From owner-pcp@oss.sgi.com Tue Dec 7 20:59:55 1999 Received: by oss.sgi.com id ; Tue, 7 Dec 1999 20:59:45 -0800 Received: from sgi.SGI.COM ([192.48.153.1]:52815 "EHLO sgi.com") by oss.sgi.com with ESMTP id ; Tue, 7 Dec 1999 20:59:26 -0800 Received: from larry.melbourne.sgi.com ([134.14.52.130]) by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam: SGI does not authorize the use of its proprietary systems or networks for unsolicited or bulk email from the Internet.) via SMTP id VAA05606 for ; Tue, 7 Dec 1999 21:06:48 -0800 (PST) mail_from (markgw@sandpit.melbourne.sgi.com) Received: from sandpit.melbourne.sgi.com (sandpit.melbourne.sgi.com [134.14.55.132]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id QAA14624; Wed, 8 Dec 1999 16:05:31 +1100 Received: (from markgw@localhost) by sandpit.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id QAA15601; Wed, 8 Dec 1999 16:05:31 +1100 (EST) From: markgw@sandpit.melbourne.sgi.com (Mark Goodwin) Message-Id: <9912081605.ZM317353@sandpit.melbourne.sgi.com> Date: Wed, 8 Dec 1999 16:05:30 -0500 Reply-To: markgw@sgi.com X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: pcp@oss.sgi.com Subject: test message Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pcp@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pcp-outgoing testing mail alias From owner-pcp@oss.sgi.com Tue Dec 7 22:45:25 1999 Received: by oss.sgi.com id ; Tue, 7 Dec 1999 22:45:16 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:33613 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Tue, 7 Dec 1999 22:45:08 -0800 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via SMTP id WAA10793 for ; Tue, 7 Dec 1999 22:48:16 -0800 (PST) mail_from (markgw@sandpit.melbourne.sgi.com) Received: from sandpit.melbourne.sgi.com (sandpit.melbourne.sgi.com [134.14.55.132]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id RAA15341; Wed, 8 Dec 1999 17:51:14 +1100 Received: (from markgw@localhost) by sandpit.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id RAA16394; Wed, 8 Dec 1999 17:51:15 +1100 (EST) From: markgw@sandpit.melbourne.sgi.com (Mark Goodwin) Message-Id: <9912081751.ZM318606@sandpit.melbourne.sgi.com> Date: Wed, 8 Dec 1999 17:51:14 -0500 Reply-To: markgw@sgi.com X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: pcp@oss.sgi.com Subject: test message (number two) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pcp@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pcp-outgoing test From owner-pcp@oss.sgi.com Wed Dec 8 11:19:29 1999 Received: by oss.sgi.com id ; Wed, 8 Dec 1999 11:19:19 -0800 Received: from [209.163.47.193] ([209.163.47.193]:18448 "EHLO www.mltgroup.com") by oss.sgi.com with ESMTP id ; Wed, 8 Dec 1999 11:19:08 -0800 Received: (from dtj@localhost) by www.mltgroup.com (8.9.3/8.9.3) id NAA25413 for pcp@oss.sgi.com; Wed, 8 Dec 1999 13:25:29 -0600 Date: Wed, 8 Dec 1999 13:25:29 -0600 From: dtj@www.mltgroup.com Message-Id: <199912081925.NAA25413@www.mltgroup.com> To: pcp@oss.sgi.com Subject: testing mailing list Sender: owner-pcp@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pcp-outgoing Howdy, I am just testing the mailing list stuff working from the outside. -Dean From owner-pcp@oss.sgi.com Thu Dec 9 12:15:54 1999 Received: by oss.sgi.com id ; Thu, 9 Dec 1999 12:15:45 -0800 Received: from sgi.SGI.COM ([192.48.153.1]:44299 "EHLO sgi.com") by oss.sgi.com with ESMTP id ; Thu, 9 Dec 1999 12:15:19 -0800 Received: from larry.melbourne.sgi.com ([134.14.52.130]) by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam: SGI does not authorize the use of its proprietary systems or networks for unsolicited or bulk email from the Internet.) via SMTP id MAA04801 for ; Thu, 9 Dec 1999 12:22:39 -0800 (PST) mail_from (markgw@sandpit.melbourne.sgi.com) Received: from sandpit.melbourne.sgi.com (sandpit.melbourne.sgi.com [134.14.55.132]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id HAA25795; Fri, 10 Dec 1999 07:21:21 +1100 Received: (from markgw@localhost) by sandpit.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id HAA24281; Fri, 10 Dec 1999 07:21:20 +1100 (EST) From: markgw@sandpit.melbourne.sgi.com (Mark Goodwin) Message-Id: <9912100721.ZM420330@sandpit.melbourne.sgi.com> Date: Fri, 10 Dec 1999 07:21:19 -0500 In-Reply-To: Douglas Eadline "Re: SGI releases Performance Co-Pilot open source version" (Dec 9, 7:49) References: Reply-To: markgw@sgi.com X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: Douglas Eadline Subject: Re: SGI releases Performance Co-Pilot open source version Cc: pcp@oss.sgi.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pcp@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pcp-outgoing On Dec 9, 7:49, Douglas Eadline wrote: > Subject: Re: SGI releases Performance Co-Pilot open source version > On Thu, 9 Dec 1999, Mark Goodwin wrote: > > > Thanks for trying PCP - although it's not the full product, > > we have released the APIs and a few cmdline tools in anticipation > > of others using these to develop new distributed monitoring tools. > > In particular, the extensibility of the package and efficiency of the > > protocols should make PCP very attractive to cluster engineers .. > > Does this mean that the visualization tools will not be > included at a later date? The problem with the visualization tools is that they use third party toolkits (motif, inventor, opengl, viewkit and Xrt/Graph from KLG). Since almost nobody except SGI has all of these toolkits, there was simply no point open sourcing the vis tools parts of PCP. There is a possibility of a binary RPM for the vis tools being released in the future - this already exists and is being tested and used internally within SGI as we speak. At this point however, I can't say ('cause I don't know) whether it will be a free download or not. It is layered on the open source pcp product (libraries in pcp are LGPL, to give us and others the option of commercial products that use the pcp APIs). > > Also, perhpas I need to investigate more, but if we were to > build some vizulaization components, is ther any documnetation > on an API. For instance the pminfo displays this information, > but we would like to know how to talk to the pmcd directly. > Of course this would be present in the pminfo source - > which we did not have time to read yet. > Yes! The PCP APIs are all thoroughly documented in both man pages and on-line html books from SGI's techpubs web site. Looks at the following ... man pages: PMAPI(3) is the "intro" page (note PMAPI is uppercase) html : the PCP Users and Administrator's Guide (PCP_UAG) html : the PCP Programmer's Guide (PCP_PG) example code: the pcp open source bits provide this, but also look in the specifc example we created called "pmclient". It's src is installed in /usr/share/pcp/demos/pmclient/ The URL for the html books is in the FAQ at http://oss.sgi.com/projects/pcp/faq.html Hope all this helps, -- Mark From owner-pcp@oss.sgi.com Thu Dec 9 19:41:40 1999 Received: by oss.sgi.com id ; Thu, 9 Dec 1999 19:41:30 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:32024 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Thu, 9 Dec 1999 19:41:09 -0800 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via SMTP id TAA10832 for ; Thu, 9 Dec 1999 19:44:22 -0800 (PST) mail_from (markgw@sandpit.melbourne.sgi.com) Received: from sandpit.melbourne.sgi.com (sandpit.melbourne.sgi.com [134.14.55.132]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id OAA28369; Fri, 10 Dec 1999 14:47:19 +1100 Received: (from markgw@localhost) by sandpit.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id OAA77974; Fri, 10 Dec 1999 14:47:18 +1100 (EST) From: markgw@sandpit.melbourne.sgi.com (Mark Goodwin) Message-Id: <9912101447.ZM427774@sandpit.melbourne.sgi.com> Date: Fri, 10 Dec 1999 14:47:17 -0500 In-Reply-To: Doug Alcorn "Re: Getting system info from the kernel" (Dec 9, 22:48) References: <9912091611.ZM367503@sandpit.melbourne.sgi.com> Reply-To: markgw@sgi.com X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: Doug Alcorn Subject: Re: Getting system info from the kernel Cc: pcp@oss.sgi.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pcp@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pcp-outgoing On Dec 9, 22:48, Doug Alcorn wrote: > Subject: Re: Getting system info from the kernel > markgw@sandpit.melbourne.sgi.com (Mark Goodwin) writes: > > > > > Please check out Performance Co-Pilot (PCP), recently open sourced by SGI. > > The PCP home page is http://oss.sgi.com/projects/pcp/ > > You can download it from there too. > > > > Thanks for sending this information out. Did you announce this > anywhere like www.freshmeat.net or comp.os.linux.announce? I am using Yes we posted to linuxtoday.com freshmeat.net linuxapps.com, slashdot.org, mailed the LSM to the lsm maintainers, and mailed to the beowulf and linux-perf mailing lists. If you know anywhere else to post, please advise! > a cheesy little monitor tool swallowed in my AfterStep Wharf that is > broken somewhat. I might try to adapt it to PCP. > Great idea - that's exactly why we released this stuff - to try and get some uniformity in the all the performance tools out there (they're all more or less "broken" in some way, usually in their transport layer). Please see the PCP home page and faq at http://oss.sgi.com/projects/pcp for pointers to the documentation, example code, etc. -- Mark From owner-pcp@oss.sgi.com Thu Dec 23 14:48:57 1999 Received: by oss.sgi.com id ; Thu, 23 Dec 1999 14:48:47 -0800 Received: from ftg10.lbl.gov ([128.3.5.21]:32018 "EHLO ftg10.lbl.gov") by oss.sgi.com with ESMTP id ; Thu, 23 Dec 1999 14:48:33 -0800 Received: (from eroman@localhost) by ftg10.lbl.gov (8.9.3/8.9.3) id OAA31796 for pcp@oss.sgi.com; Thu, 23 Dec 1999 14:48:16 -0800 Date: Thu, 23 Dec 1999 14:48:16 -0800 From: Eric Roman To: pcp@oss.sgi.com Subject: Building PCP on Red Hat Alpha Message-ID: <19991223144816.B31718@ftg10.lbl.gov> Reply-To: eroman@lbl.gov Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0pre3us Sender: owner-pcp@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pcp-outgoing Hi, I'm interested in helping to develop Performance Co-Pilot for Linux. (Mainly want to do cluster software development, assuming my boss lets me... I've been looking for a tool like this for some time. ) I'm trying to port PCP to Linux Alpha. (RH 6.0) I've run into a problem: The source RPM doesn't build (builds on x86 w/ RH 6.0 and stock GCC). I get the following error message: === weblog === echo "/*" >domain.h echo " * built from stdpmid by Makefile on `date`" >>domain.h echo " */" >>domain.h gawk < ../../pmns/stdpmid >> domain.h '$1=="#define" && $2=="WEBSERVER" { print "#define WEBSERVER " $3 }' gcc -g -DPCP_DEBUG -DPCP_VERSION=\"2.1.1\" -I../../../src/include -c weblog.c -o weblog.o weblog.c:88: initializer element for `wl_metricInfo[0].m_offset' is not computable at load time weblog.c:90: initializer element for `wl_metricInfo[1].m_offset' is not computable at load time weblog.c:92: initializer element for `wl_metricInfo[2].m_offset' is not computable at load time weblog.c:94: initializer element for `wl_metricInfo[3].m_offset' is not computable at load time weblog.c:96: initializer element for `wl_metricInfo[4].m_offset' is not computable at load time weblog.c:158: initializer element for `wl_metricInfo[35].m_offset' is not computable at load time weblog.c:160: initializer element for `wl_metricInfo[36].m_offset' is not computable at load time weblog.c:162: initializer element for `wl_metricInfo[37].m_offset' is not computable at load time weblog.c:224: initializer element for `wl_metricInfo[68].m_offset' is not computable at load time weblog.c: In function `web_fetch': weblog.c:1793: warning: cast to pointer from integer of different size weblog.c:1798: warning: cast from pointer to integer of different size weblog.c:1798: warning: cast to pointer from integer of different size weblog.c:1805: warning: cast from pointer to integer of different size weblog.c:1805: warning: cast to pointer from integer of different size weblog.c:1930: warning: cast from pointer to integer of different size weblog.c:1930: warning: cast to pointer from integer of different size gmake[3]: *** [weblog.o] Error 1 gmake[2]: *** [default] Error 2 gmake[1]: *** [default] Error 2 make: *** [default] Error 2 I think that this is a problem w/ the version of GCC that we're using % gcc -v Reading specs from /usr/lib/gcc-lib/alpha-redhat-linux/egcs-2.91.66/specs gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release) gcc 2.95.2 won't compile it either. I dies with the same error. Any ideas? Eric From owner-pcp@oss.sgi.com Thu Dec 23 15:10:07 1999 Received: by oss.sgi.com id ; Thu, 23 Dec 1999 15:09:57 -0800 Received: from paris11-nas1-40-144.dial.proxad.net ([212.27.40.144]:7935 "HELO eurythro.savigny") by oss.sgi.com with SMTP id ; Thu, 23 Dec 1999 15:09:43 -0800 Received: from free.fr (localhost [127.0.0.1]) by eurythro.savigny (Postfix) with ESMTP id F1D7B2B14F for ; Thu, 23 Dec 1999 23:09:30 +0100 (CET) Message-ID: <38629D99.119AD90B@free.fr> Date: Thu, 23 Dec 1999 22:09:29 +0000 From: Luc Stepniewski Reply-To: lstepml@free.fr X-Mailer: Mozilla 4.7 [en] (X11; I; Linux 2.2.13ext3 i686) X-Accept-Language: fr-FR, fr, en MIME-Version: 1.0 To: pcp@oss.sgi.com Subject: Missing pcp.env.sh, and some questions. Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-pcp@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pcp-outgoing Hello, I installed PCP on my machine. I find it is really wonderful. I just got one problem: It is when I tried to launch pmie, it complained about not finding the file named /etc/pcp.conf.sh (tried on Debian and Redhat) I looked for it everywhere but I couldn't find it. So: I copied the pcp.conf file as pcp.conf.sh, added the '#!/bin/sh' at the beginning, and added to the /etc/init.d/pmie file the following statement (because IS_ON is nowhere :-): IS_ON=false Then I modified the first statement of /etc/init.d/pmie which seems to have a bug: I modified: if [ -fi ${PCP_CONF:-/etc/pcp.conf}.sh ] ; then to: if [ -f ${PCP_CONF:-/etc/pcp.conf}.sh ] ; then (removed the 'i'). It now works really fine (even pmie :-). There is an important variable which is missing from the standard ones, (coming from /proc/sys/fs/file-nr). It is the number of currently open files (and the maximum assigned file number, and the maximum allocatable file number). Their description is the following: "The three values in file-nr denote the number of allocated file handles, the number of used file handles, and the maximum number of file handles. When the allocated file handles come close to the maximum, but the number of actually used ones is far behind, you've encountered a peak in your usage of file handles and you don't need to increase the maximum. " (From http://www.bb-zone.com/Proc/chapter2.html#section2.1) I think I can patch pcp to add it myself, but I don't know in which hierarchy I must put it into. In filesys.*, there are already variables named filesys.maxfiles, filesys.usedfiles and filesys.freefiles, but they are totally unrelated to the /proc/sys/fs/file-nr variables. About practical use of pmie, I'm looking for an example on how to monitor the presence of a daemon on a system. For example, I have an apache daemon, and I'd like to get an alert if it dies, or if there are too many instances of it. The only near useful variable that I see is proc.psinfo.pid. I used it like this: eurythro:/home/enlight# pmie -v val = proc.psinfo.pid #'/usr/sbin/apache'; val: 254 val: 254 val: 254 val: 254 val: ? val: ? val: ? val: ? At the fifth iteration, I stopped the daemon. And it stops returning its pid, which is cool, but it never get a pid again, even when I restart apache. I don't understand why...Another problem with this, is that I don't have the number of running daemons :-( Any ideas ? Thanks, Luc -- Luc Stepniewski ICQ# 6104530 Adequat - Securite, Linux Public key: Key D93B2D2D fingerprint = 49 00 CC D1 69 03 E2 94 C8 78 ED 3C 75 89 A8 DE From owner-pcp@oss.sgi.com Thu Dec 23 15:29:57 1999 Received: by oss.sgi.com id ; Thu, 23 Dec 1999 15:29:47 -0800 Received: from sgi.SGI.COM ([192.48.153.1]:32272 "EHLO sgi.com") by oss.sgi.com with ESMTP id ; Thu, 23 Dec 1999 15:29:33 -0800 Received: from larry.melbourne.sgi.com ([134.14.52.130]) by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam: SGI does not authorize the use of its proprietary systems or networks for unsolicited or bulk email from the Internet.) via SMTP id PAA01306 for ; Thu, 23 Dec 1999 15:28:59 -0800 (PST) mail_from (markgw@sandpit.melbourne.sgi.com) Received: from sandpit.melbourne.sgi.com (sandpit.melbourne.sgi.com [134.14.55.132]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id KAA16894; Fri, 24 Dec 1999 10:27:42 +1100 Received: (from markgw@localhost) by sandpit.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id KAA62874; Fri, 24 Dec 1999 10:27:41 +1100 (EST) From: markgw@sandpit.melbourne.sgi.com (Mark Goodwin) Message-Id: <9912241027.ZM1125771@sandpit.melbourne.sgi.com> Date: Fri, 24 Dec 1999 10:27:41 -0500 In-Reply-To: Eric Roman "Building PCP on Red Hat Alpha" (Dec 23, 14:48) References: <19991223144816.B31718@ftg10.lbl.gov> Reply-To: markgw@sgi.com X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: eroman@lbl.gov, pcp@oss.sgi.com Subject: Re: Building PCP on Red Hat Alpha Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pcp@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pcp-outgoing On Dec 23, 14:48, Eric Roman wrote: > Subject: Building PCP on Red Hat Alpha > > Hi, > > I'm interested in helping to develop Performance Co-Pilot for Linux. > (Mainly want to do cluster software development, assuming my boss > lets me... I've been looking for a tool like this for some time. ) Great! > > I'm trying to port PCP to Linux Alpha. (RH 6.0) I've run into a > problem: The source RPM doesn't build (builds on x86 w/ RH 6.0 and > stock GCC). I get the following error message: > === weblog === > weblog.c:88: initializer element for `wl_metricInfo[0].m_offset' is not computable at load time > ... It looks like in configure.in, our assumption about the default for __psint_t (pointer size int) may be wrong for the alpha platform. In configure.in, near line 334: C_CHECK_TYPE(__psint_t, u_int32_t) Try changing u_int32_t in the above macro to whatever the type should be for alpha (probably __uint64_t). Then rerun configure; make Please let me know if this works and l'll formulate a new macro that will work on non-i386 linux. Also, BTW I have no idea what's in /proc on alpha linux, so you may encounter some difficulty with src/pmdas/linux (forewarning). thanks -- Mark Goodwin SGI Engineering From owner-pcp@oss.sgi.com Thu Dec 23 16:48:27 1999 Received: by oss.sgi.com id ; Thu, 23 Dec 1999 16:48:17 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:42826 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Thu, 23 Dec 1999 16:47:55 -0800 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via SMTP id QAA18285 for ; Thu, 23 Dec 1999 16:43:17 -0800 (PST) mail_from (kenmcd@gonzo.melbourne.sgi.com) Received: from gonzo.melbourne.sgi.com (gate-ptg.melbourne.sgi.com [134.14.52.131]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id LAA17233; Fri, 24 Dec 1999 11:18:12 +1100 Received: (from kenmcd@localhost) by gonzo.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) id LAA21821; Fri, 24 Dec 1999 11:18:11 +1100 Date: Fri, 24 Dec 1999 11:18:11 +1100 From: kenmcd@gonzo.melbourne.sgi.com (Ken McDonell) Message-Id: <9912241118.ZM21819@gonzo.melbourne.sgi.com> In-Reply-To: Luc Stepniewski "Missing pcp.env.sh, and some questions." (Dec 24, 10:10) References: <38629D99.119AD90B@free.fr> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: lstepml@free.fr, pcp@oss.sgi.com Subject: Re: Missing pcp.env.sh, and some questions. Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pcp@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pcp-outgoing On Dec 24, 10:10, Luc Stepniewski wrote: > Subject: Missing pcp.env.sh, and some questions. > Hello, > > I installed PCP on my machine. I find it is really wonderful. ... That's nice to hear. > ... I just got > one > problem: It is when I tried to launch pmie, it complained about not > finding the > file named /etc/pcp.conf.sh (tried on Debian and Redhat) I looked for it > everywhere but I couldn't find it. > > So: > I copied the pcp.conf file as pcp.conf.sh, added the '#!/bin/sh' at the > beginning, > and added to the /etc/init.d/pmie file the following statement (because > IS_ON > is nowhere :-): > > IS_ON=false > > Then I modified the first statement of /etc/init.d/pmie which seems to > have a bug: > I modified: > if [ -fi ${PCP_CONF:-/etc/pcp.conf}.sh ] ; then > to: > if [ -f ${PCP_CONF:-/etc/pcp.conf}.sh ] ; then > > (removed the 'i'). > > It now works really fine (even pmie :-). Sorry about this. We had tested pmie in interactive mode, but the *rc* support for starting pmie instances automagically was clearly broken (you found 3 problems immediately). We had similar issues with the pmcd *rc* script that required a lot of re-writing to migrate from Irix to work in both Irix and Linux. Similar work remains to be done on the pmie *rc* script and we'll fix it before the next release on oss.sgi.com. > There is an important variable which is missing from the standard ones, > (coming > from /proc/sys/fs/file-nr). It is the number of currently open files > (and the > maximum assigned file number, and the maximum allocatable file number). > > Their description is the following: > "The three values in file-nr denote the number of allocated file > handles, the > number of used file handles, and the maximum number of file handles. > When the > allocated file handles come close to the maximum, but the number of > actually > used ones is far behind, you've encountered a peak in your usage of file > handles > and you don't need to increase the maximum. " > (From http://www.bb-zone.com/Proc/chapter2.html#section2.1) > > I think I can patch pcp to add it myself, but I don't know in which > hierarchy I must put it into. In filesys.*, there are already variables > named > filesys.maxfiles, filesys.usedfiles and filesys.freefiles, but they are > totally > unrelated to the /proc/sys/fs/file-nr variables. Good observation. If you have the fix, we'd be glad to incorporate it. As to the name of the metric, filesys.* may not be the right place as these metrics are all enumerated with one value per local mounted filesystem. I'd suggest - filesys.kernel.openfiles and filesys.kernel.maxfiles or - kernel.openfiles and kernel.maxfiles > About practical use of pmie, I'm looking for an example on how to > monitor > the presence of a daemon on a system. For example, I have an apache > daemon, and > I'd like to get an alert if it dies, or if there are too many instances > of > it. > The only near useful variable that I see is proc.psinfo.pid. I used it > like this: > > eurythro:/home/enlight# pmie -v > val = proc.psinfo.pid #'/usr/sbin/apache'; > val: 254 > > val: 254 > > val: 254 > > val: 254 > > val: ? > > val: ? > > val: ? > > val: ? > > > At the fifth iteration, I stopped the daemon. And it stops returning its > pid, which is cool, but it never get a pid again, even when I restart > apache. I don't understand why...Another problem with this, is that I > don't have the number of running daemons :-( > Any ideas ? This is a hard problem to solve with pmie. pmie tries (but does not always succeed) to re-evaluate instance domains that change undernearth it ... I'll need a bit longer to investigate this one, because the behaviour of pmie to find the process at all from the instance "/usr/sbin/apache" is unexpected ... the proper instance name would be "254 /usr/sbin/apache". To count processes by name, this will work better ... count_inst match_inst "/usr/sbin/apache" proc.psinfo.pid > 0; Note the proc.psinfo.pid > 0 is universally true and produces a set result for all processes, to which the (regular expression) instance matching and counting predicates are applied. From owner-pcp@oss.sgi.com Fri Dec 24 12:27:02 1999 Received: by oss.sgi.com id ; Fri, 24 Dec 1999 12:26:52 -0800 Received: from adsl-63-193-244-224.dsl.snfc21.pacbell.net ([63.193.244.224]:10763 "EHLO adsl-63-193-244-224.dsl.snfc21.pacbell.net") by oss.sgi.com with ESMTP id ; Fri, 24 Dec 1999 12:26:35 -0800 Received: from adsl-63-193-244-224.dsl.snfc21.pacbell.net (IDENT:dek@localhost [127.0.0.1]) by adsl-63-193-244-224.dsl.snfc21.pacbell.net (8.9.3/8.8.7) with ESMTP id MAA03224 for ; Fri, 24 Dec 1999 12:26:48 -0800 Message-Id: <199912242026.MAA03224@adsl-63-193-244-224.dsl.snfc21.pacbell.net> To: pcp@oss.sgi.com Subject: pcp code From: dek@cgl.ucsf.edu Reply-to: dek@cgl.ucsf.edu Date: Fri, 24 Dec 1999 12:26:48 -0800 Sender: owner-pcp@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pcp-outgoing Dear PCP project people, I noticed that you working on performance counter code for linux. I have developed a small program called "PerfWatcher" which uses the performance counter package for linux to observe the hardware counters in realtime. I no longer have time to maintain & continue developing the program, if you are interested you can use the code for whatever purpose you like. Here is the address: http://picasso.nmr.ucsf.edu/~dek/PerfWatcher-1.0.tgz Dave ----------------------------------------------------------------------------- Email: dek@cgl.ucsf.edu David Konerding WWW: http://picasso.ucsf.edu/~dek -----------------------------------------------------------------------------