs you'd like to be available for PM_CONTEXT_LOCAL is > non-standard > and fixed. > > Thoughts? I was thinking something along those lines, but more system-wide than user specific. If we had a /etc/pcp.local.d/ directory, we could have the agent Install scripts drop a file in there in __pmSpecLocalPMDA format for dso agents, and we could ship a kernel and mmv one in there by default. Infiniband rpm installs could then drop one in there, and the cluster PMDA would be simplified a bit. We could also do away completely with dsotab in libpcp then... and fill that in "on the fly" based on contents of that directory. cheers. -- Nathan From kenj@internode.on.net Tue Apr 13 17:54:14 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.2 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3DMsDFo009548 for ; Tue, 13 Apr 2010 17:54:14 -0500 X-ASG-Debug-ID: 1271199366-5c7200320000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 29E462BE3D1 for ; Tue, 13 Apr 2010 15:56:06 -0700 (PDT) Received: from mail.internode.on.net (bld-mail19.adl2.internode.on.net [150.101.137.104]) by cuda.sgi.com with ESMTP id uk7NxiNTwQlyR2ny for ; Tue, 13 Apr 2010 15:56:06 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20379574-1927428 for multiple; Wed, 14 Apr 2010 08:26:05 +0930 (CST) X-ASG-Orig-Subj: Re: Local context vs dynamic namespace Subject: Re: Local context vs dynamic namespace From: Ken McDonell Reply-To: kenj@internode.on.net To: Nathan Scott Cc: pcp In-Reply-To: <378220262.614601271196142712.JavaMail.root@mail-au.aconex.com> References: <378220262.614601271196142712.JavaMail.root@mail-au.aconex.com> Content-Type: text/plain; charset="UTF-8" Date: Wed, 14 Apr 2010 08:55:38 +1000 Message-ID: <1271199338.24244.278.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail19.adl2.internode.on.net[150.101.137.104] X-Barracuda-Start-Time: 1271199368 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27420 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Wed, 2010-04-14 at 08:02 +1000, Nathan Scott wrote: > I was thinking something along those lines, but more system-wide than > user specific. If we had a /etc/pcp.local.d/ directory, we could have > the agent Install scripts drop a file in there in __pmSpecLocalPMDA > format for dso agents, and we could ship a kernel and mmv one in there > by default. > > Infiniband rpm installs could then drop one in there, and the cluster > PMDA would be simplified a bit. > > We could also do away completely with dsotab in libpcp then... and > fill that in "on the fly" based on contents of that directory. No real issue with this, but a few comments. 1. By "drop a file" I presume you're thinking one file per pmda that wants to play, and then "somewhere" in libpcp I process all of the files in the magic directory. This means update is file replacement, rather than the edit and update a common file model that is much more error prone. 2. Could the file live below $PCP_VAR_DIR? This would avoid adding another variable to /etc/pcp.conf. I'd suggest $PCP_VAR_DIR/config/local_pmdas or $PCP_VAR_DIR/pmdas/local_pmdas. 3. Making the compile-time build of dsotbl go away is probably a good idea. 4. Should we retire the $PMDA_LOCAL_PROC and $PMDA_LOCAL_SAMPLE controls that influence when the proc and sample metrics might be available for PM_CONTEXT_LOCAL? I think they are passed their use-by date, so would not worry if they were quietly forgotten about. 5. A lot cleaner than all of this would be an extended pmcd.conf format that included a "pick me for PM_CONTEXT_LOCAL" flag ... there is space to do that at the end of the dso line (no arguments for dsos). This would be - system wide - at the discretion of the pmda install script - atomic with pmda install/remove - no environment variables 6. With all of this, it is not clear how useful the CLEAR operation to __pmLocalPMDA will be ... I can still see a remote case for needing ADD and DEL, but in some ways the need for all of __pmLocalPMDA and __pmSpecLocalPMDA would almost go away if the PMDAs for PM_CONTEXT_LOCAL was driven off pmcd.conf ... except for your original requirement 8^(> to be able to use metrics from a PMDA when pmcd was not running ... if pmcd is not running then perhaps the PMDAs have not been Installed and pmcd.conf contains no reference to the PMDAs of interest. I think I've argued around a full circle so, probably scrap 5. and 6. and go with the rest of the suggestion. From nathans@aconex.com Tue Apr 13 18:22:14 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.6 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3DNMDTw012072 for ; Tue, 13 Apr 2010 18:22:13 -0500 X-ASG-Debug-ID: 1271201045-49a003d50000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id C255C12FC7B5 for ; Tue, 13 Apr 2010 16:24:06 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id wH0bBBgbzqPi5DBm for ; Tue, 13 Apr 2010 16:24:06 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 6456B8C5789; Wed, 14 Apr 2010 09:24:05 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id GYHZcIAP88tIuWAl; Wed, 14 Apr 2010 09:24:05 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id EA17EA50110; Wed, 14 Apr 2010 09:21:03 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 0A33C4884FC; Wed, 14 Apr 2010 09:24:05 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id LUx6GYlLIsbF; Wed, 14 Apr 2010 09:24:00 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 53A37488231; Wed, 14 Apr 2010 09:24:00 +1000 (EST) Date: Wed, 14 Apr 2010 09:23:58 +1000 (EST) From: Nathan Scott To: kenj@internode.on.net Cc: pcp Message-ID: <1419860990.620491271201038915.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1271199338.24244.278.camel@bozo.localdomain> X-ASG-Orig-Subj: Re: Local context vs dynamic namespace Subject: Re: Local context vs dynamic namespace MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [220.237.111.48] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - SAF3 (Mac)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271201047 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27421 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean ----- "Ken McDonell" wrote: > On Wed, 2010-04-14 at 08:02 +1000, Nathan Scott wrote: > > No real issue with this, but a few comments. > > 1. By "drop a file" I presume you're thinking one file per pmda that > wants to play, and then "somewhere" in libpcp I process all of the > files Right, that was what I was thinking. > in the magic directory. This means update is file replacement, > rather > than the edit and update a common file model that is much more error > prone. Right. Someday, pmcd could perhaps take a leaf out of this book, but way beyond scope here. > 4. Should we retire the $PMDA_LOCAL_PROC and $PMDA_LOCAL_SAMPLE > controls > that influence when the proc and sample metrics might be available > for > PM_CONTEXT_LOCAL? I think they are passed their use-by date, so > would > not worry if they were quietly forgotten about. > I'd be happy with that too, yep. > 5. A lot cleaner than all of this would be an extended pmcd.conf > format > that included a "pick me for PM_CONTEXT_LOCAL" flag ... there is > space > to do that at the end of the dso line (no arguments for dsos). This > would be > - system wide > - at the discretion of the pmda install script > - atomic with pmda install/remove > - no environment variables Mmm, yes, nice idea - do we even need to extend the format? Just say any dso pmda listed in pmcd.conf is auto-loaded for local context? With the clear/add/del a programmer can still fine tune their needs, if they really just want particular PMDAs for a specialised tool. This all means you have to parse pmcd.conf in libpcp, which is a bit more involved perhaps than walking a directory and using the existing (-K) parser code ... but, you choose, I'd be happy with either way. cheers. -- Nathan From kenj@internode.on.net Tue Apr 13 20:44:23 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.2 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3E1iNie021549 for ; Tue, 13 Apr 2010 20:44:23 -0500 X-ASG-Debug-ID: 1271209575-30ab00110000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id B8534127A8D7 for ; Tue, 13 Apr 2010 18:46:16 -0700 (PDT) Received: from mail.internode.on.net (bld-mail16.adl2.internode.on.net [150.101.137.101]) by cuda.sgi.com with ESMTP id WaAfRZfIexpOR2FZ for ; Tue, 13 Apr 2010 18:46:16 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20510231-1927428 for multiple; Wed, 14 Apr 2010 11:16:15 +0930 (CST) X-ASG-Orig-Subj: Re: [pcp] Local context vs dynamic namespace Subject: Re: [pcp] Local context vs dynamic namespace From: Ken McDonell Reply-To: kenj@internode.on.net To: Nathan Scott Cc: pcp In-Reply-To: <1271199338.24244.278.camel@bozo.localdomain> References: <378220262.614601271196142712.JavaMail.root@mail-au.aconex.com> <1271199338.24244.278.camel@bozo.localdomain> Content-Type: text/plain; charset="UTF-8" Date: Wed, 14 Apr 2010 11:45:49 +1000 Message-ID: <1271209549.24244.453.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail16.adl2.internode.on.net[150.101.137.101] X-Barracuda-Start-Time: 1271209577 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27428 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Just to go in circles once more ... PM_CONTEXT_LOCAL is only ever going to work for the (DSO) PMDAs that have been "Install"ed on the local host because this is the only way to get the correct PMNS installed as well. So I think the using pmcd.conf model would work just fine ... and Nathan's right we don't need to extend the pmcd.conf format -- if you are a DSO PMDA defined in pmcd.conf then by default you're available via PM_CONTEXT_LOCAL, end of story. On Wed, 2010-04-14 at 08:55 +1000, Ken McDonell wrote: > ... > 6. With all of this, it is not clear how useful the CLEAR operation to > __pmLocalPMDA will be ... I can still see a remote case for needing ADD > and DEL, but in some ways the need for all of __pmLocalPMDA and > __pmSpecLocalPMDA would almost go away if the PMDAs for PM_CONTEXT_LOCAL > was driven off pmcd.conf ... except for your original requirement 8^(> > to be able to use metrics from a PMDA when pmcd was not running ... if > pmcd is not running then perhaps the PMDAs have not been Installed and > pmcd.conf contains no reference to the PMDAs of interest. > > I think I've argued around a full circle so, probably scrap 5. and 6. > and go with the rest of the suggestion. From mort@sgi.com Wed Apr 14 08:12:52 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.8 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from relay.sgi.com (relay1.corp.sgi.com [137.38.102.111]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3EDCqZK072721 for ; Wed, 14 Apr 2010 08:12:52 -0500 Received: from attica.americas.sgi.com (attica.americas.sgi.com [128.162.236.44]) by relay1.corp.sgi.com (Postfix) with ESMTP id D25C38F80BE; Wed, 14 Apr 2010 06:14:42 -0700 (PDT) Received: by attica.americas.sgi.com (Postfix, from userid 44682) id 97AF6A6FD651; Wed, 14 Apr 2010 08:14:41 -0500 (CDT) Date: Wed, 14 Apr 2010 08:14:41 -0500 From: Martin Hicks To: Ken McDonell Cc: Nathan Scott , pcp Subject: Re: [pcp] Local context vs dynamic namespace Message-ID: <20100414131441.GF19316@alcatraz.americas.sgi.com> References: <378220262.614601271196142712.JavaMail.root@mail-au.aconex.com> <1271199338.24244.278.camel@bozo.localdomain> <1271209549.24244.453.camel@bozo.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1271209549.24244.453.camel@bozo.localdomain> User-Agent: Mutt/1.5.17 (2007-11-01) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Wed, Apr 14, 2010 at 11:45:49AM +1000, Ken McDonell wrote: > Just to go in circles once more ... PM_CONTEXT_LOCAL is only ever going > to work for the (DSO) PMDAs that have been "Install"ed on the local host > because this is the only way to get the correct PMNS installed as well. > > So I think the using pmcd.conf model would work just fine ... and > Nathan's right we don't need to extend the pmcd.conf format -- if you > are a DSO PMDA defined in pmcd.conf then by default you're available via > PM_CONTEXT_LOCAL, end of story. I've read all of this and I think I understand the proposal. It sounds pretty good to me. I like that we're using pmcd.conf rather than inventing a whole new config mechanism. I intend to review Ken's current iteration of the PM_CONTEXT_LOCAL patch this morning. mh PS my repetitive strain wrist injury is pretty bad, so I'm limiting typing time pretty severely these days. This is the major reason why I haven't been participating much lately. > > On Wed, 2010-04-14 at 08:55 +1000, Ken McDonell wrote: > > ... > > 6. With all of this, it is not clear how useful the CLEAR operation to > > __pmLocalPMDA will be ... I can still see a remote case for needing ADD > > and DEL, but in some ways the need for all of __pmLocalPMDA and > > __pmSpecLocalPMDA would almost go away if the PMDAs for PM_CONTEXT_LOCAL > > was driven off pmcd.conf ... except for your original requirement 8^(> > > to be able to use metrics from a PMDA when pmcd was not running ... if > > pmcd is not running then perhaps the PMDAs have not been Installed and > > pmcd.conf contains no reference to the PMDAs of interest. > > > > I think I've argued around a full circle so, probably scrap 5. and 6. > > and go with the rest of the suggestion. > > _______________________________________________ > pcp mailing list > pcp@oss.sgi.com > http://oss.sgi.com/mailman/listinfo/pcp From mort@bork.org Wed Apr 14 08:53:10 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.3 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, URIBL_BLACK autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3EDrAOk075123 for ; Wed, 14 Apr 2010 08:53:10 -0500 X-ASG-Debug-ID: 1271253304-2c69001a0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from darwin.bork.org (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 4FC3D1DD4325 for ; Wed, 14 Apr 2010 06:55:04 -0700 (PDT) Received: from darwin.bork.org (darwin.bork.org [65.49.60.145]) by cuda.sgi.com with ESMTP id dFTf9ST3eji67HxM for ; Wed, 14 Apr 2010 06:55:04 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by darwin.bork.org (Postfix) with ESMTP id 5EE73192D7; Wed, 14 Apr 2010 09:49:53 -0400 (EDT) Received: from darwin.bork.org ([127.0.0.1]) by localhost (darwin.bork.org [127.0.0.1]) (amavisd-new, port 10024) with LMTP id wMIh+L0xzs46; Wed, 14 Apr 2010 09:49:53 -0400 (EDT) Received: from laplace.bork.org (laplace.bork.org [IPv6:2001:470:1f11:499:21a:a0ff:feec:57ab]) by darwin.bork.org (Postfix) with ESMTP id 180411923D; Wed, 14 Apr 2010 09:49:53 -0400 (EDT) Received: by laplace.bork.org (Postfix, from userid 1000) id 7E82B54674; Wed, 14 Apr 2010 09:55:03 -0400 (EDT) Date: Wed, 14 Apr 2010 09:55:03 -0400 From: Martin Hicks To: Ken McDonell Cc: pcp@oss.sgi.com X-ASG-Orig-Subj: Re: [pcp] pcp updates Subject: Re: [pcp] pcp updates Message-ID: <20100414135503.GL19119@laplace.bork.org> References: <1271151360.24244.172.camel@bozo.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1271151360.24244.172.camel@bozo.localdomain> User-Agent: Mutt/1.5.20 (2009-06-14) X-Barracuda-Connect: darwin.bork.org[65.49.60.145] X-Barracuda-Start-Time: 1271253305 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27472 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Tue, Apr 13, 2010 at 07:36:00PM +1000, Ken McDonell wrote: > Nathan/Martin, > > Can you please pull these into the mainline pcp tree at your convenience? > I've pulled this into my dev tree. I'll let Nathan take a look through my changes to Makepkgs (removing SGI pcp-pro/pcp-sgi cruft) before I merge to mainline. mh > Ta. > > Changes committed to git://oss.sgi.com/kenj/pcp.git dev > > configure.in | 14 ++++++++++++++ > man/.gitignore | 1 + > src/include/builddefs.in | 13 +++++++++---- > 3 files changed, 24 insertions(+), 4 deletions(-) > > commit 10e26829ca631fcd1a03f644ebad1d9e55fb2e9d > Author: Ken McDonell > Date: Tue Apr 13 19:33:28 2010 +1000 > > Solaris man page build changes > > Patch from Greg Banks to allow PCP man pages with embedded tbl(1) > fragments to be correctly formatted on OpenSolaris. > > > _______________________________________________ > pcp mailing list > pcp@oss.sgi.com > http://oss.sgi.com/mailman/listinfo/pcp -- Martin Hicks || mort@bork.org || PGP/GnuPG: 0x4C7F2BEE From mort@sgi.com Wed Apr 14 12:14:32 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.8 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from relay.sgi.com (relay1.corp.sgi.com [137.38.102.111]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3EHEVsM086022 for ; Wed, 14 Apr 2010 12:14:32 -0500 Received: from attica.americas.sgi.com (attica.americas.sgi.com [128.162.236.44]) by relay1.corp.sgi.com (Postfix) with ESMTP id BA8898F8073; Wed, 14 Apr 2010 10:16:24 -0700 (PDT) Received: by attica.americas.sgi.com (Postfix, from userid 44682) id 26A3FA6FD654; Wed, 14 Apr 2010 12:16:24 -0500 (CDT) Date: Wed, 14 Apr 2010 12:16:24 -0500 From: Martin Hicks To: Ken McDonell Cc: pcp@oss.sgi.com Subject: Re: [pcp] pcp updates - Generalized DSO PMDA support for PM_CONTEXT_LOCAL (part 2) Message-ID: <20100414171624.GI19316@alcatraz.americas.sgi.com> References: <1271133244.24244.160.camel@bozo.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1271133244.24244.160.camel@bozo.localdomain> User-Agent: Mutt/1.5.17 (2007-11-01) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Tue, Apr 13, 2010 at 02:34:04PM +1000, Ken McDonell wrote: > Second round of changes incorporating Nathan's review comments. > The few things I noticed were all cosmetic: - pminfo manpage is malformed after the -K blurb - pmspeclocalpmda.3 mentions that fields are separated by ':'. Looks like ',' is reality. mh From kenj@internode.on.net Wed Apr 14 15:58:34 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.2 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3EKwXvT099233 for ; Wed, 14 Apr 2010 15:58:34 -0500 X-ASG-Debug-ID: 1271278827-312e01b10000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id A5BC12BF1D0 for ; Wed, 14 Apr 2010 14:00:28 -0700 (PDT) Received: from mail.internode.on.net (bld-mail19.adl2.internode.on.net [150.101.137.104]) by cuda.sgi.com with ESMTP id oi1nhCjDvQ2A7OKz for ; Wed, 14 Apr 2010 14:00:28 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20510830-1927428 for multiple; Thu, 15 Apr 2010 06:30:26 +0930 (CST) X-ASG-Orig-Subj: Re: [pcp] pcp updates - Generalized DSO PMDA support for PM_CONTEXT_LOCAL (part 2) Subject: Re: [pcp] pcp updates - Generalized DSO PMDA support for PM_CONTEXT_LOCAL (part 2) From: Ken McDonell Reply-To: kenj@internode.on.net To: Martin Hicks Cc: pcp@oss.sgi.com In-Reply-To: <20100414171624.GI19316@alcatraz.americas.sgi.com> References: <1271133244.24244.160.camel@bozo.localdomain> <20100414171624.GI19316@alcatraz.americas.sgi.com> Content-Type: text/plain; charset="UTF-8" Date: Thu, 15 Apr 2010 07:00:00 +1000 Message-ID: <1271278800.13672.8.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail19.adl2.internode.on.net[150.101.137.104] X-Barracuda-Start-Time: 1271278829 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27495 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Thanks Martin. These are both fixed in my next round of commits. On Wed, 2010-04-14 at 12:16 -0500, Martin Hicks wrote: > On Tue, Apr 13, 2010 at 02:34:04PM +1000, Ken McDonell wrote: > > Second round of changes incorporating Nathan's review comments. > > > > The few things I noticed were all cosmetic: > > - pminfo manpage is malformed after the -K blurb One too many .RE is not good. > - pmspeclocalpmda.3 mentions that fields are separated by ':'. Looks > like ',' is reality. Yep it is comma-separated. Ta. From kenj@internode.on.net Wed Apr 14 17:26:03 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-1.0 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX,J_CHICKENPOX_73 autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3EMQ3UO104404 for ; Wed, 14 Apr 2010 17:26:03 -0500 X-ASG-Debug-ID: 1271284077-03ac01680000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id C81891292628 for ; Wed, 14 Apr 2010 15:27:57 -0700 (PDT) Received: from mail.internode.on.net (bld-mail13.adl6.internode.on.net [150.101.137.98]) by cuda.sgi.com with ESMTP id oyV8KGhcI0T5ehNt for ; Wed, 14 Apr 2010 15:27:57 -0700 (PDT) Received: from [124.182.39.133] (unverified [124.182.39.133]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20748043-1927428 for ; Thu, 15 Apr 2010 07:57:56 +0930 (CST) X-ASG-Orig-Subj: pcp updates - roll to 3.1.3 and pmcd.conf driving available PMDAs for PM_CONTEXT_LOCAL Subject: pcp updates - roll to 3.1.3 and pmcd.conf driving available PMDAs for PM_CONTEXT_LOCAL From: Ken McDonell Reply-To: kenj@internode.on.net To: pcp@oss.sgi.com Content-Type: text/plain; charset="UTF-8" Date: Thu, 15 Apr 2010 08:27:56 +1000 Message-ID: <1271284076.21228.0.camel@bozo-laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail13.adl6.internode.on.net[150.101.137.98] X-Barracuda-Start-Time: 1271284078 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27500 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Changes committed to git://oss.sgi.com/kenj/pcp.git dev CHANGELOG | 5 + VERSION.pcp | 2 debian/changelog | 9 + man/man1/pcpintro.1 | 26 ++++- man/man1/pminfo.1 | 1 man/man3/pmlocalpmda.3 | 15 ++- man/man3/pmnewcontext.3 | 16 --- man/man3/pmspeclocalpmda.3 | 4 src/libpcp/src/GNUmakefile | 20 ---- src/libpcp/src/connectlocal.c | 210 ++++++++++++++++++++++++++---------------- 10 files changed, 187 insertions(+), 121 deletions(-) commit 41d23130eec50885e9e5e17cb551142e908a0c06 Author: Ken McDonell Date: Thu Apr 15 08:21:49 2010 +1000 Change how PM_CONTEXT_LOCAL determines available PMDAs Moving from the previous ugly build-time configuration of dsotbl[] to run-time determination based on pmcd.conf. To make a new PMDA available to applications using PM_CONTEXT_LOCAL simply install the PMDA as a "dso" ... this updates the pmns and pmcd.conf and the PMDA will be added to the list of PMDAs available with PM_CONTEXT_LOCAL contexts. In the process, rolled to PCP 3.1.3 to help isolate QA differences. From kenj@internode.on.net Wed Apr 14 17:37:52 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.2 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3EMbpe9105202 for ; Wed, 14 Apr 2010 17:37:52 -0500 X-ASG-Debug-ID: 1271284785-5c0b031a0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 7C6EE2C3094 for ; Wed, 14 Apr 2010 15:39:46 -0700 (PDT) Received: from mail.internode.on.net (bld-mail12.adl6.internode.on.net [150.101.137.97]) by cuda.sgi.com with ESMTP id WfvKPvUEJ9YNh3dL for ; Wed, 14 Apr 2010 15:39:46 -0700 (PDT) Received: from [124.182.39.133] (unverified [124.182.39.133]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20644398-1927428 for multiple; Thu, 15 Apr 2010 08:09:44 +0930 (CST) X-ASG-Orig-Subj: Re: [pcp] Local context vs dynamic namespace Subject: Re: [pcp] Local context vs dynamic namespace From: Ken McDonell Reply-To: kenj@internode.on.net To: Nathan Scott Cc: pcp In-Reply-To: <1271209549.24244.453.camel@bozo.localdomain> References: <378220262.614601271196142712.JavaMail.root@mail-au.aconex.com> <1271199338.24244.278.camel@bozo.localdomain> <1271209549.24244.453.camel@bozo.localdomain> Content-Type: text/plain; charset="UTF-8" Date: Thu, 15 Apr 2010 08:39:46 +1000 Message-ID: <1271284786.21228.2.camel@bozo-laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail12.adl6.internode.on.net[150.101.137.97] X-Barracuda-Start-Time: 1271284787 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27500 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Wed, 2010-04-14 at 11:45 +1000, Ken McDonell wrote: > Just to go in circles once more ... PM_CONTEXT_LOCAL is only ever going > to work for the (DSO) PMDAs that have been "Install"ed on the local host > because this is the only way to get the correct PMNS installed as well. > > So I think the using pmcd.conf model would work just fine ... and > Nathan's right we don't need to extend the pmcd.conf format -- if you > are a DSO PMDA defined in pmcd.conf then by default you're available via > PM_CONTEXT_LOCAL, end of story. Only caveat to this (now that I've done the work) is the pmcd pmda ... it is excluded from PM_CONTEXT_LOCAL use because it only makes sense when attached to the address space of a running pmcd (not an arbitrary PCP client application). From kenj@internode.on.net Wed Apr 14 17:42:12 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.1 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3EMgBuv105442 for ; Wed, 14 Apr 2010 17:42:12 -0500 X-ASG-Debug-ID: 1271285045-5c0a031a0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 4B7A22C3345 for ; Wed, 14 Apr 2010 15:44:06 -0700 (PDT) Received: from mail.internode.on.net (bld-mail19.adl2.internode.on.net [150.101.137.104]) by cuda.sgi.com with ESMTP id UfLfdQgAuFJ3ysIx for ; Wed, 14 Apr 2010 15:44:06 -0700 (PDT) Received: from [124.182.39.133] (unverified [124.182.39.133]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20519933-1927428 for multiple; Thu, 15 Apr 2010 08:14:05 +0930 (CST) X-ASG-Orig-Subj: Re: Local context vs dynamic namespace Subject: Re: Local context vs dynamic namespace From: Ken McDonell Reply-To: kenj@internode.on.net To: Nathan Scott Cc: pcp In-Reply-To: <1419860990.620491271201038915.JavaMail.root@mail-au.aconex.com> References: <1419860990.620491271201038915.JavaMail.root@mail-au.aconex.com> Content-Type: text/plain; charset="UTF-8" Date: Thu, 15 Apr 2010 08:44:07 +1000 Message-ID: <1271285047.21228.6.camel@bozo-laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail19.adl2.internode.on.net[150.101.137.104] X-Barracuda-Start-Time: 1271285047 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27500 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Wed, 2010-04-14 at 09:23 +1000, Nathan Scott wrote: > ... > This all means you have to parse pmcd.conf in libpcp, which is a bit > more involved perhaps than walking a directory and using the existing > (-K) parser code ... but, you choose, I'd be happy with either way. The parsing is pretty simple ... I opted to not do a heap of error checking because (a) pmcd is already doing that at the time pmcd.conf is updated and reread, and (b) the parsing happens during the first call to pmNewContext(PM_CONTEXT_LOCAL, NULL) or pmLocalPMDA(...) or pmSpecLocalPMDA() and this is a bad place to be trying to return "helpful" parser error messages. I can revisit this if someone sees a problem with the approach I've taken. From kenj@internode.on.net Wed Apr 14 17:47:02 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.3 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, J_CHICKENPOX_35,J_CHICKENPOX_36 autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3EMl1Of105811 for ; Wed, 14 Apr 2010 17:47:02 -0500 X-ASG-Debug-ID: 1271285335-3a4200770000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 6E58D2C3420 for ; Wed, 14 Apr 2010 15:48:56 -0700 (PDT) Received: from mail.internode.on.net (bld-mail18.adl2.internode.on.net [150.101.137.103]) by cuda.sgi.com with ESMTP id xs5C6HA0bUPA6mOo for ; Wed, 14 Apr 2010 15:48:56 -0700 (PDT) Received: from [124.182.39.133] (unverified [124.182.39.133]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20619703-1927428 for multiple; Thu, 15 Apr 2010 08:18:54 +0930 (CST) X-ASG-Orig-Subj: Re: [pcp] Local context vs dynamic namespace Subject: Re: [pcp] Local context vs dynamic namespace From: Ken McDonell Reply-To: kenj@internode.on.net To: Nathan Scott Cc: pcp In-Reply-To: <1927427945.612971271190242960.JavaMail.root@mail-au.aconex.com> References: <1927427945.612971271190242960.JavaMail.root@mail-au.aconex.com> Content-Type: text/plain; charset="UTF-8" Date: Thu, 15 Apr 2010 08:48:55 +1000 Message-ID: <1271285335.21228.10.camel@bozo-laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail18.adl2.internode.on.net[150.101.137.103] X-Barracuda-Start-Time: 1271285337 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27500 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On reflection, I think they ALL need to produce the same result (as the last one). -n pmns does not really change the expectation of output (assuming the pmns you feed it is at least similar to the installed one, and it is _exactly_ similar in the examples below) ... the presence or absence of the -n option might change the available metric names, it is unlikely to change the name <--> pmid mapping. Independent of how you arrive at 70.*.* as the initial pmid for the metric named mmv, what you do next should be the same. So I'm off to add some pmcd-based bloat into libpcp ... On Wed, 2010-04-14 at 06:24 +1000, Nathan Scott wrote: > ----- "Ken McDonell" wrote: > > > OK, this looks like getting pretty ugly. > > > > Consider the following 4 outputs from pminfo ... > > > > kenj@bozo:~/src/pcp$ pminfo -n /var/lib/pcp/pmns/root -L -m mmv > > mmv PMID: 70.*.* > > > > kenj@bozo:~/src/pcp$ pminfo -n /var/lib/pcp/pmns/root -m mmv > > mmv PMID: 70.*.* > > > > kenj@bozo:~/src/pcp$ pminfo -L -m mmv > > mmv PMID: 70.*.* > > > > kenj@bozo:~/src/pcp$ pminfo -m mmv > > mmv.test.strings PMID: 70.1.6 > > mmv.test.string PMID: 70.1.5 > > mmv.test.interval PMID: 70.1.4 > > mmv.test.indom PMID: 70.1.3 > > mmv.test.discrete PMID: 70.1.2 > > mmv.test.counter PMID: 70.1.1 > > mmv.debug PMID: 70.0.1 > > mmv.reload PMID: 70.0.0 > > > > According to my original scope, these are all correct. > > > > Nathan would like the third one to produce the same output as the > > last > > one. > > > > If this was to happen, what is the "correct" output for the first and > > second cases? > > > > I'd like to see some justification, not just votes please. > > I think first two are OK as is. I'm wondering also if a pminfo flag > is needed for both cases 3 and 4, so that by default they do the same > thing (case 4), but with the new flag report on dynamic nodes (AIUI, > there's no way currently to tell which remote pmcd nodes are dynamic?) > > The justification is the local context should not behave differently > to using host context for no real reason, it'll result in bugs & some > surprises for users (like I got!). > > cheers. > From nathans@aconex.com Wed Apr 14 18:42:59 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3ENgwtx109723 for ; Wed, 14 Apr 2010 18:42:59 -0500 X-ASG-Debug-ID: 1271288693-78ff03210000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 3F32512940E4 for ; Wed, 14 Apr 2010 16:44:53 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id xDBEldscWW8pBEtt for ; Wed, 14 Apr 2010 16:44:53 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id B9156478B02; Thu, 15 Apr 2010 09:44:48 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id zltu0lKmaaDEJFJY; Thu, 15 Apr 2010 09:44:48 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 7BEFDA5011D; Thu, 15 Apr 2010 09:41:46 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 1DF984885B8; Thu, 15 Apr 2010 09:44:48 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id pD7AVXzRcd-E; Thu, 15 Apr 2010 09:44:43 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 6539F488231; Thu, 15 Apr 2010 09:44:43 +1000 (EST) Date: Thu, 15 Apr 2010 09:44:43 +1000 (EST) From: Nathan Scott To: kenj@internode.on.net Cc: pcp Message-ID: <165140555.666661271288683177.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1271285047.21228.6.camel@bozo-laptop> X-ASG-Orig-Subj: Re: Local context vs dynamic namespace Subject: Re: Local context vs dynamic namespace MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271288694 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0280 1.0000 -1.8394 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.84 X-Barracuda-Spam-Status: No, SCORE=-1.84 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27503 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean ----- "Ken McDonell" wrote: > ... > I can revisit this if someone sees a problem with the approach I've > taken. All sounds fine to me. I came across what looks like a day-1 bug in local context mode when dealing with help text this morning (so if you hit a free-of-non-malloc memory in there, I've got a fix for that already in my local workarea). cheers. -- Nathan From makc@iinet.net.au Wed Apr 14 23:21:51 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.5 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3F4LojY125937 for ; Wed, 14 Apr 2010 23:21:50 -0500 X-ASG-Debug-ID: 1271305423-1ec900390000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from outbound.icp-qv1-irony-out3.iinet.net.au (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id D38B212959E1 for ; Wed, 14 Apr 2010 21:23:44 -0700 (PDT) Received: from outbound.icp-qv1-irony-out3.iinet.net.au (outbound.icp-qv1-irony-out3.iinet.net.au [203.59.1.148]) by cuda.sgi.com with ESMTP id LMP86xPQ9WXDqVNW for ; Wed, 14 Apr 2010 21:23:44 -0700 (PDT) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AjMGAIMxxkvL2Rwk/2dsb2JhbACPcYtrcb4+hQ0E X-IronPort-AV: E=Sophos;i="4.52,209,1270396800"; d="scan'208";a="584656445" Received: from unknown (HELO alma.crabbed.net) ([203.217.28.36]) by outbound.icp-qv1-irony-out3.iinet.net.au with ESMTP; 15 Apr 2010 12:23:42 +0800 Received: by alma.crabbed.net (Postfix, from userid 501) id 02DEDEE5DB; Thu, 15 Apr 2010 14:23:42 +1000 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <19398.38093.571035.528981@iinet.net.au> Date: Thu, 15 Apr 2010 14:23:41 +1000 From: Max Matveev To: pcp@oss.sgi.com X-ASG-Orig-Subj: pcp updates - mostly Solaris releated Subject: pcp updates - mostly Solaris releated X-Mailer: VM 7.17 under 21.4 (patch 19) "Constant Variable" XEmacs Lucid X-Barracuda-Connect: outbound.icp-qv1-irony-out3.iinet.net.au[203.59.1.148] X-Barracuda-Start-Time: 1271305425 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27521 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean I've got a few small changes I'd like to push into dev - mostly related to OpenSolaris. I can push them myself or someone can shepherd them in. The following changes since commit ebcf83e8d8a9c7bb8789c57f7d1a35172e8d704b: Nathan Scott (1): Bump version for recent bug fixes (3.1.2). are available in the git repository at: git://oss.sgi.com/makc/pcp solaris-misc Max Matveev (4): Teach pmdaproc about Solaris' ping. Wait for pmcd to stop before removing files. Once again increase table size for dbpmda lexer Reserve domain for IPMI pmda build/sun/preremove | 4 ++-- src/dbpmda/src/lex.l | 2 +- src/pmcd/pmdaproc.sh | 3 +++ src/pmns/stdpmid.pcp | 1 + 4 files changed, 7 insertions(+), 3 deletions(-) max From nathans@aconex.com Thu Apr 15 00:28:06 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3F5S5re131908 for ; Thu, 15 Apr 2010 00:28:05 -0500 X-ASG-Debug-ID: 1271309396-43ed01c10000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 88D952C3CC2 for ; Wed, 14 Apr 2010 22:29:56 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id I4BTr1FqGnIuef00 for ; Wed, 14 Apr 2010 22:29:56 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 5C6E3480F21 for ; Thu, 15 Apr 2010 15:29:54 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id hjc3jG9Byk1VqWTC for ; Thu, 15 Apr 2010 15:29:54 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 75A24A5011D for ; Thu, 15 Apr 2010 15:26:52 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 37F0C4885B4 for ; Thu, 15 Apr 2010 15:29:54 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id dEK6kLaFgO8j for ; Thu, 15 Apr 2010 15:29:50 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 34FA24885CD for ; Thu, 15 Apr 2010 15:29:50 +1000 (EST) Date: Thu, 15 Apr 2010 15:29:50 +1000 (EST) From: nathans@aconex.com Sender: nscott@aconex.com To: pcp@oss.sgi.com Message-ID: <930378675.686141271309390084.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1842579133.685801271309053905.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: pcp updates (+mega-merge) Subject: pcp updates (+mega-merge) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271309398 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.52 X-Barracuda-Spam-Status: No, SCORE=-1.52 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=BSF_RULE7568M, NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27525 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.00 NO_REAL_NAME From: does not include a real name 0.50 BSF_RULE7568M Custom Rule 7568M X-Virus-Status: Clean Changes committed to git://oss.sgi.com/nathans/pcp.git [ Merge branch 'solaris-misc' of ../makc/pcp into dev ] [ Merge branch 'dev' of ../kenj into dev ] [ Merge branch 'dev' of ../mort/pcp into dev ] commit 94cfb60c1e0b7693f57933a952604a174b876b8f Author: Nathan Scott Date: Thu Apr 15 14:27:17 2010 +1000 Fix a memory leak in pmdammv help text handling. commit 395ed7b4066d42b96a7ac216e94db1727b661f4b Author: Nathan Scott Date: Thu Apr 15 14:21:02 2010 +1000 Add anonymous huge page Linux memory utilisation metric. Fixed up missing help text for vmalloc space while at it, and added in couple of other obscure memory metrics which we were not reporting yet (from auditing /proc/meminfo kernel sources for a current kernel - 2.6.34-rc4). commit ee4db6282c29c9b0dd07822a0545ed20ccbcc0c0 Author: Nathan Scott Date: Thu Apr 15 14:17:34 2010 +1000 Fix local context reporting of help text. Help text coming out of the direct PMDA calls to get help text is not dynamically allocated, so we cannot free it. pmLookupText callers (like pminfo) are expecting the buffer to be allocated however, so we need to strdup the text before passing it back. MMV and Windows PMDAs need updating as they are leaking memory in this area at the moment (will be separate commits). commit 4ac19ec1da0dc725aac37b4dd11693b1495f70f1 Author: Nathan Scott Date: Thu Apr 15 08:00:35 2010 +1000 Cleanout no-longer-useful IRIX pmns remnant. commit ebaab5614e33de12b0ff2724935fdba13c86d3e7 Author: Nathan Scott Date: Tue Apr 13 16:18:55 2010 +1000 Add exe files to the set we ignore with git commands. From nathans@aconex.com Thu Apr 15 00:56:28 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3F5uRf8133549 for ; Thu, 15 Apr 2010 00:56:28 -0500 X-ASG-Debug-ID: 1271311100-5d4a03a90000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 568A5125D1B1 for ; Wed, 14 Apr 2010 22:58:20 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id eYK35bZRpBLgOF1j for ; Wed, 14 Apr 2010 22:58:20 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 94C3082DAD9; Thu, 15 Apr 2010 15:58:19 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id F76HboekAtnMwaxj; Thu, 15 Apr 2010 15:58:19 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id A116FA5011D; Thu, 15 Apr 2010 15:55:17 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 678854885CD; Thu, 15 Apr 2010 15:58:19 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ZzEEYJn2OVpm; Thu, 15 Apr 2010 15:58:15 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id AECA94885B4; Thu, 15 Apr 2010 15:58:15 +1000 (EST) Date: Thu, 15 Apr 2010 15:58:15 +1000 (EST) From: Nathan Scott To: kenj@internode.on.net Cc: pcp@oss.sgi.com Message-ID: <9360372.687511271311095473.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1271284076.21228.0.camel@bozo-laptop> X-ASG-Orig-Subj: Re: [pcp] pcp updates - roll to 3.1.3 and pmcd.conf driving available PMDAs for PM_CONTEXT_LOCAL Subject: Re: [pcp] pcp updates - roll to 3.1.3 and pmcd.conf driving available PMDAs for PM_CONTEXT_LOCAL MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271311103 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27525 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean ----- "Ken McDonell" wrote: > Changes committed to git://oss.sgi.com/kenj/pcp.git dev Hmm, somethings not quite right here... 15:42 nathans@verge /source/git/pcp/src/libpcp/src 253> pminfo -Dcontext -L -f mmv build_dsotab: parsing /etc/pmcd/pmcd.conf [6] domain=60, name=/var/lib/pcp/pmdas/linux/pmda_linux.so, init=linux_init __pmLocalPMDA(op=ADD, domain=60, name=/var/lib/pcp/pmdas/linux/pmda_linux.so, init=linux_init) Local Context PMDA Table 0x9c049a0 [0] domain=60 name=/var/lib/pcp/pmdas/linux/pmda_linux.so init=linux_init handle=(nil) [7] domain=70, name=/var/lib/pcp/pmdas/mmv/pmda_mmv.so, init=mmv_init __pmLocalPMDA(op=ADD, domain=70, name=/var/lib/pcp/pmdas/mmv/pmda_mmv.so, init=mmv_init) Local Context PMDA Table 0x9c04a18 [0] domain=60 name=/var/lib/pcp/pmdas/linux/pmda_linux.so init=linux_init handle=(nil) 0x9c04a5c [1] domain=70 name=/var/lib/pcp/pmdas/mmv/pmda_mmv.so init=mmv_init handle=(nil) [9] domain=30, name=/var/lib/pcp/pmdas/sample/pmda_sample.so, init=sample_init __pmLocalPMDA(op=ADD, domain=30, name=/var/lib/pcp/pmdas/sample/pmda_sample.so, init=sample_init) Local Context PMDA Table 0x9c04ad0 [0] domain=60 name=/var/lib/pcp/pmdas/linux/pmda_linux.so init=linux_init handle=(nil) 0x9c04b14 [1] domain=70 name=/var/lib/pcp/pmdas/mmv/pmda_mmv.so init=mmv_init handle=(nil) 0x9c04b58 [2] domain=30 name=/var/lib/pcp/pmdas/sample/pmda_sample.so init=sample_init handle=(nil) pmNewContext(3, (null)) -> 0 So, its finding mmv alright but failing to fetch from it? Fetching mmv metrics from pmcd locally works as expected. cheers. -- Nathan From nathans@aconex.com Thu Apr 15 01:25:14 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.6 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, J_CHICKENPOX_44,J_CHICKENPOX_66 autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3F6PE42134951 for ; Thu, 15 Apr 2010 01:25:14 -0500 X-ASG-Debug-ID: 1271312829-1ecb02270000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 669371ADEB70 for ; Wed, 14 Apr 2010 23:27:09 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id LONXPp2OYN7mGzom for ; Wed, 14 Apr 2010 23:27:09 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 3354182DC4F; Thu, 15 Apr 2010 16:26:51 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id BvBJNzuWVCJG4M45; Thu, 15 Apr 2010 16:26:51 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 32260A5011D; Thu, 15 Apr 2010 16:23:49 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id ED3924885D0; Thu, 15 Apr 2010 16:26:50 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Zb6EcRreVA1a; Thu, 15 Apr 2010 16:26:47 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 3EBDB4885B4; Thu, 15 Apr 2010 16:26:47 +1000 (EST) Date: Thu, 15 Apr 2010 16:26:46 +1000 (EST) From: nathans@aconex.com Sender: nscott@aconex.com To: kenj@internode.on.net Cc: pcp@oss.sgi.com Message-ID: <336247698.688811271312806323.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1937029719.688641271312712797.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: Re: [pcp] pcp updates - roll to 3.1.3 and pmcd.conf driving available PMDAs for PM_CONTEXT_LOCAL Subject: Re: [pcp] pcp updates - roll to 3.1.3 and pmcd.conf driving available PMDAs for PM_CONTEXT_LOCAL MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_50745_242022388.1271312806322" X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271312830 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27528 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.00 NO_REAL_NAME From: does not include a real name X-Virus-Status: Clean ------=_Part_50745_242022388.1271312806322 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit ----- "Nathan Scott" wrote: > ----- "Ken McDonell" wrote: > > > Changes committed to git://oss.sgi.com/kenj/pcp.git dev > > Hmm, somethings not quite right here... > > 15:42 nathans@verge /source/git/pcp/src/libpcp/src 253> pminfo > -Dcontext -L -f mmv > build_dsotab: parsing /etc/pmcd/pmcd.conf > [6] domain=60, name=/var/lib/pcp/pmdas/linux/pmda_linux.so, > init=linux_init > __pmLocalPMDA(op=ADD, domain=60, > name=/var/lib/pcp/pmdas/linux/pmda_linux.so, init=linux_init) > Local Context PMDA Table > 0x9c049a0 [0] domain=60 name=/var/lib/pcp/pmdas/linux/pmda_linux.so > init=linux_init handle=(nil) > [7] domain=70, name=/var/lib/pcp/pmdas/mmv/pmda_mmv.so, init=mmv_init > __pmLocalPMDA(op=ADD, domain=70, > name=/var/lib/pcp/pmdas/mmv/pmda_mmv.so, init=mmv_init) > Local Context PMDA Table > 0x9c04a18 [0] domain=60 name=/var/lib/pcp/pmdas/linux/pmda_linux.so > init=linux_init handle=(nil) > 0x9c04a5c [1] domain=70 name=/var/lib/pcp/pmdas/mmv/pmda_mmv.so > init=mmv_init handle=(nil) > [9] domain=30, name=/var/lib/pcp/pmdas/sample/pmda_sample.so, > init=sample_init > __pmLocalPMDA(op=ADD, domain=30, > name=/var/lib/pcp/pmdas/sample/pmda_sample.so, init=sample_init) > Local Context PMDA Table > 0x9c04ad0 [0] domain=60 name=/var/lib/pcp/pmdas/linux/pmda_linux.so > init=linux_init handle=(nil) > 0x9c04b14 [1] domain=70 name=/var/lib/pcp/pmdas/mmv/pmda_mmv.so > init=mmv_init handle=(nil) > 0x9c04b58 [2] domain=30 name=/var/lib/pcp/pmdas/sample/pmda_sample.so > init=sample_init handle=(nil) > pmNewContext(3, (null)) -> 0 > Also, you can't call mmap() on Windows - theres no entry point. The Win32 API equivalent is wrapped in __pmMemoryMap(...), but it wont do private writable mappings (dunno if thats possible on Windows, we do not have any other code in PCP doing that...!). So, I started to convert this to the portable interface but it doesn't work cos we're writing a null into the mapping in a couple of places to terminate strings. Can we rework that bit? (copy the string out a bit earlier & null terminate the on-stack copy or something?) It got a bit hairy at that point, so decided to chuck the attached patch back over the fence to you to sort out ... :-) cheers. -- Nathan ------=_Part_50745_242022388.1271312806322 Content-Type: text/x-diff; name=memorymap.patch Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename=memorymap.patch diff --git a/src/libpcp/src/connectlocal.c b/src/libpcp/src/connectlocal.c index de84213..aa39407 100644 --- a/src/libpcp/src/connectlocal.c +++ b/src/libpcp/src/connectlocal.c @@ -39,12 +39,11 @@ build_dsotab(void) * mmv 70 dso mmv_init /var/lib/pcp/pmdas/mmv/pmda_mmv.so * */ - char configFileName[MAXPATHLEN]; - FILE *configFile; char *config; char *p; char *q; struct stat sbuf; + int configfd; int lineno = 1; int domain; char *init; @@ -54,23 +53,23 @@ build_dsotab(void) numdso = 0; dsotab = NULL; - strcpy(configFileName, pmGetConfig("PCP_PMCDCONF_PATH")); #ifdef PCP_DEBUG if (pmDebug & DBG_TRACE_CONTEXT) { - fprintf(stderr, "build_dsotab: parsing %s\n", configFileName); + fprintf(stderr, "build_dsotab: parsing %s\n", pmGetConfig("PCP_PMCDCONF_PATH")); } #endif - if (stat(configFileName, &sbuf) < 0) { + + if ((configfd = open(pmGetConfig("PCP_PMCDCONF_PATH"), O_RDONLY)) < 0) return -errno; - } - configFile = fopen(configFileName, "r"); - if (configFile == NULL) { + if (fstat(configfd, &sbuf) < 0) { + close(configfd); return -errno; } - if ((config = mmap(NULL, sbuf.st_size, PROT_READ|PROT_WRITE, MAP_PRIVATE, fileno(configFile), 0)) == MAP_FAILED) { - fclose(configFile); + + config = __pmMemoryMap(configfd, sbuf.st_size, 0); + close(configfd); + if (config == NULL) return -errno; - } p = config; while (*p != '\0') { @@ -136,8 +135,7 @@ eatline: } } - fclose(configFile); - munmap(config, sbuf.st_size); + __pmMemoryUnmap(config, sbuf.st_size); return 0; } ------=_Part_50745_242022388.1271312806322-- From makc@iinet.net.au Thu Apr 15 01:30:39 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.5 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3F6UcjW135271 for ; Thu, 15 Apr 2010 01:30:39 -0500 X-ASG-Debug-ID: 1271313152-4436034e0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from outbound.icp-qv1-irony-out2.iinet.net.au (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id EE84D2C3DDD for ; Wed, 14 Apr 2010 23:32:32 -0700 (PDT) Received: from outbound.icp-qv1-irony-out2.iinet.net.au (outbound.icp-qv1-irony-out2.iinet.net.au [203.59.1.107]) by cuda.sgi.com with ESMTP id 8rD2bmzYp7GuRWN0 for ; Wed, 14 Apr 2010 23:32:32 -0700 (PDT) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AhgHAP1PxkvL2Rwk/2dsb2JhbACPcYtwcb4xhQ4E X-IronPort-AV: E=Sophos;i="4.52,211,1270396800"; d="scan'208";a="637571011" Received: from unknown (HELO alma.crabbed.net) ([203.217.28.36]) by outbound.icp-qv1-irony-out2.iinet.net.au with ESMTP; 15 Apr 2010 14:32:30 +0800 Received: by alma.crabbed.net (Postfix, from userid 501) id C3161EE600; Thu, 15 Apr 2010 16:32:29 +1000 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <19398.45821.479291.285254@iinet.net.au> Date: Thu, 15 Apr 2010 16:32:29 +1000 From: Max Matveev To: kenj@internode.on.net Cc: pcp@oss.sgi.com X-ASG-Orig-Subj: Re: [pcp] pmlogger_check is fooled by localhost Subject: Re: [pcp] pmlogger_check is fooled by localhost In-Reply-To: <1271030318.24244.54.camel@bozo.localdomain> References: <19387.53537.727368.123975@iinet.net.au> <1270603621.15121.79.camel@bozo.localdomain> <19387.65322.68301.468516@iinet.net.au> <1270642137.15121.125.camel@bozo.localdomain> <19389.24907.135424.300701@iinet.net.au> <1270704927.8807.7.camel@bozo.localdomain> <19390.53020.92570.731051@iinet.net.au> <1270847750.18793.8.camel@bozo-laptop> <19393.8355.508754.477098@iinet.net.au> <1271030318.24244.54.camel@bozo.localdomain> X-Mailer: VM 7.17 under 21.4 (patch 19) "Constant Variable" XEmacs Lucid X-Barracuda-Connect: outbound.icp-qv1-irony-out2.iinet.net.au[203.59.1.107] X-Barracuda-Start-Time: 1271313153 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27528 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Mon, 12 Apr 2010 09:58:38 +1000, Ken McDonell wrote: kenj> I'm afraid it is more than just GetPort() ... we try and get the fqdn of kenj> the pmcd host in several places ... kenj> * GetPort() - to be written to the port map file (/tmp/pmlogger/* kenj> in unix-land) kenj> * sendstatus() - for response to a pmlc "status" request That would work just fine - it will try to get fqdn, fail and return the name which was fed into the __pmLogCreate (in my case it will be the name which was used in -h). kenj> * do_preamble() - for the "fake" initial pmResult that contains kenj> the metrics pmcd.pmlogger.host, pmcd.pmlogger.port and kenj> pmcd.pmlogger.archive from pmlogger (not pmcd) IMO it's worth using the same name which was used in GetPort - just pass it in instead of trying to get it every time. kenj> So, one option could be for -H to disable all of the gethostname() and kenj> gethostbyname() calls and use the -h from the command line (or kenj> "localhost" by default) without translation. That would work for me. Ideally I'd put all that stuff into a single place, do it once at startup and declare that in the environment where gethostbyname(foo) returns different fqdn results depending on the time of day or the phase of the moon is just nuts and cannot possible care about the names, so I can do whatever. kenj> This would still leave the messy question of what to do with -P? You kenj> cannot have -h with -P, and -P needs to be able to talk to pmcd on the kenj> local host, so should -P imply "localhost" with -H, or the result of kenj> gethostname() without -H? I think it should be status quo - if you're using -P then you must have a resolvable hostname, -H should apply to -h and be incompatible with -P. kenj> But before exploring that, in your dns-free environment, what does kenj> gethostbyname("localhost") return? It returns "localhost" as fqdn, 127.0.0.1 and, possibly, any other entires from the 127.0.0.1 line in /etc/hosts. kenj> And I assume pcp -h localhost works as expected, correct? It does - gives me the loggers with the name I've fed into -h. $ pcp -h localhost Performance Co-Pilot configuration on localhost: platform: SunOS sonya 5.11 snv_111b i86pc hardware: 2 cpus, 4 disks, 1015MB RAM timezone: EST-10 pmcd: Version 3.1.0-20100128-1, 4 agents, 2 clients pmda: pmcd mmv solaris emmv pmlogger: localhost: /export/home/makc/src/pcp/oss/src/pmlogger/mumble pmie: sonya: /var/log/pcp/pmie/sonya/pmie.log max From kenj@internode.on.net Thu Apr 15 01:30:51 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.2 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3F6Uolv135293 for ; Thu, 15 Apr 2010 01:30:51 -0500 X-ASG-Debug-ID: 1271313164-28de016a0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 6AD61125D337 for ; Wed, 14 Apr 2010 23:32:45 -0700 (PDT) Received: from mail.internode.on.net (bld-mail14.adl6.internode.on.net [150.101.137.99]) by cuda.sgi.com with ESMTP id CPa1wIhjrmjP3eTA for ; Wed, 14 Apr 2010 23:32:45 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20942349-1927428 for multiple; Thu, 15 Apr 2010 16:02:43 +0930 (CST) X-ASG-Orig-Subj: Re: [pcp] pcp updates - roll to 3.1.3 and pmcd.conf driving available PMDAs for PM_CONTEXT_LOCAL Subject: Re: [pcp] pcp updates - roll to 3.1.3 and pmcd.conf driving available PMDAs for PM_CONTEXT_LOCAL From: Ken McDonell Reply-To: kenj@internode.on.net To: Nathan Scott Cc: pcp@oss.sgi.com In-Reply-To: <9360372.687511271311095473.JavaMail.root@mail-au.aconex.com> References: <9360372.687511271311095473.JavaMail.root@mail-au.aconex.com> Content-Type: text/plain; charset="UTF-8" Date: Thu, 15 Apr 2010 16:32:15 +1000 Message-ID: <1271313135.8748.6.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail14.adl6.internode.on.net[150.101.137.99] X-Barracuda-Start-Time: 1271313166 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27527 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean I don't think this is a regression ... this is the same as your bash completion not working. If you turn on -D pmns also, I think you'll see it is using the local PMNS and returning pmid 70:*:* for mmv (no descendent metrics below the root of the dynamic substree of the pmns). This is the second part of the changes that I'm currently working on. The commits so far will mean we build the dsotbl on the fly to be the same as the old compile-time one for the mmv pmda, but noting else has changed. On Thu, 2010-04-15 at 15:58 +1000, Nathan Scott wrote: > ----- "Ken McDonell" wrote: > > > Changes committed to git://oss.sgi.com/kenj/pcp.git dev > > Hmm, somethings not quite right here... > > 15:42 nathans@verge /source/git/pcp/src/libpcp/src 253> pminfo -Dcontext -L -f mmv > build_dsotab: parsing /etc/pmcd/pmcd.conf > [6] domain=60, name=/var/lib/pcp/pmdas/linux/pmda_linux.so, init=linux_init > __pmLocalPMDA(op=ADD, domain=60, name=/var/lib/pcp/pmdas/linux/pmda_linux.so, init=linux_init) > Local Context PMDA Table > 0x9c049a0 [0] domain=60 name=/var/lib/pcp/pmdas/linux/pmda_linux.so init=linux_init handle=(nil) > [7] domain=70, name=/var/lib/pcp/pmdas/mmv/pmda_mmv.so, init=mmv_init > __pmLocalPMDA(op=ADD, domain=70, name=/var/lib/pcp/pmdas/mmv/pmda_mmv.so, init=mmv_init) > Local Context PMDA Table > 0x9c04a18 [0] domain=60 name=/var/lib/pcp/pmdas/linux/pmda_linux.so init=linux_init handle=(nil) > 0x9c04a5c [1] domain=70 name=/var/lib/pcp/pmdas/mmv/pmda_mmv.so init=mmv_init handle=(nil) > [9] domain=30, name=/var/lib/pcp/pmdas/sample/pmda_sample.so, init=sample_init > __pmLocalPMDA(op=ADD, domain=30, name=/var/lib/pcp/pmdas/sample/pmda_sample.so, init=sample_init) > Local Context PMDA Table > 0x9c04ad0 [0] domain=60 name=/var/lib/pcp/pmdas/linux/pmda_linux.so init=linux_init handle=(nil) > 0x9c04b14 [1] domain=70 name=/var/lib/pcp/pmdas/mmv/pmda_mmv.so init=mmv_init handle=(nil) > 0x9c04b58 [2] domain=30 name=/var/lib/pcp/pmdas/sample/pmda_sample.so init=sample_init handle=(nil) > pmNewContext(3, (null)) -> 0 > > > So, its finding mmv alright but failing to fetch from it? Fetching mmv metrics > from pmcd locally works as expected. > > cheers. > From nathans@aconex.com Thu Apr 15 01:32:55 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3F6WtOF135419 for ; Thu, 15 Apr 2010 01:32:55 -0500 X-ASG-Debug-ID: 1271313289-1eda025e0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id E1BC21ADECB0 for ; Wed, 14 Apr 2010 23:34:50 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id 1B7ZGxl8aBiWfX4n for ; Wed, 14 Apr 2010 23:34:50 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 0BC0082DFE0; Thu, 15 Apr 2010 16:34:49 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id 7rSXwAML1FhBXxzu; Thu, 15 Apr 2010 16:34:49 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 07691A5011D; Thu, 15 Apr 2010 16:31:47 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id C73CA4885CD; Thu, 15 Apr 2010 16:34:48 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id joaBz3VPrPkE; Thu, 15 Apr 2010 16:34:44 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 1D92D4885B4; Thu, 15 Apr 2010 16:34:44 +1000 (EST) Date: Thu, 15 Apr 2010 16:34:44 +1000 (EST) From: Nathan Scott To: kenj@internode.on.net Cc: pcp@oss.sgi.com Message-ID: <1515237939.689181271313284080.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1271313135.8748.6.camel@bozo.localdomain> X-ASG-Orig-Subj: Re: [pcp] pcp updates - roll to 3.1.3 and pmcd.conf driving available PMDAs for PM_CONTEXT_LOCAL Subject: Re: [pcp] pcp updates - roll to 3.1.3 and pmcd.conf driving available PMDAs for PM_CONTEXT_LOCAL MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271313290 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27528 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean ----- "Ken McDonell" wrote: > I don't think this is a regression ... this is the same as your bash > completion not working. > ... > This is the second part of the changes that I'm currently working on. Ah, right - sorry for the noise. cheers. -- Nathan From kenj@internode.on.net Thu Apr 15 01:58:12 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.2 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3F6wBU9136748 for ; Thu, 15 Apr 2010 01:58:11 -0500 X-ASG-Debug-ID: 1271314803-28db02540000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id A29AB125DDB4 for ; Thu, 15 Apr 2010 00:00:04 -0700 (PDT) Received: from mail.internode.on.net (bld-mail12.adl6.internode.on.net [150.101.137.97]) by cuda.sgi.com with ESMTP id XUXNmgGCmrHkFBLI for ; Thu, 15 Apr 2010 00:00:04 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.168.7]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 20715075-1927428 for ; Thu, 15 Apr 2010 16:30:02 +0930 (CST) X-ASG-Orig-Subj: pcp updates - replace troublesome mmap() use Subject: pcp updates - replace troublesome mmap() use From: Ken McDonell Reply-To: kenj@internode.on.net To: pcp@oss.sgi.com Content-Type: text/plain; charset="UTF-8" Date: Thu, 15 Apr 2010 16:59:35 +1000 Message-ID: <1271314775.8748.12.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail12.adl6.internode.on.net[150.101.137.97] X-Barracuda-Start-Time: 1271314805 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27529 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Nathan, try this out ... Changes committed to git://oss.sgi.com/kenj/pcp.git dev src/libpcp/src/connectlocal.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) commit a1413f7770ea4136dcd4d3f943d629f5954aba27 Author: Ken McDonell Date: Thu Apr 15 16:56:20 2010 +1000 Replace mmap() use in parsing pmcd.conf To assist portability, the recently introduced mmap() use has been replaced by a malloc() and fread() combination. The pmcd.conf file is small, so this should not be an issue for all platforms. From notify-return-pcp=oss.sgi.com@returns.groups.yahoo.com Sun Apr 18 10:08:27 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: *** X-Spam-Status: No, score=3.2 required=5.0 tests=BAYES_50,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3IF8QDC143910 for ; Sun, 18 Apr 2010 10:08:27 -0500 X-ASG-Debug-ID: 1271603423-013703be0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from n43d.bullet.mail.sp1.yahoo.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with SMTP id CD543102E0CE for ; Sun, 18 Apr 2010 08:10:23 -0700 (PDT) Received: from n43d.bullet.mail.sp1.yahoo.com (n43d.bullet.mail.sp1.yahoo.com [66.163.169.157]) by cuda.sgi.com with SMTP id 3pBgTpHo000F03aG for ; Sun, 18 Apr 2010 08:10:23 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoogroups.com; s=lima; t=1271603423; bh=/UvBgCzxN+Q3aVNrVm+oUQsZzopCWwPERpG9/XYxdos=; h=Received:Received:Date:Message-ID:X-Yahoo-Newman-Property:From:Reply-To:To:Subject:MIME-Version:Content-Type:Content-Transfer-Encoding; b=A6aR90uTC/fZlVCUqtC7T7qtfy0PoJez6H3lWmGHDuJlpw2IxpXNTuBiGjlC8NCZUAqMYTZVTCZqypUJf3o2brmXbVIWlLHfWNkShn30PVxT42vS3ngHTcsYrbnD0iCG DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=lima; d=yahoogroups.com; b=rYd3oS/JrXoBfFb/6azNq0pRH3kM4DN9ar6i0A2LUILZxzKUWoVbB2BSBJaQJuxL+rYca5vPzUjlCZAzgf6DxWLnRDlfijT64/9KOGCqDZqgMQOK0EXOL5POTPuEPLc5; Received: from [69.147.65.151] by n43.bullet.mail.sp1.yahoo.com with NNFMP; 18 Apr 2010 15:10:23 -0000 Received: from [98.137.34.35] by t5.bullet.mail.sp1.yahoo.com with NNFMP; 18 Apr 2010 15:10:23 -0000 Date: 18 Apr 2010 15:10:22 -0000 Message-ID: <1271603422.415.83023.w4@yahoogroups.com> X-Yahoo-Newman-Property: groups-notify From: homqqy Moderator Reply-To: homqqy-unsubscribe@yahoogroups.com To: pcp@oss.sgi.com X-ASG-Orig-Subj: Welcome to the homqqy group Subject: Welcome to the homqqy group MIME-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit X-Barracuda-Connect: n43d.bullet.mail.sp1.yahoo.com[66.163.169.157] X-Barracuda-Start-Time: 1271603424 X-Barracuda-Bayes: INNOCENT GLOBAL 0.4904 1.0000 0.0000 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: 0.00 X-Barracuda-Spam-Status: No, SCORE=0.00 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=DKIM_SIGNED, DKIM_VERIFIED X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27806 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- -0.00 DKIM_VERIFIED Domain Keys Identified Mail: signature passes verification 0.00 DKIM_SIGNED Domain Keys Identified Mail: message has a signature X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Hello, I've added you to my homqqy group at Yahoo! Groups, a free, easy-to-use service. Yahoo! Groups makes it easy to send and receive group messages, coordinate events, share photos and files, and more. Description of the group: ------------------------------------------------------------------------ j00u8430xjro62m2ydh Complete your Yahoo! Groups account: ---------------------------------------------------------------------- Your email address has been added to the email list of a Yahoo! Group. To gain access to all of your group's web features (previous messages, photos, files, calendar, etc.) and easier control of your message delivery options, we highly recommend that you complete your account by connecting your email address to a Yahoo account. It is easy and free. Please visit: http://groups.yahoo.com/convacct?email=pcp%40oss.sgi.com&list=homqqy Important information about the homqqy group ------------------------------------------------------------------------ * To send a message to the members of this group, send an email to: homqqy@yahoogroups.com * To leave the group, you can unsubscribe by replying to this message, or by sending an email to: homqqy-unsubscribe@yahoogroups.com Regards, Moderator, homqqy Report abuse: ------------------------------------------------------------------------ Because Yahoo! Groups values your privacy, it is a violation of our service rules for moderators to add subscribers to a group against their wishes. If you feel this has happened, please notify us: http://help.yahoo.com/l/us/yahoo/groups/original/members/forms/abuse.html You may also change your email preferences to prevent group owners from adding you to their groups. To do so, please go here: http://groups.yahoo.com/s?tag=V9o_YRGxo_ibM3KuRB_wbmnJ6kajauvF04iVo_VHEO_3q506VzYnzlReOffqNoOLB3mS2DerZtbhhw Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/ From nathans@aconex.com Tue Apr 20 18:14:19 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3KNEJuf072147 for ; Tue, 20 Apr 2010 18:14:19 -0500 X-ASG-Debug-ID: 1271805377-470500220000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id C04722DF82F for ; Tue, 20 Apr 2010 16:16:17 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id NPGs9eJHsK6leqaP for ; Tue, 20 Apr 2010 16:16:17 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 2DA768C8E07 for ; Wed, 21 Apr 2010 09:16:15 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id uu5xD6TqBbBqNWPs for ; Wed, 21 Apr 2010 09:16:15 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 3F667A503DB for ; Wed, 21 Apr 2010 09:13:10 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 26CE84885F8 for ; Wed, 21 Apr 2010 09:16:15 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EquS8FkLvxXL for ; Wed, 21 Apr 2010 09:16:11 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 294ED4885F0 for ; Wed, 21 Apr 2010 09:16:11 +1000 (EST) Date: Wed, 21 Apr 2010 09:16:11 +1000 (EST) From: Nathan Scott To: pcp@oss.sgi.com Message-ID: <512358113.119491271805371038.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: pcp updates Subject: pcp updates MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271805378 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.27980 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean Changes committed to git://oss.sgi.com/nathans/pcp.git man/man1/pcpintro.1 | 30 ++++++++++++++++++++++++++++++ man/man1/pmproxy.1 | 21 ++++++++++++++------- man/man3/pmparsehostspec.3 | 26 +++++--------------------- 3 files changed, 49 insertions(+), 28 deletions(-) commit 6acc7760e16d71edfb74c035f92868f7fdacdeba Author: Nathan Scott Date: Wed Apr 21 08:48:19 2010 +1000 Add details of the extended hostname syntax to pcpintro(1). Greg was asking about proxy host environment variables, and we noticed the documentation was in a bit of a mess. Until recently we didn't even ship the pmparsehostspec.3 man page (which Ken slipped in recently with a local context commit)! From gnb@evostor.com Tue Apr 20 23:39:47 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: * X-Spam-Status: No, score=1.1 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3L4dlc5088023 for ; Tue, 20 Apr 2010 23:39:47 -0500 X-ASG-Debug-ID: 1271824906-6f2703140000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from relay.ihostexchange.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id C739D2E020C for ; Tue, 20 Apr 2010 21:41:46 -0700 (PDT) Received: from relay.ihostexchange.net (relay.ihostexchange.net [66.46.182.55]) by cuda.sgi.com with ESMTP id qPP3yA9pTCvM3q8m for ; Tue, 20 Apr 2010 21:41:46 -0700 (PDT) Received: from mailhost.internal (203.206.165.193) by smtp.ihostexchange.net (66.46.182.50) with Microsoft SMTP Server (TLS) id 8.1.393.1; Wed, 21 Apr 2010 00:41:35 -0400 Received: from [172.29.254.198] (unknown [172.29.254.198]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mailhost.internal (Postfix) with ESMTPSA id 1D5D0309E4 for ; Wed, 21 Apr 2010 14:41:37 +1000 (EST) Message-ID: <4BCE8203.8090909@evostor.com> Date: Wed, 21 Apr 2010 14:41:39 +1000 From: Greg Banks User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: PCP Mailing List X-ASG-Orig-Subj: Dumb question about disk.dev.read_bytes Subject: Dumb question about disk.dev.read_bytes Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: 8bit X-Barracuda-Connect: relay.ihostexchange.net[66.46.182.55] X-Barracuda-Start-Time: 1271824906 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0047 1.0000 -1.9900 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.99 X-Barracuda-Spam-Status: No, SCORE=-1.99 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28000 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean G'day, I just noticed that the raw values of this metric have different units on Linux and Solaris. solaris> pminfo -d disk.dev.read_bytes disk.dev.read_bytes Data Type: 64-bit unsigned int InDom: 75.0 0x12c00000 Semantics: counter Units: byte ^^^^ linux> pminfo -d disk.dev.read_bytes disk.dev.read_bytes Data Type: 32-bit unsigned int InDom: 60.1 0xf000001 Semantics: counter Units: Kbyte ^^^^^ Both PMDAs are reporting in the units they claim to, so I guess there's no bug there. But I've been using pmdumptext, whose manpage says UNIT CONVERSION All metrics that have the semantics of counters are automatically con‐ verted to rates over the sample time interval. In interactive mode, pmdumptext will also change the units of some metrics so that they are easier to comprehend: o All metrics with space units (bytes to terabytes) are scaled to bytes. Note that 1024 bytes with be represented as 1.02K, not 1.00K. Yet I see pmdumptext reporting the Linux numbers in Kbyte/sec not byte/sec like I was expecting. So what's wrong here: the manpage, the code, or my reading of the manpage? -- Greg. From nathans@aconex.com Wed Apr 21 00:22:52 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3L5Mpuo092087 for ; Wed, 21 Apr 2010 00:22:52 -0500 X-ASG-Debug-ID: 1271827489-088402cb0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 13AE72DF14C for ; Tue, 20 Apr 2010 22:24:49 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id NKqsBi5nbufGT6eg for ; Tue, 20 Apr 2010 22:24:49 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 5858D4AEE46; Wed, 21 Apr 2010 15:24:47 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id tKab3nPjst5lOVgP; Wed, 21 Apr 2010 15:24:47 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 3B5D6A50110; Wed, 21 Apr 2010 15:21:42 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 44C894885FD; Wed, 21 Apr 2010 15:24:47 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5As4fpp7lSh2; Wed, 21 Apr 2010 15:24:43 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 45C214885F9; Wed, 21 Apr 2010 15:24:43 +1000 (EST) Date: Wed, 21 Apr 2010 15:24:41 +1000 (EST) From: nathans@aconex.com Sender: nscott@aconex.com To: Greg Banks Cc: PCP Mailing List Message-ID: <1174883142.140001271827481469.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1040297306.139951271827353441.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: Re: [pcp] Dumb question about disk.dev.read_bytes Subject: Re: [pcp] Dumb question about disk.dev.read_bytes MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271827491 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28004 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.00 NO_REAL_NAME From: does not include a real name X-Virus-Status: Clean ----- "Greg Banks" wrote: >=20 > UNIT CONVERSION > All metrics that have the semantics of counters are automatically con=E2= =80=90 > verted to rates over the sample time interval. In interactive mode, > pmdumptext will also change the units of some metrics so that they are > easier to comprehend: So, somehow you're getting into "interactive mode", it looks like. > o All metrics with space units (bytes to terabytes) are scaled to > bytes. Note that 1024 bytes with be represented as 1.02K, not > 1.00K. >=20 >=20 > Yet I see pmdumptext reporting the Linux numbers in Kbyte/sec not > byte/sec like I was expecting. So what's wrong here: the manpage, the > code, or my reading of the manpage? What options are you using? I see this... (Linux) $ pmdumptext disk.dev.total_bytes -s 2 Wed Apr 21 15:20:34 ? ? ? ? ? ? Wed Apr 21 15:20:35 0.000 518.341 0.000 0.000 0.000 0.000 $ pmdumptext -i disk.dev.total_bytes -s 2 Wed Apr 21 15:20:38 ? ? ? ? ? ? Wed Apr 21 15:20:39 0.00 0.23M 0.00 0.00 0.00 0.00 The first function (checkUnits) here... http://oss.sgi.com/cgi-bin/gitweb.cgi?p=3Dpcp/pcp-gui.git;a=3Dblob_plain;f= =3Dsrc/dumptext/pmdumptext.cpp;hb=3DHEAD ...has a comment "// Only scale units if interactive and not raw" which looks like the underlying mechanism (the test on the next line). cheers. --=20 Nathan From gnb@evostor.com Wed Apr 21 01:07:45 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.9 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3L67jKx093859 for ; Wed, 21 Apr 2010 01:07:45 -0500 X-ASG-Debug-ID: 1271830184-050f02900000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from relay.ihostexchange.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 25F0810485C3 for ; Tue, 20 Apr 2010 23:09:44 -0700 (PDT) Received: from relay.ihostexchange.net (relay.ihostexchange.net [66.46.182.55]) by cuda.sgi.com with ESMTP id KuGidf39uJRK8Kkb for ; Tue, 20 Apr 2010 23:09:44 -0700 (PDT) Received: from mailhost.internal (203.206.165.193) by smtp.ihostexchange.net (66.46.182.50) with Microsoft SMTP Server (TLS) id 8.1.393.1; Wed, 21 Apr 2010 02:09:34 -0400 Received: from [172.29.254.198] (unknown [172.29.254.198]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mailhost.internal (Postfix) with ESMTPSA id 19F01309E4; Wed, 21 Apr 2010 16:09:36 +1000 (EST) Message-ID: <4BCE96A2.80004@evostor.com> Date: Wed, 21 Apr 2010 16:09:38 +1000 From: Greg Banks User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: "nathans@aconex.com" CC: PCP Mailing List X-ASG-Orig-Subj: Re: [pcp] Dumb question about disk.dev.read_bytes Subject: Re: [pcp] Dumb question about disk.dev.read_bytes References: <1174883142.140001271827481469.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1174883142.140001271827481469.JavaMail.root@mail-au.aconex.com> Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: 7bit X-Barracuda-Connect: relay.ihostexchange.net[66.46.182.55] X-Barracuda-Start-Time: 1271830185 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28007 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean nathans@aconex.com wrote: > ----- "Greg Banks" wrote: > > > What options are you using? I see this... (Linux) > > [...] > The first function (checkUnits) here... > http://oss.sgi.com/cgi-bin/gitweb.cgi?p=pcp/pcp-gui.git;a=blob_plain;f=src/dumptext/pmdumptext.cpp;hb=HEAD > ...has a comment "// Only scale units if interactive and not raw" which > looks like the underlying mechanism (the test on the next line). > > Ah. What I was trying to do was output units with full precision and as units that are consistent across platforms, for input into a script which does mathematics on the values. After your comments and re-reading the manpage properly I tried using -i. Unfortunately I get two things I don't want: reduced precision and unit suffixes, e.g. "0.11G" instead of "114184632.76". I need unit conversion and rate conversion, but no pretty formatting. I don't need that "human-readable" stuff from pmdumptext; kmchart does a much better job at humanifying the data. -- Greg. From nathans@aconex.com Wed Apr 21 01:22:32 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3L6MVCG094578 for ; Wed, 21 Apr 2010 01:22:32 -0500 X-ASG-Debug-ID: 1271831069-74e700e10000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 4D3E82E04C2 for ; Tue, 20 Apr 2010 23:24:30 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id BXcbyCnMLXi6YQzw for ; Tue, 20 Apr 2010 23:24:30 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 93360758892; Wed, 21 Apr 2010 16:24:28 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id DcJfzPOGPBnXpWgc; Wed, 21 Apr 2010 16:24:28 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 6CF00A5028E; Wed, 21 Apr 2010 16:21:23 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 7C446488602; Wed, 21 Apr 2010 16:24:28 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0Axi0nXXbNqV; Wed, 21 Apr 2010 16:24:24 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 7251B488606; Wed, 21 Apr 2010 16:24:24 +1000 (EST) Date: Wed, 21 Apr 2010 16:24:24 +1000 (EST) From: Nathan Scott To: Greg Banks Cc: PCP Mailing List Message-ID: <1369736478.142471271831064269.JavaMail.root@mail-au.aconex.com> In-Reply-To: <4BCE96A2.80004@evostor.com> X-ASG-Orig-Subj: Re: [pcp] Dumb question about disk.dev.read_bytes Subject: Re: [pcp] Dumb question about disk.dev.read_bytes MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271831071 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28007 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean ----- "Greg Banks" wrote: > nathans@aconex.com wrote: > > ----- "Greg Banks" wrote: > > > > > > What options are you using? I see this... (Linux) > > > > [...] > > The first function (checkUnits) here... > > > http://oss.sgi.com/cgi-bin/gitweb.cgi?p=pcp/pcp-gui.git;a=blob_plain;f=src/dumptext/pmdumptext.cpp;hb=HEAD > > ...has a comment "// Only scale units if interactive and not raw" > which > > looks like the underlying mechanism (the test on the next line). > > > > > > Ah. > > What I was trying to do was output units with full precision and as > units that are consistent across platforms, for input into a script > which does mathematics on the values. > > After your comments and re-reading the manpage properly I tried using > > -i. Unfortunately I get two things I don't want: reduced precision and > > unit suffixes, e.g. "0.11G" instead of "114184632.76". I need unit > conversion and rate conversion, but no pretty formatting. > > I don't need that "human-readable" stuff from pmdumptext; kmchart does > a much better job at humanifying the data. Hmm - well, if there's no way to coerce the current dazzling array of command line options to giving that, just whack a --gnb option in there and send a patch (you know where the code is now anyway, from that last url). cheers. -- Nathan From nathans@aconex.com Wed Apr 21 22:28:24 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3M3SOB7159755 for ; Wed, 21 Apr 2010 22:28:24 -0500 X-ASG-Debug-ID: 1271907022-072f01b20000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 2AD4912F3B18 for ; Wed, 21 Apr 2010 20:30:22 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id jkASLoISx17vKnIn for ; Wed, 21 Apr 2010 20:30:22 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 202E04BE303; Thu, 22 Apr 2010 13:30:20 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id 4lRANydXHIP2G5bH; Thu, 22 Apr 2010 13:30:20 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 817B5A5011D; Thu, 22 Apr 2010 13:27:14 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 0806A4885FD; Thu, 22 Apr 2010 13:30:20 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id mR2BL7Mn7sWe; Thu, 22 Apr 2010 13:30:16 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 7DEEE488606; Thu, 22 Apr 2010 13:30:16 +1000 (EST) Date: Thu, 22 Apr 2010 13:30:16 +1000 (EST) From: nathans@aconex.com Sender: nscott@aconex.com To: kenj Cc: pcp Message-ID: <211107923.177101271907016239.JavaMail.root@mail-au.aconex.com> In-Reply-To: <376201228.176691271906735393.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: Dynamic names issue Subject: Dynamic names issue MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271907024 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28075 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.00 NO_REAL_NAME From: does not include a real name X-Virus-Status: Clean Hi Ken, Just came across an issue when using the "magic-dynamic-namespace name:pmid-auto-remapper-thanks-max" logic. This is quite critical for us, so I gotta get to the bottom of this one pronto - insight of any kind would be much appreciated. The problem can be seen as follows (obviously, in our case its our production PMDAs and not pmdasample) ... $ pminfo -m sample.secret.foo.bar.max.redirect pmcd.agent.status sample.secret.foo.bar.max.redirect PMID: 2.4.1 pmcd.agent.status PMID: 2.4.1 $ cat config log advisory on 1sec { sample.secret.foo.bar.max.redirect } $ pmlogger -c config -s 10 foo $ pminfo -a foo pmcd.agent.status pmcd.pmlogger.archive pmcd.pmlogger.port pmcd.pmlogger.host I'm quite surprised that the name stored is "pmcd.agent.status", when I asked for a sample metric. Its not clear how this is happening but its a real problem (think: scripts that were working would no longer, as soon as this moves from QA to production ... which fortunately it has not yet). Any clues as to how pmlogger is managing this impressive act? It must be doing a reverse PMID to name lookup somewhere? (can we avoid that, somehow?!?) Many thanks! -- Nathan From makc@iinet.net.au Thu Apr 22 02:46:55 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, URIBL_BLACK autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3M7ks4K177902 for ; Thu, 22 Apr 2010 02:46:55 -0500 X-ASG-Debug-ID: 1271922533-1e61030f0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from outbound.icp-qv1-irony-out3.iinet.net.au (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 22E6A2E55A6 for ; Thu, 22 Apr 2010 00:48:53 -0700 (PDT) Received: from outbound.icp-qv1-irony-out3.iinet.net.au (outbound.icp-qv1-irony-out3.iinet.net.au [203.59.1.148]) by cuda.sgi.com with ESMTP id bAavnuhNaKom1uOH for ; Thu, 22 Apr 2010 00:48:53 -0700 (PDT) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AoUGAMOcz0vL2Rwk/2dsb2JhbACQIIwAcb08gliCNgQ X-IronPort-AV: E=Sophos;i="4.52,254,1270396800"; d="scan'208";a="589270347" Received: from unknown (HELO alma.crabbed.net) ([203.217.28.36]) by outbound.icp-qv1-irony-out3.iinet.net.au with ESMTP; 22 Apr 2010 15:48:51 +0800 Received: by alma.crabbed.net (Postfix, from userid 501) id 7968CF054D; Thu, 22 Apr 2010 17:48:49 +1000 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <19407.65377.191175.378824@iinet.net.au> Date: Thu, 22 Apr 2010 17:48:49 +1000 From: Max Matveev To: pcp@oss.sgi.com X-ASG-Orig-Subj: pcp updates Subject: pcp updates X-Mailer: VM 7.17 under 21.4 (patch 19) "Constant Variable" XEmacs Lucid X-Barracuda-Connect: outbound.icp-qv1-irony-out3.iinet.net.au[203.59.1.148] X-Barracuda-Start-Time: 1271922535 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28088 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Nathan, Martin, Consider folding the following lot into dev: The following changes since commit 6acc7760e16d71edfb74c035f92868f7fdacdeba: Nathan Scott (1): Add details of the extended hostname syntax to pcpintro(1). are available in the git repository at: git://oss.sgi.com/makc/pcp sunbuild Max Matveev (4): Use git hashes to identify solaris builds Remove debris left after splitting IB PMDA Explictly include sys/stat in connectlocal.c Wait for local filesystems to mount before starting pcp build/sun/GNUmakefile | 19 +++++++++++-------- build/sun/pcp.xml | 11 +++++++---- configure.in | 4 ++++ src/include/builddefs.in | 6 ++---- src/libpcp/src/connectlocal.c | 1 + 5 files changed, 25 insertions(+), 16 deletions(-) From kenj@internode.on.net Thu Apr 22 08:19:42 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.3 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3MDJgBC201677 for ; Thu, 22 Apr 2010 08:19:42 -0500 X-ASG-Debug-ID: 1271942500-1d1f02ee0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id D54452E65AA for ; Thu, 22 Apr 2010 06:21:41 -0700 (PDT) Received: from mail.internode.on.net (bld-mail12.adl6.internode.on.net [150.101.137.97]) by cuda.sgi.com with ESMTP id l6vOczN4IcgmMC1X for ; Thu, 22 Apr 2010 06:21:41 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.216.113]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 21659652-1927428 for multiple; Thu, 22 Apr 2010 22:51:39 +0930 (CST) X-ASG-Orig-Subj: Re: Dynamic names issue Subject: Re: Dynamic names issue From: Ken McDonell Reply-To: kenj@internode.on.net To: nathans@aconex.com Cc: pcp In-Reply-To: <211107923.177101271907016239.JavaMail.root@mail-au.aconex.com> References: <211107923.177101271907016239.JavaMail.root@mail-au.aconex.com> Content-Type: text/plain; charset="UTF-8" Date: Thu, 22 Apr 2010 23:21:05 +1000 Message-ID: <1271942465.9821.6.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail12.adl6.internode.on.net[150.101.137.97] X-Barracuda-Start-Time: 1271942502 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28109 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean The first time a pmid appears in a pmFetch, pmlogger calls pmNameAll to get all of the names of the metric from the pmns to build the pmns fragment in the archive ... at this point the only name that can be found for pmid 2.4.1 is pmcd.agent.status, because the other "name" is not really in the pmns that is available to pmcd, and is not even "owned" by the PMDA with domain 2. I don't see a simple way to prevent this happening ... Changing the name->pmid mapping using dynamic pmns entries that cross PMDA boundaries is probably not a great idea in this case ... sigh. Of course you could use a derived metric to recover from the first problem, but that's probably throwing good money after bad. On Thu, 2010-04-22 at 13:30 +1000, nathans@aconex.com wrote: > Hi Ken, > > Just came across an issue when using the "magic-dynamic-namespace > name:pmid-auto-remapper-thanks-max" logic. This is quite critical > for us, so I gotta get to the bottom of this one pronto - insight > of any kind would be much appreciated. > > The problem can be seen as follows (obviously, in our case its our > production PMDAs and not pmdasample) ... > > $ pminfo -m sample.secret.foo.bar.max.redirect pmcd.agent.status > sample.secret.foo.bar.max.redirect PMID: 2.4.1 > pmcd.agent.status PMID: 2.4.1 > $ cat config > log advisory on 1sec { > sample.secret.foo.bar.max.redirect > } > $ pmlogger -c config -s 10 foo > $ pminfo -a foo > pmcd.agent.status > pmcd.pmlogger.archive > pmcd.pmlogger.port > pmcd.pmlogger.host > > I'm quite surprised that the name stored is "pmcd.agent.status", when > I asked for a sample metric. Its not clear how this is happening but > its a real problem (think: scripts that were working would no longer, > as soon as this moves from QA to production ... which fortunately it > has not yet). > > Any clues as to how pmlogger is managing this impressive act? It must > be doing a reverse PMID to name lookup somewhere? (can we avoid that, > somehow?!?) > > Many thanks! > From nathans@aconex.com Thu Apr 22 22:16:03 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3N3G2Fd008805 for ; Thu, 22 Apr 2010 22:16:03 -0500 X-ASG-Debug-ID: 1271992680-568a004c0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id A7705130BEBF for ; Thu, 22 Apr 2010 20:18:01 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id 3l8czstvcuGx2xno for ; Thu, 22 Apr 2010 20:18:01 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 0EB6675B931; Fri, 23 Apr 2010 13:17:59 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id e0VKTNzbcHpuuB5T; Fri, 23 Apr 2010 13:17:59 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id D632CA5017F; Fri, 23 Apr 2010 13:14:52 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id E30754885FF; Fri, 23 Apr 2010 13:17:58 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fMYfxGaIcm+7; Fri, 23 Apr 2010 13:17:54 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 273544885FD; Fri, 23 Apr 2010 13:17:54 +1000 (EST) Date: Fri, 23 Apr 2010 13:17:54 +1000 (EST) From: nathans@aconex.com Sender: nscott@aconex.com To: kenj@internode.on.net Cc: pcp Message-ID: <1988458083.206861271992673963.JavaMail.root@mail-au.aconex.com> In-Reply-To: <2077613103.206821271992571566.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: Re: Dynamic names issue Subject: Re: Dynamic names issue MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_15295_957658181.1271992673960" X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1271992683 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28145 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.00 NO_REAL_NAME From: does not include a real name X-Virus-Status: Clean ------=_Part_15295_957658181.1271992673960 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit ----- "Ken McDonell" wrote: > The first time a pmid appears in a pmFetch, pmlogger calls pmNameAll > to get all of the names of the metric from the pmns to build the pmns > fragment in the archive ... at this point the only name that can be > found for pmid 2.4.1 is pmcd.agent.status, because the other "name" > is not really in the pmns that is available to pmcd, and is not even > "owned" by the PMDA with domain 2. > > I don't see a simple way to prevent this happening ... How's this patch look? It has the added bonus that it removes two round trip PDUs per metric logged (hmmmm ... wonder if this is why some of my remote loggers take so long to start!). The main open question is whether the (existing) task cache is ever invalid at the time that initial callback happens... there's some hairy code in there, but fortunately the guy who wrote it is right here. :) $ cat tmp/config log advisory on 1sec { sample.secret.foo.bar.max.redirect } $ pminfo -a tmp/archive sample.secret.foo.bar.max.redirect pmcd.pmlogger.archive pmcd.pmlogger.port pmcd.pmlogger.host $ Thoughts? Seems like its much improved this way, I'm wondering why it wasn't done like this before!? (IOW, what am I missing?) cheers. -- Nathan ------=_Part_15295_957658181.1271992673960 Content-Type: text/x-patch; name=pmlogger.patch Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename=pmlogger.patch diff --git a/src/pmlogger/callback.c b/src/pmlogger/callback.c index f857d40..4eec34a 100644 --- a/src/pmlogger/callback.c +++ b/src/pmlogger/callback.c @@ -294,6 +294,16 @@ check_inst(pmValueSet *vsp, int hint, pmResult *lrp) return 0; } +static int +lookupTaskCache(task_t *tp, pmID pmid) +{ + int i; + for (i = 0; i < tp->t_numpmid; i++) { + if (tp->t_pmidlist[i] == pmid) + return i; + } + return -1; +} void log_callback(int afid, void *data) @@ -466,27 +476,28 @@ log_callback(int afid, void *data) for (i = 0; i < resp->numpmid; i++) { pmValueSet *vsp = resp->vset[i]; pmDesc desc; - char **names = NULL; + char *names = NULL; int numnames = 0; + sts = __pmLogLookupDesc(&logctl, vsp->pmid, &desc); if (sts < 0) { - if (archive_version == PM_LOG_VERS02) { - if ((numnames = pmNameAll(vsp->pmid, &names)) < 0) { - fprintf(stderr, "pmNameAll: %s\n", pmErrStr(numnames)); - exit(1); - } - } - if ((sts = pmLookupDesc(vsp->pmid, &desc)) < 0) { - fprintf(stderr, "pmLookupDesc: %s\n", pmErrStr(sts)); + /* lookup name and descriptor in task cache */ + int taskindex = lookupTaskCache(tp, vsp->pmid); + if (taskindex == -1) { + fprintf(stderr, "lookupTaskCache cannot find PMID %s\n", + pmIDStr(vsp->pmid)); exit(1); } - if ((sts = __pmLogPutDesc(&logctl, &desc, numnames, names)) < 0) { + if (archive_version == PM_LOG_VERS02) { + numnames = 1; + names = tp->t_namelist[taskindex]; + } + desc = tp->t_desclist[taskindex]; + + if ((sts = __pmLogPutDesc(&logctl, &desc, numnames, &names)) < 0) { fprintf(stderr, "__pmLogPutDesc: %s\n", pmErrStr(sts)); exit(1); } - if (names != NULL) { - free(names); - } } if (desc.indom != PM_INDOM_NULL && vsp->numval > 0) { /* @@ -605,7 +616,7 @@ log_callback(int afid, void *data) } if (rflag && tp->t_size == 0 && pdu_metrics > 0) { - char **names = NULL; + char *names = NULL; tp->t_size = pdu_bytes; @@ -614,27 +625,26 @@ log_callback(int afid, void *data) else fprintf(stderr, "\nMetric "); if (archive_version == PM_LOG_VERS02) { - if (pmNameAll(pdu_first_pmid, &names) < 0) - names = NULL; - } - if (names != NULL) { - fprintf(stderr, "%s", names[0]); - free(names); + int taskindex = lookupTaskCache(tp, pdu_first_pmid); + if (tp->t_namelist[taskindex] != NULL) + names = tp->t_namelist[taskindex]; } + if (names != NULL) + fprintf(stderr, "%s", names); else fprintf(stderr, "%s", pmIDStr(pdu_first_pmid)); if (pdu_metrics > 1) { fprintf(stderr, "\n\t"); if (pdu_metrics > 2) fprintf(stderr, "...\n\t"); + names = NULL; if (archive_version == PM_LOG_VERS02) { - if (pmNameAll(pdu_last_pmid, &names) < 0) - names = NULL; - } - if (names != NULL) { - fprintf(stderr, "%s", names[0]); - free(names); + int taskindex = lookupTaskCache(tp, pdu_last_pmid); + if (tp->t_namelist[taskindex] != NULL) + names = tp->t_namelist[taskindex]; } + if (names != NULL) + fprintf(stderr, "%s", names); else fprintf(stderr, "%s", pmIDStr(pdu_last_pmid)); fprintf(stderr, "\n}"); ------=_Part_15295_957658181.1271992673960-- From nathans@aconex.com Fri Apr 23 06:26:38 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.6 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3NBQcsT045361 for ; Fri, 23 Apr 2010 06:26:38 -0500 X-ASG-Debug-ID: 1272022117-5ef802d90000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id C44732E98A6 for ; Fri, 23 Apr 2010 04:28:38 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id cdEzXBSjkKtJCy8b for ; Fri, 23 Apr 2010 04:28:38 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id C06954CB7BC for ; Fri, 23 Apr 2010 21:28:36 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id 8YkGEmwLHlDPFVPd for ; Fri, 23 Apr 2010 21:28:36 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 70D17A5017F for ; Fri, 23 Apr 2010 21:25:30 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id A6395488615 for ; Fri, 23 Apr 2010 21:28:36 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Rj6As3Uah6Yi for ; Fri, 23 Apr 2010 21:28:32 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 4A837488610 for ; Fri, 23 Apr 2010 21:28:32 +1000 (EST) Date: Fri, 23 Apr 2010 21:28:28 +1000 (EST) From: Nathan Scott To: pcp@oss.sgi.com Message-ID: <355262156.221041272022108831.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: pcp updates (Mac) Subject: pcp updates (Mac) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [220.237.111.48] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - SAF3 (Mac)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1272022119 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28176 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean Changes committed to git://oss.sgi.com/nathans/pcp.git build/mac/GNUmakefile | 1 + src/include/builddefs.in | 1 + src/include/buildrules | 1 + src/pmdas/mmv/GNUmakefile | 13 +++++++++---- 4 files changed, 12 insertions(+), 4 deletions(-) commit 43bb7b205ecc149170e45e3d2e0e209efacbcd5a Author: Nathan Scott Date: Fri Apr 23 21:27:04 2010 +1000 Resolve an MMV issue evident in the Mac installer. Since we install the root_mmv namespace file into two locations, the Mac installer has been spweing errors. Resolve this by using two different files for Mac install. commit 2efa2a790bfbfdb07abd61080dccbd91ecf39938 Author: Nathan Scott Date: Fri Apr 23 21:25:44 2010 +1000 Add DIRTDIR clean macro and use it for mac build cleanup. From nathans@aconex.com Fri Apr 23 06:28:15 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.6 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3NBSFYu045484 for ; Fri, 23 Apr 2010 06:28:15 -0500 X-ASG-Debug-ID: 1272022215-192f00280000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 974E62E98C0 for ; Fri, 23 Apr 2010 04:30:16 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id QpdLaAuQcBlzNbCV for ; Fri, 23 Apr 2010 04:30:16 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 6A75675CE08; Fri, 23 Apr 2010 21:30:14 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id KjYkHyuTZMBVAanm; Fri, 23 Apr 2010 21:30:14 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 19DC2A5017F; Fri, 23 Apr 2010 21:27:08 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 4F5E1488610; Fri, 23 Apr 2010 21:30:14 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tWJahE8N4Cxf; Fri, 23 Apr 2010 21:30:09 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 8999448861A; Fri, 23 Apr 2010 21:30:09 +1000 (EST) Date: Fri, 23 Apr 2010 21:30:09 +1000 (EST) From: Nathan Scott To: Max Matveev Cc: pcp@oss.sgi.com Message-ID: <1308259982.221071272022209484.JavaMail.root@mail-au.aconex.com> In-Reply-To: <19407.65377.191175.378824@iinet.net.au> X-ASG-Orig-Subj: Re: [pcp] pcp updates Subject: Re: [pcp] pcp updates MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [220.237.111.48] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - SAF3 (Mac)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1272022216 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0017 1.0000 -2.0102 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.01 X-Barracuda-Spam-Status: No, SCORE=-2.01 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28176 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean ----- "Max Matveev" wrote: > Nathan, Martin, > > Consider folding the following lot into dev: > All merged in, thanks Max. -- Nathan From makc@iinet.net.au Fri Apr 23 18:18:36 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.4 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3NNIZWc097201 for ; Fri, 23 Apr 2010 18:18:36 -0500 X-ASG-Debug-ID: 1272064535-2aff02a00000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from outbound.icp-qv1-irony-out5.iinet.net.au (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 1857F2EC55B for ; Fri, 23 Apr 2010 16:15:35 -0700 (PDT) Received: from outbound.icp-qv1-irony-out5.iinet.net.au (outbound.icp-qv1-irony-out5.iinet.net.au [203.59.1.105]) by cuda.sgi.com with ESMTP id J0ATyq4OWIHS11Ia for ; Fri, 23 Apr 2010 16:15:35 -0700 (PDT) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AvsEAIbH0UvL2Rwk/2dsb2JhbACcL3G9UYULBA X-IronPort-AV: E=Sophos;i="4.52,264,1270396800"; d="scan'208";a="138848399" Received: from unknown (HELO alma.crabbed.net) ([203.217.28.36]) by outbound.icp-qv1-irony-out5.iinet.net.au with ESMTP; 24 Apr 2010 07:15:33 +0800 Received: by alma.crabbed.net (Postfix, from userid 501) id 9AFD2F0B93; Sat, 24 Apr 2010 09:15:32 +1000 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <19410.10772.320579.154452@iinet.net.au> Date: Sat, 24 Apr 2010 09:15:32 +1000 From: Max Matveev To: Nathan Scott Cc: pcp@oss.sgi.com X-ASG-Orig-Subj: Re: [pcp] pcp updates Subject: Re: [pcp] pcp updates In-Reply-To: <1308259982.221071272022209484.JavaMail.root@mail-au.aconex.com> References: <19407.65377.191175.378824@iinet.net.au> <1308259982.221071272022209484.JavaMail.root@mail-au.aconex.com> X-Mailer: VM 7.17 under 21.4 (patch 19) "Constant Variable" XEmacs Lucid X-Barracuda-Connect: outbound.icp-qv1-irony-out5.iinet.net.au[203.59.1.105] X-Barracuda-Start-Time: 1272064537 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28214 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Fri, 23 Apr 2010 21:30:09 +1000 (EST), Nathan Scott wrote: nathans> ----- "Max Matveev" wrote: >> Consider folding the following lot into dev: nathans> All merged in, thanks Max. I see you've also wrapped a small change to pmie_daily which I've made yesterday - I was going to send a formal pull request but got distracted. max From nathans@aconex.com Mon Apr 26 22:37:18 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.5 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3R3bHAp125731 for ; Mon, 26 Apr 2010 22:37:18 -0500 X-ASG-Debug-ID: 1272339559-21e902ff0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 07F561333D5D for ; Mon, 26 Apr 2010 20:39:20 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id 1BONwXa80Fz59oLj for ; Mon, 26 Apr 2010 20:39:20 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 6308B4D717A for ; Tue, 27 Apr 2010 13:39:18 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id 305cRVGsqWx0nNnn for ; Tue, 27 Apr 2010 13:39:18 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 1348AA5028D for ; Tue, 27 Apr 2010 13:36:10 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 56BFD48862B for ; Tue, 27 Apr 2010 13:39:18 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Zp74A0qw0C1V for ; Tue, 27 Apr 2010 13:39:14 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 5547A488629 for ; Tue, 27 Apr 2010 13:39:14 +1000 (EST) Date: Tue, 27 Apr 2010 13:39:14 +1000 (EST) From: Nathan Scott To: pcp@oss.sgi.com Message-ID: <520507232.255581272339554137.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: pcpqa updates Subject: pcpqa updates MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1272339561 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28455 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean Changes committed to git://oss.sgi.com/pcp/pcpqa.git 650 | 103 ++++++++++++++++++ 650.out | 360 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ group | 1 owner | 1 4 files changed, 465 insertions(+) commit 341abd9e480dc25539fa456362282876edb81dc0 Author: Nathan Scott Date: Tue Apr 27 13:36:22 2010 +1000 Add test 650 exercises pmlogger with aliased metric names New test (650) which exercises pmlogger with aliased metric names (same PMID, different metric names) under many of the potential scenarios which might affect operation when using config files with such metrics. Passes with the pmlogger binary from pcp-3.1.3+ only (has code updates in this area). From nathans@aconex.com Mon Apr 26 22:45:25 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.6 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3R3jOt8126255 for ; Mon, 26 Apr 2010 22:45:25 -0500 X-ASG-Debug-ID: 1272340046-51bc02e80000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 4D49059479E for ; Mon, 26 Apr 2010 20:47:26 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id 6gsVa9tbYgqLeTY8 for ; Mon, 26 Apr 2010 20:47:26 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 7AFB54D7397 for ; Tue, 27 Apr 2010 13:47:25 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id eZ3EEAaZmGIbVJGl for ; Tue, 27 Apr 2010 13:47:25 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 26870A5028D for ; Tue, 27 Apr 2010 13:44:17 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 6C73948862B for ; Tue, 27 Apr 2010 13:47:25 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iNttgZNKIRUo for ; Tue, 27 Apr 2010 13:47:21 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 126FB488622 for ; Tue, 27 Apr 2010 13:47:21 +1000 (EST) Date: Tue, 27 Apr 2010 13:47:20 +1000 (EST) From: Nathan Scott To: pcp@oss.sgi.com Message-ID: <1705282373.255831272340040962.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: pcp updates Subject: pcp updates MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1272340048 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28456 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean Changes committed to git://oss.sgi.com/nathans/pcp.git src/pmlogger/callback.c | 100 ++++++++++++++++++++++++++++++++++-------------- 1 file changed, 72 insertions(+), 28 deletions(-) commit 09bddb1f278d40f543c0959adbdaffa52f57ae06 Author: Nathan Scott Date: Tue Apr 27 13:45:17 2010 +1000 Improve pmlogger handling of alias names (same PMID, different names) Change the way pmlogger writes out metric names to ensure that the requested metric name (in configuration file) is recorded in the archive always. This change also removes the protocol exchange between pmlogger and pmcd on receipt of a first fetch for a metric, since that information (names and descriptor) is actually already cached and was unnecessary. From nathans@aconex.com Mon Apr 26 22:52:37 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.6 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3R3qa8u126578 for ; Mon, 26 Apr 2010 22:52:37 -0500 X-ASG-Debug-ID: 1272340479-51bd03400000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 648BB8D0549 for ; Mon, 26 Apr 2010 20:54:39 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id vuJRqzQuF4Cmz1ro for ; Mon, 26 Apr 2010 20:54:39 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 857914D6A56; Tue, 27 Apr 2010 13:54:38 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id x1UOdBHjTtIddJZy; Tue, 27 Apr 2010 13:54:38 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 2D15DA50110; Tue, 27 Apr 2010 13:51:30 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 708EF488637; Tue, 27 Apr 2010 13:54:38 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id PRL+miDGwPQ1; Tue, 27 Apr 2010 13:54:33 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id BA17C488622; Tue, 27 Apr 2010 13:54:33 +1000 (EST) Date: Tue, 27 Apr 2010 13:54:33 +1000 (EST) From: Nathan Scott To: kenj@internode.on.net Cc: pcp Message-ID: <1050111450.256351272340473591.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1988458083.206861271992673963.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: Re: Dynamic names issue Subject: Re: Dynamic names issue MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1272340480 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0209 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28456 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean ----- nathans@aconex.com wrote: > ----- "Ken McDonell" wrote: > > > The first time a pmid appears in a pmFetch, pmlogger calls pmNameAll > > to get all of the names of the metric from the pmns to build the > pmns > > fragment in the archive ... at this point the only name that can be > > found for pmid 2.4.1 is pmcd.agent.status, because the other "name" > > is not really in the pmns that is available to pmcd, and is not even > > "owned" by the PMDA with domain 2. > > > > I don't see a simple way to prevent this happening ... > > How's this patch look? It has the added bonus that it removes two (just closing the loop on this one) Ken and I discussed this further in person recently, and Ken pointed out a lack of dealing with the multiple-names to one PMID case ... I updated the patch, wrote several test cases, and committed it a short while ago - plan is to release it in pcp-3.1.3 (which should be soon). cheers. -- Nathan From kenj@internode.on.net Tue Apr 27 05:10:25 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.3 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3RAAPFf151897 for ; Tue, 27 Apr 2010 05:10:25 -0500 X-ASG-Debug-ID: 1272363145-3f39014a0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 903B21D47DE4 for ; Tue, 27 Apr 2010 03:12:26 -0700 (PDT) Received: from mail.internode.on.net (bld-mail18.adl2.internode.on.net [150.101.137.103]) by cuda.sgi.com with ESMTP id RUIF1lbp6mfkd9JC for ; Tue, 27 Apr 2010 03:12:26 -0700 (PDT) Received: from [192.168.1.66] (unverified [118.209.216.113]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 22128628-1927428 for multiple; Tue, 27 Apr 2010 19:42:24 +0930 (CST) X-ASG-Orig-Subj: Re: Dynamic names issue Subject: Re: Dynamic names issue From: Ken McDonell Reply-To: kenj@internode.on.net To: Nathan Scott Cc: pcp In-Reply-To: <1050111450.256351272340473591.JavaMail.root@mail-au.aconex.com> References: <1050111450.256351272340473591.JavaMail.root@mail-au.aconex.com> Content-Type: text/plain; charset="UTF-8" Date: Tue, 27 Apr 2010 20:12:23 +1000 Message-ID: <1272363143.20720.5.camel@bozo-laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail18.adl2.internode.on.net[150.101.137.103] X-Barracuda-Start-Time: 1272363148 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0001 1.0000 -2.0206 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28473 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Tue, 2010-04-27 at 13:54 +1000, Nathan Scott wrote: > ... the plan is to release it in pcp-3.1.3 (which should be soon). Just a related update from me ... I'm close to being done with the changes to make PM_CONTEXT_LOCAL work with both dynamic and derived metrics ... that's the only outstanding work item I'd consider a "yet to come" 3.1.3 release item from me. From nathans@aconex.com Tue Apr 27 05:55:23 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.6 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3RAtNYo153924 for ; Tue, 27 Apr 2010 05:55:23 -0500 X-ASG-Debug-ID: 1272365845-4171027c0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id D344E154CBA9 for ; Tue, 27 Apr 2010 03:57:25 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id hVerAcpRGUGXwfhC for ; Tue, 27 Apr 2010 03:57:25 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 67A764DA0D8; Tue, 27 Apr 2010 20:57:24 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id Q1IDHCA4My1bgsNP; Tue, 27 Apr 2010 20:57:24 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id E0BA3A5028D; Tue, 27 Apr 2010 20:54:15 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 5207E488646; Tue, 27 Apr 2010 20:57:24 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id PMf-H6lNPDbQ; Tue, 27 Apr 2010 20:57:19 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 949AA488640; Tue, 27 Apr 2010 20:57:19 +1000 (EST) Date: Tue, 27 Apr 2010 20:57:19 +1000 (EST) From: Nathan Scott To: kenj@internode.on.net Cc: pcp Message-ID: <976993831.271531272365839069.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1272363143.20720.5.camel@bozo-laptop> X-ASG-Orig-Subj: Re: Dynamic names issue Subject: Re: Dynamic names issue MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [220.237.111.48] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - SAF3 (Mac)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1272365846 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28473 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Status: Clean ----- "Ken McDonell" wrote: > On Tue, 2010-04-27 at 13:54 +1000, Nathan Scott wrote: > > ... the plan is to release it in pcp-3.1.3 (which should be soon). > > Just a related update from me ... > > I'm close to being done with the changes to make PM_CONTEXT_LOCAL > work > with both dynamic and derived metrics ... that's the only outstanding > work item I'd consider a "yet to come" 3.1.3 release item from me. OK, will wait on that before doing 3.1.3 ... thanks! -- Nathan From mort@sgi.com Tue Apr 27 08:04:57 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.8 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from relay.sgi.com (relay1.corp.sgi.com [137.38.102.111]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3RD4v3b160932 for ; Tue, 27 Apr 2010 08:04:57 -0500 Received: from attica.americas.sgi.com (attica.americas.sgi.com [128.162.236.44]) by relay1.corp.sgi.com (Postfix) with ESMTP id 5CE698F809C; Tue, 27 Apr 2010 06:06:57 -0700 (PDT) Received: by attica.americas.sgi.com (Postfix, from userid 44682) id EFE78A5E942D; Tue, 27 Apr 2010 08:06:56 -0500 (CDT) Date: Tue, 27 Apr 2010 08:06:56 -0500 From: Martin Hicks To: Nathan Scott Cc: kenj@internode.on.net, pcp Subject: Re: [pcp] Dynamic names issue Message-ID: <20100427130656.GA13793@alcatraz.americas.sgi.com> References: <1272363143.20720.5.camel@bozo-laptop> <976993831.271531272365839069.JavaMail.root@mail-au.aconex.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <976993831.271531272365839069.JavaMail.root@mail-au.aconex.com> User-Agent: Mutt/1.5.17 (2007-11-01) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Tue, Apr 27, 2010 at 08:57:19PM +1000, Nathan Scott wrote: > > ----- "Ken McDonell" wrote: > > > On Tue, 2010-04-27 at 13:54 +1000, Nathan Scott wrote: > > > ... the plan is to release it in pcp-3.1.3 (which should be soon). > > > > Just a related update from me ... > > > > I'm close to being done with the changes to make PM_CONTEXT_LOCAL > > work > > with both dynamic and derived metrics ... that's the only outstanding > > work item I'd consider a "yet to come" 3.1.3 release item from me. > > OK, will wait on that before doing 3.1.3 ... thanks! I know we haven't helped in fixing up the dynamic pmns stuff, but we were really hoping for a 3.1.3 release before next Tuesday. If it isn't ready then so be it, but it would be nice :) thanks, mh From kenj@internode.on.net Tue Apr 27 21:02:29 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.3 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, J_CHICKENPOX_33 autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3S22Sm1194357 for ; Tue, 27 Apr 2010 21:02:28 -0500 X-ASG-Debug-ID: 1272420267-2ad303100000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 3215F126F4F8 for ; Tue, 27 Apr 2010 19:04:28 -0700 (PDT) Received: from mail.internode.on.net (bld-mail16.adl2.internode.on.net [150.101.137.101]) by cuda.sgi.com with ESMTP id RDajm9f3GqWchPcD for ; Tue, 27 Apr 2010 19:04:28 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.216.113]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 22220847-1927428 for ; Wed, 28 Apr 2010 11:34:26 +0930 (CST) X-ASG-Orig-Subj: pcp updates - ready for 3.1.3 Subject: pcp updates - ready for 3.1.3 From: Ken McDonell Reply-To: kenj@internode.on.net To: pcp@oss.sgi.com Content-Type: text/plain; charset="UTF-8" Date: Wed, 28 Apr 2010 12:03:47 +1000 Message-ID: <1272420227.4037.30.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail16.adl2.internode.on.net[150.101.137.101] X-Barracuda-Start-Time: 1272420272 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28520 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Most of these are to allow normal PMNS services to work with PM_CONTEXT_LOCAL (aka -L as a command line option) in the presence of both dynamic metrics and derived metrics. Some review of the pmns.c changes would be welcome ... and good luck with that one. I've snuck in some other minor things as well ... bash completion, constifying fixup, pmlogreduce -T fix My script for sending this mail seems to have become confused and there is at least one other pmns fix before all of these that has just been pushed to my git tree, so trust the tree rather than this email! Changes committed to git://oss.sgi.com/kenj/pcp.git dev debian/changelog | 5 src/bashrc/pcp_completion.sh | 2 src/include/pmda.h | 8 src/libpcp/src/connectlocal.c | 13 src/libpcp/src/derive_fetch.c | 6 src/libpcp/src/logutil.c | 4 src/libpcp/src/pmns.c | 539 ++++++++++++++++++++++++----------------- src/libpcp_pmda/src/callback.c | 4 src/pmdas/mmv/mmv.c | 4 src/pmdas/sample/src/sample.c | 8 src/pmlogreduce/pmlogreduce.c | 15 + 11 files changed, 370 insertions(+), 238 deletions(-) commit 46b6496e39766185d442bcdd3d59119e6e3348f8 Merge: a1413f7 490f1ef Author: Ken McDonell Date: Wed Apr 28 11:54:11 2010 +1000 Merge branch 'localpmda' into dev commit 490f1ef5141c4c4559ed355e7f8811a087091354 Author: Ken McDonell Date: Wed Apr 28 11:47:22 2010 +1000 pcp_completion.sh - hide stderr output Now that things are working again with our mmv friend, $ pminfo mmv produces $ pminfo mmv[Wed Apr 28 11:48:58] pminfo(11563) Info: MMV: pminfo: 8 metrics and 2 indoms after reload . This change assigns stderr from the nested pminfo -L to /dev/null (I did not want to spank the info message from the mmv pmda, because it is indeed useful in other contexts). commit 0fc07c78b8d9faa395f958b09ba5b326f9172f76 Author: Ken McDonell Date: Wed Apr 28 11:35:33 2010 +1000 pmlogreduce - fix minor compilation whining from last change commit d29698c4bc4765a155437266721d51b24230d8cf Author: Ken McDonell Date: Wed Apr 28 11:31:04 2010 +1000 Tweak error reporting for PMNS operations With the PMNS, derived metrics and dynamic metrics all acting as hiding locations for metrics, one may need to seach in several places for a metric, and some of these searches may fail and some may work ... untangling the error code spaghetti is a little challenging. commit 16b262231197380f1305f80e6b270bb6ee1ab49d Author: Ken McDonell Date: Wed Apr 28 07:52:04 2010 +1000 Make sure PM_ERR_NAME is returned when all options exhausted. Build failed making help text for indom entries after last round of changes. commit 99549e65800d49094c63d175f480b8982a35a58a Author: Ken McDonell Date: Wed Apr 28 07:29:36 2010 +1000 Rework for PM_CONTEXT_LOCAL PMNS services Make PM_CONTEXT_LOCAL work like PM_CONTEXT_HOST for metrics that are visible, and in particular ensure that dynamic metrics and derived metrics are visible for the PMNS services when using PM_CONTEXT_LOCAL. This should be enough to make pminfo -L mmv work again after mmv was converted to using dynamic names in the PMNS. commit 1ff8afac4f1b63b1fafdcbf7b805936d5fe3382b Author: Ken McDonell Date: Wed Apr 28 07:21:10 2010 +1000 Dynamic names rework after derived metrics added. The PMID for derived metrics and for the root of a dynamic subtree in the PMNS both use a domain of DYNAMIC_PMID ... tighten up tests to be sure that we're not confused and consistently using the following tests: dynamic root of pmns pmid_domain(pmid) == DYNAMIC_PMID From nathans@aconex.com Wed Apr 28 18:39:03 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.6 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3SNd3J4018077 for ; Wed, 28 Apr 2010 18:39:03 -0500 X-ASG-Debug-ID: 1272498065-5c9e01eb0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 989751AEFAD9 for ; Wed, 28 Apr 2010 16:41:06 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id X1N707CWtlVXFyjb for ; Wed, 28 Apr 2010 16:41:06 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 00FB94E256B for ; Thu, 29 Apr 2010 09:41:03 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id 2RSm6tdk4ThlutcS for ; Thu, 29 Apr 2010 09:41:03 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 9D0DDA5017F for ; Thu, 29 Apr 2010 09:37:54 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id DE606488649 for ; Thu, 29 Apr 2010 09:41:03 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id p3BNKIg77pos for ; Thu, 29 Apr 2010 09:40:59 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id DEE48488647 for ; Thu, 29 Apr 2010 09:40:59 +1000 (EST) Date: Thu, 29 Apr 2010 09:40:59 +1000 (EST) From: nathans@aconex.com Sender: nscott@aconex.com To: pcp@oss.sgi.com Message-ID: <473398713.314391272498059779.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1771169571.307051272477266797.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: pcpqa updates Subject: pcpqa updates MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1272498067 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28589 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.00 NO_REAL_NAME From: does not include a real name X-Virus-Status: Clean Changes committed to git://oss.sgi.com/pcp/pcpqa.git 633.out | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) commit 46f94bc04b988dde518bf56e671b36183bdabc3b Author: Nathan Scott Date: Thu Apr 29 03:38:54 2010 +1000 Update test 633 output as pmlogger now sends even fewer PDUs. From nathans@aconex.com Wed Apr 28 18:39:07 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.6 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3SNd68k018096 for ; Wed, 28 Apr 2010 18:39:07 -0500 X-ASG-Debug-ID: 1272498069-1b7303d70000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 9BEDF2FB20F for ; Wed, 28 Apr 2010 16:41:10 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id Xs2BfR7XdSFJNu5g for ; Wed, 28 Apr 2010 16:41:10 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id E132B4E256D for ; Thu, 29 Apr 2010 09:41:07 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id FOLqGrCL32FOBu7b for ; Thu, 29 Apr 2010 09:41:07 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 89C87A5017F for ; Thu, 29 Apr 2010 09:37:58 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id CE31D488647 for ; Thu, 29 Apr 2010 09:41:07 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 07QER0mfcmaE for ; Thu, 29 Apr 2010 09:41:03 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id CE7C8488646 for ; Thu, 29 Apr 2010 09:41:03 +1000 (EST) Date: Thu, 29 Apr 2010 09:41:03 +1000 (EST) From: nathans@aconex.com Sender: nscott@aconex.com To: pcp Message-ID: <658660708.314421272498063793.JavaMail.root@mail-au.aconex.com> In-Reply-To: <659346517.307071272477353312.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: pcp updates Subject: pcp updates MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1272498071 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28589 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.00 NO_REAL_NAME From: does not include a real name X-Virus-Status: Clean Changes committed to git://oss.sgi.com/nathans/git/pcp.git src/pmlogger/dopdu.c | 62 ++++++++++++++++++++++++++++++--------------------- 1 files changed, 37 insertions(+), 25 deletions(-) commit b553ce07468d3795e4b94bff526f04f7b5cc9003 Author: Nathan Scott Date: Thu Apr 29 03:51:57 2010 +1000 Resolve fallout from pmlogger handling of alias names (pmlc) Several QA tests using pmlc were now failing after the recent metric name cache lookup changes (corrupt archives generated) - the pmlogger control interface now works properly once more. From daw@sgi.com Wed Apr 28 20:24:20 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.4 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX,J_CHICKENPOX_34 autolearn=no version=3.3.0-rupdated Received: from relay.sgi.com (relay3.corp.sgi.com [198.149.34.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3T1OKfc025111 for ; Wed, 28 Apr 2010 20:24:20 -0500 Received: from zion.americas.sgi.com (zion.americas.sgi.com [128.162.236.17]) by relay3.corp.sgi.com (Postfix) with ESMTP id 897D1AC005 for ; Wed, 28 Apr 2010 18:26:22 -0700 (PDT) Received: (from daw@localhost) by zion.americas.sgi.com (8.12.9/ASC-news-1.4) id o3T1QMmZ38550206 for pcp@oss.sgi.com; Wed, 28 Apr 2010 20:26:22 -0500 (CDT) Message-Id: <201004290126.o3T1QMmZ38550206@zion.americas.sgi.com> Subject: pmie rules To: pcp@oss.sgi.com Date: Wed, 28 Apr 2010 20:26:22 -0500 (CDT) From: David Wright X-Mailer: ELM [version 2.5 PL2] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Hello, What is the best way to enable the full list of pmie rules. By default only load level is on. In the past I went into pmie and disabled the cpu.util and then when quiting the other default rules were added to the pmie/config.default file and then students would turn it back on. It would be nice to have a pmie> save to add the default rules without modifying one. Thanks, Dave Wright From kenj@internode.on.net Wed Apr 28 22:55:17 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.4 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, J_CHICKENPOX_34,J_CHICKENPOX_62 autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3T3tGir034015 for ; Wed, 28 Apr 2010 22:55:17 -0500 X-ASG-Debug-ID: 1272513433-315f00c80000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id BD509120C6F1 for ; Wed, 28 Apr 2010 20:57:13 -0700 (PDT) Received: from mail.internode.on.net (bld-mail18.adl2.internode.on.net [150.101.137.103]) by cuda.sgi.com with ESMTP id iLJwOQxRSVfEt8CC for ; Wed, 28 Apr 2010 20:57:13 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.216.113]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 22382516-1927428 for ; Thu, 29 Apr 2010 13:27:12 +0930 (CST) X-ASG-Orig-Subj: pcpqa updates Subject: pcpqa updates From: Ken McDonell Reply-To: kenj@internode.on.net To: pcp@oss.sgi.com Content-Type: text/plain; charset="UTF-8" Date: Thu, 29 Apr 2010 13:56:31 +1000 Message-ID: <1272513391.4037.46.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail18.adl2.internode.on.net[150.101.137.103] X-Barracuda-Start-Time: 1272513440 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.02 X-Barracuda-Spam-Status: No, SCORE=-1.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=BSF_RULE7568M, BSF_RULE_7582B X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28602 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.50 BSF_RULE7568M Custom Rule 7568M 0.50 BSF_RULE_7582B Custom Rule 7582B X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean My QA changes to accommodate latest pcp changes. Changes committed to git://oss.sgi.com/pcp/pcpqa.git .gitignore | 1 003 | 21 - 019 | 67 ++- 019.linux | 701 --------------------------------------- 019.linux.1 | 701 +++++++++++++++++++++++++++++++++++++++ 019.linux.2 | 873 +++++++++++++++++++++++++++++++++++++++++++++++++ 023 | 2 024 | 54 ++- 024.out.4 | 161 ++++++++- 031 | 2 051 | 2 066 | 19 - 067 | 2 069 | 2 075.out.7 | 2 081 | 6 083 | 6 083.out.linux | 12 178 | 46 +- 208.out | 15 213.out | 4 225.out | 2 226.out | 50 -- 241 | 5 243 | 2 244 | 2 255 | 2 256 | 66 ++- 256.out | 561 ------------------------------- 256.out.1 | 562 +++++++++++++++++++++++++++++++ 256.out.2 | 570 +++++++++++++++++++++++++++++++ 283 | 2 325 | 7 343 | 76 ++++ 343.out | 134 +++++++ 344 | 177 +++++++++ 344.out | 266 ++++++++++++++ 345 | 103 +++++ 345.out | 63 +++ 365 | 2 569.out | 8 646 | 2 647 | 8 647.out | 20 + 649 | 8 649.out | 20 + common.rc | 14 group | 3 owner | 3 src-oss/torture_api.c | 2 src-oss/torture_pmns.c | 23 + 51 files changed, 3999 insertions(+), 1463 deletions(-) commit 8492f45436a935426f35c0c9b80cd101bfceb035 Author: Ken McDonell Date: Thu Apr 29 13:52:34 2010 +1000 225 - remade commit 61dc15f7049ac2f00967e65a38be69fbb3e1822c Author: Ken McDonell Date: Thu Apr 29 13:52:04 2010 +1000 024 - fixup after pmdaLocal changes commit 71454909f859f2ab3033a57a05d196003ab270e7 Merge: 328f533 46f94bc Author: Ken McDonell Date: Thu Apr 29 06:33:42 2010 +1000 Merge branch 'master' of git://oss.sgi.com/pcp/pcpqa commit 328f533e62401d5fe17d1b5a0b287b0ff2198ecb Merge: a968049 341abd9 Author: Ken McDonell Date: Wed Apr 28 12:04:52 2010 +1000 Merge branch 'master' of git://oss.sgi.com/pcp/pcpqa commit a96804967166e6e39f6f30808791f28525024f33 Author: Ken McDonell Date: Wed Apr 28 07:33:07 2010 +1000 345 Exerise pminfo with -L and dynamic and derived metrics Checks that pminfo, pminfo -h localhost and pminfo -L all return the same results with pmns-only queries, -m and -f ... modulo those metrics that are not available with the DSO PMDAs available using PM_CONTEXT_LOCAL. commit c1f3b5a98d94f49ae489ea0a157af9593e79de43 Author: Ken McDonell Date: Tue Apr 27 06:22:21 2010 +1000 343 remade output Not sure how the previous version got there ... it included some diagnostic output that is not evident in the pmlogreduce source history. commit d08de3089a6342829618e5fd59b2824bbe46487d Author: Ken McDonell Date: Tue Apr 27 06:21:29 2010 +1000 081 if cannot find a test host, not run rather than fail commit 362f01935f01d9cccd0b3891a8a8b7f42027aa3f Author: Ken McDonell Date: Mon Apr 26 19:35:13 2010 +1000 646 refine filtering commit bfb1b3b10dccf61a9a65c8135b92f7b082631ef4 Author: Ken McDonell Date: Mon Apr 26 19:32:13 2010 +1000 226 output updated Test now removes duplicate help calls and additional tests for newly added "open socket" feature. commit c8bea3b94fbc5492777fa19c6bb3b0ed1076c4fd Author: Ken McDonell Date: Mon Apr 26 19:30:18 2010 +1000 208 output updated Following torture_pmns.c changes to exercise pmNameAll(). commit 2acf5937101aaa75a5336f3c2e3873eec6375a74 Author: Ken McDonell Date: Mon Apr 26 18:02:25 2010 +1000 647 649 Don't muck with mmv pmda install type This time, check in the .out files, bozo. commit 9647ba8fefb6c7953cf37d7b27c17e4ff15a9aa3 Author: Ken McDonell Date: Mon Apr 26 17:54:57 2010 +1000 325 Scratch long-standing itch Finally found and filtered out the non-deterministic and totally unhelpful kill: 91: Process not found message that sometimes spoiled the party. [Hint: 91 is _not_ a pid] commit 3a11edf3672e177e33f64ead9bf994fb876dccd8 Author: Ken McDonell Date: Mon Apr 26 06:48:48 2010 +1000 344 more extensive testing of pmLocalPMDA and friends commit 7c99a24b743f54f11d7b1d65cbd1de92edbcd30d Author: Ken McDonell Date: Mon Apr 26 06:43:01 2010 +1000 torture_pmns.c add pmNameAll() testing commit ad6c42e89c0c9a0aedd403f66246721ba3e7c7da Author: Ken McDonell Date: Mon Apr 26 06:40:45 2010 +1000 241 647 649 Don't muck with mmv pmda install type mmv pmda is now installed by default and as a DSO PMDA, so test this and (more importantly) make sure it is left installed as a DSO after the tests are run. commit 0b428dcaacf2e370965c36fce0e3dfacc56d68ac Author: Ken McDonell Date: Mon Apr 26 06:35:31 2010 +1000 213 some mmv diags now hidden A couple of the reloading diags are now guarded by PM_DEBUG_APPL0 meaning they are not visible by default in the dbpmda output. commit 1155156c1086c3a7c8eb14b2c8b8caabbc7ee1a5 Author: Ken McDonell Date: Mon Apr 26 06:31:18 2010 +1000 178 extra diags commit 59fdadaabc6f2ffd58ca229f5eaefd8983f7a0d5 Author: Ken McDonell Date: Mon Apr 26 06:30:30 2010 +1000 031 extra diags commit 81c1ee8cdaf54960774c31f72109984fce849812 Author: Ken McDonell Date: Mon Apr 26 06:27:02 2010 +1000 torture_api.c correction after pmNameLookup returns correct error Recent changes in libpcp meant pmNameLookup was correctly returning the first error, not the second error ... amend the logic in the test program to expect the right error status. This logic would appear to have been wrong and undetected from day one! commit 4f128674dab377f4306be699a786e7d2a86bd0fc Author: Ken McDonell Date: Mon Apr 26 06:23:21 2010 +1000 256 better diags and recent dynamic metrics changes With PM_CONTEXT_LOCAL now handling dynamic metrics, there are some additional sample metrics found. commit 8181fabd1657e59a16648d27d83a6403bed6f011 Author: Ken McDonell Date: Wed Apr 14 19:52:15 2010 +1000 024 Changes to building of dsotbl for PM_CONTEXT_LOCAL Changes introduced in PCP 3.1.3 to build dsotbl on the fly from pmcd.conf. commit bccc14033821ca025fec6044a78d21e87dc2c3e0 Author: Ken McDonell Date: Wed Apr 14 19:02:25 2010 +1000 common.rc set $DSO_PREFIX commit 3b575d1658c8551c62910d285054e43faa4875ec Author: Ken McDonell Date: Wed Apr 14 19:01:21 2010 +1000 075, 569 -K option added for __pmLocalPMDA() commit b3952ce86eaa360bbbab482297fce78f4001d7c8 Author: Ken McDonell Date: Wed Apr 14 18:59:01 2010 +1000 083 change in pmlogger error message format commit bbae421d0b4724f5c90b0aec68df418bc4c10559 Author: Ken McDonell Date: Wed Apr 14 18:57:27 2010 +1000 003 additional mem.util metrics may not be present commit 3a28f95dae597261b760cc973bb1e5be35b8114b Author: Ken McDonell Date: Wed Apr 14 18:56:02 2010 +1000 019 PMDA_LOCAL_SAMPLE changes As of PCP 3.1.3, $PMDA_LOCAL_* is no longer having any effect. commit 18359e19761fc8908d2672c6be01e7d32dc85453 Author: Ken McDonell Date: Wed Apr 14 18:54:50 2010 +1000 019 PMDA_LOCAL_SAMPLE changes As of PCP 3.1.3, $PMDA_LOCAL_* is no longer having any effect. commit b61fe0036fe1cbfdcdd060a76a607882e54e0ce4 Author: Ken McDonell Date: Fri Apr 9 14:33:02 2010 +1000 assorted - fixing spelling error causes qa failures Someone fixed the usage message for pmcd so that postive (for -p) was spelled correctly as positive ... unfortunately a bunch of QA scripts were depending on the bad spelling to pick what flavour of pmcd was running ... sigh. commit 65eb7705c9cff7f8a37b93ab938aa97ded2ec664 Author: Ken McDonell Date: Sun Mar 28 14:22:18 2010 +1100 344 exercise __pmAddLocalPMDA New capabilities to add DSO PMDAs to be used with PM_CONTEXT_LOCAL. commit e3e036f2a71d4dc287ddfe65a181333d7405798c Author: Ken McDonell Date: Sun Mar 28 14:21:20 2010 +1100 343 command line arg handling for pmlogreduce From kenj@internode.on.net Thu Apr 29 03:50:49 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.4 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, J_CHICKENPOX_34,J_CHICKENPOX_53 autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3T8om69056084 for ; Thu, 29 Apr 2010 03:50:49 -0500 X-ASG-Debug-ID: 1272531171-4d7301d90000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 2DD671D48D73 for ; Thu, 29 Apr 2010 01:52:51 -0700 (PDT) Received: from mail.internode.on.net (bld-mail17.adl2.internode.on.net [150.101.137.102]) by cuda.sgi.com with ESMTP id d6DjrdkhYWHGt42p for ; Thu, 29 Apr 2010 01:52:51 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.216.113]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 22264032-1927428 for multiple; Thu, 29 Apr 2010 18:22:49 +0930 (CST) X-ASG-Orig-Subj: Help text and malloc snarfoo Subject: Help text and malloc snarfoo From: Ken McDonell Reply-To: kenj@internode.on.net To: Nathan Scott Cc: pcp@oss.sgi.com Content-Type: multipart/mixed; boundary="=-alljihw01MzEU9iMCquJ" Date: Thu, 29 Apr 2010 18:52:09 +1000 Message-ID: <1272531129.4037.132.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 X-Barracuda-Connect: bld-mail17.adl2.internode.on.net[150.101.137.102] X-Barracuda-Start-Time: 1272531173 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28617 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean --=-alljihw01MzEU9iMCquJ Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Continuation of a discussion from IRC, with conclusions and a patch ... The current state of play is described as follows ... things noted as WRONG are indeed so, and the attached patch addresses all of them. Nathan, unless you want otherwise, I'll leave you to merge this patch with your changes (that I suspect address only the dbpmda issue) and I'll keep my changes local and uncommitted awaiting the true gospel from oss.sgi.com. ------------------- src/pmcd/src/dopdus.c DoText() if AGENT_DSO calls version.*.text() (separate cases for one.text, two.text and four.text) otherwise buffer is malloc'd via __pmDecodeText() if (DSO AND PMDA_INTERFACE_1) OR (not DSO) free(buffer) src/dbpmda/src/pmda.c calls __pmDecodeText() & free(buffer) src/dbpmda/src/dso.c calls version.*.text() depending on PMDA_INTERFACE_x & free(buffer) WRONG - should not call free() unless using PMDA_INTERFACE_1 src/libpcp/src/p_text.c in __pmDecodeText(), buffer is malloc'd src/libpcp/src/help.c below pmLookupText() if PM_CONTEXT_HOST buffer is malloc'd via __pmDecodeText() if PM_CONTEXT_LOCAL calls calls version.*.text() then strdup's buffer WRONG ... should not strdup for PMDA_INTERFACE_1 so in both cases PMAPI contract for client is honoured and man/man3/pmdatext.3 is just plain WRONG ... the returned buffer should never be free'd client must free() libpcp_pmda/src/help.c pmdaGetHelp() and pmdaGetInDomHelp() return a pointer into the mmap'd help file libpcp_pmda/src/callback.c pmdaText() calls pmdaGetHelp() or pmdaGetInDomHelp(), and so returns a non-malloc'd buffer libpcp_pmda/src/mainloop.c __pmdaMainPDU() calls version.*.text() then free's the result buffer WRONG - should not call free() unless using PMDA_INTERFACE_1 pmdas/mmv/mmv.c mmv_text() returns static buffer else pointer into mmap area pmdas/sample/src/sample.c sample_text() calls pmdaText() pmdas/summary/mainloop.c calls version.two.text() --=-alljihw01MzEU9iMCquJ Content-Disposition: attachment; filename="patch.pcp" Content-Type: text/x-patch; name="patch.pcp"; charset="UTF-8" Content-Transfer-Encoding: 7bit diff --git a/man/man3/pmdatext.3 b/man/man3/pmdatext.3 index 503e5f3..a4700cf 100644 --- a/man/man3/pmdatext.3 +++ b/man/man3/pmdatext.3 @@ -89,12 +89,10 @@ for the full help text. .PP The .I buffer -is allocated using -.BR malloc (3) -and should be released by the caller of -.B pmdaText -with a call to -.BR free (3). +is managed internally and should +.B not +be released or freed by the caller of +.BR pmdaText . .SH DIAGNOSTICS If the requested help text could not be obtained, diff --git a/src/dbpmda/src/dso.c b/src/dbpmda/src/dso.c index 7c52296..4393447 100644 --- a/src/dbpmda/src/dso.c +++ b/src/dbpmda/src/dso.c @@ -464,7 +464,9 @@ dodso(int pdu) printf("%s\n", buffer); else printf("\n"); - free(buffer); + /* only PMDA_INTERFACE_1 uses a malloc'd buffer */ + if (dispatch.comm.pmda_interface == PMDA_INTERFACE_1) + free(buffer); } else { if (dispatch.comm.pmapi_version == PMAPI_VERSION_1) diff --git a/src/libpcp/src/help.c b/src/libpcp/src/help.c index 5046603..eebce9d 100644 --- a/src/libpcp/src/help.c +++ b/src/libpcp/src/help.c @@ -105,8 +105,11 @@ again_local: type |= PM_TEXT_ONELINE; goto again_local; } - if (n == 0) { - /* PMDAs don't allocate this space, but the caller will free it */ + if (n == 0 && dp->dispatch.comm.pmda_interface != PMDA_INTERFACE_1) { + /* + * PMDAs after PMDA_INTERFACE_1 don't malloc the buffer + * but the caller will free it + */ *buffer = strdup(*buffer); } else if (n < 0 && dp->dispatch.comm.pmapi_version == PMAPI_VERSION_1) { diff --git a/src/libpcp_pmda/src/libdefs.h b/src/libpcp_pmda/src/libdefs.h index e91f703..91546fd 100644 --- a/src/libpcp_pmda/src/libdefs.h +++ b/src/libpcp_pmda/src/libdefs.h @@ -19,6 +19,7 @@ #ifndef LIBDEFS_H #define LIBDEFS_H +#define HAVE_V_ONE(interface) (interface == PMDA_INTERFACE_1) #define HAVE_V_TWO(interface) (interface == PMDA_INTERFACE_2 || interface == PMDA_INTERFACE_3) #define HAVE_V_FOUR(interface) (interface == PMDA_INTERFACE_4) diff --git a/src/libpcp_pmda/src/mainloop.c b/src/libpcp_pmda/src/mainloop.c index b6333b6..445be7f 100644 --- a/src/libpcp_pmda/src/mainloop.c +++ b/src/libpcp_pmda/src/mainloop.c @@ -363,8 +363,12 @@ __pmdaMainPDU(pmdaInterface *dispatch) } if (sts < 0) __pmSendError(pmda->e_outfd, PDU_BINARY, sts); - else + else { __pmSendText(pmda->e_outfd, PDU_BINARY, ident, buffer); + /* only PMDA_INTERFACE_1 malloc's the buffer */ + if (HAVE_V_ONE(dispatch->comm.pmda_interface)) + free(buffer); + } break; case PDU_RESULT: --=-alljihw01MzEU9iMCquJ-- From kenj@internode.on.net Thu Apr 29 03:57:05 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.3 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3T8v5CH056723 for ; Thu, 29 Apr 2010 03:57:05 -0500 X-ASG-Debug-ID: 1272531548-0a4e004d0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 3901D2FCE9D for ; Thu, 29 Apr 2010 01:59:08 -0700 (PDT) Received: from mail.internode.on.net (bld-mail16.adl2.internode.on.net [150.101.137.101]) by cuda.sgi.com with ESMTP id iU32t3qEqPxsEbkQ for ; Thu, 29 Apr 2010 01:59:08 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.216.113]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 22432590-1927428 for ; Thu, 29 Apr 2010 18:29:07 +0930 (CST) X-ASG-Orig-Subj: pcpqa updates Subject: pcpqa updates From: Ken McDonell Reply-To: kenj@internode.on.net To: pcp@oss.sgi.com Content-Type: text/plain; charset="UTF-8" Date: Thu, 29 Apr 2010 18:58:27 +1000 Message-ID: <1272531507.4037.134.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail16.adl2.internode.on.net[150.101.137.101] X-Barracuda-Start-Time: 1272531550 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28617 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean free() snarfoo in libpcp_pmda ... this test kills the simple pmda without my latest patch. Changes committed to git://oss.sgi.com/pcp/pcpqa.git 350 | 39 ++++++++++++++++++++++++++ 350.out | 95 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ group | 1 owner | 1 4 files changed, 136 insertions(+) commit 32babcde8f17cdcac3e25ee6da5dc4b8bbdb45fb Merge: 7766877 e9a8d37 Author: Ken McDonell Date: Thu Apr 29 18:56:39 2010 +1000 Merge branch 'master' of /home/kenj/src/pcpqa into gpush_staging commit e9a8d37706c581de9dff2a45f629f7df1b335bbf Author: Ken McDonell Date: Thu Apr 29 18:55:12 2010 +1000 350 Another help text problem, this time in libpcp_pmda From wwwrun@oss.sgi.com Thu Apr 29 12:38:17 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.4 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from oss.sgi.com (localhost [127.0.0.1]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3THcHIp089527 for ; Thu, 29 Apr 2010 12:38:17 -0500 Received: (from wwwrun@localhost) by oss.sgi.com (8.14.3/8.14.3/Submit) id o3THcHgw089526; Thu, 29 Apr 2010 12:38:17 -0500 Date: Thu, 29 Apr 2010 12:38:17 -0500 From: bugzilla-daemon@oss.sgi.com To: pcp@oss.sgi.com Subject: [Bug 879] New: pmtime use hangs playback X-Bugzilla-Reason: CC X-Bugzilla-Type: newchanged X-Bugzilla-Watch-Reason: None X-Bugzilla-Classification: Unclassified X-Bugzilla-Product: pcp X-Bugzilla-Component: pcp-gui X-Bugzilla-Keywords: X-Bugzilla-Severity: major X-Bugzilla-Who: daw@sgi.com X-Bugzilla-Status: NEW X-Bugzilla-Priority: P5 X-Bugzilla-Assigned-To: nathans@debian.org X-Bugzilla-Target-Milestone: --- X-Bugzilla-Changed-Fields: Message-ID: Auto-Submitted: auto-generated Content-Type: text/plain; charset="UTF-8" MIME-Version: 1.0 http://oss.sgi.com/bugzilla/show_bug.cgi?id=879 Summary: pmtime use hangs playback Product: pcp Version: unspecified Platform: All OS/Version: Linux Status: NEW Severity: major Priority: P5 Component: pcp-gui AssignedTo: nathans@debian.org ReportedBy: daw@sgi.com CC: pcp@oss.sgi.com Estimated Hours: 0.0 Classification: Unclassified When looking at an archive, if speed is played with the playback will stop working. You have to leave pmchart and reopen the archive. So I have archives, I open pmchart -a, click on archive and then slide around fine. Press play and it plays back fine. slide around and it does not stop. But once I start changing the speed of playback, the playback will get stuck/hung/frozen. -- Configure bugmail: http://oss.sgi.com/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug. From wwwrun@oss.sgi.com Thu Apr 29 15:31:51 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.4 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from oss.sgi.com (localhost [127.0.0.1]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3TKVp2r099323 for ; Thu, 29 Apr 2010 15:31:51 -0500 Received: (from wwwrun@localhost) by oss.sgi.com (8.14.3/8.14.3/Submit) id o3TKVp6a099322; Thu, 29 Apr 2010 15:31:51 -0500 Date: Thu, 29 Apr 2010 15:31:51 -0500 From: bugzilla-daemon@oss.sgi.com To: pcp@oss.sgi.com Subject: [Bug 880] New: Preserve Window Geometry in pmchart not working X-Bugzilla-Reason: CC X-Bugzilla-Type: newchanged X-Bugzilla-Watch-Reason: None X-Bugzilla-Classification: Unclassified X-Bugzilla-Product: pcp X-Bugzilla-Component: pcp-gui X-Bugzilla-Keywords: X-Bugzilla-Severity: major X-Bugzilla-Who: daw@sgi.com X-Bugzilla-Status: NEW X-Bugzilla-Priority: P5 X-Bugzilla-Assigned-To: nathans@debian.org X-Bugzilla-Target-Milestone: --- X-Bugzilla-Changed-Fields: Message-ID: Auto-Submitted: auto-generated Content-Type: text/plain; charset="UTF-8" MIME-Version: 1.0 http://oss.sgi.com/bugzilla/show_bug.cgi?id=880 Summary: Preserve Window Geometry in pmchart not working Product: pcp Version: unspecified Platform: All OS/Version: Linux Status: NEW Severity: major Priority: P5 Component: pcp-gui AssignedTo: nathans@debian.org ReportedBy: daw@sgi.com CC: pcp@oss.sgi.com Estimated Hours: 0.0 Classification: Unclassified pmchart has Preserve Window Geometry button. It does save in the file but when the user view is opened the geometry info is not used. #kmchart version 1 global width 912 global height 677 global points 338 global xpos 4294967293 global ypos 4294967272 is ignored when the view is opened again -- Configure bugmail: http://oss.sgi.com/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug. From wwwrun@oss.sgi.com Thu Apr 29 15:35:06 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.4 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from oss.sgi.com (localhost [127.0.0.1]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3TKZ6ba099436 for ; Thu, 29 Apr 2010 15:35:06 -0500 Received: (from wwwrun@localhost) by oss.sgi.com (8.14.3/8.14.3/Submit) id o3TKZ6ZX099435; Thu, 29 Apr 2010 15:35:06 -0500 Date: Thu, 29 Apr 2010 15:35:06 -0500 From: bugzilla-daemon@oss.sgi.com To: pcp@oss.sgi.com Subject: [Bug 881] New: Need a pmcd_check script X-Bugzilla-Reason: CC X-Bugzilla-Type: newchanged X-Bugzilla-Watch-Reason: None X-Bugzilla-Classification: Unclassified X-Bugzilla-Product: pcp X-Bugzilla-Component: pcp X-Bugzilla-Keywords: X-Bugzilla-Severity: major X-Bugzilla-Who: daw@sgi.com X-Bugzilla-Status: NEW X-Bugzilla-Priority: P5 X-Bugzilla-Assigned-To: mort@sgi.com X-Bugzilla-Target-Milestone: --- X-Bugzilla-Changed-Fields: Message-ID: Auto-Submitted: auto-generated Content-Type: text/plain; charset="UTF-8" MIME-Version: 1.0 http://oss.sgi.com/bugzilla/show_bug.cgi?id=881 Summary: Need a pmcd_check script Product: pcp Version: unspecified Platform: All OS/Version: Linux Status: NEW Severity: major Priority: P5 Component: pcp AssignedTo: mort@sgi.com ReportedBy: daw@sgi.com CC: pcp@oss.sgi.com Estimated Hours: 0.0 Classification: Unclassified Need a pmcd_check script. I have the old irix pmcd_check still, but pmcd is dropping out and there is not daemon to check and restart it. ESP does this now but if no ESP then no restart if it drops out. There is a pmlogger_check that is choking when no pmcd. -- Configure bugmail: http://oss.sgi.com/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug. From wwwrun@oss.sgi.com Thu Apr 29 15:47:30 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.4 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from oss.sgi.com (localhost [127.0.0.1]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3TKlTIQ100100 for ; Thu, 29 Apr 2010 15:47:29 -0500 Received: (from wwwrun@localhost) by oss.sgi.com (8.14.3/8.14.3/Submit) id o3TKlT72100099; Thu, 29 Apr 2010 15:47:29 -0500 Date: Thu, 29 Apr 2010 15:47:29 -0500 Message-Id: <201004292047.o3TKlT72100099@oss.sgi.com> From: bugzilla-daemon@oss.sgi.com To: pcp@oss.sgi.com Subject: [Bug 881] Need a pmcd_check script X-Bugzilla-Reason: CC X-Bugzilla-Type: newchanged X-Bugzilla-Watch-Reason: None X-Bugzilla-Classification: Unclassified X-Bugzilla-Product: pcp X-Bugzilla-Component: pcp X-Bugzilla-Keywords: X-Bugzilla-Severity: major X-Bugzilla-Who: daw@sgi.com X-Bugzilla-Status: NEW X-Bugzilla-Priority: P5 X-Bugzilla-Assigned-To: mort@sgi.com X-Bugzilla-Target-Milestone: --- X-Bugzilla-Changed-Fields: In-Reply-To: References: Auto-Submitted: auto-generated Content-Type: text/plain; charset="UTF-8" MIME-Version: 1.0 http://oss.sgi.com/bugzilla/show_bug.cgi?id=881 --- Comment #1 from David Wright 2010-04-29 15:47:29 CST --- I have had pmcd killed by OOM. Can we get /etc/init.d/pcp to put -17 in /proc/pid/oom_adj -- Configure bugmail: http://oss.sgi.com/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug. From kenj@internode.on.net Thu Apr 29 15:54:59 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.3 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3TKsw6l100459 for ; Thu, 29 Apr 2010 15:54:59 -0500 X-ASG-Debug-ID: 1272574622-5f43000b0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 3575E2FF9DA for ; Thu, 29 Apr 2010 13:57:02 -0700 (PDT) Received: from mail.internode.on.net (bld-mail19.adl2.internode.on.net [150.101.137.104]) by cuda.sgi.com with ESMTP id W9fBuyfsYK6VBiJe for ; Thu, 29 Apr 2010 13:57:02 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.216.113]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 22344791-1927428 for multiple; Fri, 30 Apr 2010 06:26:59 +0930 (CST) X-ASG-Orig-Subj: pcp updates - pmda help text & malloc/free issues Subject: pcp updates - pmda help text & malloc/free issues From: Ken McDonell Reply-To: kenj@internode.on.net To: pcp@oss.sgi.com Cc: Nathan Scott Content-Type: text/plain; charset="UTF-8" Date: Fri, 30 Apr 2010 06:56:18 +1000 Message-ID: <1272574578.4037.141.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail19.adl2.internode.on.net[150.101.137.104] X-Barracuda-Start-Time: 1272574624 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.42 X-Barracuda-Spam-Status: No, SCORE=-1.42 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=MARKETING_SUBJECT X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28655 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.60 MARKETING_SUBJECT Subject contains popular marketing words X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean This commit should be applied after Nathan's 754246f409d52cb1964de0299ad0618d7adfccf2 commit. This is my last bunch of changes to be considered for 3.2 ... I'm doing a full QA run now. Changes committed to git://oss.sgi.com/kenj/pcp.git dev man/man3/pmdatext.3 | 12 +++++------- src/dbpmda/src/dso.c | 3 +++ src/libpcp/src/help.c | 7 +++++-- src/libpcp_pmda/src/libdefs.h | 1 + src/libpcp_pmda/src/mainloop.c | 6 +++++- src/pmdas/mmv/.gititgnore | 1 + 6 files changed, 20 insertions(+), 10 deletions(-) commit 8c68d9285fb29aed87c49b6fbd8b7c15e7815510 Author: Ken McDonell Date: Fri Apr 30 06:49:00 2010 +1000 pmda help text and malloc/free issues These changes are a refinement of Nathan's last changes in this area to fix the related issue in libpcp_pmda, tweak the wording the in pmdaText man page, re-instate the free() for PMDA_INTERFACE_1 DSO pmdas in dpmda and avoid a most unlikely memory leak from pmLookupText() and pmLookupInDomText(). From wwwrun@oss.sgi.com Thu Apr 29 16:04:33 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.4 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from oss.sgi.com (localhost [127.0.0.1]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3TL4XVL100881 for ; Thu, 29 Apr 2010 16:04:33 -0500 Received: (from wwwrun@localhost) by oss.sgi.com (8.14.3/8.14.3/Submit) id o3TL4XeA100880; Thu, 29 Apr 2010 16:04:33 -0500 Date: Thu, 29 Apr 2010 16:04:33 -0500 Message-Id: <201004292104.o3TL4XeA100880@oss.sgi.com> From: bugzilla-daemon@oss.sgi.com To: pcp@oss.sgi.com Subject: [Bug 881] Need a pmcd_check script X-Bugzilla-Reason: CC X-Bugzilla-Type: newchanged X-Bugzilla-Watch-Reason: None X-Bugzilla-Classification: Unclassified X-Bugzilla-Product: pcp X-Bugzilla-Component: pcp X-Bugzilla-Keywords: X-Bugzilla-Severity: major X-Bugzilla-Who: kenj@internode.on.net X-Bugzilla-Status: NEW X-Bugzilla-Priority: P5 X-Bugzilla-Assigned-To: mort@sgi.com X-Bugzilla-Target-Milestone: --- X-Bugzilla-Changed-Fields: CC In-Reply-To: References: Auto-Submitted: auto-generated Content-Type: text/plain; charset="UTF-8" MIME-Version: 1.0 http://oss.sgi.com/bugzilla/show_bug.cgi?id=881 Ken McDonell changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |kenj@internode.on.net --- Comment #2 from Ken McDonell 2010-04-29 16:04:32 CST --- Dave, I'd be much keener to fix the problem of pmcd, than provide a band-aid in the form of pmcd_check. pmcd should never exit via oom or segv or ... any other error condition. Do you have any idea what's the cause of the oom condition? How reproducible is it? Could you attach pmcd.conf and pmcd.log files to this bug? -- Configure bugmail: http://oss.sgi.com/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug. From kenj@internode.on.net Thu Apr 29 17:34:20 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.3 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX, J_CHICKENPOX_34 autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3TMYJeH106053 for ; Thu, 29 Apr 2010 17:34:20 -0500 X-ASG-Debug-ID: 1272580582-6b2d03ca0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id A03042FFF23 for ; Thu, 29 Apr 2010 15:36:23 -0700 (PDT) Received: from mail.internode.on.net (bld-mail12.adl6.internode.on.net [150.101.137.97]) by cuda.sgi.com with ESMTP id jwLOzYgeOheGNiLS for ; Thu, 29 Apr 2010 15:36:23 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.216.113]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 22516490-1927428 for multiple; Fri, 30 Apr 2010 08:06:22 +0930 (CST) X-ASG-Orig-Subj: Re: [pcp] pmie rules Subject: Re: [pcp] pmie rules From: Ken McDonell Reply-To: kenj@internode.on.net To: David Wright Cc: pcp@oss.sgi.com In-Reply-To: <201004290126.o3T1QMmZ38550206@zion.americas.sgi.com> References: <201004290126.o3T1QMmZ38550206@zion.americas.sgi.com> Content-Type: text/plain; charset="UTF-8" Date: Fri, 30 Apr 2010 08:35:36 +1000 Message-ID: <1272580536.4037.193.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail12.adl6.internode.on.net[150.101.137.97] X-Barracuda-Start-Time: 1272580584 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28660 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Dave, I suspect you're referring to pmieconf, rather than pmie. If so, pmieconf never made it into open source land from sgi. The config.default that ships in the open source version in fact came from pmieconf, but we don't have the tool nor the other rules in the open source distribution. If sgi wants to open source pmieconf, I'd be willing to look after it in open source land. On Wed, 2010-04-28 at 20:26 -0500, David Wright wrote: > Hello, > > What is the best way to enable the full list of pmie rules. By default only > load level is on. In the past I went into pmie and disabled the cpu.util > and then when quiting the other default rules were added to the > pmie/config.default file and then students would turn it back on. It would > be nice to have a pmie> save to add the default rules without modifying > one. > > Thanks, > Dave Wright > > _______________________________________________ > pcp mailing list > pcp@oss.sgi.com > http://oss.sgi.com/mailman/listinfo/pcp From kenj@internode.on.net Thu Apr 29 17:36:55 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.3 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3TMas3v106194 for ; Thu, 29 Apr 2010 17:36:55 -0500 X-ASG-Debug-ID: 1272580738-481d028a0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 772391AFCB39 for ; Thu, 29 Apr 2010 15:38:59 -0700 (PDT) Received: from mail.internode.on.net (bld-mail15.adl6.internode.on.net [150.101.137.100]) by cuda.sgi.com with ESMTP id HMGJAlbfli320K8r for ; Thu, 29 Apr 2010 15:38:59 -0700 (PDT) Received: from [192.168.1.100] (unverified [118.209.216.113]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 10891234-1927428 for ; Fri, 30 Apr 2010 08:08:57 +0930 (CST) X-ASG-Orig-Subj: duplicate mails from this list? Subject: duplicate mails from this list? From: Ken McDonell Reply-To: kenj@internode.on.net To: pcp@oss.sgi.com Content-Type: text/plain; charset="UTF-8" Date: Fri, 30 Apr 2010 08:38:17 +1000 Message-ID: <1272580697.4037.195.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: bld-mail15.adl6.internode.on.net[150.101.137.100] X-Barracuda-Start-Time: 1272580740 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28660 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean Is anyone else seeing duplicate mails from the pcp list? Either oss is sending mail twice, or I am somehow on the list twice. This has been going on for a while, but just started to annoy me enough to enquire. From nathans@aconex.com Thu Apr 29 18:33:13 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=0.6 required=5.0 tests=AWL,BAYES_00,FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3TNXD0p110378 for ; Thu, 29 Apr 2010 18:33:13 -0500 X-ASG-Debug-ID: 1272584116-76d601d80000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 075E01AFD2D2 for ; Thu, 29 Apr 2010 16:35:17 -0700 (PDT) Received: from postoffice2.aconex.com (mail.aconex.com [203.89.202.182]) by cuda.sgi.com with ESMTP id cKacfU5t4Vv1Rt4k for ; Thu, 29 Apr 2010 16:35:17 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 19127687884; Fri, 30 Apr 2010 09:35:13 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id LIqe76NO0O0EsjN9; Fri, 30 Apr 2010 09:35:13 +1000 (EST) Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 0CB1CA5028D; Fri, 30 Apr 2010 09:32:04 +1000 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id CC502488640; Fri, 30 Apr 2010 09:35:13 +1000 (EST) X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id YyuLbZp78GLU; Fri, 30 Apr 2010 09:35:09 +1000 (EST) Received: from mail-au.aconex.com (mail-au.aconex.com [192.168.102.12]) by gatekeeper.aconex.com (Postfix) with ESMTP id 1E8DF488645; Fri, 30 Apr 2010 09:35:09 +1000 (EST) Date: Fri, 30 Apr 2010 09:35:09 +1000 (EST) From: nathans@aconex.com Sender: nscott@aconex.com To: kenj@internode.on.net Cc: pcp@oss.sgi.com Message-ID: <1063111995.352721272584109039.JavaMail.root@mail-au.aconex.com> In-Reply-To: <1669634868.352701272584051360.JavaMail.root@mail-au.aconex.com> X-ASG-Orig-Subj: Re: [pcp] duplicate mails from this list? Subject: Re: [pcp] duplicate mails from this list? MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [203.89.192.141] X-Mailer: Zimbra 5.0.18_GA_3011.RHEL5_64 (ZimbraWebClient - FF3.0 (Linux)/5.0.18_GA_3011.RHEL5_64) X-Barracuda-Connect: mail.aconex.com[203.89.202.182] X-Barracuda-Start-Time: 1272584118 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28664 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.00 NO_REAL_NAME From: does not include a real name X-Virus-Status: Clean ----- "Ken McDonell" wrote: > Is anyone else seeing duplicate mails from the pcp list? > > Either oss is sending mail twice, or I am somehow on the list twice. > > This has been going on for a while, but just started to annoy me > enough to enquire. I get just one copy of each message FWIW. cheers. -- Nathan From mort@bork.org Thu Apr 29 19:43:17 2010 X-Spam-Checker-Version: SpamAssassin 3.3.0-rupdated (updated) on oss.sgi.com X-Spam-Level: X-Spam-Status: No, score=-0.5 required=5.0 tests=AWL,BAYES_00, FH_DATE_PAST_20XX autolearn=no version=3.3.0-rupdated Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o3U0hHHs114584 for ; Thu, 29 Apr 2010 19:43:17 -0500 X-ASG-Debug-ID: 1272588320-354e03690000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from darwin.bork.org (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 72F11300467 for ; Thu, 29 Apr 2010 17:45:20 -0700 (PDT) Received: from darwin.bork.org (darwin.bork.org [65.49.60.145]) by cuda.sgi.com with ESMTP id YbgUE4NDlkPN8PWQ for ; Thu, 29 Apr 2010 17:45:20 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by darwin.bork.org (Postfix) with ESMTP id 002C4192D7; Thu, 29 Apr 2010 20:45:10 -0400 (EDT) Received: from darwin.bork.org ([127.0.0.1]) by localhost (darwin.bork.org [127.0.0.1]) (amavisd-new, port 10024) with LMTP id qV0jplSoORK9; Thu, 29 Apr 2010 20:45:09 -0400 (EDT) Received: from socrates.bork.org (localhost [127.0.0.1]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client CN "socrates.bork.org", Issuer "Bork Consulting CA" (not verified)) by darwin.bork.org (Postfix) with ESMTPS id 4B52619260; Thu, 29 Apr 2010 20:45:09 -0400 (EDT) Received: by socrates.bork.org (Postfix, from userid 1000) id E65B7CBAE7; Thu, 29 Apr 2010 20:45:17 -0400 (EDT) Date: Thu, 29 Apr 2010 20:45:17 -0400 From: Martin Hicks To: Ken McDonell Cc: David Wright , pcp@oss.sgi.com X-ASG-Orig-Subj: Re: [pcp] pmie rules Subject: Re: [pcp] pmie rules Message-ID: <20100430004517.GA4984@socrates.bork.org> References: <201004290126.o3T1QMmZ38550206@zion.americas.sgi.com> <1272580536.4037.193.camel@bozo.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1272580536.4037.193.camel@bozo.localdomain> User-Agent: Mutt/1.5.20 (2009-06-14) X-Barracuda-Connect: darwin.bork.org[65.49.60.145] X-Barracuda-Start-Time: 1272588322 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.28668 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-Virus-Scanned: ClamAV version 0.94.2, clamav-milter version 0.94.2 on oss.sgi.com X-Virus-Status: Clean On Fri, Apr 30, 2010 at 08:35:36AM +1000, Ken McDonell wrote: > > If sgi wants to open source pmieconf, I'd be willing to look after it in > open source land. I'll try to organize that. mh