From owner-pcp@oss.sgi.com Fri Jan 4 11:10:01 2002 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id g04JA1L18075 for pcp-outgoing; Fri, 4 Jan 2002 11:10:01 -0800 Received: from e3.ny.us.ibm.com (e3.ny.us.ibm.com [32.97.182.103]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id g04J9vg18070 for ; Fri, 4 Jan 2002 11:09:57 -0800 Received: from southrelay01.raleigh.ibm.com (southrelay01.raleigh.ibm.com [9.37.3.208]) by e3.ny.us.ibm.com (8.9.3/8.9.3) with ESMTP id NAA120102 for ; Fri, 4 Jan 2002 13:06:52 -0500 Received: from WMMASON2 (sig-9-15-94-145.mts.ibm.com [9.15.94.145]) by southrelay01.raleigh.ibm.com (8.11.1m3/NCO v5.01) with SMTP id g04I9mB177470 for ; Fri, 4 Jan 2002 13:09:48 -0500 Message-ID: <014401c1954a$c735e0c0$ad7ba8c0@beaverton.ibm.com> From: "Mike Mason" To: Subject: kernel.all.uptime units Date: Fri, 4 Jan 2002 10:08:13 -0800 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 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: owner-pcp@oss.sgi.com Precedence: bulk Why is the kernel.all.uptime metric provided in hour units rather than minutes or seconds? I'm looking into converting "top" to use PCP. Top displays uptime in days, hours and minutes, but I can't get an accurate minute count out of PCP. Can kernel.all.uptime be changed to minute units or do I need to add another uptime metric for that? It seems that, as a general rule, metrics should be provided at the smallest reasonable units, then let the client programs convert to larger units if desired. Is there a general rule like this that PCP developers follow? Regards, Mike --- Mike Mason Software Engineer IBM Linux Technology Center, RAS Group Beaverton, Oregon, USA phone: (503) 578-4123 tie line: 775-4123 mmlnx@us.ibm.com fax: (503) 578-4700 From owner-pcp@oss.sgi.com Fri Jan 4 12:36:42 2002 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id g04Kagc21500 for pcp-outgoing; Fri, 4 Jan 2002 12:36:42 -0800 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 g04Kabg21492 for ; Fri, 4 Jan 2002 12:36:37 -0800 Received: from rattle.melbourne.sgi.com (rattle.melbourne.sgi.com [134.14.55.145]) by zok.sgi.com (8.11.4/8.11.4/linux-outbound_gateway-1.1) with ESMTP id g04JaTA10615 for ; Fri, 4 Jan 2002 11:36:29 -0800 Received: from localhost (kenmcd@localhost) by rattle.melbourne.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id GAA61489; Sat, 5 Jan 2002 06:35:10 +1100 (EST) X-Authentication-Warning: rattle.melbourne.sgi.com: kenmcd owned process doing -bs Date: Sat, 5 Jan 2002 06:35:10 +1100 From: Ken McDonell Reply-To: kenmcd@sgi.com To: Mike Mason cc: pcp@oss.sgi.com Subject: Re: kernel.all.uptime units In-Reply-To: <014401c1954a$c735e0c0$ad7ba8c0@beaverton.ibm.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Fri, 4 Jan 2002, Mike Mason wrote: > Why is the kernel.all.uptime metric provided in hour units rather than > minutes or seconds? No good reason that I can see ... gilly@exanet.com provided the code and markgw@sgi.com did the integration into the open source code base. It seems to divide by 60 * 60, so the base data is obviously in seconds. The only issue I can think of is overflow ... if the system is up more than 136 years, the 32-bit unsigned long will overflow when counting in seconds ... 8^)> > I'm looking into converting "top" to use PCP. Top > displays uptime in days, hours and minutes, but I can't get an accurate > minute count out of PCP. Can kernel.all.uptime be changed to minute units > or do I need to add another uptime metric for that? Unless someone can contribute a good reason, I'd suggest just change it and send us the patch. > It seems that, as a general rule, metrics should be provided at the smallest > reasonable units, then let the client programs convert to larger units if > desired. ... Agreed. > ... Is there a general rule like this that PCP developers follow? For the most part we have followed this rule. The one caveat I'd make is that I am not going to export metrics in bozo units like ticks and blocks because they are meaningless in the larger context of multiple OSes and multiple OS versions and the passage of time. We have always converted these to msec and Kbyte (in both cases this does not _reduce_ the precision, but it does bring on some overflow problems that need to be addressed through careful arithmetic or extending the precision of the exported metric). From owner-pcp@oss.sgi.com Fri Jan 4 18:50:42 2002 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id g052og428800 for pcp-outgoing; Fri, 4 Jan 2002 18:50:42 -0800 Received: from e1.ny.us.ibm.com (e1.ny.us.ibm.com [32.97.182.101]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id g052oag28797 for ; Fri, 4 Jan 2002 18:50:36 -0800 Received: from southrelay01.raleigh.ibm.com (southrelay01.raleigh.ibm.com [9.37.3.208]) by e1.ny.us.ibm.com (8.9.3/8.9.3) with ESMTP id UAA499824; Fri, 4 Jan 2002 20:47:31 -0500 Received: from WMMASON2 (sig-9-15-94-145.mts.ibm.com [9.15.94.145]) by southrelay01.raleigh.ibm.com (8.11.1m3/NCO v5.01) with SMTP id g051oRB61812; Fri, 4 Jan 2002 20:50:28 -0500 Message-ID: <019d01c1958b$215eac90$ad7ba8c0@beaverton.ibm.com> From: "Mike Mason" To: Cc: References: Subject: Re: kernel.all.uptime units Date: Fri, 4 Jan 2002 17:48:52 -0800 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 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: owner-pcp@oss.sgi.com Precedence: bulk On Fri, 4 Jan 2002, Ken McDonell wrote: >> I'm looking into converting "top" to use PCP. Top >> displays uptime in days, hours and minutes, but I can't get an accurate >> minute count out of PCP. Can kernel.all.uptime be changed to minute units >> or do I need to add another uptime metric for that? > >Unless someone can contribute a good reason, I'd suggest just change it >and send us the patch. OK, here's a patch that changes the units for kernel.all.uptime from hours to seconds. It applies against pcp-2.2.2. Regards, Mike --- src/pmdas/linux/pmda.c.orig Fri Jan 4 17:01:43 2002 +++ src/pmdas/linux/pmda.c Fri Jan 4 16:21:24 2002 @@ -427,7 +427,7 @@ /* kernel.all.uptime */ { NULL, { PMDA_PMID(CLUSTER_STAT,49), PM_TYPE_U32, PM_INDOM_NULL, PM_SEM_INSTANT, - PMDA_PMUNITS(0,1,1,0,PM_TIME_HOUR,PM_COUNT_ONE) }, }, + PMDA_PMUNITS(0,1,1,0,PM_TIME_SEC,PM_COUNT_ONE) }, }, /* @@ -2641,10 +2641,10 @@ case 49: /* - * kernel.all.uptime in hours + * kernel.all.uptime in seconds * contributed by 'gilly' */ - atom->ul = (u_long)(times(NULL)) / (proc_stat.hz * 60 * 60); + atom->ul = (u_long)(times(NULL)) / (proc_stat.hz); break; default: --- Mike Mason Software Engineer IBM Linux Technology Center, RAS Group Beaverton, Oregon, USA phone: (503) 578-4123 tie line: 775-4123 mmlnx@us.ibm.com fax: (503) 578-4700 From owner-pcp@oss.sgi.com Sun Jan 6 17:30:47 2002 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id g071UlQ22665 for pcp-outgoing; Sun, 6 Jan 2002 17:30:47 -0800 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 g071Ujg22662 for ; Sun, 6 Jan 2002 17:30:45 -0800 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-1.1) with SMTP id g070UaY23349 for ; Sun, 6 Jan 2002 16:30:36 -0800 Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.232]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id LAA26484; Mon, 7 Jan 2002 11:29:14 +1100 Date: Mon, 7 Jan 2002 11:29:14 +1100 (EST) From: Mark Goodwin X-X-Sender: To: Mike Mason cc: Subject: Re: kernel.all.uptime units In-Reply-To: <014401c1954a$c735e0c0$ad7ba8c0@beaverton.ibm.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Fri, 4 Jan 2002, Mike Mason wrote: > Why is the kernel.all.uptime metric provided in hour units ... Thanks Mike, I have added your patch. Are you converting top (as shipped in the procps package), or libgtop (as used by gtop, shipped in the gtop package)? -- Mark From owner-pcp@oss.sgi.com Mon Jan 7 11:19:45 2002 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id g07JJje21752 for pcp-outgoing; Mon, 7 Jan 2002 11:19:45 -0800 Received: from e3.ny.us.ibm.com (e3.ny.us.ibm.com [32.97.182.103]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id g07JJfg21748 for ; Mon, 7 Jan 2002 11:19:41 -0800 Received: from northrelay01.pok.ibm.com (northrelay01.pok.ibm.com [9.117.200.21]) by e3.ny.us.ibm.com (8.9.3/8.9.3) with ESMTP id NAA14938; Mon, 7 Jan 2002 13:16:35 -0500 Received: from WMMASON2 (dyn9-47-24-207.wil.beaverton.ibm.com [9.47.24.207]) by northrelay01.pok.ibm.com (8.11.1m3/NCO v5.01) with SMTP id g07IJUW112300; Mon, 7 Jan 2002 13:19:30 -0500 Message-ID: <001001c197a7$a120f9b0$cf182f09@beaverton.ibm.com> From: "Mike Mason" To: "Mark Goodwin" Cc: References: Subject: Re: kernel.all.uptime units Date: Mon, 7 Jan 2002 10:17:54 -0800 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 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: owner-pcp@oss.sgi.com Precedence: bulk I'm looking at both top and libgtop. I haven't decided yet if I'm going to convert one or both of them. ----- Original Message ----- From: "Mark Goodwin" To: "Mike Mason" Cc: Sent: Sunday, January 06, 2002 4:29 PM Subject: Re: kernel.all.uptime units > On Fri, 4 Jan 2002, Mike Mason wrote: > > > Why is the kernel.all.uptime metric provided in hour units ... > > Thanks Mike, I have added your patch. > > Are you converting top (as shipped in the procps package), or libgtop > (as used by gtop, shipped in the gtop package)? > > -- Mark > > From owner-pcp@oss.sgi.com Tue Jan 15 21:21:10 2002 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id g0G5LAU15026 for pcp-outgoing; Tue, 15 Jan 2002 21:21:10 -0800 Received: from sgi.com (sgi.SGI.COM [192.48.153.1]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id g0G5L0P15023 for ; Tue, 15 Jan 2002 21:21:00 -0800 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 UAA01120 for ; Tue, 15 Jan 2002 20:20:38 -0800 (PST) mail_from (markgw@sgi.com) Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.232]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id PAA20687; Wed, 16 Jan 2002 15:19:36 +1100 Date: Wed, 16 Jan 2002 15:19:36 +1100 (EST) From: Mark Goodwin X-X-Sender: To: Mike Mason cc: Subject: Re: Representing task memory maps in PCP In-Reply-To: <001e01c19e19$10ccc1f0$9e7ba8c0@papabear> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Tue, 15 Jan 2002, Mike Mason wrote: > One of the things that libgtop wants for each task is a table of the > task's memory maps. I can get this from /proc//maps. Here's an > example of the maps from a task on my system. The list for most tasks > is longer than this one. > > 8048000-080ad000 r-xp 00000000 03:02 1240437 /sbin/dhcpcd > 080ad000-080b0000 rw-p 00064000 03:02 1240437 /sbin/dhcpcd > 080b0000-080b5000 rwxp 00000000 00:00 0 > 40011000-4001a000 r-xp 00000000 03:02 65339 /lib/libnss_files-2.2.4.so > 4001a000-4001c000 rw-p 00008000 03:02 65339 /lib/libnss_files-2.2.4.so > 4001c000-40148000 r-xp 00000000 03:02 65314 /lib/libc-2.2.4.so > 40148000-4014d000 rw-p 0012b000 03:02 65314 /lib/libc-2.2.4.so > 4014d000-40151000 rw-p 00000000 00:00 0 > 40151000-40166000 r-xp 00000000 03:02 65297 /lib/ld-2.2.4.so > 40166000-40167000 rw-p 00014000 03:02 65297 /lib/ld-2.2.4.so > bfffd000-c0000000 rwxp ffffe000 00:00 0 > > PCP doesn't provide task memory maps right now. About the only way I > can see to represent this within the PCP namespace is too include the > whole table as a single binary or string metric. Seems a bit kludgy, > but I don't see any other way to do it. > > Any suggestions? If you really have to, you can export the whole table for each process - there is no technical reason preventing embedded new lines in metrics with string values, and the length limit is 2^24 (I think). How does libgtop use the fields in the table? Most likely it wants to work out per-process virtual and rss mem usage. In that case, you could do the sums in PCP, and then export a suitable set of metrics, e.g. very similar to the following metrics that are exported by PCP on IRIX : proc.memory.virtual.txt [per-process executable text virtual memory usage] proc.memory.virtual.dat [per-process initialized data virtual memory usage] proc.memory.virtual.bss [per-process uninitialized data and break segment virtual memory usage] proc.memory.virtual.stack [per-process stack segment virtual memory usage] proc.memory.virtual.shm [per-process shared-memory segment virtual memory usage] proc.memory.physical.txt [per-process executable text physical memory usage] proc.memory.physical.dat [per-process initialized data physical memory usage] proc.memory.physical.bss [per-process uninitialized data and break segment physical memory usage] proc.memory.physical.stack [per-process stack segment physical memory usage] proc.memory.physical.shm [per-process shared-memory segment physical memory usage] With a suitable heuristic you can tell the mapping type (e.g. bss, txt, dat, etc) from the mode. The virtual size can be inferred from the high-low addresss of the mappings and the physical size can be inferred from the resident page count. Unfortunately there doesn't seem to be any refcnt field that would allow you to determine pro-rata rss (i.e. the real rss after accounting for sharing). > BTW -- would you prefer that I post this type of question to the mailing > list or email you directly? Also, should I communicate primarily with > just one of you? If so, who should that be? This is open source so posting to the pcp@oss list is fine. thanks -- Mark From owner-pcp@oss.sgi.com Wed Jan 30 15:28:13 2002 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id g0UNSDU26998 for pcp-outgoing; Wed, 30 Jan 2002 15:28:13 -0800 Received: from e31.co.us.ibm.com (e31.co.us.ibm.com [32.97.110.129]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id g0UNS8d26993 for ; Wed, 30 Jan 2002 15:28:09 -0800 Received: from westrelay01.boulder.ibm.com (westrelay01.boulder.ibm.com [9.99.140.22]) by e31.co.us.ibm.com (8.9.3/8.9.3) with ESMTP id RAA22810 for ; Wed, 30 Jan 2002 17:24:59 -0500 Received: from papabear (sig-9-15-111-236.mts.ibm.com [9.15.111.236]) by westrelay01.boulder.ibm.com (8.11.1m3/NCO v5.01) with SMTP id g0UMS4k78928 for ; Wed, 30 Jan 2002 15:28:05 -0700 Message-ID: <00c601c1a9dd$64224a80$9e7ba8c0@papabear> From: "Mike Mason" To: Subject: A few PCP questions... Date: Wed, 30 Jan 2002 14:28:07 -0800 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 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: owner-pcp@oss.sgi.com Precedence: bulk Here are a few questions that I've been asked lately... - Do any of the distros include PCP by default? - What tools (proprietary or open source) use PCP's PMAPI for data gathering on Linux? - Any idea who's using the open source version of PCP outside of SGI? Help with answers would be greatly appreciated. Thanks, Mike --- Mike Mason Software Engineer IBM Linux Technology Center, RAS Group Beaverton, Oregon, USA phone: (503) 578-4123 tie line: 775-4123 mmlnx@us.ibm.com fax: (503) 578-4700 From owner-pcp@oss.sgi.com Wed Jan 30 16:21:36 2002 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id g0V0La129401 for pcp-outgoing; Wed, 30 Jan 2002 16:21:36 -0800 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 g0V0LRd29380 for ; Wed, 30 Jan 2002 16:21:27 -0800 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 AAA109672 for ; Thu, 31 Jan 2002 00:19:05 +0100 (CET) mail_from (markgw@sgi.com) Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.232]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id KAA18239; Thu, 31 Jan 2002 10:20:00 +1100 Date: Thu, 31 Jan 2002 10:20:00 +1100 (EST) From: Mark Goodwin X-X-Sender: To: Mike Mason cc: Subject: Re: A few PCP questions... In-Reply-To: <00c601c1a9dd$64224a80$9e7ba8c0@papabear> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pcp@oss.sgi.com Precedence: bulk On Wed, 30 Jan 2002, Mike Mason wrote: > Here are a few questions that I've been asked lately... > > - Do any of the distros include PCP by default? Suse 7.x does (or used to). Redhat does not. Don't know about others, > - What tools (proprietary or open source) use PCP's PMAPI for data gathering on Linux? Proprietary: SGI has a large mature suite of 2D and 3D GUI monitoring tools built over the PMAPI, see http://www.sgi.com/software/co-pilot/ This has been ported to Linux, but you can't buy it without buying SGI hardware. Exactly the same tools are available for IRIX. GPL: pcpmon provides a graphical stripchart. I think pcpmon's home page is http://k332.feld.cvut.cz/~lemming/projects/pcpmon.html There's other stuff and we'd like to see more GPL tools ported over to using PCP (e.g. libgtop). Unfortunately, SGI doesn't have the resources to spend time porting existing tools to use PCP, so it really has to be a community effort. There have been some great contributions in terms of bug fixes and new functionality (e.g. new PMDA agents), but not much in terms of converting existing GPL performance tools. > - Any idea who's using the open source version of PCP outside of SGI? > Well, we seem to get about 30 downloads from oss.sgi.com per week. This varies a lot, depending on release timeframes, etc. AFAICT, most people are using the open source PCP tools in scripts for monitoring system availability and performance, and archive logging for performance audits and/or capacity planning kinda stuff. Mostly using cmdline tools such as pmprobe, pminfo and pmie in assorted homebrew scripts, etc., especially popular with Beowulf style clusters. > Help with answers would be greatly appreciated. > it would be good if others chipped in with their experiences. -- Mark From owner-pcp@oss.sgi.com Wed Jan 30 16:51:38 2002 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id g0V0pcx30194 for pcp-outgoing; Wed, 30 Jan 2002 16:51:38 -0800 Received: from mail11.speakeasy.net (mail11.speakeasy.net [216.254.0.211]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id g0V0pad30190 for ; Wed, 30 Jan 2002 16:51:36 -0800 Received: (qmail 6571 invoked from network); 30 Jan 2002 23:51:28 -0000 Received: from unknown (HELO PATCHES) ([216.254.47.170]) (envelope-sender ) by mail11.speakeasy.net (qmail-ldap-1.03) with SMTP for ; 30 Jan 2002 23:51:29 -0000 From: "Corey Cole" To: "SGI-PCP" Subject: PCP on Solaris Date: Wed, 30 Jan 2002 16:51:37 -0700 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: owner-pcp@oss.sgi.com Precedence: bulk Having nothing better to do with my time, I'm starting on a port of PCP to Solaris. So far I've figured out how to modify configure.in to create a Solaris friendly configure script. What's the next available pmda number? From earlier messages it looks like 75 is reserved for Tru64. Anybody got 76? I'll mail a patch once I get something that runs. Regards, Corey Cole From owner-pcp@oss.sgi.com Wed Jan 30 17:05:00 2002 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id g0V150930526 for pcp-outgoing; Wed, 30 Jan 2002 17:05:00 -0800 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 g0V14vd30523 for ; Wed, 30 Jan 2002 17:04:57 -0800 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-1.1) with SMTP id g0V04mY19494 for ; Wed, 30 Jan 2002 16:04:48 -0800 Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.232]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id LAA18467; Thu, 31 Jan 2002 11:03:29 +1100 Date: Thu, 31 Jan 2002 11:03:29 +1100 (EST) From: Mark Goodwin X-X-Sender: To: Corey Cole cc: SGI-PCP Subject: Re: PCP on Solaris 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, 30 Jan 2002, Corey Cole wrote: > Having nothing better to do with my time, I'm starting on a port > of PCP to Solaris. So far I've figured out how to modify configure.in > to create a Solaris friendly configure script. > > What's the next available pmda number? From earlier messages it > looks like 75 is reserved for Tru64. Anybody got 76? use 76 for solaris, thanks. --- /usr/tmp/TmpDir.6702491-0/src/pmns/stdpmid.pcp_1.6 Thu Jan 31 11:01:56 2002 +++ src/pmns/stdpmid.pcp Thu Jan 31 11:01:04 2002 @@ -31,6 +31,8 @@ MOUNTS 72 SNIFF 73 LMSENSORS 74 +TRU64 75 +SOLARIS 76 COUNTD 245 LAB 246 TXMON 248 From owner-pcp@oss.sgi.com Wed Jan 30 19:13:16 2002 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id g0V3DGN01494 for pcp-outgoing; Wed, 30 Jan 2002 19:13:16 -0800 Received: from blaise.ens-lyon.fr (mail@dhcp20-127.wirelesstest.ucsb.edu [128.111.20.127]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id g0V3DBd01491 for ; Wed, 30 Jan 2002 19:13:11 -0800 Received: from mquinson by blaise.ens-lyon.fr with local (Exim 3.34 #1 (Debian)) id 16W6fg-0000CI-00 for ; Wed, 30 Jan 2002 18:09:52 -0800 Date: Wed, 30 Jan 2002 18:09:52 -0800 To: pcp@oss.sgi.com Subject: Re: A few PCP questions... Message-ID: <20020131020952.GX21688@blaise.ens-lyon.fr> References: <00c601c1a9dd$64224a80$9e7ba8c0@papabear> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <00c601c1a9dd$64224a80$9e7ba8c0@papabear> User-Agent: Mutt/1.3.27i From: Martin Quinson Sender: owner-pcp@oss.sgi.com Precedence: bulk On Wed, Jan 30, 2002 at 02:28:07PM -0800, Mike Mason wrote: > Here are a few questions that I've been asked lately... > > - Do any of the distros include PCP by default? It is definitively not in the stable version of Debian. Some packages are underway for the developper version, but nothing available yet, AFAIK. More details under: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=105587&repeatmerged=yes > - What tools (proprietary or open source) use PCP's PMAPI for data gathering on Linux? No idea > - Any idea who's using the open source version of PCP outside of SGI? icluster is a cluster based in france, with about 200 PC. It uses PCP, and you can query it at this address: http://www-id.imag.fr/cgi-bin/pcp Bye, Mt. From owner-pcp@oss.sgi.com Thu Jan 31 07:27:13 2002 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id g0VFRDt21004 for pcp-outgoing; Thu, 31 Jan 2002 07:27:13 -0800 Received: from fnal.gov (heffalump.fnal.gov [131.225.9.20]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id g0VFR6d21000 for ; Thu, 31 Jan 2002 07:27:07 -0800 Received: from fnal.gov ([131.225.80.75]) by smtp.fnal.gov (PMDF V6.0-24 #37519) with ESMTP id <0GQT00LKF44QFM@smtp.fnal.gov> for pcp@oss.sgi.com; Thu, 31 Jan 2002 08:26:50 -0600 (CST) Date: Thu, 31 Jan 2002 08:27:01 -0600 From: Troy Dawson Subject: Re: A few PCP questions... To: Mike Mason Cc: pcp@oss.sgi.com Message-id: <3C595435.5070907@fnal.gov> MIME-version: 1.0 Content-type: text/plain; format=flowed; charset=us-ascii Content-transfer-encoding: 7bit X-Accept-Language: en-us User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.7) Gecko/20011226 References: <00c601c1a9dd$64224a80$9e7ba8c0@papabear> Sender: owner-pcp@oss.sgi.com Precedence: bulk Mike Mason wrote: > Here are a few questions that I've been asked lately... > > - Do any of the distros include PCP by default? FermiLinux does (it's not exactly as large a distribution as redhat but ... hey ... it's technically a distribution) > - What tools (proprietary or open source) use PCP's PMAPI for data gathering on Linux? > - Any idea who's using the open source version of PCP outside of SGI? There are areas at Fermilab that use it. Depending on what group you talk to you either don't know about it, or use it alot. My group mainly use it for troubleshooting our linux clusters. > > Help with answers would be greatly appreciated. > > Thanks, > Mike > -- __________________________________________________ Troy Dawson dawson@fnal.gov (630)840-6468 Fermilab ComputingDivision/OSS CSI Group __________________________________________________ From owner-pcp@oss.sgi.com Thu Jan 31 17:32:04 2002 Received: (from majordomo@localhost) by oss.sgi.com (8.11.2/8.11.3) id g111W4r18712 for pcp-outgoing; Thu, 31 Jan 2002 17:32:04 -0800 Received: from e31.co.us.ibm.com (e31.co.us.ibm.com [32.97.110.129]) by oss.sgi.com (8.11.2/8.11.3) with SMTP id g111W1d18709 for ; Thu, 31 Jan 2002 17:32:01 -0800 Received: from westrelay01.boulder.ibm.com (westrelay01.boulder.ibm.com [9.99.140.22]) by e31.co.us.ibm.com (8.9.3/8.9.3) with ESMTP id TAA63108 for ; Thu, 31 Jan 2002 19:28:17 -0500 Received: from w-mmason3.beaverton.ibm.com (dyn9-47-24-207.wil.beaverton.ibm.com [9.47.24.207]) by westrelay01.boulder.ibm.com (8.11.1m3/NCO v5.01) with ESMTP id g110VCa91650 for ; Thu, 31 Jan 2002 17:31:12 -0700 Subject: Units wrong for proc.psinfo.rss_rlim From: Mike Mason To: pcp@oss.sgi.com Content-Type: text/plain Content-Transfer-Encoding: 7bit X-Mailer: Evolution/1.0.0.99+cvs.2001.12.18.08.57 (Preview Release) Date: 31 Jan 2002 16:29:38 -0800 Message-Id: <1012523379.1439.64.camel@w-mmason3> Mime-Version: 1.0 Sender: owner-pcp@oss.sgi.com Precedence: bulk The units for proc.psinfo.rss_rlim should be defined as Kbytes, but are currently undefined even though rss_rlim is clearly converted from bytes to Kbytes further down in pmda.c I've attached a patch that fixes the problem. Mike Mason --- pcp-2.2.2.orig/src/pmdas/linux/pmda.c Mon Dec 10 17:52:09 2001 +++ pcp-2.2.2.new/src/pmdas/linux/pmda.c Thu Jan 31 15:27:13 2002 @@ -1096,7 +1096,7 @@ /* proc.psinfo.rss_rlim */ { NULL, { PMDA_PMID(CLUSTER_PID_STAT,24), PM_TYPE_U32, PROC_INDOM, PM_SEM_INSTANT, - PMDA_PMUNITS(0,0,0,0,0,0) } }, + PMDA_PMUNITS(1,0,0,PM_SPACE_KBYTE,0,0) } }, /* proc.psinfo.start_code */ { NULL,