From owner-pcp@oss.sgi.com Thu Aug 2 03:14:40 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f72AEeY24051 for pcp-outgoing; Thu, 2 Aug 2001 03:14:40 -0700 Received: from sgi.com (sgi.SGI.COM [192.48.153.1]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f72AEdV24046 for ; Thu, 2 Aug 2001 03:14:39 -0700 Received: from mailgate3.cinetic.de (mailgate3.cinetic.de [212.227.116.80]) 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 ESMTP id DAA08022 for ; Thu, 2 Aug 2001 03:14:03 -0700 (PDT) mail_from (sarmbruster@web.de) Received: from smtp.web.de (smtp01.web.de [194.45.170.210]) by mailgate3.cinetic.de (8.11.2/8.11.2/SuSE Linux 8.11.0-0.4) with SMTP id f72ADF631382 for ; Thu, 2 Aug 2001 12:13:15 +0200 Received: from smtp by smtp.web.de with smtp (freemail 4.2.2.2 #11) id m15SFTe-007oRMC; Thu, 2 Aug 2001 12:13 +0200 Date: Thu, 02 Aug 2001 12:12 To: "pcp" From: "Stefan Armbruster" Subject: pcp on linux - newbie question Message-ID: MDAyMD.U3RlZmFuIEFybWJydXN0ZX@smtp.web.de X-Mailer: Aethera [0.9.0] 17 Jan 2000 X-Aethera-Generated: header (rfc822), reference id idx000000200 MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Transfer-Encoding: 7bit Sender: owner-pcp@oss.sgi.com Precedence: bulk Hi, I'm trying to establish performance monitoring on some linux (SuSE7.2) boxes. First I had problems starting the pmlogger. After some investigation of the startup-scripts I removed /sbin/chkconfig, now it works. Now my question: I managed to collect and archive my performance data. Is there a way to export the archived data to any ASCII format for further processing with gnuplot, excel or something else. PCPMon is pretty nice for online-monitoring, but raw ASCII values would be more valuable for me. Thanks in advance, Stefan Armbruste From owner-pcp@oss.sgi.com Thu Aug 2 05:03:22 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f72C3MR08709 for pcp-outgoing; Thu, 2 Aug 2001 05:03:22 -0700 Received: from mailgate3.cinetic.de (mailgate3.cinetic.de [212.227.116.80]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f72C3LV08706 for ; Thu, 2 Aug 2001 05:03:21 -0700 Received: from smtp.web.de (smtp01.web.de [194.45.170.210]) by mailgate3.cinetic.de (8.11.2/8.11.2/SuSE Linux 8.11.0-0.4) with SMTP id f72C2Y609691 for ; Thu, 2 Aug 2001 14:02:34 +0200 Received: from smtp by smtp.web.de with smtp (freemail 4.2.2.2 #11) id m15SHBS-007oOiC; Thu, 2 Aug 2001 14:02 +0200 Date: Thu, 02 Aug 2001 14:02 To: "pcp" From: "Stefan Armbruster" Subject: pcp on linux - newbie question Message-ID: MDAyMT.U3RlZmFuIEFybWJydXN0ZX@smtp.web.de X-Mailer: Aethera [0.9.0] 17 Jan 2000 X-Aethera-Generated: header (rfc822), reference id idx000000215 MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Transfer-Encoding: 8bit Sender: owner-pcp@oss.sgi.com Precedence: bulk Hi, I'm trying to establish performance monitoring on some linux (SuSE7.2) boxes. First I had problems starting the pmlogger. After some investigation of the startup-scripts I removed /sbin/chkconfig, now it works. Is this a bug in pcp or a Suse-related problem? Now my question: I managed to collect and archive my performance data. Is there a way to export the archived data to any ASCII format for further processing with gnuplot, excel or something else. PCPMon is pretty nice for online-monitoring, but raw ASCII values would be more valuable for me. Thanks in advance, Stefan Armbruster From owner-pcp@oss.sgi.com Thu Aug 2 05:27:01 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f72CR1m12245 for pcp-outgoing; Thu, 2 Aug 2001 05:27:01 -0700 Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f72CQxV12235 for ; Thu, 2 Aug 2001 05:26:59 -0700 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via SMTP id OAA492401 for ; Thu, 2 Aug 2001 14:26:06 +0200 (CEST) mail_from (markgw@sgi.com) Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.175]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id WAA08910; Thu, 2 Aug 2001 22:24:56 +1000 Date: Thu, 2 Aug 2001 22:24:56 +1000 (EST) From: Mark Goodwin X-Sender: markgw@sherman.melbourne.sgi.com To: Stefan Armbruster cc: pcp Subject: Re: pcp on linux - newbie question In-Reply-To: MDAyMD.U3RlZmFuIEFybWJydXN0ZX@smtp.web.de Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Thu, 2 Aug 2001, Stefan Armbruster wrote: > Now my question: I managed to collect and archive my performance > data. Is there a way to export the archived data to any ASCII format > for further processing with gnuplot, excel or something else. PCPMon > is pretty nice for online-monitoring, but raw ASCII values would be > more valuable for me. Use the -a archive option with pmprobe, pmie, pminfo, pmval or pmstat. See the respective man pages for details. -- Mark From owner-pcp@oss.sgi.com Thu Aug 2 16:55:27 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f72NtRb11787 for pcp-outgoing; Thu, 2 Aug 2001 16:55:27 -0700 Received: from zok.corp.sgi.com (zok.sgi.com [204.94.215.101]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f72NtQV11784 for ; Thu, 2 Aug 2001 16:55:26 -0700 Received: from rattle.melbourne.sgi.com (rattle.melbourne.sgi.com [134.14.55.145]) by zok.corp.sgi.com (8.11.4/8.11.4/linux-outbound_gateway-1.0) with ESMTP id f7300Xm04459 for ; Thu, 2 Aug 2001 17:00:33 -0700 Received: from localhost (kenmcd@localhost) by rattle.melbourne.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id JAA89124; Fri, 3 Aug 2001 09:54:01 +1000 (AEST) X-Authentication-Warning: rattle.melbourne.sgi.com: kenmcd owned process doing -bs Date: Fri, 3 Aug 2001 09:54:01 +1000 From: Ken McDonell Reply-To: To: Stefan Armbruster cc: pcp Subject: Re: pcp on linux - newbie question In-Reply-To: MDAyMD.U3RlZmFuIEFybWJydXN0ZX@smtp.web.de Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Thu, 2 Aug 2001, Stefan Armbruster wrote: > Hi, > > I'm trying to establish performance monitoring on some linux > (SuSE7.2) boxes. First I had problems starting the pmlogger. After some > investigation of the startup-scripts I removed /sbin/chkconfig, now it > works. > > Now my question: I managed to collect and archive my performance > data. Is there a way to export the archived data to any ASCII format > for further processing with gnuplot, excel or something else. PCPMon is > pretty nice for online-monitoring, but raw ASCII values would be more > valuable for me. >From the open source pieces, your best bet is likely to be pmie -v with the output post-processed ... check the mail archives, I posted a recipe and simple script to do this sometime ago. From owner-pcp@oss.sgi.com Fri Aug 3 10:42:56 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f73Hguk16826 for pcp-outgoing; Fri, 3 Aug 2001 10:42:56 -0700 Received: from dfw-gate4.raytheon.com (dfw-gate4.raytheon.com [199.46.199.233]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f73HgsV16820 for ; Fri, 3 Aug 2001 10:42:54 -0700 Received: from ds02c00.directory.ray.com (ds02c00.rsc.raytheon.com [147.25.138.118]) by dfw-gate4.raytheon.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id f73HgrQ08426 for ; Fri, 3 Aug 2001 12:42:53 -0500 (CDT) Received: from rtshou-ds01.hou.us.ray.com (localhost [127.0.0.1]) by ds02c00.directory.ray.com (8.9.3/8.9.3) with ESMTP id MAA20195 for ; Fri, 3 Aug 2001 12:41:28 -0500 (CDT) Subject: Query on cluster measurement To: pcp@oss.sgi.com X-Mailer: Lotus Notes Release 5.0.4 June 8, 2000 Message-ID: From: Mark_H_Johnson@Raytheon.com Date: Fri, 3 Aug 2001 12:43:15 -0500 X-MIMETrack: Serialize by Router on RTSHOU-DS01/RTS/Raytheon/US(Release 5.0.6a |January 17, 2001) at 08/03/2001 12:43:22 PM MIME-Version: 1.0 Content-type: text/plain; charset=us-ascii Sender: owner-pcp@oss.sgi.com Precedence: bulk We are looking at using PCP for measuring information on our cluster of PC's and have a few questions... To set the stage, our network looks something like... Workstation(s) | | | | ---------+---------- | Head Node | Switch (private LAN) | ---------+---------- | | | | Compute Nodes | | | | Other Equipment The head node is NOT a router - workstations can't see the compute nodes (nor the other equipment) with TCP/IP. We would prefer to run the monitoring tools on one or more workstations. We would prefer to run the agents on both the compute nodes and head node. We would prefer to collect the data at the head node for distribution to the workstations. [I think I got the terminology right...] All the machines are running Linux, and we have PCP 2.2.1 downloaded and installed on all of the machines that will be doing this. (1) In a few places, the documentation says that the collector works with local agents. But in the man page for pmcd(1), it indicates that socket connections are supported. Is there some way we can gather key data items from the compute node, send them to the head node [socket connection?] & include them in the head node's name space? If not, do you have suggestions for implementing such a capability? (2) In lieu of an elegant solution to (1) - could we use remote shell to the compute nodes, use pminfo to the dump data & import w/ the ASCII interface to pmcd? (3) We want to measure data transfer rates to the other equipment. We were looking at getting data out of /proc, but we have function interfaces available as well. Should we just filter the /proc output similar to that done by the Linux agent or use code instead? (4) Was there additional work done in ACE (Advanced Cluster Environment) that may have implemented this already? If so, who should we contact at SGI for more information? Thanks. --Mark H Johnson From owner-pcp@oss.sgi.com Tue Aug 14 03:47:24 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7EAlOf05633 for pcp-outgoing; Tue, 14 Aug 2001 03:47:24 -0700 Received: from rj.sgi.com (rj.sgi.com [204.94.215.100]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7EAlMj05630 for ; Tue, 14 Aug 2001 03:47:22 -0700 Received: from nodin.corp.sgi.com (nodin.corp.sgi.com [192.26.51.193]) by rj.sgi.com (8.11.4/8.11.4/linux-outbound_gateway-1.0) with ESMTP id f7EAq5S29489 for ; Tue, 14 Aug 2001 03:52:05 -0700 Received: from yog-sothoth.sgi.com (eugate.sgi.com [144.253.131.5]) by nodin.corp.sgi.com (8.11.4/8.11.2/nodin-1.0) with ESMTP id f7EAkEF34644127 for ; Tue, 14 Aug 2001 03:46:14 -0700 (PDT) Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via SMTP id MAA09586 for ; Tue, 14 Aug 2001 12:43:37 +0200 (CEST) mail_from (kenmcd@melbourne.sgi.com) From: kenmcd@melbourne.sgi.com Received: from kenj-ppp.melbourne.sgi.com (kenj-ppp.melbourne.sgi.com [134.14.55.215]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id UAA16357; Tue, 14 Aug 2001 20:41:56 +1000 Date: Tue, 14 Aug 2001 20:42:55 +1000 (EST) Reply-To: kenmcd@melbourne.sgi.com To: Madhu Sudhan R Tera cc: pcp@oss.sgi.com Subject: Re: getting the currently running process metrics In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Thu, 19 Jul 2001, Madhu Sudhan R Tera wrote: > Hello ! > There is no metric that gives the per process percentage of CPU cycles > used.The "pminfo -dT proc.*" command (ofcourse not *).But i didnt find > the one i required.Can anyone suggest me a solution. Looks like I did not answer this one. The Linux metrics you need are: proc.psinfo.utime and proc.psinfo.stime both are counters in units of msec ... PCP tools like pmval, pmie or pcpmon will rate convert this to produce a number in the range 0 to 1 representing the CPU utilization for user time and system time for each (or selected) processes. From owner-pcp@oss.sgi.com Wed Aug 15 09:02:48 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7FG2mk28834 for pcp-outgoing; Wed, 15 Aug 2001 09:02:48 -0700 Received: from boondoggle.office.aol.com (pix-fw.wan.aol.com [152.163.190.1]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7FG2kj28821 for ; Wed, 15 Aug 2001 09:02:46 -0700 Received: (from harvell@localhost) by boondoggle.office.aol.com (8.11.4/8.11.4) id f7FG2eY13010; Wed, 15 Aug 2001 12:02:40 -0400 (EDT) Date: Wed, 15 Aug 2001 12:02:40 -0400 (EDT) From: Brian Harvell X-X-Sender: To: Subject: --prefix does not work. Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk Looks like the package isn't very friendly for locating it outside of the default installation directories. Passing in --prefix to configure doesn't do anything. Brian -- Brian Harvell harvell@aol.net http://ToolBoy.com/ echo '[q]sa[ln0=aln256%Pln256/snlbx]sb3135071790101768542287578439snlbxq'|dc From owner-pcp@oss.sgi.com Wed Aug 15 15:06:06 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7FM66015289 for pcp-outgoing; Wed, 15 Aug 2001 15:06:06 -0700 Received: from rj.sgi.com (rj.sgi.com [204.94.215.100]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7FM65j15286 for ; Wed, 15 Aug 2001 15:06:05 -0700 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by rj.sgi.com (8.11.4/8.11.4/linux-outbound_gateway-1.0) with SMTP id f7FMAoS26864 for ; Wed, 15 Aug 2001 15:10:50 -0700 Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.175]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id IAA25247; Thu, 16 Aug 2001 08:04:37 +1000 Date: Thu, 16 Aug 2001 08:04:37 +1000 (EST) From: Mark Goodwin X-Sender: markgw@sherman.melbourne.sgi.com To: Brian Harvell cc: pcp@oss.sgi.com Subject: Re: --prefix does not work. In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Wed, 15 Aug 2001, Brian Harvell wrote: > > Looks like the package isn't very friendly for locating it outside of the > default installation directories. Passing in --prefix to configure doesn't > do anything. > PCP is not a relocatable RPM because relocatable RPMs seem to need a single point of install (so you can move, for e.g. /usr to /usr/local or something). Since PCP has multiple install dirs, it can't be relocated in this way - it's not the only RPM package with this problem. Instead, you need to edit src/include/pcp.conf.in and change whatever directories you need relocated, and then build your own RPM. -- Mark From owner-pcp@oss.sgi.com Wed Aug 15 15:40:54 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7FMesK21497 for pcp-outgoing; Wed, 15 Aug 2001 15:40:54 -0700 Received: from boondoggle.office.aol.com (pix-fw.wan.aol.com [152.163.190.1]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7FMeqj21492 for ; Wed, 15 Aug 2001 15:40:53 -0700 Received: (from harvell@localhost) by boondoggle.office.aol.com (8.11.4/8.11.4) id f7FMeOD14994; Wed, 15 Aug 2001 18:40:24 -0400 (EDT) Date: Wed, 15 Aug 2001 18:40:24 -0400 (EDT) From: Brian Harvell X-X-Sender: To: Mark Goodwin cc: Subject: Re: --prefix does not work. In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Thu, 16 Aug 2001, Mark Goodwin wrote: > On Wed, 15 Aug 2001, Brian Harvell wrote: > > > > > Looks like the package isn't very friendly for locating it outside of the > > default installation directories. Passing in --prefix to configure doesn't > > do anything. > > > > PCP is not a relocatable RPM because relocatable RPMs seem to need a > single point of install (so you can move, for e.g. /usr to /usr/local > or something). Since PCP has multiple install dirs, it can't be > relocated in this way - it's not the only RPM package with this problem. I'm talking about when I compile it myself and specify --prefix to configure like any other gnu/autoconf'd package. tia[root][18:37][/opt/src/pcp-2.2.1]# ./configure --help | grep prefix --prefix=PREFIX install architecture-independent files in PREFIX --exec-prefix=EPREFIX install architecture-dependent files in EPREFIX [same as prefix] --program-prefix=PREFIX prepend PREFIX to installed program names > Instead, you need to edit src/include/pcp.conf.in and change whatever > directories you need relocated, and then build your own RPM. yes I did hack up the pcp.conf.in and it ended up giving me an error during the install because it couldn't copy somewhere. I didn't have time to dig into the problem. Brian -- Brian Harvell harvell@aol.net http://ToolBoy.com/ echo '[q]sa[ln0=aln256%Pln256/snlbx]sb3135071790101768542287578439snlbxq'|dc From owner-pcp@oss.sgi.com Fri Aug 17 14:58:18 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7HLwID14130 for pcp-outgoing; Fri, 17 Aug 2001 14:58:18 -0700 Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7HLwGj14127 for ; Fri, 17 Aug 2001 14:58:16 -0700 Received: from rattle.melbourne.sgi.com (rattle.melbourne.sgi.com [134.14.55.145]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id OAA00560 for ; Fri, 17 Aug 2001 14:56:20 -0700 (PDT) mail_from (kenmcd@melbourne.sgi.com) From: kenmcd@melbourne.sgi.com Received: from localhost (kenmcd@localhost) by rattle.melbourne.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id HAA12899; Sat, 18 Aug 2001 07:56:56 +1000 (AEST) X-Authentication-Warning: rattle.melbourne.sgi.com: kenmcd owned process doing -bs Date: Sat, 18 Aug 2001 07:56:56 +1000 Reply-To: To: Brian Harvell cc: Subject: Re: --prefix does not work. In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk In addition to what Mark said ... PCP needs to be started as part of the system reboot, so this means stuffing things in /etc for most Linuxes with no alternatives supported, so this cannot be relocatable, and hence the whole RPM is not relocatable. I don't think config --prefix=... is ever going to work for the PCP RPM ... use pcp.conf.in to selectively move _parts_ of the PCP installation before you build. From owner-pcp@oss.sgi.com Fri Aug 17 20:58:24 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7I3wOO21464 for pcp-outgoing; Fri, 17 Aug 2001 20:58:24 -0700 Received: from boondoggle.office.aol.com (x98A3A234.pix.aol.com [152.163.162.52]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7I3wLj21460 for ; Fri, 17 Aug 2001 20:58:22 -0700 Received: (from harvell@localhost) by boondoggle.office.aol.com (8.11.4/8.11.4) id f7I3wE429212; Fri, 17 Aug 2001 23:58:14 -0400 (EDT) Date: Fri, 17 Aug 2001 23:58:14 -0400 (EDT) From: Brian Harvell X-X-Sender: To: Ken McDonell cc: Subject: Re: --prefix does not work. In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Sat, 18 Aug 2001, Ken McDonell wrote: > In addition to what Mark said ... PCP needs to be started as part of the > system reboot, so this means stuffing things in /etc for most Linuxes > with no alternatives supported, so this cannot be relocatable, and hence > the whole RPM is not relocatable. > > I don't think config --prefix=... is ever going to work for the PCP RPM > ... use pcp.conf.in to selectively move _parts_ of the PCP installation > before you build. > Again I wasn't trying to install an rpm I was trying to build from scratch by running configure. Brian -- Brian Harvell harvell@aol.net http://ToolBoy.com/ echo '[q]sa[ln0=aln256%Pln256/snlbx]sb3135071790101768542287578439snlbxq'|dc From owner-pcp@oss.sgi.com Sun Aug 19 03:09:53 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7JA9rT16414 for pcp-outgoing; Sun, 19 Aug 2001 03:09:53 -0700 Received: from mailgw3.netvision.net.il (mailgw.netvision.net.il [194.90.1.11]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7JA9lj16410 for ; Sun, 19 Aug 2001 03:09:47 -0700 Received: from mail.exanet.co.il ([212.143.73.99]) by mailgw3.netvision.net.il (8.9.3/8.9.3) with ESMTP id NAA00788 for ; Sun, 19 Aug 2001 13:07:47 +0300 (IDT) Received: from cat (fw1.exanet.co.il [212.143.73.98]) by mail.exanet.co.il (8.11.0/8.11.0) with SMTP id f7JA7nb22744 for ; Sun, 19 Aug 2001 13:07:49 +0300 Reply-To: From: "gilly" To: Subject: 2 pmcd's on one machine Date: Sun, 19 Aug 2001 13:08:20 +0200 Message-ID: <00e201c1289f$423894c0$2a04000a@exanet> MIME-Version: 1.0 Content-Type: text/plain; charset="windows-1255" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2910.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Sender: owner-pcp@oss.sgi.com Precedence: bulk Is it possible to have 2 pmcd's running on 2 different ports, on one machine? If so - can one pmcd fetch metrics from the other pmcd? How can I control the 2 PMCD_PORT variables? thanks in advance, gilly From owner-pcp@oss.sgi.com Sun Aug 19 17:14:36 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7K0EaE01572 for pcp-outgoing; Sun, 19 Aug 2001 17:14:36 -0700 Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7K0EYj01569 for ; Sun, 19 Aug 2001 17:14:34 -0700 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 RAA15531 for ; Sun, 19 Aug 2001 17:14:16 -0700 (PDT) mail_from (markgw@sgi.com) Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.175]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id KAA16178; Mon, 20 Aug 2001 10:13:10 +1000 Date: Mon, 20 Aug 2001 10:13:10 +1000 (EST) From: Mark Goodwin X-Sender: markgw@sherman.melbourne.sgi.com To: Brian Harvell cc: Ken McDonell , pcp@oss.sgi.com Subject: Re: --prefix does not work. In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Fri, 17 Aug 2001, Brian Harvell wrote: > On Sat, 18 Aug 2001, Ken McDonell wrote: > > > In addition to what Mark said ... PCP needs to be started as part of the > > system reboot, so this means stuffing things in /etc for most Linuxes > > with no alternatives supported, so this cannot be relocatable, and hence > > the whole RPM is not relocatable. > > > > I don't think config --prefix=... is ever going to work for the PCP RPM > > ... use pcp.conf.in to selectively move _parts_ of the PCP installation > > before you build. > > > > Again I wasn't trying to install an rpm I was trying to build from > scratch by running configure. > Perhaps we weren't clear enough: PCP does not use --prefix in the configure part of the build because PCP requires multiple points of relocation but --prefix only provides one, rendering --prefix useless. The correct way to relocate PCP directories is to change src/include/pcp.conf.in. If you still disagree, then you must know something about --prefix that we are missing, so fix it and send me a patch! thanks -- Mark From owner-pcp@oss.sgi.com Tue Aug 21 00:53:23 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7L7rNk12834 for pcp-outgoing; Tue, 21 Aug 2001 00:53:23 -0700 Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7L7rL912831 for ; Tue, 21 Aug 2001 00:53:21 -0700 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via SMTP id AAA05829 for ; Tue, 21 Aug 2001 00:51:28 -0700 (PDT) mail_from (makc@kuku.melbourne.sgi.com) Received: from kuku.melbourne.sgi.com (kuku.melbourne.sgi.com [134.14.55.163]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id RAA25167; Tue, 21 Aug 2001 17:52:03 +1000 Received: (from makc@localhost) by kuku.melbourne.sgi.com (SGI-8.9.3/8.9.3) id RAA83094; Tue, 21 Aug 2001 17:52:01 +1000 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <15234.4897.705297.231219@kuku.melbourne.sgi.com> Date: Tue, 21 Aug 2001 17:52:01 +1000 (EST) From: Max Matveev To: Cc: Subject: Re: 2 pmcd's on one machine In-Reply-To: <00e201c1289f$423894c0$2a04000a@exanet> References: <00e201c1289f$423894c0$2a04000a@exanet> X-Mailer: VM 6.72 under 21.4 (patch 3) "Academic Rigor" XEmacs Lucid Sender: owner-pcp@oss.sgi.com Precedence: bulk >>>>> "g" == gilly writes: g> Is it possible to have 2 pmcd's running on 2 different ports, on one g> machine? Yes. g> If so - can one pmcd fetch metrics from the other pmcd? No - pmcd can only get metrics from PMDAs, not from other pmcd. g> How can I control the 2 PMCD_PORT variables? Start first one as usual. Start second on with smth like PMCD_PORT=4322 /usr/etc/pmcd & max From owner-pcp@oss.sgi.com Tue Aug 21 14:33:13 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7LLXDD31074 for pcp-outgoing; Tue, 21 Aug 2001 14:33:13 -0700 Received: from rj.sgi.com (rj.sgi.com [204.94.215.100]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7LLXC931071 for ; Tue, 21 Aug 2001 14:33:12 -0700 Received: from yog-sothoth.sgi.com (eugate.neu.sgi.com [144.253.131.5]) by rj.sgi.com (8.11.4/8.11.4/linux-outbound_gateway-1.0) with ESMTP id f7LLcJ807223 for <@rj.corp.sgi.com:pcp@oss.sgi.com>; Tue, 21 Aug 2001 14:38:19 -0700 Received: from rattle.melbourne.sgi.com (rattle.melbourne.sgi.com [134.14.55.145]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id XAA313833 for ; Tue, 21 Aug 2001 23:33:03 +0200 (CEST) mail_from (kenmcd@melbourne.sgi.com) Received: from localhost (kenmcd@localhost) by rattle.melbourne.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id HAA91902; Wed, 22 Aug 2001 07:31:43 +1000 (AEST) X-Authentication-Warning: rattle.melbourne.sgi.com: kenmcd owned process doing -bs Date: Wed, 22 Aug 2001 07:31:43 +1000 From: Ken McDonell Reply-To: To: Max Matveev cc: , Subject: Re: 2 pmcd's on one machine In-Reply-To: <15234.4897.705297.231219@kuku.melbourne.sgi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Tue, 21 Aug 2001, Max Matveev wrote: > >>>>> "g" == gilly writes: > > g> Is it possible to have 2 pmcd's running on 2 different ports, on one > g> machine? > Yes. > > g> If so - can one pmcd fetch metrics from the other pmcd? > No - pmcd can only get metrics from PMDAs, not from other pmcd. > > g> How can I control the 2 PMCD_PORT variables? > Start first one as usual. Start second on with smth like > > PMCD_PORT=4322 /usr/etc/pmcd & You'll probably also need/want to set $PCP_PMCDCONF_PATH so the two pmcds have different pmcd configuration files. From owner-pcp@oss.sgi.com Wed Aug 22 07:45:09 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7MEj9n30681 for pcp-outgoing; Wed, 22 Aug 2001 07:45:09 -0700 Received: from mail.teraport.de (mail.teraport.de [195.143.8.72]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7MEj7930678 for ; Wed, 22 Aug 2001 07:45:07 -0700 Received: from TeraPort.de ([10.10.12.32]) by mail.teraport.de (Lotus Domino Release 5.0.7) with ESMTP id 2001082216450128:5346 ; Wed, 22 Aug 2001 16:45:01 +0200 Message-ID: <3B83C567.564F8053@TeraPort.de> Date: Wed, 22 Aug 2001 16:44:55 +0200 From: Martin Knoblauch Organization: TeraPort GmbH X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.4.8-ac8 i686) X-Accept-Language: en, de MIME-Version: 1.0 To: pcp@oss.sgi.com Subject: GEtting number and names of network interfaces (Linux) X-MIMETrack: Itemize by SMTP Server on lotus/Teraport/de(Release 5.0.7 |March 21, 2001) at 08/22/2001 04:45:01 PM, Serialize by Router on lotus/Teraport/de(Release 5.0.7 |March 21, 2001) at 08/22/2001 04:45:07 PM, Serialize complete at 08/22/2001 04:45:07 PM Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Sender: owner-pcp@oss.sgi.com Precedence: bulk Hi, I am currently evaluating performance monitoring/measuring tools for a mostly Linux based cluster environment. Due to some history, PCP came to my mind :-) I installed the binary package and it seems to do a lot of the things we need. One thing I am missing is the number and the names of the network interfaces. A quick look at the sources seems to indicate that the information itself is available, but not exported. Am I correct with this? If I wanted to add the fileds to the network metrics, what cautions dio I need to observe. I don't want to break things ... Thanks Martin -- ------------------------------------------------------------------ Martin Knoblauch | email: Martin.Knoblauch@TeraPort.de TeraPort GmbH | Phone: +49-89-510857-309 C+ITS | Fax: +49-89-510857-111 http://www.teraport.de | Mobile: +49-170-4904759 From owner-pcp@oss.sgi.com Wed Aug 22 08:26:38 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7MFQcc31521 for pcp-outgoing; Wed, 22 Aug 2001 08:26:38 -0700 Received: from mail.teraport.de (mail.teraport.de [195.143.8.72]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7MFQa931518 for ; Wed, 22 Aug 2001 08:26:36 -0700 Received: from TeraPort.de ([10.10.12.32]) by mail.teraport.de (Lotus Domino Release 5.0.7) with ESMTP id 2001082217262886:5376 ; Wed, 22 Aug 2001 17:26:28 +0200 Message-ID: <3B83CF1E.13FD12F6@TeraPort.de> Date: Wed, 22 Aug 2001 17:26:22 +0200 From: Martin Knoblauch Organization: TeraPort GmbH X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.4.8-ac8 i686) X-Accept-Language: en, de MIME-Version: 1.0 To: pcp@oss.sgi.com Subject: Re: GEtting number and names of network interfaces (Linux) References: <3B83C567.564F8053@TeraPort.de> X-MIMETrack: Itemize by SMTP Server on lotus/Teraport/de(Release 5.0.7 |March 21, 2001) at 08/22/2001 05:26:28 PM, Serialize by Router on lotus/Teraport/de(Release 5.0.7 |March 21, 2001) at 08/22/2001 05:26:36 PM, Serialize complete at 08/22/2001 05:26:36 PM Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Sender: owner-pcp@oss.sgi.com Precedence: bulk Martin Knoblauch wrote: > > Hi, > > I am currently evaluating performance monitoring/measuring tools for a > mostly Linux based cluster environment. Due to some history, PCP came to > my mind :-) > > I installed the binary package and it seems to do a lot of the things > we need. One thing I am missing is the number and the names of the > network interfaces. > OK, just had a look at the source to pmprobe and it seems the info is available to clients. Sorry for the stupid question :-) Martin -- ------------------------------------------------------------------ Martin Knoblauch | email: Martin.Knoblauch@TeraPort.de TeraPort GmbH | Phone: +49-89-510857-309 C+ITS | Fax: +49-89-510857-111 http://www.teraport.de | Mobile: +49-170-4904759 From owner-pcp@oss.sgi.com Wed Aug 22 10:53:47 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7MHrlf01641 for pcp-outgoing; Wed, 22 Aug 2001 10:53:47 -0700 Received: from rigel.cis.ksu.edu (rigel.cis.ksu.edu [129.130.10.65]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7MHre901638 for ; Wed, 22 Aug 2001 10:53:40 -0700 Received: from pollux.cis.ksu.edu (mst9696@pollux.cis.ksu.edu [129.130.10.33]) by rigel.cis.ksu.edu (8.9.1/8.9.1/000517) with ESMTP id MAA11573 for ; Wed, 22 Aug 2001 12:53:30 -0500 (CDT) Received: from localhost (mst9696@localhost) by pollux.cis.ksu.edu (8.9.1/8.9.1/000517) with ESMTP id MAA08537 for ; Wed, 22 Aug 2001 12:53:30 -0500 (CDT) X-Authentication-Warning: pollux.cis.ksu.edu: mst9696 owned process doing -bs Date: Wed, 22 Aug 2001 12:53:30 -0500 (CDT) From: Madhu Sudhan R Tera To: pcp@oss.sgi.com Subject: Process network bandwidth usage info Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk Hello all! I am currently invovled in load balancing in a beowlf cluster. we got stuck at one point where we must know the amount of network bandwidth( or number of bytes transferred ) by a process(or to a process). Network bandwidth usage is one of the parameter that we use in our load balancing algorithms. Can anyone throw more light on how we can get info for per-process network usage( with socket number) using PCP Thank you, With regards, MadhuSudhan Reddy Tera -------------------------------------------------- |Department of Computing and Information Sciences, | |Kansas State University, | |Manhattan, KS 66502. | |ph no:785-537-3276 | -------------------------------------------------- From owner-pcp@oss.sgi.com Wed Aug 22 17:11:53 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7N0Brh09543 for pcp-outgoing; Wed, 22 Aug 2001 17:11:53 -0700 Received: from rj.sgi.com (rj.sgi.com [204.94.215.100]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7N0Bpd09540 for ; Wed, 22 Aug 2001 17:11:51 -0700 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by rj.sgi.com (8.11.4/8.11.4/linux-outbound_gateway-1.0) with SMTP id f7N0H1823766 for ; Wed, 22 Aug 2001 17:17:01 -0700 Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.175]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id KAA06226; Thu, 23 Aug 2001 10:10:22 +1000 Date: Thu, 23 Aug 2001 10:10:22 +1000 (EST) From: Mark Goodwin X-Sender: markgw@sherman.melbourne.sgi.com To: Madhu Sudhan R Tera cc: pcp@oss.sgi.com Subject: Re: Process network bandwidth usage info In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Wed, 22 Aug 2001, Madhu Sudhan R Tera wrote: > Hello all! > I am currently invovled in load balancing in a beowlf cluster. > > we got stuck at one point where we must know the amount of > network bandwidth( or number of bytes transferred ) by a process(or > to a process). Network bandwidth usage is one of the parameter that > we use in our load balancing algorithms. AFAIK the linux kernel does not export (nor even does it keep counters for) per-process network/socket stats. > > Can anyone throw more light on how we can get info for per-process network > usage( with socket number) using PCP You'll have to patch the kernel, then add something to the linux PMDA. This will be non-trivial - you'll have to implement the counters on a per socket basis, export the per-socket counters to /proc/net/{tcp,udp,raw} and then in the linux PMDA, correlate this with the sockets open for each process as reported in /proc//fd. Of course there may be other ways to achieve what you need ... -- Mark From owner-pcp@oss.sgi.com Wed Aug 22 18:06:47 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7N16lu10410 for pcp-outgoing; Wed, 22 Aug 2001 18:06:47 -0700 Received: from smile.idiom.com (inreach-gw1.idiom.com [209.209.13.26]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7N16jd10407 for ; Wed, 22 Aug 2001 18:06:45 -0700 Received: from mdaxke (adsl-63-196-0-240.dsl.snfc21.pacbell.net [63.196.0.240]) by smile.idiom.com (8.9.1/8.8.5) with SMTP id SAA28137 for ; Wed, 22 Aug 2001 18:06:44 -0700 (PDT) Message-ID: <014401c12b6f$d7ed1e50$6c456420@mdaxke> From: "Mark D. Anderson" To: References: Subject: Re: Process network bandwidth usage info Date: Wed, 22 Aug 2001 18:05:59 -0700 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Sender: owner-pcp@oss.sgi.com Precedence: bulk if you can make do with per user and/or per port number accounting, look at http://www.comlink.apc.org/~moritz/ipac.html http://ramses.smeyers.be/homepage/useripacct/ it might also be possible to play games with ipchains to make ipac do what you want, but that depends on your app. -mda From owner-pcp@oss.sgi.com Wed Aug 22 23:22:44 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7N6MiW16020 for pcp-outgoing; Wed, 22 Aug 2001 23:22:44 -0700 Received: from mailgw2.netvision.net.il (mailgw.netvision.net.il [194.90.1.9]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7N6Mfd16017 for ; Wed, 22 Aug 2001 23:22:41 -0700 Received: from mail.exanet.co.il ([212.143.73.99]) by mailgw2.netvision.net.il (8.9.3/8.9.3) with ESMTP id JAA21818 for ; Thu, 23 Aug 2001 09:25:15 +0300 (IDT) Received: from cat (fw1.exanet.co.il [212.143.73.98]) by mail.exanet.co.il (8.11.0/8.11.0) with SMTP id f7N6KCb30217 for ; Thu, 23 Aug 2001 09:20:12 +0300 Reply-To: From: "gilly" To: Subject: RE: 2 pmcd's on one machine Date: Thu, 23 Aug 2001 09:21:13 +0200 Message-ID: <016501c12ba4$3195db20$2a04000a@exanet> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2910.0) In-Reply-To: <15234.8929.652576.763283@kuku.melbourne.sgi.com> Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Sender: owner-pcp@oss.sgi.com Precedence: bulk Now I have the 2 PMCDs running on one machine, but I dont know how to tell the PMDA to connect to PMCD2 (on the other port - 4322). thanks, gilly -----Original Message----- From: Max Matveev [mailto:makc@sgi.com] Sent: Tuesday, August 21, 2001 10:59 AM To: gilly@exanet.com Subject: RE: 2 pmcd's on one machine >>>>> "g" == gilly writes: g> of course metrics can be fetched from PMDAs only, but it passes g> through the PMCD. The flow control as I see it is: g> monitor <--> PMCD1 <--> total-PMDA1 *<--> PMCD2 <--> PMDA2 g> (* in the phase of calling PMCD2, the total-PMDA1 might also call g> other PMCDs - on remote hosts). It could, there is summary pmda on Irix which does smth like this. But you have to watch for time delays in this case. max From owner-pcp@oss.sgi.com Thu Aug 23 19:57:45 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7O2vjT08522 for pcp-outgoing; Thu, 23 Aug 2001 19:57:45 -0700 Received: from boondoggle.office.aol.com (x98A3A2D1.pix.aol.com [152.163.162.209]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7O2vXd08514 for ; Thu, 23 Aug 2001 19:57:33 -0700 Received: (from harvell@localhost) by boondoggle.office.aol.com (8.11.4/8.11.4) id f7O2vKO13058; Thu, 23 Aug 2001 22:57:20 -0400 (EDT) Date: Thu, 23 Aug 2001 22:57:20 -0400 (EDT) From: Brian Harvell X-X-Sender: To: Mark Goodwin cc: Ken McDonell , Subject: Re: --prefix does not work. In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="-559023410-684387517-998621840=:12692" Sender: owner-pcp@oss.sgi.com Precedence: bulk This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. Send mail to mime@docserver.cac.washington.edu for more info. ---559023410-684387517-998621840=:12692 Content-Type: TEXT/PLAIN; charset=US-ASCII On Mon, 20 Aug 2001, Mark Goodwin wrote: > > Perhaps we weren't clear enough: PCP does not use --prefix in the > configure part of the build because PCP requires multiple points of > relocation but --prefix only provides one, rendering --prefix useless. > > The correct way to relocate PCP directories is to change > src/include/pcp.conf.in. If you still disagree, then you must know > something about --prefix that we are missing, so fix it and send me > a patch! > If you just use prefix it uses the defaults and you can't split it. However, you can have a more fine grained control by specifying libdir localstatedir etc. I've attached a patch that will allow you to do this. It's a little bit of a kludge but it works. If you don't specify anything it works as before. This patch also fixes a few typos that prevent even modifying pcp.conf.in from working correctly (hard coded paths and pulling the wrong config var) Personally I think the old directory structure could just go away. I didn't like it in the Irix day and it's even worse in a gnu environment. On another note anyone know if there are any hardware monitoring pmdas? I'm looking to be able to detect when things like disk, memory, cpus etc go bad. Brian -- Brian Harvell harvell@aol.net http://ToolBoy.com/ echo '[q]sa[ln0=aln256%Pln256/snlbx]sb3135071790101768542287578439snlbxq'|dc ---559023410-684387517-998621840=:12692 Content-Type: TEXT/PLAIN; charset=US-ASCII; name="pcp-2.2.1.patch" Content-Transfer-Encoding: BASE64 Content-ID: Content-Description: Content-Disposition: attachment; filename="pcp-2.2.1.patch" ZGlmZiAtdU5yIHBjcC0yLjIuMS5vcmlnL2NvbmZpZ3VyZS5pbiBwY3AtMi4y LjEvY29uZmlndXJlLmluDQotLS0gcGNwLTIuMi4xLm9yaWcvY29uZmlndXJl LmluCVdlZCBKdW4gMjAgMjI6Mzc6MDkgMjAwMQ0KKysrIHBjcC0yLjIuMS9j b25maWd1cmUuaW4JVGh1IEF1ZyAyMyAxNToyMDozMiAyMDAxDQpAQCAtNjc4 LDQzICs2NzgsNTkgQEANCiBwY3BfbWFnaWNfZmlsZT0kUENQX01BR0lDX0ZJ TEUNCiBBQ19TVUJTVChwY3BfbWFnaWNfZmlsZSkNCiANCit0ZXN0ICJ4JGV4 ZWNfcHJlZml4IiA9IHhOT05FICYmIGV4ZWNfcHJlZml4PSR7cHJlZml4fQ0K Kw0KIGRubCBzaGFyZWQgUENQIGZpbGVzIChzaGFyZWFibGUgZm9yIGRpc2ts ZXNzKQ0KLWlmIHRlc3QgJHRhcmdldF9vcyA9IGlyaXgNCitpZiBldmFsIHRl c3QgJHtkYXRhZGlyJS8qfSA9IE5PTkUNCiB0aGVuDQotICAgIHBjcF9zaGFy ZV9kaXI9L3Vzci9wY3ANCisgICBpZiB0ZXN0ICR0YXJnZXRfb3MgPSBpcml4 DQorICAgdGhlbg0KKyAgICAgICBwY3Bfc2hhcmVfZGlyPS91c3IvcGNwDQor ICAgZWxzZQ0KKyAgICAgICBpZiB0ZXN0IC1kIC91c3Ivc2hhcmUNCisgICAg ICAgdGhlbg0KKwkgICBwY3Bfc2hhcmVfZGlyPS91c3Ivc2hhcmUvcGNwDQor ICAgICAgIGVsaWYgdGVzdCAtZCAvdXNyL2xvY2FsDQorICAgICAgIHRoZW4N CisJICAgcGNwX3NoYXJlX2Rpcj0vdXNyL2xvY2FsL3BjcA0KKyAgICAgICBl bHNlDQorCSAgIHBjcF9zaGFyZV9kaXI9L3Vzci9wY3ANCisgICAgICAgZmkN CisgICBmaQ0KIGVsc2UNCi0gICAgaWYgdGVzdCAtZCAvdXNyL3NoYXJlDQot ICAgIHRoZW4NCi0JcGNwX3NoYXJlX2Rpcj0vdXNyL3NoYXJlL3BjcA0KLSAg ICBlbGlmIHRlc3QgLWQgL3Vzci9sb2NhbA0KLSAgICB0aGVuDQotCXBjcF9z aGFyZV9kaXI9L3Vzci9sb2NhbC9wY3ANCi0gICAgZWxzZQ0KLQlwY3Bfc2hh cmVfZGlyPS91c3IvcGNwDQotICAgIGZpDQorICAgcGNwX3NoYXJlX2Rpcj1g ZXZhbCBlY2hvICRkYXRhZGlyYA0KIGZpDQogQUNfU1VCU1QocGNwX3NoYXJl X2RpcikNCiANCiBkbmwgcHJpdmF0ZSBQQ1AgZXhlY3V0YWJsZXMNCi1pZiB0 ZXN0ICR0YXJnZXRfb3MgPSBsaW51eA0KLXRoZW4NCi0gICAgcGNwX2JpbmFk bV9kaXI9JHBjcF9zaGFyZV9kaXIvYmluDQotZWxpZiB0ZXN0ICR0YXJnZXRf b3MgPSBpcml4DQoraWYgZXZhbCB0ZXN0ICR7c2JpbmRpciUvKn0gPSBOT05F DQogdGhlbg0KLSAgICBwY3BfYmluYWRtX2Rpcj0vdXNyL3BjcC9iaW4NCisg ICBpZiB0ZXN0ICR0YXJnZXRfb3MgPSBsaW51eA0KKyAgIHRoZW4NCisgICAg ICAgcGNwX2JpbmFkbV9kaXI9JHBjcF9zaGFyZV9kaXIvYmluDQorICAgZWxp ZiB0ZXN0ICR0YXJnZXRfb3MgPSBpcml4DQorICAgdGhlbg0KKyAgICAgICBw Y3BfYmluYWRtX2Rpcj0vdXNyL3BjcC9iaW4NCisgICBmaQ0KK2Vsc2UNCisg ICBwY3BfYmluYWRtX2Rpcj1gZXZhbCBlY2hvICRzYmluZGlyYA0KIGZpDQot DQogQUNfU1VCU1QocGNwX2JpbmFkbV9kaXIpDQogDQogZG5sIG5vbi1zaGFy ZWQgKGkuZS4gc3lzdGVtIGxvY2FsKSBQQ1AgZmlsZXMNCi1pZiB0ZXN0IC1k IC92YXINCitpZiBldmFsIHRlc3QgJHtsb2NhbHN0YXRlZGlyJS8qfSA9IE5P TkUNCiB0aGVuDQotICAgIHBjcF92YXJfZGlyPS92YXIvcGNwDQotZWxpZiB0 ZXN0IC1kIC91c3IvbG9jYWwNCi10aGVuDQotICAgIHBjcF92YXJfZGlyPS91 c3IvbG9jYWwvcGNwDQorICAgaWYgdGVzdCAtZCAvdmFyDQorICAgdGhlbg0K KyAgICAgICBwY3BfdmFyX2Rpcj0vdmFyL3BjcA0KKyAgIGVsaWYgdGVzdCAt ZCAvdXNyL2xvY2FsDQorICAgdGhlbg0KKyAgICAgICBwY3BfdmFyX2Rpcj0v dXNyL2xvY2FsL3BjcA0KKyAgIGVsc2UNCisgICAgICAgcGNwX3Zhcl9kaXI9 L3Vzcg0KKyAgIGZpDQogZWxzZQ0KLSAgICBwY3BfdmFyX2Rpcj0vdXNyDQor ICAgcGNwX3Zhcl9kaXI9YGV2YWwgZWNobyAkbG9jYWxzdGF0ZWRpcmANCiBm aQ0KIEFDX1NVQlNUKHBjcF92YXJfZGlyKQ0KIA0KQEAgLTcyMiwyNSArNzM4 LDM1IEBADQogQUNfU1VCU1QocGNwX3BtZGFzX2RpcikNCiANCiBkbmwgcnVu dGltZSBzaGFyZWQgbGlicmFyaWVzDQotaWYgdGVzdCAtZCAvdXNyL2xpYg0K LXRoZW4NCi0gICAgcGNwX2xpYl9kaXI9L3Vzci9saWINCi1lbGlmIHRlc3Qg LWQgL3Vzci9sb2NhbC9saWINCitpZiBldmFsIHRlc3QgJHtsaWJkaXIlLyp9 ID0gTk9ORQ0KIHRoZW4NCi0gICAgcGNwX2xpYl9kaXI9L3Vzci9sb2NhbC9s aWINCisgICBpZiB0ZXN0IC1kIC91c3IvbGliDQorICAgdGhlbg0KKyAgICAg ICBwY3BfbGliX2Rpcj0vdXNyL2xpYg0KKyAgIGVsaWYgdGVzdCAtZCAvdXNy L2xvY2FsL2xpYg0KKyAgIHRoZW4NCisgICAgICAgcGNwX2xpYl9kaXI9L3Vz ci9sb2NhbC9saWINCisgICBlbHNlDQorICAgICAgIHBjcF9saWJfZGlyPS9s aWINCisgICBmaQ0KIGVsc2UNCi0gICAgcGNwX2xpYl9kaXI9L2xpYg0KKyAg IHBjcF9saWJfZGlyPWBldmFsIGVjaG8gJGxpYmRpcmANCiBmaQ0KIEFDX1NV QlNUKHBjcF9saWJfZGlyKQ0KIA0KIGRubCBkaXJlY3RvcnkgZm9yIGNvbXBh dGliaWxpdHkgbGlicmFyaWVzLiBBdCB0aGUgbW9tZW50IHdlIGhpZGUNCiBk bmwgeDg2IHNoYXJlZCBsaWJyYXJpZXMgYmVsb3cgUENQX1NIQVJFX0RJUi9j b21wYXQgb24gaWE2NCBidXQgaXQNCiBkbmwgbWF5IGNoYW5nZSBpbiB0aGUg ZnV0dXJlLg0KLWlmIHRlc3QgJHRhcmdldF9jcHUgPSAiaWE2NCINCitpZiBl dmFsIHRlc3QgJHtsaWJkaXIlLyp9ID0gTk9ORQ0KIHRoZW4NCi0gICAgcGNw X2xpYl9jb21wYXRfZGlyPSRwY3Bfc2hhcmVfZGlyL2NvbXBhdA0KKyAgIGlm IHRlc3QgJHRhcmdldF9jcHUgPSAiaWE2NCINCisgICB0aGVuDQorICAgICAg IHBjcF9saWJfY29tcGF0X2Rpcj0kcGNwX3NoYXJlX2Rpci9jb21wYXQNCisg ICBlbHNlDQorICAgICAgIHBjcF9saWJfY29tcGF0X2Rpcj0kcGNwX2xpYl9k aXINCisgICBmaQ0KIGVsc2UNCi0gICAgcGNwX2xpYl9jb21wYXRfZGlyPSRw Y3BfbGliX2Rpcg0KKyAgIHBjcF9saWJfY29tcGF0X2Rpcj1gZXZhbCBlY2hv ICRsaWJkaXJgDQogZmkNCiBBQ19TVUJTVChwY3BfbGliX2NvbXBhdF9kaXIp DQogDQpAQCAtNzY1LDMxICs3OTEsNDMgQEANCiAJZmkNCiAgICAgZG9uZQ0K IGZpDQotDQoraWYgZXZhbCB0ZXN0ICR7bWFuZGlyJS8qfSAhPSBOT05FDQor dGhlbg0KKyAgIHBjcF9tYW5fZGlyPWBldmFsIGVjaG8gJG1hbmRpcmANCitm aQ0KIGlmIHRlc3QgLXogIiRwY3BfbWFuX2RpciINCiB0aGVuDQogICAgIGVj aG8gIkZBVEFMIEVSUk9SOiBjb3VsZCBub3QgZmluZCBmaW5kIHBjcF9tYW5f ZGlyIGRpcmVjdG9yeS4iDQogICAgIGV4aXQgMQ0KIGZpDQotDQogQUNfU1VC U1QocGNwX21hbl9kaXIpDQogQUNfU1VCU1QoaGF2ZV9nemlwcGVkX21hbnBh Z2VzKQ0KIA0KIGRubCBwdWJsaWMgYmluYXJpZXMNCi1pZiB0ZXN0IC1kIC91 c3IvYmluDQoraWYgZXZhbCB0ZXN0ICR7YmluZGlyJS8qfSA9IE5PTkUNCiB0 aGVuDQotICAgIHBjcF9iaW5fZGlyPS91c3IvYmluDQorICAgaWYgdGVzdCAt ZCAvdXNyL2Jpbg0KKyAgIHRoZW4NCisgICAgICAgcGNwX2Jpbl9kaXI9L3Vz ci9iaW4NCisgICBlbHNlDQorICAgICAgIHBjcF9iaW5fZGlyPS9iaW4NCisg ICBmaQ0KIGVsc2UNCi0gICAgcGNwX2Jpbl9kaXI9L2Jpbg0KKyAgIHBjcF9i aW5fZGlyPWBldmFsIGVjaG8gJGJpbmRpcmANCiBmaQ0KIEFDX1NVQlNUKHBj cF9iaW5fZGlyKQ0KIA0KIGRubCBpbmNsdWRlIGZpbGVzDQotaWYgdGVzdCAt ZCAvdXNyL2luY2x1ZGUNCitpZiBldmFsIHRlc3QgJHtpbmNsdWRlZGlyJS8q fSA9IE5PTkUNCiB0aGVuDQotICAgIHBjcF9pbmNfZGlyPS91c3IvaW5jbHVk ZS9wY3ANCisgICBpZiB0ZXN0IC1kIC91c3IvaW5jbHVkZQ0KKyAgIHRoZW4N CisgICAgICAgcGNwX2luY19kaXI9L3Vzci9pbmNsdWRlL3BjcA0KKyAgIGVs c2UNCisgICAgICAgcGNwX2luY19kaXI9JHBjcF9zaGFyZV9kaXIvaW5jbHVk ZQ0KKyAgIGZpDQogZWxzZQ0KLSAgICBwY3BfaW5jX2Rpcj0kcGNwX3NoYXJl X2Rpci9pbmNsdWRlDQorICAgcGNwX2luY19kaXI9YGV2YWwgZWNobyAkaW5j bHVkZWRpcmANCiBmaQ0KIEFDX1NVQlNUKHBjcF9pbmNfZGlyKQ0KIA0KQEAg LTg0MCwzMCArODc4LDQwIEBADQogQUNfU1VCU1QocGNwX3RtcF9kaXIpDQog DQogZG5sIGRvYyBkaXJlY3RvcnkNCi1pZiB0ZXN0ICR0YXJnZXRfb3MgPSBp cml4DQoraWYgZXZhbCB0ZXN0ICR7cHJlZml4JS8qfSA9IE5PTkUNCiB0aGVu DQotICAgIHBjcF9kb2NfZGlyPS92YXIvcGNwDQorICAgaWYgdGVzdCAkdGFy Z2V0X29zID0gaXJpeA0KKyAgIHRoZW4NCisgICAgICAgcGNwX2RvY19kaXI9 L3Zhci9wY3ANCisgICBlbHNlDQorICAgICAgIGlmIHRlc3QgLWQgL3Vzci9k b2MNCisgICAgICAgdGhlbg0KKwkgICBwY3BfZG9jX2Rpcj0vdXNyL2RvYy9w Y3AtJHtQQUNLQUdFX1ZFUlNJT059DQorICAgICAgIGVsc2UNCisJICAgcGNw X2RvY19kaXI9L3Vzci9zaGFyZS9wY3AvZG9jDQorICAgICAgIGZpDQorICAg ZmkNCiBlbHNlDQotICAgIGlmIHRlc3QgLWQgL3Vzci9kb2MNCi0gICAgdGhl bg0KLQlwY3BfZG9jX2Rpcj0vdXNyL2RvYy9wY3AtJHtQQUNLQUdFX1ZFUlNJ T059DQotICAgIGVsc2UNCi0JcGNwX2RvY19kaXI9L3Vzci9zaGFyZS9wY3Av ZG9jDQotICAgIGZpDQorICAgcGNwX2RvY19kaXI9YGV2YWwgZWNobyAkcHJl Zml4L2RvY2ANCiBmaQ0KIEFDX1NVQlNUKHBjcF9kb2NfZGlyKQ0KIA0KIGRu bCBkZW1vcyBkaXJlY3RvcnkNCi1pZiB0ZXN0ICR0YXJnZXRfb3MgPSBpcml4 DQoraWYgZXZhbCB0ZXN0ICR7cHJlZml4JS8qfSA9IE5PTkUNCiB0aGVuDQot ICAgIHBjcF9kZW1vc19kaXI9L3Zhci9wY3AvZGVtb3MNCisgICBpZiB0ZXN0 ICR0YXJnZXRfb3MgPSBpcml4DQorICAgdGhlbg0KKyAgICAgICBwY3BfZGVt b3NfZGlyPS92YXIvcGNwL2RlbW9zDQorICAgZWxzZQ0KKyAgICAgICBpZiB0 ZXN0IC1kIC91c3Ivc2hhcmUvZGVtb3MNCisgICAgICAgdGhlbg0KKwkgICBw Y3BfZGVtb3NfZGlyPS91c3Ivc2hhcmUvZGVtb3MvcGNwDQorICAgICAgIGVs c2UNCisJICAgcGNwX2RlbW9zX2Rpcj0vdXNyL3NoYXJlL3BjcC9kZW1vcw0K KyAgICAgICBmaQ0KKyAgIGZpDQogZWxzZQ0KLSAgICBpZiB0ZXN0IC1kIC91 c3Ivc2hhcmUvZGVtb3MNCi0gICAgdGhlbg0KLQlwY3BfZGVtb3NfZGlyPS91 c3Ivc2hhcmUvZGVtb3MvcGNwDQotICAgIGVsc2UNCi0JcGNwX2RlbW9zX2Rp cj0vdXNyL3NoYXJlL3BjcC9kZW1vcw0KLSAgICBmaQ0KKyAgIHBjcF9kZW1v c19kaXI9YGV2YWwgZWNobyAkcGNwX2RvY19kaXIvZGVtb3NgDQogZmkNCiBB Q19TVUJTVChwY3BfZGVtb3NfZGlyKQ0KIA0KZGlmZiAtdU5yIHBjcC0yLjIu MS5vcmlnL3NyYy9wbWNkL0dOVW1ha2VmaWxlIHBjcC0yLjIuMS9zcmMvcG1j ZC9HTlVtYWtlZmlsZQ0KLS0tIHBjcC0yLjIuMS5vcmlnL3NyYy9wbWNkL0dO VW1ha2VmaWxlCVdlZCBKdW4gMjAgMjI6Mzc6MTcgMjAwMQ0KKysrIHBjcC0y LjIuMS9zcmMvcG1jZC9HTlVtYWtlZmlsZQlUaHUgQXVnIDIzIDE1OjE5OjEw IDIwMDENCkBAIC01Miw3ICs1Miw3IEBADQogaW5zdGFsbCA6OiBkZWZhdWx0 DQogCSQoSU5TVEFMTCkgLUQgLW0gNTU1IHJjX3BjcCAkKFBDUF9SQ19ESVIp L3BjcA0KIAkkKElOU1RBTEwpIC1EIC1tIDQ0NCBwbWNkLm9wdGlvbnMgJChQ Q1BfUE1DRE9QVElPTlNfUEFUSCkNCi0JJChJTlNUQUxMKSAtbSA3NTUgcmNf bG9jYWwgJChQQ1BfVkFSX0RJUikvY29uZmlnL3BtY2QvcmMubG9jYWwNCisJ JChJTlNUQUxMKSAtRCAtbSA3NTUgcmNfbG9jYWwgJChQQ1BfVkFSX0RJUikv Y29uZmlnL3BtY2QvcmMubG9jYWwNCiAJJChJTlNUQUxMKSAtbSA3NTUgLWQg JChQQ1BfU0hBUkVfRElSKS9saWINCiAJJChJTlNUQUxMKSAtbSA0NDQgcG1k YXByb2Muc2ggJChQQ1BfU0hBUkVfRElSKS9saWIvcG1kYXByb2Muc2gNCiAJ JChJTlNUQUxMKSAtbSA0NDQgcmMtcHJvYy5zaCAkKFBDUF9TSEFSRV9ESVIp L2xpYi9yYy1wcm9jLnNoDQpkaWZmIC11TnIgcGNwLTIuMi4xLm9yaWcvc3Jj L3BtaWUvc3JjL3BtaWUuYyBwY3AtMi4yLjEvc3JjL3BtaWUvc3JjL3BtaWUu Yw0KLS0tIHBjcC0yLjIuMS5vcmlnL3NyYy9wbWllL3NyYy9wbWllLmMJV2Vk IEp1biAyMCAyMjozNzoyMCAyMDAxDQorKysgcGNwLTIuMi4xL3NyYy9wbWll L3NyYy9wbWllLmMJVGh1IEF1ZyAyMyAxNTo0NTowNiAyMDAxDQpAQCAtNzks NyArNzksNyBAQA0KIHN0YXRpYyBjaGFyICpwcm9tcHQgPSAicG1pZT4gIjsN CiBzdGF0aWMgY2hhciAqaW50cm8gID0gIlBlcmZvcm1hbmNlIENvLVBpbG90 IEluZmVyZW5jZSBFbmdpbmUgKHBtaWUpLCAiDQogCQkgICAgICAiVmVyc2lv biAlc1xuXG4lcyVzIjsNCi1zdGF0aWMgY2hhciAqY3BhdGggID0gIi92YXIv cGNwL2NvbmZpZy9wbWllLyI7DQorY2hhciBjcGF0aFtNQVhQQVRITEVOXTsN CiANCiBzdGF0aWMgY2hhciBsb2dmaWxlW01BWFBBVEhMRU4rMV07DQogc3Rh dGljIGNoYXIgcGVyZmZpbGVbUE1JRV9QQVRIU0laRV07CS8qIC92YXIvdG1w LzxwaWQ+IGZpbGUgbmFtZSAqLw0KQEAgLTIxNiw2ICsyMTYsNyBAQA0KIAkg ICAgZnByaW50ZihzdGRlcnIsICJsb2FkOiBjYW5ub3QgYWNjZXNzIGNvbmZp ZyBmaWxlICVzOiAlc1xuIiwgZm5hbWUsIHN0cmVycm9yKHN0cykpOw0KIAl9 DQogI2VuZGlmDQorICAgICAgICBzcHJpbnRmKGNwYXRoLCAiJXMvY29uZmln L3BtaWUvIiwgcG1HZXRDb25maWcoIlBDUF9WQVJfRElSIikpOw0KIAlzcHJp bnRmKGNvbmZpZywgIiVzJXMiLCBjcGF0aCwgZm5hbWUpOw0KIAlpZiAoYWNj ZXNzKGNvbmZpZywgRl9PSykgIT0gMCkgew0KIAkgICAgZnByaW50ZihzdGRl cnIsICIlczogY2Fubm90IGFjY2VzcyBjb25maWcgZmlsZSBhcyBlaXRoZXIg JXMgb3IgJXM6ICVzXG4iLA0KZGlmZiAtdU5yIHBjcC0yLjIuMS5vcmlnL3Ny Yy9wbW5zL3BtbnNkZWwuYyBwY3AtMi4yLjEvc3JjL3BtbnMvcG1uc2RlbC5j DQotLS0gcGNwLTIuMi4xLm9yaWcvc3JjL3BtbnMvcG1uc2RlbC5jCVdlZCBK dW4gMjAgMjI6Mzc6MTQgMjAwMQ0KKysrIHBjcC0yLjIuMS9zcmMvcG1ucy9w bW5zZGVsLmMJVGh1IEF1ZyAyMyAxNToxOToxMCAyMDAxDQpAQCAtMjExLDgg KzIxMSw4IEBADQogICAgIHBtbnNfb3V0cHV0KHJvb3QsIG91dGYpOw0KICAg ICBmY2xvc2Uob3V0Zik7DQogDQotICAgIHNwcmludGYoY21kLCAiJXMvYmlu L3BtbnNjb21wICVzIC1mIC1uICVzICVzLmJpbiIsDQotCXBtR2V0Q29uZmln KCJQQ1BfU0hBUkVfRElSIiksIGR1cG9rID8gIi1kIiA6ICIiLCBwbW5zZmls ZSwgcG1uc2ZpbGUpOw0KKyAgICBzcHJpbnRmKGNtZCwgIiVzL3BtbnNjb21w ICVzIC1mIC1uICVzICVzLmJpbiIsDQorCXBtR2V0Q29uZmlnKCJQQ1BfQklO QURNX0RJUiIpLCBkdXBvayA/ICItZCIgOiAiIiwgcG1uc2ZpbGUsIHBtbnNm aWxlKTsNCiAgICAgc3RzID0gc3lzdGVtKGNtZCk7DQogDQogICAgIGV4aXQo c3RzKTsNCmRpZmYgLXVOciBwY3AtMi4yLjEub3JpZy9zcmMvcG1ucy9wbW5z bWVyZ2UuYyBwY3AtMi4yLjEvc3JjL3BtbnMvcG1uc21lcmdlLmMNCi0tLSBw Y3AtMi4yLjEub3JpZy9zcmMvcG1ucy9wbW5zbWVyZ2UuYwlXZWQgSnVuIDIw IDIyOjM3OjE0IDIwMDENCisrKyBwY3AtMi4yLjEvc3JjL3BtbnMvcG1uc21l cmdlLmMJVGh1IEF1ZyAyMyAxNToxOToxMCAyMDAxDQpAQCAtMjIyLDcgKzIy Miw3IEBADQogICAgIGNoYXIJY21kWzEwMjRdOw0KICAgICBjaGFyCXBtbnNj b21wW01BWFBBVEhMRU5dOw0KICAgICANCi0gICAgc3ByaW50ZihwbW5zY29t cCwgIiVzL2Jpbi9wbW5zY29tcCIsIHBtR2V0Q29uZmlnKCJQQ1BfU0hBUkVf RElSIikpOw0KKyAgICBzcHJpbnRmKHBtbnNjb21wLCAiJXMvcG1uc2NvbXAi LCBwbUdldENvbmZpZygiUENQX0JJTkFETV9ESVIiKSk7DQogDQogICAgIC8q IHRyaW0gY29tbWFuZCBuYW1lIG9mIGxlYWRpbmcgZGlyZWN0b3J5IGNvbXBv bmVudHMgKi8NCiAgICAgcG1Qcm9nbmFtZSA9IGFyZ3ZbMF07DQo= ---559023410-684387517-998621840=:12692-- From owner-pcp@oss.sgi.com Thu Aug 23 21:46:10 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7O4kA602490 for pcp-outgoing; Thu, 23 Aug 2001 21:46:10 -0700 Received: from sgi.com (sgi.SGI.COM [192.48.153.1]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7O4k6d02487 for ; Thu, 23 Aug 2001 21:46:06 -0700 Received: from larry.melbourne.sgi.com (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 VAA02746 for ; Thu, 23 Aug 2001 21:46:05 -0700 (PDT) mail_from (markgw@sgi.com) Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.175]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id OAA14514; Fri, 24 Aug 2001 14:44:43 +1000 Date: Fri, 24 Aug 2001 14:44:43 +1000 (EST) From: Mark Goodwin X-Sender: markgw@sherman.melbourne.sgi.com To: Brian Harvell cc: Mark Goodwin , Ken McDonell , pcp@oss.sgi.com Subject: Re: --prefix does not work. In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Thu, 23 Aug 2001, Brian Harvell wrote: > On Mon, 20 Aug 2001, Mark Goodwin wrote: > > > > Perhaps we weren't clear enough: PCP does not use --prefix in the > > configure part of the build because PCP requires multiple points of > > relocation but --prefix only provides one, rendering --prefix useless. > > > > The correct way to relocate PCP directories is to change > > src/include/pcp.conf.in. If you still disagree, then you must know > > something about --prefix that we are missing, so fix it and send me > > a patch! > > > > If you just use prefix it uses the defaults and you can't split it. However, > you can have a more fine grained control by specifying libdir localstatedir > etc. I've attached a patch that will allow you to do this. It's a little bit of > a kludge but it works. If you don't specify anything it works as before. This > patch also fixes a few typos that prevent even modifying pcp.conf.in from > working correctly (hard coded paths and pulling the wrong config var) thanks for the patch. I'm going to split it into two : first to fix the incorrect/hardcoded paths, and then second for your --prefix changes. The existing pcp.conf mechanism cannot be deprecated without breaking IRIX compatibility and we're not willing to do that. However, we should be able to accommodate your changes for rewriting src/include/pcp.conf.in. > > Personally I think the old directory structure could just go away. I didn't > like it in the Irix day and it's even worse in a gnu environment. Why you don't like it? > > On another note anyone know if there are any hardware monitoring pmdas? I'm > looking to be able to detect when things like disk, memory, cpus etc go bad. > try the "lmsensors" PMDA, contributed by Troy Dawson (dawson@fnal.gov). This will allow you to monitor fans, voltages, etc on your motherboard (if your h/w supports it). Also, the "roomtemp" PMDA exports the room temperature from one or more sensors built using the DS2480 and DS1280 chipsets and MicroLAN technology from Dallas Semiconductor. As for detecting failed h/w, you might get somewhere with a syslog dredger of some kind (we have a PMDA for this called "logtail", but it's not part of the open source package). The lmsensor and roomtemp PMDAs were both shipped with pcp-2.2.1-3, get it from ftp://oss.sgi.com/projects/pcp/download/ -- Mark From owner-pcp@oss.sgi.com Fri Aug 24 07:20:59 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7OEKxe15238 for pcp-outgoing; Fri, 24 Aug 2001 07:20:59 -0700 Received: from boondoggle.office.aol.com (pix-fw.wan.aol.com [152.163.190.1]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7OEKud15235 for ; Fri, 24 Aug 2001 07:20:56 -0700 Received: (from harvell@localhost) by boondoggle.office.aol.com (8.11.4/8.11.4) id f7OEKih15099; Fri, 24 Aug 2001 10:20:44 -0400 (EDT) Date: Fri, 24 Aug 2001 10:20:44 -0400 (EDT) From: Brian Harvell X-X-Sender: To: Mark Goodwin cc: Ken McDonell , Subject: Re: --prefix does not work. In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Fri, 24 Aug 2001, Mark Goodwin wrote: > > thanks for the patch. I'm going to split it into two : first to fix the > incorrect/hardcoded paths, and then second for your --prefix changes. > The existing pcp.conf mechanism cannot be deprecated without breaking > IRIX compatibility and we're not willing to do that. However, we should > be able to accommodate your changes for rewriting src/include/pcp.conf.in. > cool. > > Personally I think the old directory structure could just go away. I didn't > > like it in the Irix day and it's even worse in a gnu environment. > > Why you don't like it? The main reason is it breaks the flow of gnu setups. The second is that it's so hard to find where everything is unless your intimately familiar with the package. If --prefix works though it really doesn't matter since the user can decide. > try the "lmsensors" PMDA, contributed by Troy Dawson (dawson@fnal.gov). > This will allow you to monitor fans, voltages, etc on your motherboard > (if your h/w supports it). Also, the "roomtemp" PMDA exports the room > temperature from one or more sensors built using the DS2480 and DS1280 > chipsets and MicroLAN technology from Dallas Semiconductor. As for detecting > failed h/w, you might get somewhere with a syslog dredger of some kind > (we have a PMDA for this called "logtail", but it's not part of the > open source package). > > The lmsensor and roomtemp PMDAs were both shipped with pcp-2.2.1-3, > get it from ftp://oss.sgi.com/projects/pcp/download/ > thanks I'll give those a try. Brian -- Brian Harvell harvell@aol.net http://ToolBoy.com/ echo '[q]sa[ln0=aln256%Pln256/snlbx]sb3135071790101768542287578439snlbxq'|dc From owner-pcp@oss.sgi.com Sat Aug 25 14:57:58 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7PLvwk16515 for pcp-outgoing; Sat, 25 Aug 2001 14:57:58 -0700 Received: from zok.sgi.com (zok.sgi.com [204.94.215.101]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7PLvtd16512 for ; Sat, 25 Aug 2001 14:57:55 -0700 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by zok.sgi.com (8.11.4/8.11.4/linux-outbound_gateway-1.0) with SMTP id f7PM4Ia06659 for ; Sat, 25 Aug 2001 15:04:18 -0700 Received: from kenj-ppp.melbourne.sgi.com (kenj-ppp.melbourne.sgi.com [134.14.55.215]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id IAA21835; Sun, 26 Aug 2001 08:56:28 +1100 Date: Sun, 26 Aug 2001 07:57:46 +1000 (EST) From: kenmcd@melbourne.sgi.com Reply-To: kenmcd@melbourne.sgi.com To: Martin Knoblauch cc: pcp@oss.sgi.com Subject: Re: GEtting number and names of network interfaces (Linux) In-Reply-To: <3B83C567.564F8053@TeraPort.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk Looks like this has been skirted around some, without necessarily being completely answered. Martin, what _exactly_ are you after? The names of the network interfaces are included in the instance names for the network.interface.* metrics, e.g. $ pminfo -f network.interface.total.bytes network.interface.total.bytes inst [0 or "lo"] value 7967920 inst [1 or "eth0"] value 0 We don't have the IP address(es) for each network interface, but looks like we could get that by using the same SIOCGIF* ioctl()s that ioconfig(1) uses. On Wed, 22 Aug 2001, Martin Knoblauch wrote: > Hi, > > I am currently evaluating performance monitoring/measuring tools for a > mostly Linux based cluster environment. Due to some history, PCP came to > my mind :-) > > I installed the binary package and it seems to do a lot of the things > we need. One thing I am missing is the number and the names of the > network interfaces. > > A quick look at the sources seems to indicate that the information > itself is available, but not exported. Am I correct with this? Not sure what you mean here, perhaps more detail would help. > If I wanted to add the fileds to the network metrics, what cautions dio > I need to observe. I don't want to break things ... Brieflly (refer to the PCP Programmer's Guide, available from techpubs.sgi.com for more detail), you need to: - choose a name - choose a PMID - define the metadata - integrate the new metric, its descriptor (metabata), fetch method and any associated instance domain information into the PMDA - add help text to describe the new metric From owner-pcp@oss.sgi.com Sat Aug 25 14:58:29 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7PLwTX16543 for pcp-outgoing; Sat, 25 Aug 2001 14:58:29 -0700 Received: from sgi.com (sgi.SGI.COM [192.48.153.1]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7PLwJd16531 for ; Sat, 25 Aug 2001 14:58:19 -0700 Received: from larry.melbourne.sgi.com (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 OAA03261 for ; Sat, 25 Aug 2001 14:58:19 -0700 (PDT) mail_from (kenmcd@melbourne.sgi.com) From: kenmcd@melbourne.sgi.com Received: from kenj-ppp.melbourne.sgi.com (kenj-ppp.melbourne.sgi.com [134.14.55.215]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id IAA21841; Sun, 26 Aug 2001 08:56:58 +1100 Date: Sun, 26 Aug 2001 07:58:17 +1000 (EST) Reply-To: kenmcd@melbourne.sgi.com To: gilly cc: pcp@oss.sgi.com Subject: RE: 2 pmcd's on one machine In-Reply-To: <016501c12ba4$3195db20$2a04000a@exanet> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Thu, 23 Aug 2001, gilly wrote: > Now I have the 2 PMCDs running on one machine, but I dont know how to tell > the PMDA to connect to PMCD2 (on the other port - 4322). Set PMCD_PORT in the environment to 4322 before calling pmNewContext(). From owner-pcp@oss.sgi.com Sun Aug 26 23:57:32 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7R6vW715871 for pcp-outgoing; Sun, 26 Aug 2001 23:57:32 -0700 Received: from mail.teraport.de (mail.teraport.de [195.143.8.72]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7R6vTd15867 for ; Sun, 26 Aug 2001 23:57:29 -0700 Received: from TeraPort.de ([10.10.12.32]) by mail.teraport.de (Lotus Domino Release 5.0.7) with ESMTP id 2001082708572075:6214 ; Mon, 27 Aug 2001 08:57:20 +0200 Message-ID: <3B89EF51.7E96D0CB@TeraPort.de> Date: Mon, 27 Aug 2001 08:57:21 +0200 From: Martin Knoblauch Organization: TeraPort GmbH X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.4.8-ac10 i686) X-Accept-Language: en, de MIME-Version: 1.0 To: kenmcd@melbourne.sgi.com CC: pcp@oss.sgi.com Subject: Re: GEtting number and names of network interfaces (Linux) References: X-MIMETrack: Itemize by SMTP Server on lotus/Teraport/de(Release 5.0.7 |March 21, 2001) at 08/27/2001 08:57:20 AM, Serialize by Router on lotus/Teraport/de(Release 5.0.7 |March 21, 2001) at 08/27/2001 08:57:29 AM, Serialize complete at 08/27/2001 08:57:29 AM Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Sender: owner-pcp@oss.sgi.com Precedence: bulk kenmcd@melbourne.sgi.com wrote: > > Looks like this has been skirted around some, without necessarily being > completely answered. > > Martin, what _exactly_ are you after? > > The names of the network interfaces are included in the instance > names for the network.interface.* metrics, e.g. > Hi Ken, I found out about the instance names shortly after posting my question, so basically I am happy although I think that explicitely exporting the number of interfaces and the names might be useful to some. Thanks Martin > > > > A quick look at the sources seems to indicate that the information > > itself is available, but not exported. Am I correct with this? > > Not sure what you mean here, perhaps more detail would help. > Now, I found that the info is stored, but I did not see the "instance" concept :-) -- ------------------------------------------------------------------ Martin Knoblauch | email: Martin.Knoblauch@TeraPort.de TeraPort GmbH | Phone: +49-89-510857-309 C+ITS | Fax: +49-89-510857-111 http://www.teraport.de | Mobile: +49-170-4904759 From owner-pcp@oss.sgi.com Mon Aug 27 11:33:47 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7RIXl831757 for pcp-outgoing; Mon, 27 Aug 2001 11:33:47 -0700 Received: from ztxmail05.ztx.compaq.com (ztxmail05.ztx.compaq.com [161.114.1.209]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7RIXRd31744 for ; Mon, 27 Aug 2001 11:33:27 -0700 Received: by ztxmail05.ztx.compaq.com (Postfix, from userid 12345) id 09E6C1730; Mon, 27 Aug 2001 13:33:22 -0500 (CDT) Received: from mailrelay01.cce.cpqcorp.net (mailrelay01.cce.cpqcorp.net [16.47.68.171]) by ztxmail05.ztx.compaq.com (Postfix) with ESMTP id D48FA14EB for ; Mon, 27 Aug 2001 13:33:21 -0500 (CDT) Received: by mailrelay01.cce.cpqcorp.net (Postfix, from userid 12345) id A9E1186E; Mon, 27 Aug 2001 13:33:21 -0500 (CDT) Received: from perf.zko.dec.com (aseng.zko.dec.com [16.31.80.86]) by mailrelay01.cce.cpqcorp.net (Postfix) with ESMTP id 01622917; Mon, 27 Aug 2001 13:33:20 -0500 (CDT) Received: from localhost by perf.zko.dec.com (8.9.3/1.1.10.5/07Mar97-1056AM) id OAA0000492414; Mon, 27 Aug 2001 14:33:20 -0400 (EDT) Date: Mon, 27 Aug 2001 14:33:20 -0400 (EDT) From: Phillip Ezolt To: Cc: "Stanley, Dave" , Bill French Subject: Performance Co-Pilot patch for Compaq's Tru64 Message-ID: MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="1401954064-1937414519-998937200=:482480" Sender: owner-pcp@oss.sgi.com Precedence: bulk This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. Send mail to mime@docserver.cac.washington.edu for more info. --1401954064-1937414519-998937200=:482480 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi All, I've patched the Performance Co-Pilot infrastructure to work with Tru64. All of the clients (except for pmstat) that I've tested work. The included PMDAs all compile except for cisco and shping. Known issues: 1) cisco and pmdas do not compile. (Missing sys/prctl) 2) No test for whether to use "hostname -f" or "hostname". 3) Magic file format is not compatible with Tru64. 4) Testing for the "runlevel" command is not done properly in the shell scripts. 5) No Tru64 specific PMDA. Questions/Comments: 1) What pmda number should I use for Tru64? (Will 74 work?) 2) The memory values in the Linux pmda should be 64-bit, not 32-bit. Problems show up when a machine has more than 4-gig of memory. /* mem.util.used */ { &proc_meminfo.mem[1], { PMDA_PMID(CLUSTER_MEMINFO,1), PM_TYPE_U32, PM_INDOM_NULL, PM_SEM_INSTANT, PMDA_PMUNITS(1,0,0,PM_SPACE_BYTE,0,0) }, }, Hopefully, the patch is self explanitory. I had to add some automake checks in for things that incorrectly pcp assumed. If you need me to test or explain anything, just tell me! --Phil Compaq: High Performance Server Systems Quality & Performance Engineering ---------------- Alpha, The Fastest Processor on Earth -------------------- Phillip.Ezolt@compaq.com Performance Tools/Analysis ------------------- See the results at www.spec.org ----------------------- --1401954064-1937414519-998937200=:482480 Content-Type: TEXT/PLAIN; charset=US-ASCII; name=tru64_pcp_patch Content-Transfer-Encoding: BASE64 Content-ID: Content-Description: Content-Disposition: attachment; filename=tru64_pcp_patch SW5kZXg6IEdOVW1ha2VmaWxlDQo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09DQpS Q1MgZmlsZTogL3Vzci91c2Vycy9lem9sdC9jdnMvcmVwb3NpdG9yeS9wY3Av R05VbWFrZWZpbGUsdg0KcmV0cmlldmluZyByZXZpc2lvbiAxLjEuMS4xDQpk aWZmIC1yMS4xLjEuMSBHTlVtYWtlZmlsZQ0KNTNjNTMNCjwgCQltYWtlIC1D ICQkZCAkQCB8fCBleGl0ICQkPzsgXA0KLS0tDQo+IAkJJChNQUtFKSAtQyAk JGQgJEAgfHwgZXhpdCAkJD87IFwNCjU1YzU1DQo8ICAJZG9uZQ0KLS0tDQo+ IAlkb25lDQo3MGM3MA0KPCAJCW1ha2UgLUMgJCRkICRAIHx8IGV4aXQgJCQ/ OyBcDQotLS0NCj4gCQkkKE1BS0UpIC1DICQkZCAkQCB8fCBleGl0ICQkPzsg XA0KNzJjNzINCjwgIAlkb25lDQotLS0NCj4gCWRvbmUNCkluZGV4OiBjb25m aWd1cmUuaW4NCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0NClJDUyBmaWxlOiAv dXNyL3VzZXJzL2V6b2x0L2N2cy9yZXBvc2l0b3J5L3BjcC9jb25maWd1cmUu aW4sdg0KcmV0cmlldmluZyByZXZpc2lvbiAxLjEuMS4xDQpkaWZmIC1yMS4x LjEuMSBjb25maWd1cmUuaW4NCjMyYTMzLDM2DQo+IA0KPiAgICAgI0NsZWFy IGF3YXkgdGhlIGV4Y2VzcyBPU0YgdmVyc2lvbiBpbmZvcm1hdGlvbi4gDQo+ ICAgICBidWlsZF9vcz1gZWNobyAkYnVpbGRfb3MgfCBzZWQgJ1tzL29zZi4q L29zZi9dJ2ANCj4gICAgIHRhcmdldF9vcz1gZWNobyAkYnVpbGRfb3MgfCBz ZWQgJ1tzL29zZi4qL29zZi9dJ2ANCjQyYTQ3LDQ5DQo+IGVsaWYgdGVzdCAk dGFyZ2V0X29zID0gb3NmDQo+IHRoZW4NCj4gICAgIDoNCjUzYzYwDQo8ICAg ICAuIFZFUlNJT04ucGNwDQotLS0NCj4gICAgIC4gLi9WRVJTSU9OLnBjcA0K NjNjNzAsNzQNCjwgICAgIHBhY2thZ2VfYnVpbGRlcj1gaWQgLXUgLW5gQGBo b3N0bmFtZSAtZmANCi0tLQ0KPiAJaWYgdGVzdCAkdGFyZ2V0X29zID0gb3Nm OyB0aGVuDQo+IAkgICAgcGFja2FnZV9idWlsZGVyPWBpZCAtdSAtbmBAYGhv c3RuYW1lIGANCj4gCWVsc2UNCj4gCSAgICBwYWNrYWdlX2J1aWxkZXI9YGlk IC11IC1uYEBgaG9zdG5hbWUgLWZgDQo+IAlmaQ0KMTU5YTE3MSwxODgNCj4g ZG5sIFRydTY0IG5lZWRzIGNyVCB0byB0cnVuY2F0ZSB0aGUgYXJjaGl2ZSBu YW1lcywgd2hpY2ggTGludXggdXNlICJmIg0KPiBpZiB0ZXN0ICR0YXJnZXRf b3MgPSBvc2YNCj4gdGhlbg0KPiAgICAgYXJmbGFncz1jclQNCj4gZWxzZQ0K PiAgICAgYXJmbGFncz1jcmYNCj4gZmkNCj4gQUNfU1VCU1QoYXJmbGFncykN Cj4gDQo+IA0KPiBkbmwgVHJ1NjQgbmVlZHMgdGhpcyBmb3IgInBtaWUiIHdo aWNoIGRvZXMgdHJpY2tzIHdpdGggIm5hbnMiIGFuZCByZXF1aXJlcyBJRUVF IGNvbXBsaWFudCBtYXRoLg0KPiBpZiB0ZXN0ICR0YXJnZXRfb3MgPSBvc2YN Cj4gdGhlbg0KPiAgICAgaWVlZV9tYXRoX2ZsYWc9LW1pZWVlDQo+IGZpDQo+ IEFDX1NVQlNUKGllZWVfbWF0aF9mbGFnKQ0KPiANCj4gDQozMDJjMzMxDQo8 IHBzICR7cGNwX3BzX2FsbF9mbGFnc30gMj4mMSB8IGdyZXAgJ1sgCV0xWyAJ XS4qWyAvXWluaXQnID4vZGV2L251bGwgJiYgcGNwX3BzX2FsbF9mbGFncz0k e3BjcF9wc19hbGxfZmxhZ3N9dw0KLS0tDQo+IHBzICR7cGNwX3BzX2FsbF9m bGFnc313IDI+JjEgfCBncmVwICdbIAldMVsgCV0uKlsgL11pbml0JyA+L2Rl di9udWxsICYmIHBjcF9wc19hbGxfZmxhZ3M9JHtwY3BfcHNfYWxsX2ZsYWdz fXcNCjMwNGEzMzQNCj4gDQozMjBjMzUwDQo8IEFDX0NIRUNLX0hFQURFUlMo ZmNudGwuaCBsaW1pdHMuaCBtYWxsb2MuaCBzdHJpbmdzLmggc3lzL3RpbWUu aCBzeXNsb2cuaCB1bmlzdGQuaCBzY2hlZC5oIHRlcm1pb3MuaCBkbGZjbi5o IGllZWVmcC5oIGVuZGlhbi5oIHN5cy9lbmRpYW4uaCBwdGhyZWFkLmgpDQot LS0NCj4gQUNfQ0hFQ0tfSEVBREVSUyhmY250bC5oIGxpbWl0cy5oIG1hbGxv Yy5oIHN0cmluZ3MuaCBzeXMvdGltZS5oIHN5c2xvZy5oIHVuaXN0ZC5oIHNj aGVkLmggdGVybWlvcy5oIGRsZmNuLmggaWVlZWZwLmggZW5kaWFuLmggc3lz L2VuZGlhbi5oIHB0aHJlYWQuaCBpbnR0eXBlcy5oIGZsb2F0LmgpDQozMzRj MzY0DQo8IEFDX0NIRUNLX0ZVTkNTKGdldGN3ZCBnZXRob3N0bmFtZSBnZXR0 aW1lb2ZkYXkgbWt0aW1lIHB1dGVudiBzZWxlY3Qgc29ja2V0IHN0cmR1cCBz dHJlcnJvciBzdHJ0b2Qgc3RydG9sIHVuYW1lIHdhaXRwaWQgYXRleGl0IGJy ayBzYnJrIF9fY2xvbmUgc3Byb2MgdGNnZXRhdHRyIHNpZ25hbCBoc3RyZXJy b3IgaXNuYW5mIGlzbmFuZCBzaGxfbG9hZCBzZ2luYXAgdXNsZWVwIG9zZXJy b3Igc2V0b3NlcnJvciBzaWdob2xkIHNpZ3JlbHNlIHByY3RsIHJlZ2V4IHJl Z2NtcCByZWdleGVjIHJlZ2NvbXApDQotLS0NCj4gQUNfQ0hFQ0tfRlVOQ1Mo Z2V0Y3dkIGdldGhvc3RuYW1lIGdldHRpbWVvZmRheSBta3RpbWUgcHV0ZW52 IHNlbGVjdCBzb2NrZXQgc3RyZHVwIHN0cmVycm9yIHN0cnRvZCBzdHJ0b2wg c3RydG9sbCB1bmFtZSB3YWl0cGlkIGF0ZXhpdCBicmsgc2JyayBfX2Nsb25l IHNwcm9jIG1lbWFsaWduIHRjZ2V0YXR0ciBzaWduYWwgaHN0cmVycm9yIGlz bmFuZiBpc25hbmQgc2hsX2xvYWQgc2dpbmFwIHVzbGVlcCBvc2Vycm9yIHNl dG9zZXJyb3Igc2lnaG9sZCBzaWdyZWxzZSBwcmN0bCByZWdleCByZWdjbXAg cmVnZXhlYyByZWdjb21wKQ0KMzM5YTM3MA0KPiANCjM0MWMzNzINCjwgQUNf Q0hFQ0tfVFlQRShfX3VpbnQzMl90LCB1X2ludDMyX3QpDQotLS0NCj4gQUNf Q0hFQ0tfVFlQRShfX3VpbnQzMl90LCB1aW50MzJfdCkNCjM0NGMzNzUNCjwg QUNfQ0hFQ0tfVFlQRSh1aW50X3QsIHVfaW50MzJfdCkNCi0tLQ0KPiBBQ19D SEVDS19UWVBFKHVpbnRfdCwgdWludDMyX3QpDQo1OTVhNjI3LDYyOQ0KPiBk bmwgY2hlY2sgaWYgZGxvcGVuIGV0IGFsIGFyZSByZXNvbHZlZCB3aXRob3V0 IGxpbmtpbmcgaW4gYSBsaWJyYXJ5DQo+IEFDX1RSWV9MSU5LX0ZVTkMoZGxv cGVuLCBBQ19ERUZJTkUoSEFWRV9ETEZDTikpDQo+IA0KNTk3YTYzMg0KPiAN CjYwMWE2MzcsNjQ0DQo+IGRubCBjaGVjayBpZiBuYW5vc2xlZXAgbmVlZHMg LWxydCB0byB3b3JrDQo+IEFDX0NIRUNLX0xJQihydCwgbmFub3NsZWVwLCBs aWJfZm9yX25hbm9zbGVlcD0tbHJ0OyBBQ19ERUZJTkUoSEFWRV9OQU5PU0xF RVApKQ0KPiBBQ19TVUJTVChsaWJfZm9yX25hbm9zbGVlcCkNCj4gDQo+IGRu bCBjaGVjayBpZiBpc25hbmYgbmVlZHMgLWxtIHRvIHdvcmsNCj4gQUNfQ0hF Q0tfTElCKG0sIGlzbmFuZiwgbGliX2Zvcl9pc25hbmY9LWxtOyBBQ19ERUZJ TkUoSEFWRV9JU05BTkYpKQ0KPiBBQ19TVUJTVChsaWJfZm9yX2lzbmFuZikN Cj4gDQo2OTlhNzQzLDc0NQ0KPiB0aGVuDQo+ICAgICBwY3BfYmluYWRtX2Rp cj0kcGNwX3NoYXJlX2Rpci9iaW4NCj4gZWxpZiB0ZXN0ICR0YXJnZXRfb3Mg PSBvc2YNCkluZGV4OiBzcmMvZGJwbWRhL3NyYy9wbWRhLmMNCj09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT0NClJDUyBmaWxlOiAvdXNyL3VzZXJzL2V6b2x0L2N2 cy9yZXBvc2l0b3J5L3BjcC9zcmMvZGJwbWRhL3NyYy9wbWRhLmMsdg0KcmV0 cmlldmluZyByZXZpc2lvbiAxLjEuMS4xDQpkaWZmIC1yMS4xLjEuMSBwbWRh LmMNCjQ4YTQ5LDUyDQo+ICNpZiBkZWZpbmVkKEhBVkVfRkxPQVRfSCkNCj4g I2luY2x1ZGUgPGZsb2F0Lmg+DQo+ICNlbmRpZg0KPiANCkluZGV4OiBzcmMv aW5jbHVkZS9idWlsZGRlZnMuaW4NCj09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0N ClJDUyBmaWxlOiAvdXNyL3VzZXJzL2V6b2x0L2N2cy9yZXBvc2l0b3J5L3Bj cC9zcmMvaW5jbHVkZS9idWlsZGRlZnMuaW4sdg0KcmV0cmlldmluZyByZXZp c2lvbiAxLjEuMS4xDQpkaWZmIC1yMS4xLjEuMSBidWlsZGRlZnMuaW4NCjU3 YzU3LDU4DQo8IFBBQ0tBR0VfQlVJTERFUiA9ICQoc2hlbGwgZWNobyBgaWQg LXVuYEBgaG9zdG5hbWUgLWZgKQ0KLS0tDQo+ICNQQUNLQUdFX0JVSUxERVIg PSAkKHNoZWxsIGVjaG8gYGlkIC11bmBAYGhvc3RuYW1lIC1mYCkNCj4gUEFD S0FHRV9CVUlMREVSID0gJChzaGVsbCBlY2hvIGBpZCAtdW5gQGBob3N0bmFt ZSBgKQ0KNzVhNzcsODANCj4gQVJGTEFHUz0gQGFyZmxhZ3NADQo+IA0KPiBJ RUVFX01BVEhfRkxBRz0gQGllZWVfbWF0aF9mbGFnQA0KPiANCjEyNmExMzIs MTMzDQo+IExJQl9GT1JfTkFOT1NMRUVQID0gQGxpYl9mb3JfbmFub3NsZWVw QA0KPiBMSUJfRk9SX0lTTkFORiA9IEBsaWJfZm9yX2lzbmFuZkANCkluZGV4 OiBzcmMvaW5jbHVkZS9idWlsZHJ1bGVzDQo9PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09DQpSQ1MgZmlsZTogL3Vzci91c2Vycy9lem9sdC9jdnMvcmVwb3NpdG9y eS9wY3Avc3JjL2luY2x1ZGUvYnVpbGRydWxlcyx2DQpyZXRyaWV2aW5nIHJl dmlzaW9uIDEuMS4xLjENCmRpZmYgLXIxLjEuMS4xIGJ1aWxkcnVsZXMNCjU0 YzU0DQo8IAkkKEFSKSBjcmYgJChTVEFUSUNMSUJUQVJHRVQpICQ/DQotLS0N Cj4gCSQoQVIpICQoQVJGTEFHUykgJChTVEFUSUNMSUJUQVJHRVQpICQ/DQpJ bmRleDogc3JjL2luY2x1ZGUvcGNwLmVudg0KPT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PQ0KUkNTIGZpbGU6IC91c3IvdXNlcnMvZXpvbHQvY3ZzL3JlcG9zaXRv cnkvcGNwL3NyYy9pbmNsdWRlL3BjcC5lbnYsdg0KcmV0cmlldmluZyByZXZp c2lvbiAxLjEuMS4xDQpkaWZmIC1yMS4xLjEuMSBwY3AuZW52DQo3OGE3OSw4 MQ0KPiAgICAgIyBBbGwgcHMncyBleGNlcHQgVHJ1NjQuLi4uIEl0IGhhcyB0 aGUgZm9ybTogSEg6TU06U1MuVFQuDQo+ICAgICAjIFRoZSBzZWQgc2NyaXB0 cyBoYXZlIGJlZW4gdXBkYXRlZCB0byBjb3BlIHdpdGggdGhpcy4gIC0tUEdF DQo+ICAgICAjDQo4NGM4Nw0KPCAgICAgfCBzZWQgLW4gXA0KLS0tDQo+ICAg ICAgfCBzZWQgLW4gXA0KODhjOTENCjwgCS1lICdzL15bXiBdKiAvLycgXA0K LS0tDQo+ICAJLWUgJ3MvXlteIF0qIC8vJyBcDQo5MGM5Myw5NQ0KPCAJLWUg Ii9bMC05XTpbMC05XVswLTldICAqJDEgL3MvIC4qLy9wIg0KLS0tDQo+IAkt ZSAiL1swLTldOlswLTldWzAtOV0gICokMSAvcy8gLiovL3AiIFwNCj4gCS1l ICIvWzAtOV06WzAtOV1bMC05XS5bMC05XVswLTldICAqW14gXSpcLyQxIC9z LyAuKi8vcCIgXA0KPiAJLWUgIi9bMC05XTpbMC05XVswLTldLlswLTldWzAt OV0gICokMSAvcy8gLiovL3AiDQpJbmRleDogc3JjL2luY2x1ZGUvcGxhdGZv cm1fZGVmcy5oLmluDQo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09DQpSQ1MgZmls ZTogL3Vzci91c2Vycy9lem9sdC9jdnMvcmVwb3NpdG9yeS9wY3Avc3JjL2lu Y2x1ZGUvcGxhdGZvcm1fZGVmcy5oLmluLHYNCnJldHJpZXZpbmcgcmV2aXNp b24gMS4xLjEuMQ0KZGlmZiAtcjEuMS4xLjEgcGxhdGZvcm1fZGVmcy5oLmlu DQoxMjBhMTIxLDEyMg0KPiAjdW5kZWYgSEFWRV9JTlRUWVBFU19IDQo+ICN1 bmRlZiBIQVZFX0ZMT0FUX0gNCjE0NWExNDgNCj4gI3VuZGVmIEhBVkVfU1RS VE9MTA0KMjA5YTIxMywyMTcNCj4gLyogVGhpcyBkZWZpbmVzIHVpbnQzMl90 IGFuZCBpbnQzMl90IGZvciBUcnU2NCAqLw0KPiAjaWYgZGVmaW5lZChIQVZF X0lOVFRZUEVTX0gpDQo+ICNpbmNsdWRlIDxpbnR0eXBlcy5oPg0KPiAjZW5k aWYNCj4gDQoyMzRhMjQzLDI1Mw0KPiAvKiBJZiB3ZSBkb24ndCBoYXZlIGEg c3RydG9sbCBmdW5jdGlvbiwgYnV0IHdlIGhhdmUgNjQgYml0IGxvbmdzLCAN Cj4gICogc3RydG9sIGlzIHRoZSBlcXVpdmFsZW50IG9mIHN0cnRvbGwuIA0K PiAgKi8NCj4gDQo+ICNpZmRlZiBIQVZFXzY0QklUX0xPTkcNCj4gI2lmbmRl ZiBIQVZFX1NUUlRPTEwNCj4gI2RlZmluZSBzdHJ0b2xsIHN0cnRvbA0KPiAj ZGVmaW5lIHN0cnRvdWxsIHN0cnRvdWwNCj4gI2VuZGlmDQo+ICNlbmRpZg0K PiANCjIzN2EyNTcsMjYwDQo+ICNlbmRpZg0KPiANCj4gI2lmbmRlZiBIQVZF X01FTUFMSUdODQo+ICNkZWZpbmUgbWVtYWxpZ24oeCx5KSBtYWxsb2MoeSkN CkluZGV4OiBzcmMvaW5jbHVkZS90cmFjZV9kZXYuaA0KPT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PQ0KUkNTIGZpbGU6IC91c3IvdXNlcnMvZXpvbHQvY3ZzL3Jl cG9zaXRvcnkvcGNwL3NyYy9pbmNsdWRlL3RyYWNlX2Rldi5oLHYNCnJldHJp ZXZpbmcgcmV2aXNpb24gMS4xLjEuMQ0KZGlmZiAtcjEuMS4xLjEgdHJhY2Vf ZGV2LmgNCjcyYTczDQo+ICNpZmRlZiBVU0VfU0dJX0lOVDMyDQo3M2E3NSw3 Nw0KPiAjZWxzZQ0KPiB0eXBlZGVmIHVpbnQJX19wbVRyYWNlUERVOw0KPiAj ZW5kaWYNCkluZGV4OiBzcmMvbGlicGNwL3NyYy9HTlVtYWtlZmlsZQ0KPT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PQ0KUkNTIGZpbGU6IC91c3IvdXNlcnMvZXpv bHQvY3ZzL3JlcG9zaXRvcnkvcGNwL3NyYy9saWJwY3Avc3JjL0dOVW1ha2Vm aWxlLHYNCnJldHJpZXZpbmcgcmV2aXNpb24gMS4xLjEuMQ0KZGlmZiAtcjEu MS4xLjEgR05VbWFrZWZpbGUNCjUwYTUxLDUyDQo+IExMRExJQlMgPSAkKExJ Ql9GT1JfSVNOQU5GKQ0KPiANCjkxYzkzDQo8IAljZCAkKEBEKTsgbWFrZSAk KEBGKQ0KLS0tDQo+IAljZCAkKEBEKTsgJChNQUtFKSAkKEBGKQ0KSW5kZXg6 IHNyYy9wbWNkL3BtZGFwcm9jLnNoDQo9PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 DQpSQ1MgZmlsZTogL3Vzci91c2Vycy9lem9sdC9jdnMvcmVwb3NpdG9yeS9w Y3Avc3JjL3BtY2QvcG1kYXByb2Muc2gsdg0KcmV0cmlldmluZyByZXZpc2lv biAxLjEuMS4xDQpkaWZmIC1yMS4xLjEuMSBwbWRhcHJvYy5zaA0KMTc4YTE3 OSwxOTgNCj4gIyBUaGlzIGZ1bmN0aW9uIHdpbGwgc2VuZCBzaWduYWwgJDEg a2lsbCBhbGwgcHJvY2Vzc29ycyB0aGF0IGFyZSBuYW1lZCAkMg0KPiAjIExp bnV4IGhhcyBhIG5pY2UgdXRpbGl0eSAoa2lsbGFsbCkgdGhhdCB3aWxsIGRv IHRoaXMuICBJdCBoYXMgdG8gYmUgc2ltdWxhdGVkDQo+ICMgb24gVHJ1NjQu DQo+IA0KPiBfX2tpbGxhbGwoKQ0KPiB7DQo+IA0KPiAgICAgaWYgWyAkUENQ X1BMQVRGT1JNID0gIm9zZiIgXQ0KPiAgICAgdGhlbg0KPiAJX2dldF9waWRz X2J5X25hbWUgJDIgPiR0bXAua2lsbA0KPiAJaWYgWyAtcyAkdG1wLmtpbGwg XQ0KPiAJdGhlbg0KPiAJICAgIGtpbGwgJDEgYGNhdCAkdG1wLmtpbGxgIA0K PiAJZmkNCj4gICAgIGVsc2UNCj4gICAgICAgICBraWxsYWxsICQxICQyID4v ZGV2L251bGwgMj4mMQ0KPiAgICAgZmkNCj4gDQo+IH0NCj4gDQoyNzZjMjk2 DQo8IAkgICAga2lsbGFsbCAtSFVQIHBtY2QgPi9kZXYvbnVsbCAyPiYxDQot LS0NCj4gCSAgICBfX2tpbGxhbGwgLUhVUCBwbWNkDQozNTBjMzcwDQo8IAlr aWxsYWxsIC1IVVAgcG1jZCA+L2Rldi9udWxsIDI+JjENCi0tLQ0KPiAJX19r aWxsYWxsIC1IVVAgcG1jZA0KODAyYzgyMg0KPCAJCWtpbGxhbGwgLUhVUCBw bWNkDQotLS0NCj4gCQlfX2tpbGxhbGwgLUhVUCBwbWNkDQo4NDJjODYyDQo8 IAkgICAga2lsbGFsbCAtSFVQIHBtY2QNCi0tLQ0KPiAJICAgIF9fa2lsbGFs bCAtSFVQIHBtY2QNCjkzM2M5NTMNCjwgCSAgICBpZiBtYWtlIGluc3RhbGwN Ci0tLQ0KPiAJICAgIGlmIGdtYWtlIGluc3RhbGwNCjk3OGM5OTgNCjwgCSAg ICBraWxsYWxsIC1IVVAgcG1jZA0KLS0tDQo+IAkgICAgX19raWxsYWxsIC1I VVAgcG1jZA0KMTAwN2MxMDI3DQo8IAkgICAgbWFrZSBjbG9iYmVyID4vZGV2 L251bGwNCi0tLQ0KPiAJICAgIGdtYWtlIGNsb2JiZXIgPi9kZXYvbnVsbA0K SW5kZXg6IHNyYy9wbWNkL3JjX3BjcA0KPT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PQ0KUkNTIGZpbGU6IC91c3IvdXNlcnMvZXpvbHQvY3ZzL3JlcG9zaXRvcnkv cGNwL3NyYy9wbWNkL3JjX3BjcCx2DQpyZXRyaWV2aW5nIHJldmlzaW9uIDEu MS4xLjENCmRpZmYgLXIxLjEuMS4xIHJjX3BjcA0KODdhODgsMTA4DQo+ICMg VGhpcyBmdW5jdGlvbiB3aWxsIHNlbmQgc2lnbmFsICQxIGtpbGwgYWxsIHBy b2Nlc3NvcnMgdGhhdCBhcmUgbmFtZWQgJDINCj4gIyBMaW51eCBoYXMgYSBu aWNlIHV0aWxpdHkgKGtpbGxhbGwpIHRoYXQgd2lsbCBkbyB0aGlzLiAgSXQg aGFzIHRvIGJlIHNpbXVsYXRlZA0KPiAjIG9uIFRydTY0Lg0KPiANCj4gX19r aWxsYWxsKCkNCj4gew0KPiANCj4gICAgIGlmIFsgJFBDUF9QTEFURk9STSA9 ICJvc2YiIF0NCj4gICAgIHRoZW4NCj4gCV9nZXRfcGlkc19ieV9uYW1lICQy ID4kdG1wLmtpbGwNCj4gCWlmIFsgLXMgJHRtcC5raWxsIF0NCj4gCXRoZW4N Cj4gCSAgICBraWxsICQxIGBjYXQgJHRtcC5raWxsYCANCj4gCWZpDQo+ICAg ICBlbHNlDQo+ICAgICAgICAga2lsbGFsbCAkMSAkMiA+L2Rldi9udWxsIDI+ JjENCj4gICAgIGZpDQo+IA0KPiB9DQo+IA0KPiANCjMzOSwzNDBjMzYwLDM2 MQ0KPCBXYXJuaW5nOiBQTUNEIGNvbnRyb2wgZmlsZSAiJFBDUF9QTUNEQ09O Rl9QQVRIIiBpcyBtaXNzaW5nLCBjYW5ub3QgaWRlbnRpZnkgUE1EQXMNCjwg CSB0byBiZSB0ZXJtaW5hdGVkLicNCi0tLQ0KPiBXYXJuaW5nOiBQTUNEIGNv bnRyb2wgZmlsZSAiJFBDUF9QTUNEQ09ORl9QQVRIIiBpcyBtaXNzaW5nLCBj YW5ub3QgaWRlbnRpZnkgUE1EQXMgDQo+ICAgICAgICB0byBiZSB0ZXJtaW5h dGVkLicNCjM1M2MzNzQNCjwgCWtpbGxhbGwgLVRFUk0gYGJhc2VuYW1lICRw bWRhYCA+L2Rldi9udWxsIDI+JjEgJg0KLS0tDQo+IAlfX2tpbGxhbGwgLVRF Uk0gYGJhc2VuYW1lICRwbWRhYA0KMzY1YzM4Ng0KPCAJa2lsbGFsbCAtS0lM TCBgYmFzZW5hbWUgJHBtZGFgID4vZGV2L251bGwgMj4mMSAmDQotLS0NCj4g CV9fa2lsbGFsbCAtS0lMTCBgYmFzZW5hbWUgJHBtZGFgDQozNzhjMzk5DQo8 ICAgICBraWxsYWxsIC1URVJNIHBtY2QgPi9kZXYvbnVsbCAyPiYxDQotLS0N Cj4gICAgIF9fa2lsbGFsbCAtVEVSTSBwbWNkDQo0MDZjNDI3DQo8IAkgICAg a2lsbGFsbCAtS0lMTCBwbWNkID4vZGV2L251bGwgMj4mMQ0KLS0tDQo+IAlf X2tpbGxhbGwgLUtJTEwgcG1jZA0KNDI3LDQyOGM0NDgsNDQ5DQo8IA0KPCB3 aGlsZSBnZXRvcHRzIHYgYw0KLS0tDQo+ICAgICANCj4gd2hpbGUgZ2V0b3B0 cyB2IGMgDQo0ODFhNTAzDQo+IAkgICAgUE1DRF9DVEw9b24NCkluZGV4OiBz cmMvcG1kYXMvcG1jZC9HTlVtYWtlZmlsZQ0KPT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PQ0KUkNTIGZpbGU6IC91c3IvdXNlcnMvZXpvbHQvY3ZzL3JlcG9zaXRv cnkvcGNwL3NyYy9wbWRhcy9wbWNkL0dOVW1ha2VmaWxlLHYNCnJldHJpZXZp bmcgcmV2aXNpb24gMS4xLjEuMQ0KZGlmZiAtcjEuMS4xLjEgR05VbWFrZWZp bGUNCjI3LDI4YzI3LDI4DQo8IAlAaWYgISBncmVwIC1xICQoQ09ORl9MSU5F KSAuLi9wbWNkLmNvbmYgOyB0aGVuIFwNCjwgCSAgIGVjaG8gJChDT05GX0xJ TkUpID4+IC4uL3BtY2QuY29uZiA7IFwNCi0tLQ0KPiAJQGlmIGdyZXAgLXEg JChDT05GX0xJTkUpIC4uL3BtY2QuY29uZiA7IHRoZW4gdHJ1ZTtcDQo+IAll bHNlIGVjaG8gJChDT05GX0xJTkUpID4+IC4uL3BtY2QuY29uZiA7IFwNCklu ZGV4OiBzcmMvcG1kYXMvcG1jZC9zcmMvcG1jZC5jDQo9PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09DQpSQ1MgZmlsZTogL3Vzci91c2Vycy9lem9sdC9jdnMvcmVw b3NpdG9yeS9wY3Avc3JjL3BtZGFzL3BtY2Qvc3JjL3BtY2QuYyx2DQpyZXRy aWV2aW5nIHJldmlzaW9uIDEuMS4xLjENCmRpZmYgLXIxLjEuMS4xIHBtY2Qu Yw0KMTAwMGExMDAxLDEwMDQNCj4gCQkJCSAgICAvKiBJJ20gbm90IHN1cmUg d2hhdCB0aGlzIHNob3VsZCBiZS4uLiAtLVBHRSAqLw0KPiAjZWxpZiBkZWZp bmVkKF9fb3NmX18pDQo+ICNpZiBkZWZpbmVkKF9fYWxwaGEpDQo+IAkJCQkg ICAgImNvZmYiOw0KMTAwMmExMDA3LDEwMDgNCj4gI2VuZGlmDQo+IA0KSW5k ZXg6IHNyYy9wbWRhcy9yb29tdGVtcC9HTlVtYWtlZmlsZQ0KPT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PQ0KUkNTIGZpbGU6IC91c3IvdXNlcnMvZXpvbHQvY3Zz L3JlcG9zaXRvcnkvcGNwL3NyYy9wbWRhcy9yb29tdGVtcC9HTlVtYWtlZmls ZSx2DQpyZXRyaWV2aW5nIHJldmlzaW9uIDEuMS4xLjENCmRpZmYgLXIxLjEu MS4xIEdOVW1ha2VmaWxlDQoxNmMxNg0KPCBMTERMSUJTID0gLWxtbGFuIC1s cGNwIC1scGNwX3BtZGENCi0tLQ0KPiBMTERMSUJTID0gJChMSUJfRk9SX05B Tk9TTEVFUCkgLWxtbGFuIC1scGNwIC1scGNwX3BtZGENCjU1YTU2DQo+IAly bSAtZiBwY3ANCkluZGV4OiBzcmMvcG1kYXMvc2ltcGxlL0dOVW1ha2VmaWxl DQo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09DQpSQ1MgZmlsZTogL3Vzci91c2Vy cy9lem9sdC9jdnMvcmVwb3NpdG9yeS9wY3Avc3JjL3BtZGFzL3NpbXBsZS9H TlVtYWtlZmlsZSx2DQpyZXRyaWV2aW5nIHJldmlzaW9uIDEuMS4xLjENCmRp ZmYgLXIxLjEuMS4xIEdOVW1ha2VmaWxlDQo3NmE3Nw0KPiAJcm0gLWYgcGNw DQpJbmRleDogc3JjL3BtZGFzL3RyYWNlL0dOVW1ha2VmaWxlDQo9PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09DQpSQ1MgZmlsZTogL3Vzci91c2Vycy9lem9sdC9j dnMvcmVwb3NpdG9yeS9wY3Avc3JjL3BtZGFzL3RyYWNlL0dOVW1ha2VmaWxl LHYNCnJldHJpZXZpbmcgcmV2aXNpb24gMS4xLjEuMQ0KZGlmZiAtcjEuMS4x LjEgR05VbWFrZWZpbGUNCjI2YzI2DQo8IExMRExJQlMgPSAtbHBjcF90cmFj ZQ0KLS0tDQo+IExMRExJQlMgPSAkKFBUSFJFQURfTElCKSAtbHBjcF90cmFj ZQ0KSW5kZXg6IHNyYy9wbWRhcy90cmFjZS9wbXRyYWNlLmMNCj09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT0NClJDUyBmaWxlOiAvdXNyL3VzZXJzL2V6b2x0L2N2 cy9yZXBvc2l0b3J5L3BjcC9zcmMvcG1kYXMvdHJhY2UvcG10cmFjZS5jLHYN CnJldHJpZXZpbmcgcmV2aXNpb24gMS4xLjEuMQ0KZGlmZiAtcjEuMS4xLjEg cG10cmFjZS5jDQozM2MzMw0KPCAjaWRlbnQgIiRJZDogcG10cmFjZS5jLHYg MS4xLjEuMSAyMDAxLzA4LzAxIDEzOjMzOjAyIGV6b2x0IEV4cCAkIg0KLS0t DQo+ICNpZGVudCAiJElkOiBwbXRyYWNlLmMsdiAxLjEuMS4xIDIwMDEvMDgv MDEgMTM6MzM6MDEgZXpvbHQgRXhwICQiDQpJbmRleDogc3JjL3BtZGFzL3Ry aXZpYWwvR05VbWFrZWZpbGUNCj09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0NClJD UyBmaWxlOiAvdXNyL3VzZXJzL2V6b2x0L2N2cy9yZXBvc2l0b3J5L3BjcC9z cmMvcG1kYXMvdHJpdmlhbC9HTlVtYWtlZmlsZSx2DQpyZXRyaWV2aW5nIHJl dmlzaW9uIDEuMS4xLjENCmRpZmYgLXIxLjEuMS4xIEdOVW1ha2VmaWxlDQo3 NmE3Nw0KPiAJcm0gLWYgcGNwDQpJbmRleDogc3JjL3BtZGFzL3dlYmxvZy9w bWRhLmMNCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0NClJDUyBmaWxlOiAvdXNy L3VzZXJzL2V6b2x0L2N2cy9yZXBvc2l0b3J5L3BjcC9zcmMvcG1kYXMvd2Vi bG9nL3BtZGEuYyx2DQpyZXRyaWV2aW5nIHJldmlzaW9uIDEuMS4xLjENCmRp ZmYgLXIxLjEuMS4xIHBtZGEuYw0KNTlhNjAsNjINCj4gI2lmbmRlZiBDTE9O RV9WTQ0KPiAjZGVmaW5lIENMT05FX1ZNICAgICAgICAweDAwMDAwMTAwDQo+ ICNlbmRpZg0KSW5kZXg6IHNyYy9wbWllL3NyYy9HTlVtYWtlZmlsZQ0KPT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PQ0KUkNTIGZpbGU6IC91c3IvdXNlcnMvZXpv bHQvY3ZzL3JlcG9zaXRvcnkvcGNwL3NyYy9wbWllL3NyYy9HTlVtYWtlZmls ZSx2DQpyZXRyaWV2aW5nIHJldmlzaW9uIDEuMS4xLjENCmRpZmYgLXIxLjEu MS4xIEdOVW1ha2VmaWxlDQo0NmM0Ng0KPCAgICAgICAgCSAgc2hvdy5jIG1h dGNoX2luc3QuYyBzeXNsb2cuYw0KLS0tDQo+IAkgIHNob3cuYyBtYXRjaF9p bnN0LmMgc3lzbG9nLmMNCjYwYTYxLDYyDQo+IA0KPiBMQ0ZMQUdTID0gJChJ RUVFX01BVEhfRkxBRykNCkluZGV4OiBzcmMvcG1ucy9HTlVtYWtlZmlsZQ0K PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PQ0KUkNTIGZpbGU6IC91c3IvdXNlcnMv ZXpvbHQvY3ZzL3JlcG9zaXRvcnkvcGNwL3NyYy9wbW5zL0dOVW1ha2VmaWxl LHYNCnJldHJpZXZpbmcgcmV2aXNpb24gMS4xLjEuMQ0KZGlmZiAtcjEuMS4x LjEgR05VbWFrZWZpbGUNCjQ3YzQ3DQo8IFNURFBNSURTPSBzdGRwbWlkLmly aXggIHN0ZHBtaWQubGludXggc3RkcG1pZC5hZGQtb24gc3RkcG1pZC5sb2Nh bCBcDQotLS0NCj4gU1REUE1JRFM9IHN0ZHBtaWQuaXJpeCBzdGRwbWlkLmxp bnV4IHN0ZHBtaWQub3NmIHN0ZHBtaWQuYWRkLW9uIHN0ZHBtaWQubG9jYWwg XA0KNTJjNTINCjwgCSAgICANCi0tLQ0KPiANCkluZGV4OiBzcmMvcG1zdGF0 L0dOVW1ha2VmaWxlDQo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09DQpSQ1MgZmls ZTogL3Vzci91c2Vycy9lem9sdC9jdnMvcmVwb3NpdG9yeS9wY3Avc3JjL3Bt c3RhdC9HTlVtYWtlZmlsZSx2DQpyZXRyaWV2aW5nIHJldmlzaW9uIDEuMS4x LjENCmRpZmYgLXIxLjEuMS4xIEdOVW1ha2VmaWxlDQo0NWM0NQ0KPCBMTERM SUJTCT0gLWxwY3ANCi0tLQ0KPiBMTERMSUJTCT0gLWxwY3AgJChMSUJfRk9S X05BTk9TTEVFUCkNCkluZGV4OiBzcmMvcG1zdG9yZS9wbXN0b3JlLmMNCj09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT0NClJDUyBmaWxlOiAvdXNyL3VzZXJzL2V6 b2x0L2N2cy9yZXBvc2l0b3J5L3BjcC9zcmMvcG1zdG9yZS9wbXN0b3JlLmMs dg0KcmV0cmlldmluZyByZXZpc2lvbiAxLjEuMS4xDQpkaWZmIC1yMS4xLjEu MSBwbXN0b3JlLmMNCjM4YTM5LDQyDQo+ICNpZiBkZWZpbmVkKEhBVkVfRkxP QVRfSCkNCj4gI2luY2x1ZGUgPGZsb2F0Lmg+DQo+ICNlbmRpZg0KPiANCg== --1401954064-1937414519-998937200=:482480-- From owner-pcp@oss.sgi.com Tue Aug 28 01:34:51 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7S8YpF15758 for pcp-outgoing; Tue, 28 Aug 2001 01:34:51 -0700 Received: from mailgw3.netvision.net.il (mailgw.netvision.net.il [194.90.1.11]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7S8Yjd15752 for ; Tue, 28 Aug 2001 01:34:46 -0700 Received: from mail.exanet.co.il ([212.143.73.99]) by mailgw3.netvision.net.il (8.9.3/8.9.3) with ESMTP id LAA00007 for ; Tue, 28 Aug 2001 11:32:46 +0300 (IDT) Received: from cat (fw1.exanet.co.il [212.143.73.98]) by mail.exanet.co.il (8.11.0/8.11.0) with SMTP id f7S8VZb15396 for ; Tue, 28 Aug 2001 11:31:37 +0300 Reply-To: From: "gilly" To: Subject: bug in /proc/stat parsing on linux 2.4... Date: Tue, 28 Aug 2001 11:33:08 +0200 Message-ID: <01c401c12fa4$73542180$2a04000a@exanet> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_NextPart_000_01C5_01C12FB5.36DCF180" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2910.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Sender: owner-pcp@oss.sgi.com Precedence: bulk This is a multi-part message in MIME format. ------=_NextPart_000_01C5_01C12FB5.36DCF180 Content-Type: text/plain; charset="windows-1255" Content-Transfer-Encoding: 7bit In the linux PMDA, in proc_stat.c file (parsing of /proc/stat file) there's a different parsing for 2.4 kernels and 2.2 ones. I'm running on 2.4.6 linux kernel and in /proc/stat the line says: "disk_io:............." (disk io stats for all devices) includes 5 parameters, and not 4 as expected in proc_stat.c: instead of - (X,Y):(RIO,RBLK,WIO,WBLK), the line is - (X,Y):(TOTALIO*,RIO,RBLK,WIO,WBLK) *(TOTALIO = RIO + WIO) Attached is an example for /proc/stat file in 2.4.6 linux kernel & a fix for this bug. Thanks, gilly ------=_NextPart_000_01C5_01C12FB5.36DCF180 Content-Type: application/octet-stream; name="stat.example" Content-Transfer-Encoding: quoted-printable Content-Disposition: attachment; filename="stat.example" cpu 383770 496 413263 80606081=0A= cpu0 179726 372 169159 40352548=0A= cpu1 204044 124 244104 40253533=0A= page 714089 1246232=0A= swap 1 0=0A= intr 47836583 40701805 37 0 4 4 0 3 0 1 0 0 0 1002 0 3920650 0 0 0 0 0 = 2226314 0 0 0 0 0 0 3365 0 0 983398 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0=0A= disk_io: (3,0):(164434,84470,1428176,79964,2492464) (3,1):(1,1,2,0,0) =0A= ctxt 6445590=0A= btime 998575658=0A= processes 723756=0A= ------=_NextPart_000_01C5_01C12FB5.36DCF180 Content-Type: application/octet-stream; name="proc_stat.c.patch" Content-Transfer-Encoding: quoted-printable Content-Disposition: attachment; filename="proc_stat.c.patch" 341c341=0A= < unsigned int k, maj, dn, rio, brio, wio, bwio;=0A= ---=0A= > unsigned int k, maj, dn, iotot, rio, brio, wio, bwio;=0A= 346c346,348=0A= < if (sscanf((const char *)s, "(%u,%u):(%u,%u,%u,%u)", &maj, &dn, = &rio, &brio, &wio, &bwio) =3D=3D 6) {=0A= ---=0A= > /* the original code doesn't fit the format of /proc/stat in the=0A= > format we use the first parameter is the sum of rio & wio */=0A= > if (sscanf((const char *)s, "(%u,%u):(%u,%u,%u,%u,%u)", &maj, &dn, = &iotot, &rio, &brio, &wio, &bwio) =3D=3D 7) {=0A= ------=_NextPart_000_01C5_01C12FB5.36DCF180-- From owner-pcp@oss.sgi.com Tue Aug 28 14:34:54 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7SLYsP03523 for pcp-outgoing; Tue, 28 Aug 2001 14:34:54 -0700 Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7SLYqd03520 for ; Tue, 28 Aug 2001 14:34:52 -0700 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via SMTP id OAA09881 for ; Tue, 28 Aug 2001 14:33:07 -0700 (PDT) mail_from (markgw@sgi.com) Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.175]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id IAA00837; Wed, 29 Aug 2001 08:32:16 +1100 Date: Wed, 29 Aug 2001 07:32:16 +1000 (EST) From: Mark Goodwin X-Sender: markgw@sherman.melbourne.sgi.com To: gilly cc: pcp@oss.sgi.com Subject: Re: bug in /proc/stat parsing on linux 2.4... In-Reply-To: <01c401c12fa4$73542180$2a04000a@exanet> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Tue, 28 Aug 2001, gilly wrote: > In the linux PMDA, in proc_stat.c file (parsing of /proc/stat file) there's > a different parsing for 2.4 kernels and 2.2 ones. I'm running on 2.4.6 linux > kernel and in /proc/stat the line says: > "disk_io:............." (disk io stats for all devices) includes 5 > parameters, and not 4 as expected in proc_stat.c: > instead of - (X,Y):(RIO,RBLK,WIO,WBLK), > the line is - (X,Y):(TOTALIO*,RIO,RBLK,WIO,WBLK) > *(TOTALIO = RIO + WIO) > Attached is an example for /proc/stat file in 2.4.6 linux kernel & a fix for > this bug. thanks, it looks like you are correct and I have applied your patch. Most of the machines around here have the "sard" patch applied (which gives extra I/O accounting in /proc/partitions) and if sard stats are available then /proc/stat is not used for disk stats. -- Mark From owner-pcp@oss.sgi.com Tue Aug 28 14:42:10 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7SLgA503606 for pcp-outgoing; Tue, 28 Aug 2001 14:42:10 -0700 Received: from rj.sgi.com (rj.SGI.COM [204.94.215.100]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7SLg8d03603 for ; Tue, 28 Aug 2001 14:42:08 -0700 Received: from yog-sothoth.sgi.com (eugate.neu.sgi.com [144.253.131.5]) by rj.sgi.com (8.11.4/8.11.4/linux-outbound_gateway-1.0) with ESMTP id f7SLlb818412 for <@rj.corp.sgi.com:pcp@oss.sgi.com>; Tue, 28 Aug 2001 14:47:37 -0700 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via SMTP id XAA620574 for ; Tue, 28 Aug 2001 23:41:59 +0200 (CEST) mail_from (markgw@sgi.com) Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.175]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id IAA00884; Wed, 29 Aug 2001 08:40:37 +1100 Date: Wed, 29 Aug 2001 07:40:37 +1000 (EST) From: Mark Goodwin X-Sender: markgw@sherman.melbourne.sgi.com To: Phillip Ezolt cc: pcp@oss.sgi.com, "Stanley, Dave" , Bill French Subject: Re: Performance Co-Pilot patch for Compaq's Tru64 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Mon, 27 Aug 2001, Phillip Ezolt wrote: > I've patched the Performance Co-Pilot infrastructure to work > with Tru64. All of the clients (except for pmstat) that I've tested work. > The included PMDAs all compile except for cisco and shping. > > Known issues: > 1) cisco and pmdas do not compile. (Missing sys/prctl) > 2) No test for whether to use "hostname -f" or "hostname". > 3) Magic file format is not compatible with Tru64. > 4) Testing for the "runlevel" command is not done properly in > the shell scripts. > 5) No Tru64 specific PMDA. > > Questions/Comments: > > 1) What pmda number should I use for Tru64? (Will 74 work?) please use 75 (74 is used by lmsensors). > > 2) The memory values in the Linux pmda should be 64-bit, not 32-bit. > Problems show up when a machine has more than 4-gig of memory. oops. I'll fix that. > > Hopefully, the patch is self explanitory. I had to add some automake > checks in for things that incorrectly pcp assumed. > > If you need me to test or explain anything, just tell me! I'm wading thru your patch and will let you know of any problems. thanks -- Mark From owner-pcp@oss.sgi.com Wed Aug 29 00:17:56 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7T7HuG16389 for pcp-outgoing; Wed, 29 Aug 2001 00:17:56 -0700 Received: from mailgw3.netvision.net.il (mailgw.netvision.net.il [194.90.1.11]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7T7Hqd16385 for ; Wed, 29 Aug 2001 00:17:52 -0700 Received: from mail.exanet.co.il ([212.143.73.99]) by mailgw3.netvision.net.il (8.9.3/8.9.3) with ESMTP id KAA26888 for ; Wed, 29 Aug 2001 10:15:51 +0300 (IDT) Received: from cat (fw1.exanet.co.il [212.143.73.98]) by mail.exanet.co.il (8.11.0/8.11.0) with SMTP id f7T7ESb25219 for ; Wed, 29 Aug 2001 10:14:29 +0300 Reply-To: From: "gilly" To: Subject: RE: bug in /proc/stat parsing on linux 2.4... Date: Wed, 29 Aug 2001 10:16:07 +0200 Message-ID: <01fc01c13062$db4dc5c0$2a04000a@exanet> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2910.0) In-Reply-To: Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Sender: owner-pcp@oss.sgi.com Precedence: bulk do you know if there's such a patch for linux kernel 2.4.6, since it seems the structures are changed and the sard patch versions I can find are not compatible to this kernel. thanks in advance, gilly -----Original Message----- From: Mark Goodwin [mailto:markgw@sgi.com] Sent: Tuesday, August 28, 2001 11:32 PM To: gilly Cc: pcp@oss.sgi.com Subject: Re: bug in /proc/stat parsing on linux 2.4... On Tue, 28 Aug 2001, gilly wrote: > In the linux PMDA, in proc_stat.c file (parsing of /proc/stat file) there's > a different parsing for 2.4 kernels and 2.2 ones. I'm running on 2.4.6 linux > kernel and in /proc/stat the line says: > "disk_io:............." (disk io stats for all devices) includes 5 > parameters, and not 4 as expected in proc_stat.c: > instead of - (X,Y):(RIO,RBLK,WIO,WBLK), > the line is - (X,Y):(TOTALIO*,RIO,RBLK,WIO,WBLK) > *(TOTALIO = RIO + WIO) > Attached is an example for /proc/stat file in 2.4.6 linux kernel & a fix for > this bug. thanks, it looks like you are correct and I have applied your patch. Most of the machines around here have the "sard" patch applied (which gives extra I/O accounting in /proc/partitions) and if sard stats are available then /proc/stat is not used for disk stats. -- Mark From owner-pcp@oss.sgi.com Wed Aug 29 04:07:05 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7TB75d20862 for pcp-outgoing; Wed, 29 Aug 2001 04:07:05 -0700 Received: from rj.sgi.com (rj.sgi.com [204.94.215.100]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7TB72d20859 for ; Wed, 29 Aug 2001 04:07:02 -0700 Received: from yog-sothoth.sgi.com (eugate.neu.sgi.com [144.253.131.5]) by rj.sgi.com (8.11.4/8.11.4/linux-outbound_gateway-1.0) with ESMTP id f7TBCT828955 for <@rj.corp.sgi.com:pcp@oss.sgi.com>; Wed, 29 Aug 2001 04:12:29 -0700 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via SMTP id NAA646952 for ; Wed, 29 Aug 2001 13:06:44 +0200 (CEST) mail_from (markgw@sgi.com) Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.175]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id WAA03388; Wed, 29 Aug 2001 22:05:24 +1100 Date: Wed, 29 Aug 2001 21:05:23 +1000 (EST) From: Mark Goodwin X-Sender: markgw@sherman.melbourne.sgi.com To: gilly cc: pcp@oss.sgi.com Subject: RE: bug in /proc/stat parsing on linux 2.4... In-Reply-To: <01fc01c13062$db4dc5c0$2a04000a@exanet> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Wed, 29 Aug 2001, gilly wrote: > do you know if there's such a patch for linux kernel 2.4.6, since it seems > the structures are changed and the sard patch versions I can find are not > compatible to this kernel. > thanks in advance, gilly The last message I know of from Stephen C. Tweedie (sard author) is more than a year old, as follows: > Date: Mon, 17 Jul 2000 17:44:12 +0100 > From: "Stephen C. Tweedie" > To: Linux Performance Monitoring , > linux-fsdevel@vger.rutgers.edu > Cc: Chris Evans , Stephen Tweedie , > Alan Cox , linux-kernel@vger.rutgers.edu > Subject: [Linux-Perf] sard-0.6 released > > Hi all, > > sard-0.6 is being uploaded to > > ftp.uk.linux.org:/pub/linux/sct/fs/profiling/ > > Included are the sard disk profiling patches for Red Hat 2.2.16-3 > kernels, plain 2.2.17pre12, and for 2.4.0test5pre1. > > This patch also includes a modified version of atsar-1.5 (the Linux > "sar" clone) which understands the newer sard kernel output format. > There are no changes in functionality in this version, although the > 2.4 patch has been cleaned up somewhat. > > Cheers, > Stephen > From owner-pcp@oss.sgi.com Wed Aug 29 10:29:31 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7THTVR28775 for pcp-outgoing; Wed, 29 Aug 2001 10:29:31 -0700 Received: from zcamail04.zca.compaq.com (zcamail04.zca.compaq.com [161.114.32.104]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7THTSd28741 for ; Wed, 29 Aug 2001 10:29:28 -0700 Received: by zcamail04.zca.compaq.com (Postfix, from userid 12345) id D7DDD1B41; Wed, 29 Aug 2001 10:31:49 -0700 (PDT) Received: from mailrelay01.cce.cpqcorp.net (mailrelay01.cce.cpqcorp.net [16.47.68.171]) by zcamail04.zca.compaq.com (Postfix) with ESMTP id A7F1C1883; Wed, 29 Aug 2001 10:31:49 -0700 (PDT) Received: by mailrelay01.cce.cpqcorp.net (Postfix, from userid 12345) id 6B9E499A; Wed, 29 Aug 2001 12:29:17 -0500 (CDT) Received: from perf.zko.dec.com (aseng.zko.dec.com [16.31.80.86]) by mailrelay01.cce.cpqcorp.net (Postfix) with ESMTP id ABAA6AF3; Wed, 29 Aug 2001 12:29:16 -0500 (CDT) Received: from localhost by perf.zko.dec.com (8.9.3/1.1.10.5/07Mar97-1056AM) id NAA0000042934; Wed, 29 Aug 2001 13:29:16 -0400 (EDT) Date: Wed, 29 Aug 2001 13:29:15 -0400 (EDT) From: Phillip Ezolt To: Mark Goodwin Cc: Subject: Re: Performance Co-Pilot patch for Compaq's Tru64 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk Mark, > please use 75 (74 is used by lmsensors). > Ok. I'll change my code here. One question for you, I am trying to implement the osf code as close as possible to the Linux code. I'm trying to keep as many metrics as possible the same. What should I do with NEW metrics? Is it ok to add them, or is there some reason not to? For example, in the network driver, Tru64 can provide the following information for Multi-casts: 0 Multicast Packets Recvd 0 Multicast Bytes Received 0 Multicast Packets Sent 0 Multicast Bytes Sent However, network.interface.out network.interface.out { bytes 60:3:8 packets 60:3:9 errors 60:3:10 drops 60:3:11 fifo 60:3:12 carrier 60:3:14 compressed 60:3:15 } doesn't say ANYTHING about: network.interface.out.mcast.packets network.interface.out.mcast.bytes Should I just add it, or is there some standard that we should agree on? --Phil Compaq: High Performance Server Systems Quality & Performance Engineering ---------------- Alpha, The Fastest Processor on Earth -------------------- Phillip.Ezolt@compaq.com Performance Tools/Analysis ------------------- See the results at www.spec.org ----------------------- From owner-pcp@oss.sgi.com Thu Aug 30 16:45:09 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7UNj9w02182 for pcp-outgoing; Thu, 30 Aug 2001 16:45:09 -0700 Received: from rj.sgi.com (rj.sgi.com [204.94.215.100]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7UNj7d02178 for ; Thu, 30 Aug 2001 16:45:07 -0700 Received: from yog-sothoth.sgi.com (eugate.neu.sgi.com [144.253.131.5]) by rj.sgi.com (8.11.4/8.11.4/linux-outbound_gateway-1.0) with ESMTP id f7UNoh803947 for <@rj.corp.sgi.com:pcp@oss.sgi.com>; Thu, 30 Aug 2001 16:50:43 -0700 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via SMTP id BAA732537 for ; Fri, 31 Aug 2001 01:44:58 +0200 (CEST) mail_from (markgw@sgi.com) Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.175]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id KAA08967; Fri, 31 Aug 2001 10:43:38 +1100 Date: Fri, 31 Aug 2001 09:43:38 +1000 (EST) From: Mark Goodwin X-Sender: markgw@sherman.melbourne.sgi.com To: Brian Harvell cc: pcp@oss.sgi.com Subject: Re: --prefix does not work. In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Thu, 30 Aug 2001, Brian Harvell wrote: > > Sounds good. Out of curiosity why wouldn't you stick it in libdir/ia64 > or level with libdir in lib64 or some other like name. > > If I recall that's what Irix does for n32 vs 64 vs 32 bit libs. yeah that's what IRIX does and IMO it's the most suitable solution to the problem, but it's not the model adopted by trillian (aka linux for ia64). The crux of the trillian model is that ia32 apps on ia64 are a hack and perform poorly. In the open source world, the src should be ported to ia64 and the apps should run native ia64, i.e. the ia32 libs problem is a non-problem. So the resulting trillian model for ia32 libs is adhoc - we had discussions about this on the ia64 lists and none of the vendors (redhat, suse, turbo) seemed to care two hoots, much less would they agree on a path/FS hierarchy for ia32 libs on ia64. As a result, PCP installs ia32 libs in a private directory (currently in $PCP_SHARE_DIR/lib/compat) and we link ia32 apps using -rpath so they'll find their libs correctly. This seems to work just fine .. for now. All this is needed for the pcp-pro RPM on ia64, which has an app linked with a 3rd party static library (for which we don't have the source). thanks -- Mark From owner-pcp@oss.sgi.com Thu Aug 30 17:28:21 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id f7V0SLE02785 for pcp-outgoing; Thu, 30 Aug 2001 17:28:21 -0700 Received: from rj.sgi.com (rj.sgi.com [204.94.215.100]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id f7V0SHd02779 for ; Thu, 30 Aug 2001 17:28:17 -0700 Received: from yog-sothoth.sgi.com (eugate.neu.sgi.com [144.253.131.5]) by rj.sgi.com (8.11.4/8.11.4/linux-outbound_gateway-1.0) with ESMTP id f7V0Xq806765 for <@rj.corp.sgi.com:pcp@oss.sgi.com>; Thu, 30 Aug 2001 17:33:52 -0700 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via SMTP id CAA738101 for ; Fri, 31 Aug 2001 02:28:06 +0200 (CEST) mail_from (markgw@sgi.com) Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.175]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id LAA09122; Fri, 31 Aug 2001 11:24:07 +1100 Date: Fri, 31 Aug 2001 10:24:07 +1000 (EST) From: Mark Goodwin X-Sender: markgw@sherman.melbourne.sgi.com To: Phillip Ezolt cc: pcp@oss.sgi.com Subject: Re: Performance Co-Pilot patch for Compaq's Tru64 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Wed, 29 Aug 2001, Phillip Ezolt wrote: > One question for you, I am trying to implement the osf code as close > as possible to the Linux code. > > I'm trying to keep as many metrics as possible the same. > > What should I do with NEW metrics? Is it ok to add them, or is there some > reason not to? You should use the same PMNS name for metrics that have the same semantics and interpretation as existing metrics on other platforms (currently IRIX and linux). Note there are much more metrics on IRIX than on Linux, so you should also be checking the IRIX name space as well as linux. For new metrics, you should add them into your PMNS in the most suitable place. That's the only rule that makes sense, and is _far_ better than restricting the PMNS to be just the set of metrics that are common on all platforms (which is the SNMP mib-II model). > > For example, in the network driver, Tru64 can provide the following information > for Multi-casts: > ... > doesn't say ANYTHING about: > > network.interface.out.mcast.packets > network.interface.out.mcast.bytes > > Should I just add it, or is there some standard that we should agree on? > In this case Linux doesn't have network.interface.out.mcast.packets (it's not reported in /proc/net/dev for some reason). IRIX does, but only packet counts are available: snort 4% pminfo -d -t network.interface.{in,out,total}.mcasts network.interface.in.mcasts [packets received via broad/multicast on network interface [MIB-II]] Data Type: 32-bit unsigned int InDom: 1.6 0x400006 Semantics: counter Units: count network.interface.out.mcasts [packets sent via broad/multicast on network interface [MIB-II]] Data Type: 32-bit unsigned int InDom: 1.6 0x400006 Semantics: counter Units: count network.interface.total.mcasts [total packets sent and received via broad/multicast on network interface [MIB-II]] Data Type: 32-bit unsigned int InDom: 1.6 0x400006 Semantics: counter Units: count As it turns out, techpubs.sgi.com is an IRIX box on the public internet and has PCP installed. So for your reference you can query it's PMNS using pminfo, e.g. # pminfo -mdt -h techpubs.sgi.com network.interface.out.mcasts network.interface.out.mcasts PMID: 1.25.13 [packets sent via broad/multicast on network interface [MIB-II]] Data Type: 32-bit unsigned int InDom: 1.6 0x400006 Semantics: counter Units: count thanks -- Mark