From ahoyt@moser-inc.com Tue Aug 5 16:18:31 2003 Received: with ECARTIS (v1.0.0; list pcp); Tue, 05 Aug 2003 16:18:35 -0700 (PDT) Received: from mcimail.moser-inc.com ([207.250.3.118]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h75NIUFl014323 for ; Tue, 5 Aug 2003 16:18:31 -0700 Received: from moser-inc.com (r9100-vpn.moser-inc.com [207.250.3.126] (may be forged)) by mcimail.moser-inc.com (8.12.9/8.12.6) with ESMTP id h75NF02c024883 for ; Tue, 5 Aug 2003 18:15:34 -0500 (EST) Message-ID: <3F303B27.4050503@moser-inc.com> Date: Tue, 05 Aug 2003 18:17:59 -0500 From: Alan Hoyt User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030617 X-Accept-Language: en-us, en MIME-Version: 1.0 To: PCP Subject: pmlogger_check.sh in 2.3.1-4 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-archive-position: 299 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: ahoyt@moser-inc.com Precedence: bulk X-list: pcp Status: O Content-Length: 287 Lines: 8 Line 64 was changed from LOCALHOSTNAME=`hostname` LOCALHOSTNAME=`hostname -s` What was the logic behind this change? If it's necessary - I can find a work around! Otherwise, there is no "short" hostname option on Solaris which resuls in the host name being set to "-s". - Alan - From DDouthitt@cuna.coop Tue Aug 5 16:31:27 2003 Received: with ECARTIS (v1.0.0; list pcp); Tue, 05 Aug 2003 16:31:30 -0700 (PDT) Received: from esfsserv.cuna.coop (cuna.com [64.73.77.8]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h75NVQFl015101 for ; Tue, 5 Aug 2003 16:31:26 -0700 Received: FROM esfsserv.cuna.coop BY esfsserv.cuna.coop ; Tue Aug 05 18:30:22 2003 -0500 Received: from Unknown [172.16.3.7] by esfsserv.cuna.coop - SurfControl E-mail Filter (4.6); Tuesday, 05 August 2003, 18:30:21 Received: from CUNACAMP-MTA by email1.cuna.coop with Novell_GroupWise; Tue, 05 Aug 2003 18:32:44 -0500 Message-ID: From: "David Douthitt" To: Date: Tue, 05 Aug 2003 18:32:41 -0500 Subject: Re: pmlogger_check.sh in 2.3.1-4 MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Disposition: inline X-Mailer: Novell GroupWise Internet Agent 6.0.2 X-archive-position: 300 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: DDouthitt@cuna.coop Precedence: bulk X-list: pcp Status: O Content-Length: 608 Lines: 27 I've run across this in other contexts... "hostname -s" is a GNU-ism that forces a "short" hostname; the closest thing in other systems seems to be "uname -n". David Douthitt CUNA & Affiliates UNIX Systems Administrator LPI Level 1, Linux+ ddouthitt@cuna.coop >>> Alan Hoyt 8/5/03 6:17PM >>> Line 64 was changed from LOCALHOSTNAME=`hostname` LOCALHOSTNAME=`hostname -s` What was the logic behind this change? If it's necessary - I can find a work around! Otherwise, there is no "short" hostname option on Solaris which resuls in the host name being set to "-s". - Alan - From kenmcd@melbourne.sgi.com Tue Aug 5 17:09:09 2003 Received: with ECARTIS (v1.0.0; list pcp); Tue, 05 Aug 2003 17:09:12 -0700 (PDT) Received: from tolkor.sgi.com ([198.149.18.6]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h76098Fl017574 for ; Tue, 5 Aug 2003 17:09:09 -0700 Received: from rattle.melbourne.sgi.com (rattle.melbourne.sgi.com [134.14.55.145]) by tolkor.sgi.com (8.12.9/8.12.9/linux-outbound_gateway-1.1) with ESMTP id h760SfsR024386 for ; Tue, 5 Aug 2003 19:28:45 -0500 Received: from rattle.melbourne.sgi.com (localhost [127.0.0.1]) by rattle.melbourne.sgi.com (SGI-8.12.5/8.12.5) with ESMTP id h7608r9t028541; Wed, 6 Aug 2003 10:08:53 +1000 (EST) Received: from localhost (kenmcd@localhost) by rattle.melbourne.sgi.com (SGI-8.12.5/8.12.5/Submit) with ESMTP id h7608qXC028533; Wed, 6 Aug 2003 10:08:52 +1000 (EST) X-Authentication-Warning: rattle.melbourne.sgi.com: kenmcd owned process doing -bs Date: Wed, 6 Aug 2003 10:08:52 +1000 From: Ken McDonell To: Alan Hoyt cc: PCP Subject: Re: pmlogger_check.sh in 2.3.1-4 In-Reply-To: <3F303B27.4050503@moser-inc.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-archive-position: 301 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: kenmcd@melbourne.sgi.com Precedence: bulk X-list: pcp Status: O Content-Length: 766 Lines: 18 On Tue, 5 Aug 2003, Alan Hoyt wrote: > Line 64 was changed from LOCALHOSTNAME=`hostname` LOCALHOSTNAME=`hostname -s` At that point we need the short form of the hostname, and on some systems hostname returns the fully qualified domain name, while hostname -s returns the short name. > What was the logic behind this change? If it's necessary - I can find a work around! > Otherwise, there is no "short" hostname option on Solaris which resuls in the host name being set to "-s". Seems like we'll need a workaround for Solaris ... bugger. Wait a moment ... Nathan's in my office and he suggested adding a -s option to pmhostname and abandoning hostname everywhere ... we'll investigate this option, but for the moment I'd suggest Alan uses a local workaround. From pj@sgi.com Tue Aug 5 17:32:55 2003 Received: with ECARTIS (v1.0.0; list pcp); Tue, 05 Aug 2003 17:32:58 -0700 (PDT) Received: from zok.sgi.com (zok.SGI.COM [204.94.215.101]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h760WsFl018368 for ; Tue, 5 Aug 2003 17:32:55 -0700 Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) by zok.sgi.com (8.12.9/8.12.9/linux-outbound_gateway-1.1) with ESMTP id h760Wnq0027423 for ; Tue, 5 Aug 2003 17:32:49 -0700 Received: from sam.engr.sgi.com (sam.engr.sgi.com [163.154.6.103]) by cthulhu.engr.sgi.com (SGI-8.12.5/8.12.5) with SMTP id h760Wl6B8724755; Tue, 5 Aug 2003 17:32:47 -0700 (PDT) Date: Tue, 5 Aug 2003 17:32:46 -0700 From: Paul Jackson To: Ken McDonell Cc: ahoyt@moser-inc.com, pcp@oss.sgi.com Subject: Re: pmlogger_check.sh in 2.3.1-4 Message-Id: <20030805173246.4dd90d5d.pj@sgi.com> In-Reply-To: References: <3F303B27.4050503@moser-inc.com> Organization: SGI X-Mailer: Sylpheed version 0.8.10claws (GTK+ 1.2.10; i686-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-archive-position: 302 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: pj@sgi.com Precedence: bulk X-list: pcp Status: O Content-Length: 644 Lines: 18 > LOCALHOSTNAME=`hostname -s` > ... > on Solaris which resuls in the host name being set to "-s". The rest of this thread makes sense, but this part, about setting hostname to "-s", baffles me. I'd expect it to get set to whatever was the output (to fd == 1) of: hostname -s This might be an error message, or, if that message went to fd == 2, the empty string. I doubt that said error message was, in its entirety, the two character string "-s". -- I won't rest till it's the best ... Programmer, Linux Scalability Paul Jackson 1.650.933.1373 From kaos@sgi.com Tue Aug 5 19:08:04 2003 Received: with ECARTIS (v1.0.0; list pcp); Tue, 05 Aug 2003 19:08:09 -0700 (PDT) Received: from tolkor.sgi.com ([198.149.18.6]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h76283Fl026995 for ; Tue, 5 Aug 2003 19:08:04 -0700 Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by tolkor.sgi.com (8.12.9/8.12.9/linux-outbound_gateway-1.1) with SMTP id h762ResR015788 for ; Tue, 5 Aug 2003 21:27:40 -0500 Received: from kao2.melbourne.sgi.com (kao2.melbourne.sgi.com [134.14.55.180]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id MAA06511; Wed, 6 Aug 2003 12:07:51 +1000 Received: by kao2.melbourne.sgi.com (Postfix, from userid 16331) id 2A3A4C21CF; Wed, 6 Aug 2003 12:07:48 +1000 (EST) Received: from kao2.melbourne.sgi.com (localhost [127.0.0.1]) by kao2.melbourne.sgi.com (Postfix) with ESMTP id 276A7140681; Wed, 6 Aug 2003 12:07:48 +1000 (EST) X-Mailer: exmh version 2.5 01/15/2001 with nmh-1.0.4 From: Keith Owens To: Ken McDonell Cc: Alan Hoyt , PCP Subject: Re: pmlogger_check.sh in 2.3.1-4 In-reply-to: Your message of "Wed, 06 Aug 2003 10:08:52 +1000." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 06 Aug 2003 12:07:47 +1000 Message-ID: <2620.1060135667@kao2.melbourne.sgi.com> X-archive-position: 303 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: kaos@sgi.com Precedence: bulk X-list: pcp Status: O Content-Length: 408 Lines: 12 On Wed, 6 Aug 2003 10:08:52 +1000, Ken McDonell wrote: >On Tue, 5 Aug 2003, Alan Hoyt wrote: > >> Line 64 was changed from LOCALHOSTNAME=`hostname` LOCALHOSTNAME=`hostname -s` > >At that point we need the short form of the hostname, and on some >systems hostname returns the fully qualified domain name, while hostname -s >returns the short name. `hostname | sed -e 's/\..*//'` From pj@sgi.com Tue Aug 5 22:15:32 2003 Received: with ECARTIS (v1.0.0; list pcp); Tue, 05 Aug 2003 22:15:38 -0700 (PDT) Received: from zok.sgi.com (zok.SGI.COM [204.94.215.101]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h765FVFl022501 for ; Tue, 5 Aug 2003 22:15:32 -0700 Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) by zok.sgi.com (8.12.9/8.12.9/linux-outbound_gateway-1.1) with ESMTP id h765FQq0019489 for ; Tue, 5 Aug 2003 22:15:26 -0700 Received: from sam.engr.sgi.com (sam.engr.sgi.com [163.154.6.103]) by cthulhu.engr.sgi.com (SGI-8.12.5/8.12.5) with SMTP id h765Er6B8806198; Tue, 5 Aug 2003 22:14:53 -0700 (PDT) Date: Tue, 5 Aug 2003 22:14:53 -0700 From: Paul Jackson To: Alan Hoyt Cc: kenmcd@melbourne.sgi.com, pcp@oss.sgi.com Subject: Re: pmlogger_check.sh in 2.3.1-4 Message-Id: <20030805221453.48ce62ad.pj@sgi.com> In-Reply-To: <3F308DE8.2030100@moser-inc.com> References: <3F303B27.4050503@moser-inc.com> <20030805173246.4dd90d5d.pj@sgi.com> <3F308DE8.2030100@moser-inc.com> Organization: SGI X-Mailer: Sylpheed version 0.8.10claws (GTK+ 1.2.10; i686-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-archive-position: 304 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: pj@sgi.com Precedence: bulk X-list: pcp Status: O Content-Length: 259 Lines: 8 > a request to set the hostname to "-s" aha - makes sense - thanks -- I won't rest till it's the best ... Programmer, Linux Scalability Paul Jackson 1.650.933.1373 From ahoyt@moser-inc.com Tue Aug 5 22:44:42 2003 Received: with ECARTIS (v1.0.0; list pcp); Tue, 05 Aug 2003 22:44:46 -0700 (PDT) Received: from flavatown.mail.pas.earthlink.net (flavatown.mail.pas.earthlink.net [207.217.120.148]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h765ifFl024315 for ; Tue, 5 Aug 2003 22:44:41 -0700 Received: from conure.mail.pas.earthlink.net (conure.mail.pas.earthlink.net [207.217.120.54]) by flavatown.mail.pas.earthlink.net (8.11.6+Sun/8.11.6) with ESMTP id h765IiF15913 for ; Tue, 5 Aug 2003 22:18:44 -0700 (PDT) Received: from user-0c93cjo.cable.mindspring.com ([24.145.178.120] helo=moser-inc.com) by conure.mail.pas.earthlink.net with esmtp (Exim 3.33 #1) id 19kGZ7-0007VE-00; Tue, 05 Aug 2003 22:10:26 -0700 Message-ID: <3F308DE8.2030100@moser-inc.com> Date: Wed, 06 Aug 2003 00:11:04 -0500 From: Alan Hoyt User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4) Gecko/20030612 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Paul Jackson CC: Ken McDonell , pcp@oss.sgi.com Subject: Re: pmlogger_check.sh in 2.3.1-4 References: <3F303B27.4050503@moser-inc.com> <20030805173246.4dd90d5d.pj@sgi.com> In-Reply-To: <20030805173246.4dd90d5d.pj@sgi.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-archive-position: 305 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: ahoyt@moser-inc.com Precedence: bulk X-list: pcp Status: O Content-Length: 554 Lines: 26 Paul Jackson wrote: >>LOCALHOSTNAME=`hostname -s` >>... >>on Solaris which resuls in the host name being set to "-s". >> >> > >The rest of this thread makes sense, but this part, about setting >hostname to "-s", baffles me. I'd expect it to get set to whatever was >the output (to fd == 1) of: > > hostname -s > > It is being set to the output of the above - and on Solaris hostname doesn't know about the -s option (or any options at all) and interprets this as a request to set the hostname to "-s" which it diligently does. - Alan - From ahoyt@moser-inc.com Tue Aug 5 22:54:41 2003 Received: with ECARTIS (v1.0.0; list pcp); Tue, 05 Aug 2003 22:54:45 -0700 (PDT) Received: from badboy.mail.pas.earthlink.net (badboy.mail.pas.earthlink.net [207.217.120.20]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h765seFl024960 for ; Tue, 5 Aug 2003 22:54:41 -0700 Received: from conure.mail.pas.earthlink.net (conure.mail.pas.earthlink.net [207.217.120.54]) by badboy.mail.pas.earthlink.net (8.11.6+Sun/8.11.6) with ESMTP id h765mpv01921 for ; Tue, 5 Aug 2003 22:48:51 -0700 (PDT) Received: from user-0c93cjo.cable.mindspring.com ([24.145.178.120] helo=moser-inc.com) by conure.mail.pas.earthlink.net with esmtp (Exim 3.33 #1) id 19kH2Q-0007En-00; Tue, 05 Aug 2003 22:40:42 -0700 Message-ID: <3F309501.9010609@moser-inc.com> Date: Wed, 06 Aug 2003 00:41:21 -0500 From: Alan Hoyt User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4) Gecko/20030612 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Ken McDonell CC: PCP Subject: Re: pmlogger_check.sh in 2.3.1-4 References: In-Reply-To: Content-Type: text/plain; charset=windows-1252; format=flowed X-MIME-Autoconverted: from 8bit to quoted-printable by badboy.mail.pas.earthlink.net id h765mpv01921 Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by oss.sgi.com id h765seFl024960 X-archive-position: 306 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: ahoyt@moser-inc.com Precedence: bulk X-list: pcp Status: O Content-Length: 579 Lines: 27 Ken McDonell wrote: >On Tue, 5 Aug 2003, Alan Hoyt wrote: > > > >>Line 64 was changed from LOCALHOSTNAME=`hostname` LOCALHOSTNAME=`hostname -s` >> >> > >At that point we need the short form of the hostname, and on some >systems hostname returns the fully qualified domain name, while hostname -s >returns the short name. > > > >>What was the logic behind this change? >> I haven’t tested it exhaustively but it seems to work correctly either way – with a fully qualified name or a short name. That's why the addition of the -s flag wasn't clear to me. - Alan - From ahoyt@moser-inc.com Wed Aug 6 09:29:23 2003 Received: with ECARTIS (v1.0.0; list pcp); Wed, 06 Aug 2003 09:29:27 -0700 (PDT) Received: from mcimail.moser-inc.com ([207.250.3.118]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h76GTMFl001812 for ; Wed, 6 Aug 2003 09:29:23 -0700 Received: from moser-inc.com (r9100-vpn.moser-inc.com [207.250.3.126] (may be forged)) by mcimail.moser-inc.com (8.12.9/8.12.6) with ESMTP id h76GPq2c005124; Wed, 6 Aug 2003 11:26:23 -0500 (EST) Message-ID: <3F312CC3.80305@moser-inc.com> Date: Wed, 06 Aug 2003 11:28:51 -0500 From: Alan Hoyt User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.5a) Gecko/20030718 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Ken McDonell CC: PCP Subject: Re: pmlogger_check.sh in 2.3.1-4 References: <3F309501.9010609@moser-inc.com> In-Reply-To: <3F309501.9010609@moser-inc.com> Content-Type: text/plain; charset=windows-1252; format=flowed X-MIME-Autoconverted: from 8bit to quoted-printable by mcimail.moser-inc.com id h76GPq2c005124 Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by oss.sgi.com id h76GTMFl001812 X-archive-position: 307 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: ahoyt@moser-inc.com Precedence: bulk X-list: pcp Status: O Content-Length: 854 Lines: 35 Alan Hoyt wrote: > Ken McDonell wrote: > >> On Tue, 5 Aug 2003, Alan Hoyt wrote: >> >> >> >>> Line 64 was changed from LOCALHOSTNAME=`hostname` >>> LOCALHOSTNAME=`hostname -s` >>> >> >> >> At that point we need the short form of the hostname, and on some >> systems hostname returns the fully qualified domain name, while >> hostname -s >> returns the short name. >> >> >> >>> What was the logic behind this change? > > > I haven’t tested it exhaustively but it seems to work correctly either > way – with a fully qualified name or a short name. That's why the > addition of the -s flag wasn't clear to me. > Ok, the logger control file specifies the logger instance (controlling pmlogger_check) and this change fixed the inability to stipulate short names in the control file - so it's necessary to work for all cases. - Alan - From kenmcd@melbourne.sgi.com Wed Aug 6 15:05:21 2003 Received: with ECARTIS (v1.0.0; list pcp); Wed, 06 Aug 2003 15:05:27 -0700 (PDT) Received: from tolkor.sgi.com ([198.149.18.6]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h76M5LFl003479 for ; Wed, 6 Aug 2003 15:05:21 -0700 Received: from rattle.melbourne.sgi.com (rattle.melbourne.sgi.com [134.14.55.145]) by tolkor.sgi.com (8.12.9/8.12.9/linux-outbound_gateway-1.1) with ESMTP id h76MP0sR005960 for ; Wed, 6 Aug 2003 17:25:01 -0500 Received: from rattle.melbourne.sgi.com (localhost [127.0.0.1]) by rattle.melbourne.sgi.com (SGI-8.12.5/8.12.5) with ESMTP id h76M5D9t030076; Thu, 7 Aug 2003 08:05:13 +1000 (EST) Received: from localhost (kenmcd@localhost) by rattle.melbourne.sgi.com (SGI-8.12.5/8.12.5/Submit) with ESMTP id h76M5CpM030007; Thu, 7 Aug 2003 08:05:12 +1000 (EST) X-Authentication-Warning: rattle.melbourne.sgi.com: kenmcd owned process doing -bs Date: Thu, 7 Aug 2003 08:05:12 +1000 From: Ken McDonell To: Alan Hoyt cc: PCP Subject: Re: pmlogger_check.sh in 2.3.1-4 In-Reply-To: <3F312CC3.80305@moser-inc.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-archive-position: 308 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: kenmcd@melbourne.sgi.com Precedence: bulk X-list: pcp Status: O Content-Length: 503 Lines: 17 Thanks Alan, for confiming this was not a gratuitous change. Keith Owens suggested this quick and robust workaround ... LOCALHOSTNAME=`hostname | sed -e 's/\..*//'` I'm planning on checking this into the code base. On Wed, 6 Aug 2003, Alan Hoyt wrote: > Alan Hoyt wrote: > ... > > Ok, the logger control file specifies the logger instance (controlling > pmlogger_check) and this change fixed the inability to stipulate short > names in the control file - so it's necessary to work for all cases. From DDouthitt@cuna.coop Wed Aug 6 15:10:24 2003 Received: with ECARTIS (v1.0.0; list pcp); Wed, 06 Aug 2003 15:10:27 -0700 (PDT) Received: from esfsserv.cuna.coop (cuna.com [64.73.77.8]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h76MANFl003535 for ; Wed, 6 Aug 2003 15:10:24 -0700 Received: FROM esfsserv.cuna.coop BY esfsserv.cuna.coop ; Wed Aug 06 17:09:19 2003 -0500 Received: from Unknown [172.16.3.7] by esfsserv.cuna.coop - SurfControl E-mail Filter (4.6); Wednesday, 06 August 2003, 17:09:18 Received: from CUNACAMP-MTA by email1.cuna.coop with Novell_GroupWise; Wed, 06 Aug 2003 17:11:38 -0500 Message-ID: From: "David Douthitt" To: , Cc: " Date: Wed, 06 Aug 2003 17:11:32 -0500 Subject: Re: pmlogger_check.sh in 2.3.1-4 MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Disposition: inline X-Mailer: Novell GroupWise Internet Agent 6.0.2 X-archive-position: 309 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: DDouthitt@cuna.coop Precedence: bulk X-list: pcp Status: O Content-Length: 823 Lines: 37 If you are using ksh, why not this? LOCALHOSTNAME=`hostname` LOCALHOSTNAME=${LOCALHOSTNAME%%.*} That removes the need for a pipe and for a sed run... David Douthitt CUNA & Affiliates UNIX Systems Administrator LPI Level 1, Linux+ ddouthitt@cuna.coop >>> Ken McDonell 8/6/03 5:05PM >>> Thanks Alan, for confiming this was not a gratuitous change. Keith Owens suggested this quick and robust workaround ... LOCALHOSTNAME=`hostname | sed -e 's/\..*//'` I'm planning on checking this into the code base. On Wed, 6 Aug 2003, Alan Hoyt wrote: > Alan Hoyt wrote: > ... > > Ok, the logger control file specifies the logger instance (controlling > pmlogger_check) and this change fixed the inability to stipulate short > names in the control file - so it's necessary to work for all cases. From kenmcd@melbourne.sgi.com Wed Aug 6 16:36:21 2003 Received: with ECARTIS (v1.0.0; list pcp); Wed, 06 Aug 2003 16:36:26 -0700 (PDT) Received: from tolkor.sgi.com ([198.149.18.6]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h76NaKFl005899 for ; Wed, 6 Aug 2003 16:36:21 -0700 Received: from rattle.melbourne.sgi.com (rattle.melbourne.sgi.com [134.14.55.145]) by tolkor.sgi.com (8.12.9/8.12.9/linux-outbound_gateway-1.1) with ESMTP id h76Nu0sR023715 for ; Wed, 6 Aug 2003 18:56:01 -0500 Received: from rattle.melbourne.sgi.com (localhost [127.0.0.1]) by rattle.melbourne.sgi.com (SGI-8.12.5/8.12.5) with ESMTP id h76NaC9t037709; Thu, 7 Aug 2003 09:36:12 +1000 (EST) Received: from localhost (kenmcd@localhost) by rattle.melbourne.sgi.com (SGI-8.12.5/8.12.5/Submit) with ESMTP id h76Na7Zg037560; Thu, 7 Aug 2003 09:36:09 +1000 (EST) X-Authentication-Warning: rattle.melbourne.sgi.com: kenmcd owned process doing -bs Date: Thu, 7 Aug 2003 09:36:07 +1000 From: Ken McDonell To: David Douthitt cc: ahoyt@moser-inc.com, " Subject: Re: pmlogger_check.sh in 2.3.1-4 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-archive-position: 310 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: kenmcd@melbourne.sgi.com Precedence: bulk X-list: pcp Status: O Content-Length: 957 Lines: 42 PCP does not use ksh, sorry. On Wed, 6 Aug 2003, David Douthitt wrote: > If you are using ksh, why not this? > > LOCALHOSTNAME=`hostname` > LOCALHOSTNAME=${LOCALHOSTNAME%%.*} > > That removes the need for a pipe and for a sed run... > > David Douthitt > CUNA & Affiliates > UNIX Systems Administrator > LPI Level 1, Linux+ > ddouthitt@cuna.coop > > > >>> Ken McDonell 8/6/03 5:05PM >>> > Thanks Alan, for confiming this was not a gratuitous change. > > Keith Owens suggested this quick and robust workaround ... > > LOCALHOSTNAME=`hostname | sed -e 's/\..*//'` > > I'm planning on checking this into the code base. > > On Wed, 6 Aug 2003, Alan Hoyt wrote: > > > Alan Hoyt wrote: > > ... > > > > Ok, the logger control file specifies the logger instance > (controlling > > pmlogger_check) and this change fixed the inability to stipulate > short > > names in the control file - so it's necessary to work for all cases. > > > > From scott_list@mischko.com Tue Aug 12 10:41:06 2003 Received: with ECARTIS (v1.0.0; list pcp); Tue, 12 Aug 2003 10:41:09 -0700 (PDT) Received: from lapdog.lund.com ([65.243.131.113]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h7CHf4Fl011358 for ; Tue, 12 Aug 2003 10:41:06 -0700 Received: from lapdog.lund.com (localhost [127.0.0.1]) by lapdog.lund.com (Postfix) with ESMTP id 20765A0B5 for ; Tue, 12 Aug 2003 13:41:03 -0400 (EDT) From: Scott Chapman Reply-To: scott_list@mischko.com To: pcp@oss.sgi.com Subject: [OT?]: Linux per-process disk and network I/O metrics? Date: Tue, 12 Aug 2003 10:41:02 -0700 User-Agent: KMail/1.5.1 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200308121041.02048.scott_list@mischko.com> X-archive-position: 311 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: scott_list@mischko.com Precedence: bulk X-list: pcp Status: O Content-Length: 250 Lines: 10 Hi all! I see that PCP gives some XFS process-level disk I/O metrics. Does anyone have any idea of the availability of disk I/O metrics for ext2/ext3 and other file systems? Any other tools that have this? Same for network traffic? TIA, Scott From kenmcd@melbourne.sgi.com Tue Aug 12 12:41:44 2003 Received: with ECARTIS (v1.0.0; list pcp); Tue, 12 Aug 2003 12:41:48 -0700 (PDT) Received: from tolkor.sgi.com (tolkor.sgi.com [198.149.18.6]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h7CJfgFl013331 for ; Tue, 12 Aug 2003 12:41:43 -0700 Received: from rattle.melbourne.sgi.com (rattle.melbourne.sgi.com [134.14.55.145]) by tolkor.sgi.com (8.12.9/8.12.9/linux-outbound_gateway-1.1) with ESMTP id h7CK1hsR025691 for ; Tue, 12 Aug 2003 15:01:44 -0500 Received: from rattle.melbourne.sgi.com (localhost [127.0.0.1]) by rattle.melbourne.sgi.com (SGI-8.12.5/8.12.5) with ESMTP id h7CJfZ9t087814; Wed, 13 Aug 2003 05:41:35 +1000 (EST) Received: from localhost (kenmcd@localhost) by rattle.melbourne.sgi.com (SGI-8.12.5/8.12.5/Submit) with ESMTP id h7CJfYhw087953; Wed, 13 Aug 2003 05:41:34 +1000 (EST) X-Authentication-Warning: rattle.melbourne.sgi.com: kenmcd owned process doing -bs Date: Wed, 13 Aug 2003 05:41:33 +1000 From: Ken McDonell To: Scott Chapman cc: pcp@oss.sgi.com Subject: Re: [OT?]: Linux per-process disk and network I/O metrics? In-Reply-To: <200308121041.02048.scott_list@mischko.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-archive-position: 312 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: kenmcd@melbourne.sgi.com Precedence: bulk X-list: pcp Status: O Content-Length: 651 Lines: 22 On Tue, 12 Aug 2003, Scott Chapman wrote: > Hi all! > > I see that PCP gives some XFS process-level disk I/O metrics. ... Which metrics are you referring to ... I don't believe there are any process-level metrics from XFS. > ... Does anyone > have any idea of the availability of disk I/O metrics for ext2/ext3 and other > file systems? Any other tools that have this? > > Same for network traffic? This is more a matter of kernel instrumentation ... in Linux this sort of process-level accounting of resource consumption is typically not available. If this data _is_ present, please let us know, as the effort to export it via PCP is minimal. From grooms@opal.cc.uky.edu Tue Aug 12 14:10:59 2003 Received: with ECARTIS (v1.0.0; list pcp); Tue, 12 Aug 2003 14:11:02 -0700 (PDT) Received: from opal.cc.uky.edu (opal.cc.uky.edu [128.163.18.75]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h7CLAxFl015693 for ; Tue, 12 Aug 2003 14:10:59 -0700 Received: by opal.cc.uky.edu (Postfix, from userid 501) id 3BC54450E; Tue, 12 Aug 2003 14:27:46 -0400 (EDT) To: pcp@oss.sgi.com Subject: cpp pre-req to start pcp/pmcd? From: Message-Id: <20030812182746.3BC54450E@opal.cc.uky.edu> Date: Tue, 12 Aug 2003 14:27:46 -0400 (EDT) X-archive-position: 313 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: jerry.mlist@spamex.com Precedence: bulk X-list: pcp Status: O Content-Length: 1092 Lines: 36 pcp version: pcp-2.3.1-4 OS: Linux I have a cluster I'd like to run pmcd collectors on. The compute nodes do not presently have cpp available, which appears to be required (see below). Is there any work-around short of installing gcc/cpp on all the compute nodes? I looked through the User/Admin Guide, FAQ, and pcp list archives and did not find any mention of this. Thanks for any feedback, Jerry --- # /etc/init.d/pcp start Performance Co-Pilot rebuilding PMNS ... pmLoadNameSpace: Unable to find an executable cpp at any of ... /lib/cpp /lib/cpp /usr/cpu/sysgen/root/lib/cpp /usr/lib/cpp /usr/cpu/sysgen/root/usr/lib/cpp Sorry, but this is fatal Rebuilding the Performance Metrics Name Space (PMNS) in /var/pcp/pmns ... Rebuild: PCP upgrade processing for "root" PMNS changes ... Rebuild: merging the following PMNS files: root root_linux root_pmcd Rebuild: pmnsmerge failed "root" and "root.bin" have not been changed. Performance Co-Pilot starting PMCD (logfile is /var/log/pcp/pmcd/pmcd.log) ... Performance Co-Pilot starting archive loggers ... From scott_list@mischko.com Tue Aug 12 14:18:45 2003 Received: with ECARTIS (v1.0.0; list pcp); Tue, 12 Aug 2003 14:18:49 -0700 (PDT) Received: from lapdog.lund.com ([65.243.131.113]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h7CLIiFl015777 for ; Tue, 12 Aug 2003 14:18:45 -0700 Received: from lapdog.lund.com (localhost [127.0.0.1]) by lapdog.lund.com (Postfix) with ESMTP id 9846FA0B5; Tue, 12 Aug 2003 17:18:42 -0400 (EDT) From: Scott Chapman Reply-To: scott_list@mischko.com To: Ken McDonell Subject: Re: [OT?]: Linux per-process disk and network I/O metrics? Date: Tue, 12 Aug 2003 14:18:41 -0700 User-Agent: KMail/1.5.1 Cc: pcp@oss.sgi.com References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200308121418.41363.scott_list@mischko.com> X-archive-position: 314 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: scott_list@mischko.com Precedence: bulk X-list: pcp Status: O Content-Length: 1296 Lines: 37 On Tuesday 12 August 2003 12:41, Ken McDonell wrote: > On Tue, 12 Aug 2003, Scott Chapman wrote: > > Hi all! > > > > I see that PCP gives some XFS process-level disk I/O metrics. ... > > Which metrics are you referring to ... I don't believe there are any > process-level metrics from XFS. My bad. :( I was reading the metric list wrong. > > ... Does anyone > > have any idea of the availability of disk I/O metrics for ext2/ext3 and > > other file systems? Any other tools that have this? > > > > Same for network traffic? > > This is more a matter of kernel instrumentation ... in Linux this sort > of process-level accounting of resource consumption is typically not > available. > > If this data _is_ present, please let us know, as the effort to export > it via PCP is minimal. I'm not aware of any such data. It seems Linux is rather short on some useful metrics that other Unix flavors have. This would be a major improvement in Linux. It could surpass other Unix flavors in this arena if there was motive force in the kernel development core. If anyone has influence on kernel development, please contact the appropriate people! I understand that Linus is not big on instrumentation. Don't know if that's true or not, just something I heard somewhere. Cordially, Scott From kenmcd@melbourne.sgi.com Tue Aug 12 14:34:05 2003 Received: with ECARTIS (v1.0.0; list pcp); Tue, 12 Aug 2003 14:34:12 -0700 (PDT) Received: from rj.sgi.com (rj.sgi.com [192.82.208.96]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h7CLY4Fl015859 for ; Tue, 12 Aug 2003 14:34:05 -0700 Received: from rattle.melbourne.sgi.com (rattle.melbourne.sgi.com [134.14.55.145]) by rj.sgi.com (8.12.9/8.12.9/linux-outbound_gateway-1.1) with ESMTP id h7CJYgQa006589 for ; Tue, 12 Aug 2003 12:34:42 -0700 Received: from rattle.melbourne.sgi.com (localhost [127.0.0.1]) by rattle.melbourne.sgi.com (SGI-8.12.5/8.12.5) with ESMTP id h7CLXv9t090248; Wed, 13 Aug 2003 07:33:57 +1000 (EST) Received: from localhost (kenmcd@localhost) by rattle.melbourne.sgi.com (SGI-8.12.5/8.12.5/Submit) with ESMTP id h7CLXvYN083210; Wed, 13 Aug 2003 07:33:57 +1000 (EST) X-Authentication-Warning: rattle.melbourne.sgi.com: kenmcd owned process doing -bs Date: Wed, 13 Aug 2003 07:33:56 +1000 From: Ken McDonell To: jerry.mlist@spamex.com cc: pcp@oss.sgi.com Subject: Re: cpp pre-req to start pcp/pmcd? In-Reply-To: <20030812182746.3BC54450E@opal.cc.uky.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-archive-position: 315 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: kenmcd@melbourne.sgi.com Precedence: bulk X-list: pcp Status: O Content-Length: 1908 Lines: 57 Hmm ... this is a documentation oversight ... on IRIX systems, you cannot relink the kernel without cpp so _every_ IRIX system has a cpp variant lurking somewhere. This is not necessarily true for other OS variants. The PCP performance metrics name space (PMNS) may use #define and #include directives ... as a work around, check your /var/pcp/pmns/root and remove directives by equivalent hand editing, and then make one of the cpp paths below a link to "cat". But you'll next run into problems with installing the PMDAs, because their local fragments of the PMNS also use cpp directives. So you will need to install cpp eventually. I'll try and figure out the right place to document this. On Tue, 12 Aug 2003 jerry.mlist@spamex.com wrote: > > pcp version: pcp-2.3.1-4 > OS: Linux > > I have a cluster I'd like to run pmcd collectors on. The compute > nodes do not presently have cpp available, which appears to be > required (see below). > > Is there any work-around short of installing gcc/cpp on all the > compute nodes? I looked through the User/Admin Guide, FAQ, and pcp > list archives and did not find any mention of this. > > Thanks for any feedback, > > Jerry > > --- > > > # /etc/init.d/pcp start > Performance Co-Pilot rebuilding PMNS ... > pmLoadNameSpace: Unable to find an executable cpp at any of ... > /lib/cpp > /lib/cpp > /usr/cpu/sysgen/root/lib/cpp > /usr/lib/cpp > /usr/cpu/sysgen/root/usr/lib/cpp > Sorry, but this is fatal > Rebuilding the Performance Metrics Name Space (PMNS) in /var/pcp/pmns ... > Rebuild: PCP upgrade processing for "root" PMNS changes ... > Rebuild: merging the following PMNS files: > root root_linux root_pmcd > Rebuild: pmnsmerge failed > "root" and "root.bin" have not been changed. > Performance Co-Pilot starting PMCD (logfile is /var/log/pcp/pmcd/pmcd.log) ... > Performance Co-Pilot starting archive loggers ... > From dawson@fnal.gov Wed Aug 13 06:15:40 2003 Received: with ECARTIS (v1.0.0; list pcp); Wed, 13 Aug 2003 06:15:47 -0700 (PDT) Received: from heffalump.fnal.gov (heffalump.fnal.gov [131.225.9.20]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h7DDFdFl003675 for ; Wed, 13 Aug 2003 06:15:40 -0700 Received: from fnal.gov (thebrain.fnal.gov [131.225.80.75]) by heffalump.fnal.gov (iPlanet Messaging Server 5.2 HotFix 1.10 (built Jan 23 2003)) with ESMTPSA id <0HJK00FZP7I20R@heffalump.fnal.gov> for pcp@oss.sgi.com; Wed, 13 Aug 2003 08:15:39 -0500 (CDT) Date: Wed, 13 Aug 2003 08:15:38 -0500 From: Troy Dawson Subject: Re: cpp pre-req to start pcp/pmcd? In-reply-to: To: Ken McDonell Cc: jerry.mlist@spamex.com, pcp@oss.sgi.com Message-id: <3F3A39FA.2090808@fnal.gov> MIME-version: 1.0 Content-type: text/plain; charset=us-ascii; format=flowed Content-transfer-encoding: 7BIT X-Accept-Language: en-us, en User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2.1) Gecko/20030225 References: X-archive-position: 316 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: dawson@fnal.gov Precedence: bulk X-list: pcp Status: O Content-Length: 3211 Lines: 92 Howdy, This has been such a part of how we use pcp here at Fermilab that I forgot about it. We have several hundred farm nodes that don't have cpp or any compiler. What we have is a seperate rpm called pcp-config. In this, besides having config files that are specific for our lab, we also have the the following three files that were made on a machine that had cpp. (To have them made, just start pcp once and they get made.) /var/pcp/pmns/root /var/pcp/pmns/root.bin /var/pcp/pmns/stdpmid So when this get's installed, we remove the .NeedRebuild file, move these files into place, and everything works. There are a couple of catches. Make sure you make the files on a similar platform. So RedHat 9's work with RedHat 9, while all of the RedHat 7's worked on each other (7.1, 7.2, 7.3) If you add more pmda's, you have to rebuild these files. Since we just use the standard out of the box pmda's, this part is easy, but it could be more difficult if you are still trying out different pmda's. Hope this helps. Troy Ken McDonell wrote: > Hmm ... this is a documentation oversight ... on IRIX systems, you > cannot relink the kernel without cpp so _every_ IRIX system has a cpp > variant lurking somewhere. This is not necessarily true for other OS > variants. > > The PCP performance metrics name space (PMNS) may use #define and > #include directives ... as a work around, check your /var/pcp/pmns/root > and remove directives by equivalent hand editing, and then make one of > the cpp paths below a link to "cat". > > But you'll next run into problems with installing the PMDAs, because > their local fragments of the PMNS also use cpp directives. > > So you will need to install cpp eventually. > > I'll try and figure out the right place to document this. > > On Tue, 12 Aug 2003 jerry.mlist@spamex.com wrote: > > >>pcp version: pcp-2.3.1-4 >>OS: Linux >> >>I have a cluster I'd like to run pmcd collectors on. The compute >>nodes do not presently have cpp available, which appears to be >>required (see below). >> >>Is there any work-around short of installing gcc/cpp on all the >>compute nodes? I looked through the User/Admin Guide, FAQ, and pcp >>list archives and did not find any mention of this. >> >>Thanks for any feedback, >> >>Jerry >> >>--- >> >> >># /etc/init.d/pcp start >>Performance Co-Pilot rebuilding PMNS ... >>pmLoadNameSpace: Unable to find an executable cpp at any of ... >> /lib/cpp >> /lib/cpp >> /usr/cpu/sysgen/root/lib/cpp >> /usr/lib/cpp >> /usr/cpu/sysgen/root/usr/lib/cpp >>Sorry, but this is fatal >>Rebuilding the Performance Metrics Name Space (PMNS) in /var/pcp/pmns ... >>Rebuild: PCP upgrade processing for "root" PMNS changes ... >>Rebuild: merging the following PMNS files: >>root root_linux root_pmcd >>Rebuild: pmnsmerge failed >> "root" and "root.bin" have not been changed. >>Performance Co-Pilot starting PMCD (logfile is /var/log/pcp/pmcd/pmcd.log) ... >>Performance Co-Pilot starting archive loggers ... >> > > > -- __________________________________________________ Troy Dawson dawson@fnal.gov (630)840-6468 Fermilab ComputingDivision/OSS CSI Group __________________________________________________ From ahoyt@moser-inc.com Mon Aug 25 11:30:10 2003 Received: with ECARTIS (v1.0.0; list pcp); Mon, 25 Aug 2003 11:30:18 -0700 (PDT) Received: from mcimail.moser-inc.com ([207.250.3.118]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h7PIU7oO015113 for ; Mon, 25 Aug 2003 11:30:10 -0700 Received: from moser-inc.com (r9100-vpn.moser-inc.com [207.250.3.126] (may be forged)) by mcimail.moser-inc.com (8.12.9/8.12.6) with ESMTP id h7PIQ62c024611; Mon, 25 Aug 2003 13:26:36 -0500 (EST) Message-ID: <3F4A5589.9020209@moser-inc.com> Date: Mon, 25 Aug 2003 13:29:29 -0500 From: Alan Hoyt User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.5a) Gecko/20030718 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Alan Hoyt CC: kenmcd@melbourne.sgi.com, Mark Boyd , Chris Howland , PCP , "'Said Maafala'" Subject: Re: DMBS Add-ons for Performance Co-Pilot References: <3F254AA0.2050207@moser-inc.com> In-Reply-To: <3F254AA0.2050207@moser-inc.com> Content-Type: text/plain; charset=windows-1252; format=flowed X-MIME-Autoconverted: from 8bit to quoted-printable by mcimail.moser-inc.com id h7PIQ62c024611 Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by oss.sgi.com id h7PIU7oO015113 X-archive-position: 317 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: ahoyt@moser-inc.com Precedence: bulk X-list: pcp Status: O Content-Length: 2458 Lines: 84 Since I haven’t heard anything in awhile, I was wondering if there is anything I can do to facilitate the open sourcing of these agents. If there are clean up issues – could you offload the work to me – would this help? - Alan - Alan Hoyt wrote: > kenmcd@melbourne.sgi.com wrote: > >> My apologies for the tardiness in responding here ... we've had several >> overlapping issues that have been demanding the attention of our open >> source engineers and our legal team. >> >> I would like to make the following proposal: >> >> 1. We bundle all of the source for the dbms PMDAs into one package >> to be known as "DBMS Add-ons for Performance Co-Pilot". This would >> include PMDA implementations for (old) verions of Oracle, Sybase, >> Informix. >> >> The name issue is important, because Performance Co-Pilot is a >> trademark of SGI, and we have to take certain actions to protect >> that trademark, and one of them is to ensure that there is no >> confusion between the _real_ PCP, and something that "works with PCP". >> Similarly adding Oracle, Informix and Sybase to the title will only >> risk incuring the wrath of a different bunch of lawyers. >> >> 2. We negotiate and select someone to act as the gatekeeper for this >> source. >> > > What would have to be negotiated? > >> SGI is not planning to maintain any down-stream responsibility for >> this code, so in particular we do not want to deal with bug reports >> or patches. >> > > > Make's sense - sometimes I don't even like maintaining my own code let > alone someone else's :-) > > >> We are willing to add a pointer to it from the PCP project >> page on oss.sgi.com. >> >> 3. We will make the source in 1. available to the person(s) selected in >> 2. under the standard GPL. >> >> Thereafter, you can do with it as you please, subject to the terms and >> conditions of the GPL. >> >> Is this acceptable? >> > > Yes > >> Could I ask for volunteers as per 2. above? >> > > > I would like to volunteer as the primary maintainer and help facilitate > the merging/consolidation of Chris's/Mark's agents with the older SGI > authored agents. > > That being said, I would prefer to use Sourceforge as the venue for > providing adjunct PCP agents to the Open Source community – i.e. those > agents that are not/will not be supported by SGI (including my Solaris > agent). > > > > - Alan - > > > > > > >