From nscott@aconex.com Tue Apr 1 18:41:47 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 01 Apr 2008 18:42:02 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m321fkGh024029 for ; Tue, 1 Apr 2008 18:41:47 -0700 X-ASG-Debug-ID: 1207100540-178703340000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 1F00712758FB for ; Tue, 1 Apr 2008 18:42:20 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id MwEBATuU1ytbZ0lI for ; Tue, 01 Apr 2008 18:42:20 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id CCB604D3D44 for ; Wed, 2 Apr 2008 11:26:53 +1100 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id GxcfWkHa3Jta1vq9 for ; Wed, 02 Apr 2008 11:26:53 +1100 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id AD66692C1F1 for ; Wed, 2 Apr 2008 11:26:53 +1100 (EST) X-ASG-Orig-Subj: pcp updates Subject: pcp updates From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Content-Type: text/plain Organization: Aconex Date: Wed, 02 Apr 2008 11:26:52 +1100 Message-Id: <1207096013.21048.107.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207100543 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.52 X-Barracuda-Spam-Status: No, SCORE=-1.52 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=BSF_RULE7568M X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46579 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.50 BSF_RULE7568M Custom Rule 7568M X-archive-position: 1567 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/pcp.git man/man1/GNUmakefile | 2 src/pmdas/jstat/README | 21 +--- src/pmdas/jstat/jstat.c | 235 ++++++++++++++++++++---------------------------- src/pmdas/jstat/jstat.h | 7 - 4 files changed, 110 insertions(+), 155 deletions(-) commit 5acc58d58052f04a2b5826142c5703a923242936 Author: Nathan Scott Date: Wed Apr 2 11:25:25 2008 +1100 Add pmloglabel man page into the list installed by the man1 makefile. commit c41147eaf01219e121257c1fc68fc1d0f2a81f0c Author: Nathan Scott Date: Wed Apr 2 10:50:11 2008 +1100 Completely rework the jstat PMDA instance domain handling. This is now setup on-the-fly using the output of jps(1). commit f6f8122beb3b5093a38b4814c5a83d1ebf731938 Author: Nathan Scott Date: Tue Apr 1 14:40:42 2008 +1100 jstat PMDA time metrics needs to be scaled by the JVM HZ value. From nscott@aconex.com Tue Apr 1 23:11:47 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 01 Apr 2008 23:11:57 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m326BkLx003935 for ; Tue, 1 Apr 2008 23:11:47 -0700 X-ASG-Debug-ID: 1207116739-468403dd0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 6A9B3727E2D for ; Tue, 1 Apr 2008 23:12:19 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id UkRx9GAe4i4prcEu for ; Tue, 01 Apr 2008 23:12:19 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id C22184D7154 for ; Wed, 2 Apr 2008 16:40:26 +1100 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id 6FOaZ02oT1AhKsgW for ; Wed, 02 Apr 2008 16:40:26 +1100 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id B31E592C1F1 for ; Wed, 2 Apr 2008 16:40:26 +1100 (EST) X-ASG-Orig-Subj: pcp updates Subject: pcp updates From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Content-Type: text/plain Organization: Aconex Date: Wed, 02 Apr 2008 16:40:26 +1100 Message-Id: <1207114826.21048.109.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207116742 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -0.47 X-Barracuda-Spam-Status: No, SCORE=-0.47 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=BSF_RULE7568M, BSF_RULE_7582B X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46596 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 1.05 BSF_RULE_7582B BODY: Custom Rule 7582B 0.50 BSF_RULE7568M Custom Rule 7568M X-archive-position: 1568 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/pcp.git VERSION.pcp | 2 debian/changelog | 6 + debian/libpcp-pmda3-dev.install | 28 ++++---- debian/libpcp3-dev.install | 136 ++++++++++++++++++++-------------------- man/man1/pmloglabel.1 | 6 - src/include/builddefs.in | 3 src/pmdas/jstat/jstat.c | 28 +++----- 7 files changed, 106 insertions(+), 103 deletions(-) commit 77f6bd2e7e414b5a737fc5abc47ae38f592f9037 Author: Nathan Scott Date: Wed Apr 2 16:30:03 2008 +1100 Bump build version number. commit 68cb0996d1a3ae641d55c9ac522906122ec30068 Author: Nathan Scott Date: Wed Apr 2 16:29:35 2008 +1100 Fix syntax issues on new pmloglabel man page. commit 3bd0eef0d4e50a6bef9fe64099af806150240f5e Author: Nathan Scott Date: Wed Apr 2 16:27:52 2008 +1100 Ensure we install the same set of man pages always (install produces this list now). commit 1f6d830f3c4d767925344ce8c920269fae833018 Author: Nathan Scott Date: Wed Apr 2 16:14:11 2008 +1100 Make sort use in man page install deterministic. commit fc9149f02d2ae5cedb305a1060848081f6d7c112 Author: Nathan Scott Date: Wed Apr 2 14:14:50 2008 +1100 Fix a silly typo. commit afcea11dcfb5c6485ed8785519433978c4fe7e96 Author: Nathan Scott Date: Wed Apr 2 14:12:22 2008 +1100 Improve handling of transient occurences of unavailable data from jps and jstat, which seems to just happen occassionally. Fix some compiler warnings on Cygwin, and further improve the handling of Windows end-of-line oddities. commit 9fb523c80b4aaaa02aa286573b2ae00697b1d25c Author: Nathan Scott Date: Wed Apr 2 12:08:31 2008 +1100 Handle output from Windows jps, where eol also has back-slash-r. From jamespeach@mac.com Wed Apr 2 13:20:39 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 02 Apr 2008 13:21:04 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m32KKcaO017740 for ; Wed, 2 Apr 2008 13:20:38 -0700 X-ASG-Debug-ID: 1207165461-593c00d70000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from smtpoutm.mac.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 58AC48EBD4F for ; Wed, 2 Apr 2008 12:44:21 -0700 (PDT) Received: from smtpoutm.mac.com (smtpoutm.mac.com [17.148.16.66]) by cuda.sgi.com with ESMTP id KYV60LVEvYBJLRir for ; Wed, 02 Apr 2008 12:44:21 -0700 (PDT) Received: from mac.com (asmtp008-s [10.150.69.71]) by smtpoutm.mac.com (Xserve/smtpout003/MantshX 4.0) with ESMTP id m32JgJfm008483; Wed, 2 Apr 2008 12:42:19 -0700 (PDT) Received: from [192.168.1.70] (adsl-75-18-195-199.dsl.pltn13.sbcglobal.net [75.18.195.199]) (authenticated bits=0) by mac.com (Xserve/asmtp008/MantshX 4.0) with ESMTP id m32JgEUG021620 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Wed, 2 Apr 2008 12:42:16 -0700 (PDT) Message-Id: From: James Peach To: nscott@aconex.com Content-Type: text/plain; charset=WINDOWS-1252; format=flowed; delsp=yes Mime-Version: 1.0 (Apple Message framework v919.2) X-ASG-Orig-Subj: missing xcode projects when building kmchart Subject: missing xcode projects when building kmchart Date: Wed, 2 Apr 2008 12:42:09 -0700 Cc: pcp@oss.sgi.com X-Mailer: Apple Mail (2.919.2) X-Barracuda-Connect: smtpoutm.mac.com[17.148.16.66] X-Barracuda-Start-Time: 1207165463 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46649 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by oss.sgi.com id m32KKcaO017740 X-archive-position: 1569 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: jamespeach@mac.com Precedence: bulk X-list: pcp Hi Nathan, I pulled the kmchart git tree, but I'm a bit stumped by the build: blacko:kmchart.git jpeach$ make === src === === include === make[2]: Nothing to be done for `default'. === libqmc === /opt/local/bin/qmake && xcodebuild xcodebuild: Error: the directory /Users/jpeach/src/kmchart.git/src/ libqmc does not contain an Xcode project. make[2]: *** [default] Error 1 make[1]: *** [default] Error 2 make: *** [default] Error 2 It looks like qmake is expected to generate a xcode project, but it doesn't actually do it. Not sure what information I can provide that would be helpful, so here's a random sampling .... blacko:kmchart.git jpeach$ qmake --version QMake version 2.01a Using Qt version 4.3.4 in /opt/local/lib blacko:kmchart.git jpeach$ sw_vers ProductName: Mac OS X ProductVersion: 10.5.2 BuildVersion: 9C7010 blacko:kmchart.git jpeach$ gcc --version i686-apple-darwin9-gcc-4.0.1 (GCC) 4.0.1 (Apple Inc. build 5470) (Aspen 5470.3) blacko:kmchart.git jpeach$ git status # On branch master # Untracked files: # (use "git add ..." to include in what will be committed) # # Logs/ # autom4te.cache/ # config.log # config.status # configure # man/html/kmchart.adp # man/html/kmtime.adp # src/include/builddefs # src/include/version.h # src/libqmc/Makefile Now, if I build the generated Makefile, I still get errors, though these look like some problem with system headers .... blacko:kmchart.git jpeach$ cd src/libqmc/ blacko:libqmc jpeach$ make -f Makefile c++ -c -pipe -Os -Wall -W -fPIC -DQT_NO_DEBUG -DQT_CORE_LIB - DQT_SHARED -I/opt/local/share/qt4/mkspecs/macx-g++ -I. -I/opt/local/ include/qt4/QtCore -I/opt/local/include/qt4/QtCore -I/opt/local/ include/qt4 -I. -I. -o qmc_context.o qmc_context.cpp /usr/lib/gcc/i686-apple-darwin9/4.0.1/include/stddef.h:152: error: declaration does not declare anything /usr/include/c++/4.0.0/cstddef:52: error: expected unqualified-id before ‘long’ /usr/include/c++/4.0.0/cstddef:52: error: expected `;' before ‘long’ /usr/include/c++/4.0.0/cstddef:52: error: declaration does not declare anything make: *** [qmc_context.o] Error 1 From nscott@aconex.com Wed Apr 2 16:19:53 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 02 Apr 2008 16:20:05 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m32NJq5W018776 for ; Wed, 2 Apr 2008 16:19:53 -0700 X-ASG-Debug-ID: 1207178425-16b300980000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 3848373113C for ; Wed, 2 Apr 2008 16:20:25 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id L9AUYq0utTrrIwo5 for ; Wed, 02 Apr 2008 16:20:25 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id D05844DF436; Thu, 3 Apr 2008 10:20:20 +1100 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id wFffKEJ5nwEG3UI1; Thu, 03 Apr 2008 10:20:20 +1100 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id B8A2D92C0DE; Thu, 3 Apr 2008 10:20:20 +1100 (EST) X-ASG-Orig-Subj: Re: missing xcode projects when building kmchart Subject: Re: missing xcode projects when building kmchart From: Nathan Scott Reply-To: nscott@aconex.com To: James Peach Cc: pcp@oss.sgi.com In-Reply-To: References: Content-Type: text/plain Organization: Aconex Date: Thu, 03 Apr 2008 10:20:20 +1100 Message-Id: <1207178420.21048.123.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207178429 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46665 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1570 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Hey James, On Wed, 2008-04-02 at 12:42 -0700, James Peach wrote: > Hi Nathan, > > I pulled the kmchart git tree, but I'm a bit stumped by the build: > > blacko:kmchart.git jpeach$ make > === src === > === include === > make[2]: Nothing to be done for `default'. > === libqmc === > /opt/local/bin/qmake && xcodebuild > xcodebuild: Error: the directory /Users/jpeach/src/kmchart.git/src/ > libqmc does not contain an Xcode project. Hmm. qmake is supposed to generate a valid xcode project, which we then just blindly build. So, I guess qmake is not doing the right thing on this version of MacOSX / xcode. Does adding "-macx" to the qmake command line help (this is coming from src/include/builddefs[.in], IIRC). cheers. -- Nathan From jamespeach@mac.com Wed Apr 2 21:17:10 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 02 Apr 2008 21:17:35 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m334H9mt001120 for ; Wed, 2 Apr 2008 21:17:10 -0700 X-ASG-Debug-ID: 1207196265-5d5402a00000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from smtpoutm.mac.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id E0082732D9A for ; Wed, 2 Apr 2008 21:17:45 -0700 (PDT) Received: from smtpoutm.mac.com (smtpoutm.mac.com [17.148.16.79]) by cuda.sgi.com with ESMTP id RFPzBRSVBEOPvzfx for ; Wed, 02 Apr 2008 21:17:45 -0700 (PDT) Received: from mac.com (asmtp006-s [10.150.69.69]) by smtpoutm.mac.com (Xserve/smtpout016/MantshX 4.0) with ESMTP id m334HSYY025547; Wed, 2 Apr 2008 21:17:28 -0700 (PDT) Received: from [192.168.1.70] (adsl-75-18-195-199.dsl.pltn13.sbcglobal.net [75.18.195.199]) (authenticated bits=0) by mac.com (Xserve/asmtp006/MantshX 4.0) with ESMTP id m334HPXO025344 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Wed, 2 Apr 2008 21:17:27 -0700 (PDT) Cc: pcp@oss.sgi.com Message-Id: <23CD9178-1413-4F82-B3AB-789A34541701@mac.com> From: James Peach To: nscott@aconex.com In-Reply-To: <1207178420.21048.123.camel@edge.scott.net.au> Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v919.2) X-ASG-Orig-Subj: Re: missing xcode projects when building kmchart Subject: Re: missing xcode projects when building kmchart Date: Wed, 2 Apr 2008 21:17:24 -0700 References: <1207178420.21048.123.camel@edge.scott.net.au> X-Mailer: Apple Mail (2.919.2) X-Barracuda-Connect: smtpoutm.mac.com[17.148.16.79] X-Barracuda-Start-Time: 1207196265 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46685 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1571 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: jamespeach@mac.com Precedence: bulk X-list: pcp On 02/04/2008, at 4:20 PM, Nathan Scott wrote: > Hey James, > > On Wed, 2008-04-02 at 12:42 -0700, James Peach wrote: >> Hi Nathan, >> >> I pulled the kmchart git tree, but I'm a bit stumped by the build: >> >> blacko:kmchart.git jpeach$ make >> === src === >> === include === >> make[2]: Nothing to be done for `default'. >> === libqmc === >> /opt/local/bin/qmake && xcodebuild >> xcodebuild: Error: the directory /Users/jpeach/src/kmchart.git/src/ >> libqmc does not contain an Xcode project. > > Hmm. qmake is supposed to generate a valid xcode project, > which we then just blindly build. So, I guess qmake is not > doing the right thing on this version of MacOSX / xcode. > > Does adding "-macx" to the qmake command line help (this is > coming from src/include/builddefs[.in], IIRC). Nope, it still generates the Makefile. so I looked at the headers issue .... /usr/lib/gcc/i686-apple-darwin9/4.0.1/include/stddef.h has this declaration: typedef __PTRDIFF_TYPE__ ptrdiff_t; But there must be some namespace pollution, because the c++ -E output shows this: # 1 "/usr/lib/gcc/i686-apple-darwin9/4.0.1/include/stddef.h" 1 3 4 # 152 "/usr/lib/gcc/i686-apple-darwin9/4.0.1/include/stddef.h" 3 4 typedef int long; which is obviously bogus. Using c++ -E -dD, it looks like the definition of __PTRDIFF_TYPE__ is a gcc builtin, but the bogus ptrdiff_t definition comes from /usr/include/pcp/platform_defs.h. platform_defs.h contains this little gem: #if defined(HAVE_MALLOC_H) #include #else /* #undef ptrdiff_t */ #define ptrdiff_t long #endif From nscott@aconex.com Wed Apr 2 21:31:13 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 02 Apr 2008 21:31:28 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m334VCeo003212 for ; Wed, 2 Apr 2008 21:31:13 -0700 X-ASG-Debug-ID: 1207197108-654902790000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 9A966732E7B for ; Wed, 2 Apr 2008 21:31:49 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id uaz1v8cZbHfWzUHV for ; Wed, 02 Apr 2008 21:31:49 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id B546F4E1FA9; Thu, 3 Apr 2008 15:31:46 +1100 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id gSnyIC3iKrv81zzH; Thu, 03 Apr 2008 15:31:46 +1100 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id A18C492C1F3; Thu, 3 Apr 2008 15:31:46 +1100 (EST) X-ASG-Orig-Subj: Re: missing xcode projects when building kmchart Subject: Re: missing xcode projects when building kmchart From: Nathan Scott Reply-To: nscott@aconex.com To: James Peach Cc: pcp@oss.sgi.com In-Reply-To: <23CD9178-1413-4F82-B3AB-789A34541701@mac.com> References: <1207178420.21048.123.camel@edge.scott.net.au> <23CD9178-1413-4F82-B3AB-789A34541701@mac.com> Content-Type: text/plain Organization: Aconex Date: Thu, 03 Apr 2008 15:31:46 +1100 Message-Id: <1207197106.21048.136.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207197109 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46685 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1572 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp On Wed, 2008-04-02 at 21:17 -0700, James Peach wrote: > > platform_defs.h contains this little gem: > > #if defined(HAVE_MALLOC_H) > #include > #else > /* #undef ptrdiff_t */ > #define ptrdiff_t long > #endif > Oh, you're not using my git tree - I fixed that one months back. Your platform_defs.h should have a ptrdiff section along the lines... /* Check if ptrdiff_t type is available */ #define HAVE_PTRDIFF_T 1 #ifndef HAVE_PTRDIFF_T #define ptrdiff_t long #endif Your earlier snippet rings bells in my distant memory as to what the original pcp header looked like. WRT, the qmake/xcode issue, I may be able to get hold of a 10.5 system tonight - I'll see if I can reproduce and will look into it a bit more. cheers. -- Nathan From nscott@aconex.com Thu Apr 3 00:01:49 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 03 Apr 2008 00:02:17 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3371k10025799 for ; Thu, 3 Apr 2008 00:01:48 -0700 X-ASG-Debug-ID: 1207206141-16de01590000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 1FD15733E8F for ; Thu, 3 Apr 2008 00:02:21 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id QBGJghdtAmHqlJat for ; Thu, 03 Apr 2008 00:02:21 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id DA7144E3CD2 for ; Thu, 3 Apr 2008 17:12:38 +1100 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id NSFoRwA11FD1aYGg for ; Thu, 03 Apr 2008 17:12:38 +1100 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id C722A92C1F7 for ; Thu, 3 Apr 2008 17:12:38 +1100 (EST) X-ASG-Orig-Subj: kmchart updates Subject: kmchart updates From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Content-Type: text/plain Organization: Aconex Date: Thu, 03 Apr 2008 17:12:38 +1100 Message-Id: <1207203158.21048.143.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207206143 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46696 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1573 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/kmchart.git build/mac/build-installer | 2 build/mac/installer-info | 20 man/man1/kmquery.1 | 1 man/man1/kmtime.1 | 1 src/dumptext/kmdumptext.cpp | 94 +-- src/libqwt/libqwt.pro | 20 src/libqwt/qwt_analog_clock.cpp | 232 ------- src/libqwt/qwt_analog_clock.h | 94 --- src/libqwt/qwt_arrow_button.cpp | 367 ----------- src/libqwt/qwt_arrow_button.h | 54 - src/libqwt/qwt_compass.cpp | 312 --------- src/libqwt/qwt_compass.h | 85 -- src/libqwt/qwt_compass_rose.cpp | 281 -------- src/libqwt/qwt_compass_rose.h | 85 -- src/libqwt/qwt_counter.cpp | 625 ------------------- src/libqwt/qwt_counter.h | 157 ----- src/libqwt/qwt_dial.cpp | 1252 ---------------------------------------- src/libqwt/qwt_dial.h | 214 ------ src/libqwt/qwt_dial_needle.cpp | 584 ------------------ src/libqwt/qwt_dial_needle.h | 198 ------ src/libqwt/qwt_knob.cpp | 543 ----------------- src/libqwt/qwt_knob.h | 100 --- src/libqwt/qwt_thermo.cpp | 835 -------------------------- src/libqwt/qwt_thermo.h | 182 ----- 24 files changed, 62 insertions(+), 6276 deletions(-) commit 0069f352ca49f6786a683a645feb318b693bbe70 Author: Nathan Scott Date: Thu Apr 3 08:51:21 2008 +1100 Remove unused widgets from our local qwt sources. commit 143a1f144353f3ebbb5da2f9e44ba631797c340d Author: Nathan Scott Date: Thu Apr 3 08:31:14 2008 +1100 Fix up copyright notice on couple of man pages. commit 17b1284871d5b0e4bd5f869d51eddb2152c5e682 Author: Nathan Scott Date: Thu Apr 3 08:30:28 2008 +1100 Resolve a dumptext formatting issue with Qt stream formatters. commit 80e368815b5bd3f3cbc53d6e28f62cabc29611f8 Author: Nathan Scott Date: Thu Apr 3 08:29:34 2008 +1100 Relax MacOSX install requirements on "next" OSX version from build system. From jamespeach@mac.com Thu Apr 3 16:06:15 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 03 Apr 2008 16:06:23 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m33N6DlQ011428 for ; Thu, 3 Apr 2008 16:06:14 -0700 X-ASG-Debug-ID: 1207254486-7ea700940000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from smtpoutm.mac.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 714BF73A8A4 for ; Thu, 3 Apr 2008 13:28:06 -0700 (PDT) Received: from smtpoutm.mac.com (smtpoutm.mac.com [17.148.16.78]) by cuda.sgi.com with ESMTP id 384CnFEV9cFABuD1 for ; Thu, 03 Apr 2008 13:28:06 -0700 (PDT) Received: from mac.com (asmtp002-s [10.150.69.65]) by smtpoutm.mac.com (Xserve/smtpout015/MantshX 4.0) with ESMTP id m33KRGw2024612 for ; Thu, 3 Apr 2008 13:27:16 -0700 (PDT) Received: from [192.168.1.70] (adsl-75-18-195-199.dsl.pltn13.sbcglobal.net [75.18.195.199]) (authenticated bits=0) by mac.com (Xserve/asmtp002/MantshX 4.0) with ESMTP id m33KQXmF023960 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO) for ; Thu, 3 Apr 2008 13:26:46 -0700 (PDT) Message-Id: <7BE130C4-F6A6-428D-82D5-A7D7EE399E87@mac.com> From: James Peach To: pcp@oss.sgi.com Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v919.2) X-ASG-Orig-Subj: [PATCH] respect configured prefix instead of hardcoded /etc Subject: [PATCH] respect configured prefix instead of hardcoded /etc Date: Thu, 3 Apr 2008 13:26:32 -0700 X-Mailer: Apple Mail (2.919.2) X-Barracuda-Connect: smtpoutm.mac.com[17.148.16.78] X-Barracuda-Start-Time: 1207254488 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46749 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1576 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: jamespeach@mac.com Precedence: bulk X-list: pcp commit f3d97e64f2774ea4414c0556e17d7fdd83e30df6 Author: James Peach Date: Thu Apr 3 13:22:11 2008 -0700 Use the prefix-relative etc directory instead of harcoding /etc. diff --git a/GNUmakefile b/GNUmakefile index a945efd..077f105 100644 --- a/GNUmakefile +++ b/GNUmakefile @@ -104,6 +104,7 @@ endif $(INSTALL) -m 755 -d $(PCP_SHARE_DIR)/lib $(INSTALL) -m 755 -d $(PCP_SHARE_DIR)/examples $(INSTALL) -m 755 -d $(PCP_INC_DIR) + $(INSTALL) -m 755 -d $(PCP_ETC_DIR) $(INSTALL) -m 755 -d $(PCP_VAR_DIR) $(INSTALL) -m 755 -d $(PCP_VAR_DIR)/config $(INSTALL) -m 755 -d $(PCP_VAR_DIR)/config/pmchart diff --git a/configure.in b/configure.in index 5587a11..d75647a 100644 --- a/configure.in +++ b/configure.in @@ -1151,15 +1151,7 @@ if test ! -z "$PCP_ETC_DIR" then pcp_etc_dir=$PCP_ETC_DIR else - if test -d /etc - then - pcp_etc_dir=/etc - else - echo "FATAL ERROR: could not find /etc directory." - echo "You need to set \$PCP_ETC_DIR in the environment" - echo "as the path to the directory containing \"pcp.conf\"." - exit 1 - fi + pcp_etc_dir=`eval echo $sysconfdir` fi AC_SUBST(pcp_etc_dir) @@ -1240,12 +1232,12 @@ then pcp_pmloggercontrol_path=$pcp_var_dir/config/pmlogger/control elif test $target_os = darwin -o $target_distro = debian then - pcp_pmcdconf_path=/etc/pmcd/pmcd.conf - pcp_pmcdrclocal_path=/etc/pmcd/rc.local - pcp_pmcdoptions_path=/etc/pmcd/pmcd.options - pcp_pmproxyoptions_path=/etc/pmproxy/pmproxy.options - pcp_pmiecontrol_path=/etc/pmie/control - pcp_pmloggercontrol_path=/etc/pmlogger/control + pcp_pmcdconf_path=$pcp_etc_dir/pmcd/pmcd.conf + pcp_pmcdrclocal_path=$pcp_etc_dir/pmcd/rc.local + pcp_pmcdoptions_path=$pcp_etc_dir/pmcd/pmcd.options + pcp_pmproxyoptions_path=$pcp_etc_dir/pmproxy/pmproxy.options + pcp_pmiecontrol_path=$pcp_etc_dir/pmie/control + pcp_pmloggercontrol_path=$pcp_etc_dir/pmlogger/control else pcp_pmcdconf_path=$pcp_var_dir/config/pmcd/pmcd.conf pcp_pmcdrclocal_path=$pcp_var_dir/config/pmcd/rc.local diff --git a/src/include/GNUmakefile b/src/include/GNUmakefile index 49309e4..3e1f163 100644 --- a/src/include/GNUmakefile +++ b/src/include/GNUmakefile @@ -43,8 +43,8 @@ include $(BUILDRULES) install :: default_pcp install_pcp install_pcp : default_pcp - $(INSTALL) -m 644 pcp.conf /etc/pcp.conf - $(INSTALL) -m 644 pcp.env /etc/pcp.env + $(INSTALL) -m 644 pcp.conf $(PCP_ETC_DIR)/pcp.conf + $(INSTALL) -m 644 pcp.env $(PCP_ETC_DIR)/pcp.env $(INSTALL) -m 644 $(HFILES) $(GENERATED_HFILES) buildrules $ (PCP_INC_DIR) $(INSTALL) -m 644 builddefs.install $(PCP_INC_DIR)/builddefs diff --git a/src/include/pcp.conf.in b/src/include/pcp.conf.in index d028f53..93ec70f 100644 --- a/src/include/pcp.conf.in +++ b/src/include/pcp.conf.in @@ -39,6 +39,10 @@ # default on the platform on which this file is installed. # +# directory for config files +# Standard path: /etc +PCP_ETC_DIR=@pcp_etc_dir@ + # directory for rc/startup scripts # Standard path: /etc/init.d PCP_RC_DIR=@pcp_rc_dir@ From makc@melbourne.sgi.com Thu Apr 3 16:22:51 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 03 Apr 2008 16:23:00 -0700 (PDT) Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with SMTP id m33NMlJG020295 for ; Thu, 3 Apr 2008 16:22:50 -0700 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 JAA03308; Fri, 4 Apr 2008 09:23:19 +1000 Received: from kuku.melbourne.sgi.com (localhost [127.0.0.1]) by kuku.melbourne.sgi.com (SGI-8.12.11.20060308/8.12.11) with ESMTP id m33NNIgY019919; Fri, 4 Apr 2008 10:23:18 +1100 (EST) Received: (from makc@localhost) by kuku.melbourne.sgi.com (SGI-8.12.11.20060308/8.12.11/Submit) id m33NNIQU019924; Fri, 4 Apr 2008 10:23:18 +1100 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <18421.26341.981959.130522@kuku.melbourne.sgi.com> Date: Fri, 4 Apr 2008 10:23:17 +1100 From: Max Matveev To: James Peach Cc: pcp@oss.sgi.com Subject: Re: [PATCH] respect configured prefix instead of hardcoded /etc In-Reply-To: <7BE130C4-F6A6-428D-82D5-A7D7EE399E87@mac.com> References: <7BE130C4-F6A6-428D-82D5-A7D7EE399E87@mac.com> X-Mailer: VM 7.07 under 21.4 (patch 15) "Security Through Obscurity" XEmacs Lucid X-archive-position: 1577 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: makc@sgi.com Precedence: bulk X-list: pcp On Thu, 3 Apr 2008 13:26:32 -0700, James Peach wrote: JP> commit f3d97e64f2774ea4414c0556e17d7fdd83e30df6 JP> Author: James Peach JP> Date: Thu Apr 3 13:22:11 2008 -0700 JP> Use the prefix-relative etc directory instead of harcoding /etc. JP> diff --git a/GNUmakefile b/GNUmakefile JP> index a945efd..077f105 100644 JP> --- a/GNUmakefile JP> +++ b/GNUmakefile JP> @@ -104,6 +104,7 @@ endif JP> $(INSTALL) -m 755 -d $(PCP_SHARE_DIR)/lib JP> $(INSTALL) -m 755 -d $(PCP_SHARE_DIR)/examples JP> $(INSTALL) -m 755 -d $(PCP_INC_DIR) JP> + $(INSTALL) -m 755 -d $(PCP_ETC_DIR) That would make pcp the owner of /etc on Linux: pcp does not have its own /etc/pcp directory and splats its files around the shared directory. Also, /etc/pcp.conf is hardcoded deliberately to avoid chicken'n'egg problem finding the source of information about relocation which can be relocatable itself. max From nscott@aconex.com Thu Apr 3 16:31:45 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 03 Apr 2008 16:31:54 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m33NVjwc022700 for ; Thu, 3 Apr 2008 16:31:45 -0700 X-ASG-Debug-ID: 1207265539-44a700580000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 2988E912149 for ; Thu, 3 Apr 2008 16:32:20 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id 9Wxfm3fngLpO96BU for ; Thu, 03 Apr 2008 16:32:20 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id C9ACF4EBA6B; Fri, 4 Apr 2008 10:18:54 +1100 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id C3KGPVHQ5QS1COl0; Fri, 04 Apr 2008 10:18:54 +1100 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id BB18F92C0E2; Fri, 4 Apr 2008 10:18:54 +1100 (EST) X-ASG-Orig-Subj: Re: [PATCH] respect configured prefix instead of hardcoded /etc Subject: Re: [PATCH] respect configured prefix instead of hardcoded /etc From: Nathan Scott Reply-To: nscott@aconex.com To: James Peach Cc: pcp@oss.sgi.com In-Reply-To: <7BE130C4-F6A6-428D-82D5-A7D7EE399E87@mac.com> References: <7BE130C4-F6A6-428D-82D5-A7D7EE399E87@mac.com> Content-Type: text/plain Organization: Aconex Date: Fri, 04 Apr 2008 10:18:54 +1100 Message-Id: <1207264734.21048.156.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207265542 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46760 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1578 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp On Thu, 2008-04-03 at 13:26 -0700, James Peach wrote: > > commit f3d97e64f2774ea4414c0556e17d7fdd83e30df6 > Author: James Peach > Date: Thu Apr 3 13:22:11 2008 -0700 > > Use the prefix-relative etc directory instead of harcoding /etc. Can you give some more details as to why? I don't think this is going to work as you'd expect - in particular pmGetConfig(3) will not like this, as it (kinda has to) hard-code the pcp.conf location. cheers. -- Nathan From nscott@aconex.com Thu Apr 3 17:01:44 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 03 Apr 2008 17:01:53 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3401iTw028029 for ; Thu, 3 Apr 2008 17:01:44 -0700 X-ASG-Debug-ID: 1207267339-3fce01cf0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id D9F61106E58E for ; Thu, 3 Apr 2008 17:02:20 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id vHHC3SwY0xKHWUGH for ; Thu, 03 Apr 2008 17:02:20 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id D48494EBE93 for ; Fri, 4 Apr 2008 10:48:15 +1100 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id fmQuGkbiMqmH7iMN for ; Fri, 04 Apr 2008 10:48:15 +1100 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id C594092C0DE for ; Fri, 4 Apr 2008 10:48:15 +1100 (EST) X-ASG-Orig-Subj: pcp updates Subject: pcp updates From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Content-Type: text/plain Organization: Aconex Date: Fri, 04 Apr 2008 10:48:15 +1100 Message-Id: <1207266495.21048.161.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207267340 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46762 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1579 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/pcp.git VERSION.pcp | 2 +- configure.in | 11 ++++++++++- debian/changelog | 6 ++++++ debian/control | 2 +- debian/libpcp3-dev.install | 7 ------- debian/rules | 2 +- man/man1/GNUmakefile | 8 ++++++-- man/man1/pcpintro.1 | 9 +++++++-- man/man1/pmdajstat.1 | 37 +++++++++++++++++++++++++++++-------- man/man1/pminfo.1 | 1 + man/man1/pmlogsummary.1 | 2 ++ man/man3/GNUmakefile | 8 ++++++-- src/GNUmakefile | 6 +++++- src/cpan/TODO | 2 -- src/include/builddefs.in | 3 +-- src/include/platform_defs.h.in | 1 + 16 files changed, 77 insertions(+), 30 deletions(-) commit 443ea791cc87fc7d9a2f62212bc213e0f1e04e6d Author: Nathan Scott Date: Fri Apr 4 10:42:18 2008 +1100 Bump build version number. commit 12fe26ac32d3218857fbae36448401224f80fbba Author: Nathan Scott Date: Fri Apr 4 08:38:33 2008 +1100 Remove dependency on libncurses-dev for Debian builds, as its no longer needed. commit 80193536f603d1c9fa1e09e4fe175d1d58c7bcfe Author: Nathan Scott Date: Fri Apr 4 08:30:11 2008 +1100 Update Perl PMDA interface TODO list to remove a completed item. commit 04f29f1c463a5498be540f97e67e432fe97d5df8 Author: Nathan Scott Date: Fri Apr 4 08:26:44 2008 +1100 Update the pmdajstat(1) manual page to reflect recent changes, particularly related to instance domain handling. commit 20f716a73831ed7de45621678c1395213f4f2d13 Author: Nathan Scott Date: Fri Apr 4 08:15:08 2008 +1100 Make the PMC library and PMC-based tools configurable. This gets PCP building in places where the deprecated iostream.h header doesn't necessarily exist (Cygwin, and several Linux distros now). Equivalent functionality, but using Qt-based IOStream functionality, is available in the kmchart package. From jamespeach@mac.com Thu Apr 3 20:41:49 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 03 Apr 2008 20:41:58 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m343fmVh007962 for ; Thu, 3 Apr 2008 20:41:49 -0700 X-ASG-Debug-ID: 1207280542-253703ce0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from smtpoutm.mac.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id DBCCD73D689; Thu, 3 Apr 2008 20:42:22 -0700 (PDT) Received: from smtpoutm.mac.com (smtpoutm.mac.com [17.148.16.80]) by cuda.sgi.com with ESMTP id GoYvwVyIE9uEkctK; Thu, 03 Apr 2008 20:42:22 -0700 (PDT) Received: from mac.com (asmtp008-s [10.150.69.71]) by smtpoutm.mac.com (Xserve/smtpout017/MantshX 4.0) with ESMTP id m343gMer001000; Thu, 3 Apr 2008 20:42:22 -0700 (PDT) Received: from [192.168.1.70] (adsl-75-18-195-199.dsl.pltn13.sbcglobal.net [75.18.195.199]) (authenticated bits=0) by mac.com (Xserve/asmtp008/MantshX 4.0) with ESMTP id m343gJYF028525 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Thu, 3 Apr 2008 20:42:20 -0700 (PDT) Cc: pcp@oss.sgi.com Message-Id: <03ADE2A4-AAD6-4C8B-850D-1352D5D4C7DC@mac.com> From: James Peach To: Max Matveev In-Reply-To: <18421.26341.981959.130522@kuku.melbourne.sgi.com> Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v919.2) X-ASG-Orig-Subj: Re: [PATCH] respect configured prefix instead of hardcoded /etc Subject: Re: [PATCH] respect configured prefix instead of hardcoded /etc Date: Thu, 3 Apr 2008 20:42:19 -0700 References: <7BE130C4-F6A6-428D-82D5-A7D7EE399E87@mac.com> <18421.26341.981959.130522@kuku.melbourne.sgi.com> X-Mailer: Apple Mail (2.919.2) X-Barracuda-Connect: smtpoutm.mac.com[17.148.16.80] X-Barracuda-Start-Time: 1207280545 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46777 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1580 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: jamespeach@mac.com Precedence: bulk X-list: pcp On 03/04/2008, at 4:23 PM, Max Matveev wrote: > On Thu, 3 Apr 2008 13:26:32 -0700, James Peach wrote: > > JP> commit f3d97e64f2774ea4414c0556e17d7fdd83e30df6 > JP> Author: James Peach > JP> Date: Thu Apr 3 13:22:11 2008 -0700 > > JP> Use the prefix-relative etc directory instead of harcoding / > etc. > > JP> diff --git a/GNUmakefile b/GNUmakefile > JP> index a945efd..077f105 100644 > JP> --- a/GNUmakefile > JP> +++ b/GNUmakefile > JP> @@ -104,6 +104,7 @@ endif > JP> $(INSTALL) -m 755 -d $(PCP_SHARE_DIR)/lib > JP> $(INSTALL) -m 755 -d $(PCP_SHARE_DIR)/examples > JP> $(INSTALL) -m 755 -d $(PCP_INC_DIR) > JP> + $(INSTALL) -m 755 -d $(PCP_ETC_DIR) > That would make pcp the owner of /etc on Linux: Then shouldn't $(INSTALL) specify the owner and group? > pcp does not have its > own /etc/pcp directory and splats its files around the shared > directory. ick :) > Also, /etc/pcp.conf is hardcoded deliberately to avoid chicken'n'egg > problem finding the source of information about relocation which can > be relocatable itself. So does this only apply to pcp.conf? It seems like the build hardcodes all the config files to go into /etc ... I'm trying to gradually teach PCP not to blat itself all over my filesystem when I give configure the --prefix option. I'm very happy to take advice .... From markgw@sgi.com Thu Apr 3 20:56:21 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 03 Apr 2008 20:56:29 -0700 (PDT) Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with SMTP id m343uG21010075 for ; Thu, 3 Apr 2008 20:56:19 -0700 Received: from [134.14.55.21] (dhcp21.melbourne.sgi.com [134.14.55.21]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id NAA10109; Fri, 4 Apr 2008 13:56:43 +1000 Message-ID: <47F5A6F5.1070604@sgi.com> Date: Fri, 04 Apr 2008 14:56:37 +1100 From: Mark Goodwin Reply-To: markgw@sgi.com Organization: SGI Engineering User-Agent: Thunderbird 1.5.0.14 (Windows/20071210) MIME-Version: 1.0 To: James Peach CC: Max Matveev , pcp@oss.sgi.com Subject: Re: [PATCH] respect configured prefix instead of hardcoded /etc References: <7BE130C4-F6A6-428D-82D5-A7D7EE399E87@mac.com> <18421.26341.981959.130522@kuku.melbourne.sgi.com> <03ADE2A4-AAD6-4C8B-850D-1352D5D4C7DC@mac.com> In-Reply-To: <03ADE2A4-AAD6-4C8B-850D-1352D5D4C7DC@mac.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-archive-position: 1581 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: markgw@sgi.com Precedence: bulk X-list: pcp James Peach wrote: > On 03/04/2008, at 4:23 PM, Max Matveev wrote: >> On Thu, 3 Apr 2008 13:26:32 -0700, James Peach wrote: >> >> JP> commit f3d97e64f2774ea4414c0556e17d7fdd83e30df6 >> JP> Author: James Peach >> JP> Date: Thu Apr 3 13:22:11 2008 -0700 >> >> JP> Use the prefix-relative etc directory instead of harcoding /etc. >> >> JP> diff --git a/GNUmakefile b/GNUmakefile >> JP> index a945efd..077f105 100644 >> JP> --- a/GNUmakefile >> JP> +++ b/GNUmakefile >> JP> @@ -104,6 +104,7 @@ endif >> JP> $(INSTALL) -m 755 -d $(PCP_SHARE_DIR)/lib >> JP> $(INSTALL) -m 755 -d $(PCP_SHARE_DIR)/examples >> JP> $(INSTALL) -m 755 -d $(PCP_INC_DIR) >> JP> + $(INSTALL) -m 755 -d $(PCP_ETC_DIR) >> That would make pcp the owner of /etc on Linux: > > Then shouldn't $(INSTALL) specify the owner and group? I don't think you want to change user/group for /etc, nor do we want /etc to be owned by both the 'filesystem' package and pcp. >> pcp does not have its >> own /etc/pcp directory and splats its files around the shared >> directory. > > ick :) We could move all this stuff into an /etc/pcp directory if that would be more palatable. But it won't be easy. >> Also, /etc/pcp.conf is hardcoded deliberately to avoid chicken'n'egg >> problem finding the source of information about relocation which can >> be relocatable itself. > > So does this only apply to pcp.conf? It seems like the build hardcodes > all the config files to go into /etc ... Actually, it only applies to /etc/pcp.env. The location of pcp.conf is controlled by $PCP_CONF in the environment. > I'm trying to gradually teach PCP not to blat itself all over my > filesystem when I give configure the --prefix option. I'm very happy to > take advice .... It's painful but doable. Keep plugging away ... :) Cheers -- Mark From nscott@aconex.com Thu Apr 3 21:21:44 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 03 Apr 2008 21:21:53 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m344Lh2x013934 for ; Thu, 3 Apr 2008 21:21:44 -0700 X-ASG-Debug-ID: 1207282939-483e01400000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 6A8BC73D745 for ; Thu, 3 Apr 2008 21:22:20 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id aHTsMcabQNfUXJvN for ; Thu, 03 Apr 2008 21:22:20 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 14D294EE610; Fri, 4 Apr 2008 15:05:50 +1100 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id 0rGtZdYsCUzBF4R4; Fri, 04 Apr 2008 15:05:50 +1100 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id 02B4D92C0DE; Fri, 4 Apr 2008 15:05:50 +1100 (EST) X-ASG-Orig-Subj: Re: [PATCH] respect configured prefix instead of hardcoded /etc Subject: Re: [PATCH] respect configured prefix instead of hardcoded /etc From: Nathan Scott Reply-To: nscott@aconex.com To: James Peach , markgw@sgi.com Cc: Max Matveev , pcp@oss.sgi.com In-Reply-To: <47F5A6F5.1070604@sgi.com> References: <7BE130C4-F6A6-428D-82D5-A7D7EE399E87@mac.com> <18421.26341.981959.130522@kuku.melbourne.sgi.com> <03ADE2A4-AAD6-4C8B-850D-1352D5D4C7DC@mac.com> <47F5A6F5.1070604@sgi.com> Content-Type: text/plain Organization: Aconex Date: Fri, 04 Apr 2008 15:05:49 +1100 Message-Id: <1207281949.21048.166.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207282940 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46781 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1582 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp On Fri, 2008-04-04 at 14:56 +1100, Mark Goodwin wrote: > > We could move all this stuff into an /etc/pcp directory if that > would be more palatable. But it won't be easy. This is largely done already - the Debian packages put things in /etc/{pmcd,pmlogger,pmie}/*. But James, why don't you just type "./Makepkgs" and use the resulting .dmg file instead of worrying about files being "splatted" on your root? (that gives you a neat little package you can install/uninstall all at once). cheers. -- Nathan From jamespeach@mac.com Thu Apr 3 21:59:39 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 03 Apr 2008 21:59:54 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m344xaES019341 for ; Thu, 3 Apr 2008 21:59:39 -0700 X-ASG-Debug-ID: 1207285213-424902ed0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from smtpoutm.mac.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 7C97F73D36B for ; Thu, 3 Apr 2008 22:00:13 -0700 (PDT) Received: from smtpoutm.mac.com (smtpoutm.mac.com [17.148.16.64]) by cuda.sgi.com with ESMTP id indOOmZdGOII3tme for ; Thu, 03 Apr 2008 22:00:13 -0700 (PDT) Received: from mac.com (asmtp009-s [10.150.69.72]) by smtpoutm.mac.com (Xserve/smtpout001/MantshX 4.0) with ESMTP id m343btDb019507; Thu, 3 Apr 2008 20:37:55 -0700 (PDT) Received: from [192.168.1.70] (adsl-75-18-195-199.dsl.pltn13.sbcglobal.net [75.18.195.199]) (authenticated bits=0) by mac.com (Xserve/asmtp009/MantshX 4.0) with ESMTP id m343blFb028911 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Thu, 3 Apr 2008 20:37:54 -0700 (PDT) Cc: pcp@oss.sgi.com Message-Id: <0189EA57-5C71-46DD-BC30-C63E54D63C50@mac.com> From: James Peach To: nscott@aconex.com In-Reply-To: <1207264734.21048.156.camel@edge.scott.net.au> Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v919.2) X-ASG-Orig-Subj: Re: [PATCH] respect configured prefix instead of hardcoded /etc Subject: Re: [PATCH] respect configured prefix instead of hardcoded /etc Date: Thu, 3 Apr 2008 20:37:47 -0700 References: <7BE130C4-F6A6-428D-82D5-A7D7EE399E87@mac.com> <1207264734.21048.156.camel@edge.scott.net.au> X-Mailer: Apple Mail (2.919.2) X-Barracuda-Connect: smtpoutm.mac.com[17.148.16.64] X-Barracuda-Start-Time: 1207285213 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46783 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1583 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: jamespeach@mac.com Precedence: bulk X-list: pcp On 03/04/2008, at 4:18 PM, Nathan Scott wrote: > On Thu, 2008-04-03 at 13:26 -0700, James Peach wrote: >> >> commit f3d97e64f2774ea4414c0556e17d7fdd83e30df6 >> Author: James Peach >> Date: Thu Apr 3 13:22:11 2008 -0700 >> >> Use the prefix-relative etc directory instead of harcoding /etc. > > Can you give some more details as to why? I just want the --prefix option to work correctly. When I run ./ configure --prefix=/foo, I expect *all* of the package's files to go under /foo. IMHO that's kinda the point. > I don't think this > is going to work as you'd expect - in particular pmGetConfig(3) > will not like this, as it (kinda has to) hard-code the pcp.conf > location. Maybe I could teach pmGetConfig how to find pcp.conf? From jamespeach@mac.com Thu Apr 3 23:28:13 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 03 Apr 2008 23:28:22 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m346SC3p031725 for ; Thu, 3 Apr 2008 23:28:13 -0700 X-ASG-Debug-ID: 1207290528-1b48031a0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from smtpoutm.mac.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 793E2918F2F; Thu, 3 Apr 2008 23:28:48 -0700 (PDT) Received: from smtpoutm.mac.com (smtpoutm.mac.com [17.148.16.82]) by cuda.sgi.com with ESMTP id 5kztXuk1rbehYXtM; Thu, 03 Apr 2008 23:28:48 -0700 (PDT) Received: from mac.com (asmtp007-s [10.150.69.70]) by smtpoutm.mac.com (Xserve/smtpout019/MantshX 4.0) with ESMTP id m344eqxx008742; Thu, 3 Apr 2008 21:40:52 -0700 (PDT) Received: from [192.168.1.70] (adsl-75-18-195-199.dsl.pltn13.sbcglobal.net [75.18.195.199]) (authenticated bits=0) by mac.com (Xserve/asmtp007/MantshX 4.0) with ESMTP id m344em1l015744 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Thu, 3 Apr 2008 21:40:50 -0700 (PDT) Cc: markgw@sgi.com, Max Matveev , pcp@oss.sgi.com Message-Id: <429B980A-B498-4D23-B25A-A7A9D1A8C181@mac.com> From: James Peach To: nscott@aconex.com In-Reply-To: <1207281949.21048.166.camel@edge.scott.net.au> Content-Type: text/plain; charset=US-ASCII; format=flowed Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v919.2) X-ASG-Orig-Subj: Re: [PATCH] respect configured prefix instead of hardcoded /etc Subject: Re: [PATCH] respect configured prefix instead of hardcoded /etc Date: Thu, 3 Apr 2008 21:40:48 -0700 References: <7BE130C4-F6A6-428D-82D5-A7D7EE399E87@mac.com> <18421.26341.981959.130522@kuku.melbourne.sgi.com> <03ADE2A4-AAD6-4C8B-850D-1352D5D4C7DC@mac.com> <47F5A6F5.1070604@sgi.com> <1207281949.21048.166.camel@edge.scott.net.au> X-Mailer: Apple Mail (2.919.2) X-Barracuda-Connect: smtpoutm.mac.com[17.148.16.82] X-Barracuda-Start-Time: 1207290529 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.46788 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1584 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: jamespeach@mac.com Precedence: bulk X-list: pcp On 03/04/2008, at 9:05 PM, Nathan Scott wrote: > > On Fri, 2008-04-04 at 14:56 +1100, Mark Goodwin wrote: >> >> We could move all this stuff into an /etc/pcp directory if that >> would be more palatable. But it won't be easy. > > This is largely done already - the Debian packages put > things in /etc/{pmcd,pmlogger,pmie}/*. > > But James, why don't you just type "./Makepkgs" and use > the resulting .dmg file instead of worrying about files > being "splatted" on your root? (that gives you a neat > little package you can install/uninstall all at once). I had no idea that I could do that :) I'll give it a shot ... From jamespeach@mac.com Sun Apr 6 14:51:32 2008 Received: with ECARTIS (v1.0.0; list pcp); Sun, 06 Apr 2008 14:51:40 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m36LpVQs003544 for ; Sun, 6 Apr 2008 14:51:32 -0700 X-ASG-Debug-ID: 1207518727-14e303b80000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from smtpoutm.mac.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 67074751E06; Sun, 6 Apr 2008 14:52:07 -0700 (PDT) Received: from smtpoutm.mac.com (smtpoutm.mac.com [17.148.16.64]) by cuda.sgi.com with ESMTP id HjqIVf8hnJT5gQDD; Sun, 06 Apr 2008 14:52:07 -0700 (PDT) Received: from mac.com (asmtp006-s [10.150.69.69]) by smtpoutm.mac.com (Xserve/smtpout001/MantshX 4.0) with ESMTP id m36Lq635025939; Sun, 6 Apr 2008 14:52:06 -0700 (PDT) Received: from [17.151.105.109] ([17.151.105.109]) (authenticated bits=0) by mac.com (Xserve/asmtp006/MantshX 4.0) with ESMTP id m36Lq53i021220 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Sun, 6 Apr 2008 14:52:05 -0700 (PDT) Cc: pcp@oss.sgi.com Message-Id: <6DF3753B-C520-47BF-A7EB-90178C2BFD6B@mac.com> From: James Peach To: Max Matveev In-Reply-To: <18421.26341.981959.130522@kuku.melbourne.sgi.com> Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v919.2) X-ASG-Orig-Subj: Re: [PATCH] respect configured prefix instead of hardcoded /etc Subject: Re: [PATCH] respect configured prefix instead of hardcoded /etc Date: Sun, 6 Apr 2008 14:52:03 -0700 References: <7BE130C4-F6A6-428D-82D5-A7D7EE399E87@mac.com> <18421.26341.981959.130522@kuku.melbourne.sgi.com> X-Mailer: Apple Mail (2.919.2) X-Barracuda-Connect: smtpoutm.mac.com[17.148.16.64] X-Barracuda-Start-Time: 1207518729 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.47040 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1585 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: jamespeach@mac.com Precedence: bulk X-list: pcp On 03/04/2008, at 4:23 PM, Max Matveev wrote: > On Thu, 3 Apr 2008 13:26:32 -0700, James Peach wrote: > > JP> commit f3d97e64f2774ea4414c0556e17d7fdd83e30df6 > JP> Author: James Peach > JP> Date: Thu Apr 3 13:22:11 2008 -0700 > > JP> Use the prefix-relative etc directory instead of harcoding / > etc. > > JP> diff --git a/GNUmakefile b/GNUmakefile > JP> index a945efd..077f105 100644 > JP> --- a/GNUmakefile > JP> +++ b/GNUmakefile > JP> @@ -104,6 +104,7 @@ endif > JP> $(INSTALL) -m 755 -d $(PCP_SHARE_DIR)/lib > JP> $(INSTALL) -m 755 -d $(PCP_SHARE_DIR)/examples > JP> $(INSTALL) -m 755 -d $(PCP_INC_DIR) > JP> + $(INSTALL) -m 755 -d $(PCP_ETC_DIR) > That would make pcp the owner of /etc on Linux: pcp does not have its > own /etc/pcp directory and splats its files around the shared > directory. > > Also, /etc/pcp.conf is hardcoded deliberately to avoid chicken'n'egg > problem finding the source of information about relocation which can > be relocatable itself. So why does the hard-coded location have to be /etc/pcp.conf? Why can't it be a build-time setting? It seems that as long at it's well-known and hard-coded it shouldn't matter much what it is .... From nscott@aconex.com Mon Apr 7 19:11:44 2008 Received: with ECARTIS (v1.0.0; list pcp); Mon, 07 Apr 2008 19:11:56 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m382Bh78011861 for ; Mon, 7 Apr 2008 19:11:44 -0700 X-ASG-Debug-ID: 1207620739-6a8802b80000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 1C3BA75CE86 for ; Mon, 7 Apr 2008 19:12:19 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id Dy2R7DNkqhSKOa5O for ; Mon, 07 Apr 2008 19:12:19 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 966A650EFDD for ; Tue, 8 Apr 2008 11:54:35 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id 5W3Q4EYbX8SqpSGl for ; Tue, 08 Apr 2008 11:54:35 +1000 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id 6D65292C1F8 for ; Tue, 8 Apr 2008 11:54:35 +1000 (EST) X-ASG-Orig-Subj: kmchart updates Subject: kmchart updates From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Content-Type: text/plain Organization: Aconex Date: Tue, 08 Apr 2008 11:54:34 +1000 Message-Id: <1207619674.21048.187.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207620741 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.47151 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1586 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/kmchart.git doc/COPYING | 27 +++++++++++++++- man/man1/kmchart.1 | 28 ++++++++++++++++ src/chart/chartdialog.ui | 1 src/chart/exportdialog.cpp | 75 +++++++++++++++++++++++++++++++++++++++------ src/chart/exportdialog.h | 5 +++ src/chart/exportdialog.ui | 3 - src/chart/kmchart.cpp | 24 +++++++++++--- src/chart/kmchart.h | 9 ++++- src/chart/main.cpp | 26 +++++++++++++-- src/chart/main.h | 3 + src/libqwt/GNUmakefile | 2 - 11 files changed, 175 insertions(+), 28 deletions(-) commit 65751d60e21b87b8dbdf018d85dc113f4dfc37b8 Author: Nathan Scott Date: Tue Apr 8 11:50:51 2008 +1000 Add options to kmchart to allow image generation from the command line. commit 17c2395f12e7a457b1440a8b6f6a053842fbf947 Author: Nathan Scott Date: Tue Apr 8 09:44:40 2008 +1000 Minor formatting tweak to docs. commit 187bc0a996d6ebf7926621302c0260e5e3579a45 Author: Nathan Scott Date: Fri Apr 4 10:34:10 2008 +1100 Fix LDIRT macro in libqwt makefile. commit 2ed8bf3f2200e78f84d807ac0a9c04e195db3f75 Author: Nathan Scott Date: Fri Apr 4 09:33:55 2008 +1100 Remove references to no-longer-used qwt_counter.h header. From nscott@aconex.com Thu Apr 10 17:01:47 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 10 Apr 2008 17:02:08 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3B01hkW011788 for ; Thu, 10 Apr 2008 17:01:47 -0700 X-ASG-Debug-ID: 1207872141-314803bf0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 5F69177B317 for ; Thu, 10 Apr 2008 17:02:21 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id erMG5bHFvnh1ElBK for ; Thu, 10 Apr 2008 17:02:21 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id A5267530EF9; Fri, 11 Apr 2008 09:51:12 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id 4tv2OeXijF9ncMaB; Fri, 11 Apr 2008 09:51:12 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id 9913292C0E0; Fri, 11 Apr 2008 09:51:12 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Fri, 11 Apr 2008 09:51:12 +1000 (EST) Message-ID: <54214.192.168.3.1.1207871472.squirrel@mail.aconex.com> In-Reply-To: <86E3161F-2C93-4C0C-B0E8-3F839D350B47@mac.com> References: <1207178420.21048.123.camel@edge.scott.net.au> <23CD9178-1413-4F82-B3AB-789A34541701@mac.com> <1207197106.21048.136.camel@edge.scott.net.au> <86E3161F-2C93-4C0C-B0E8-3F839D350B47@mac.com> Date: Fri, 11 Apr 2008 09:51:12 +1000 (EST) X-ASG-Orig-Subj: Re: missing xcode projects when building kmchart Subject: Re: missing xcode projects when building kmchart From: nscott@aconex.com To: "James Peach" Cc: pcp@oss.sgi.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207872142 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.47 X-Barracuda-Spam-Status: No, SCORE=-1.47 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.47430 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name X-archive-position: 1587 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Hi James, I haven't done a build yet (and outta time again for now) - but I did build current pcp and kmchart on Mac OS X 10.4 and have verified that these do install and run on Mac OS X 10.5 now - try the current dmg files in http://oss.sgi.com/~nathans/ Theres also a copy of the qt dmg file that I used, which seems to work (not as recent a version as yours James). cheers. > On 02/04/2008, at 9:31 PM, Nathan Scott wrote: >> On Wed, 2008-04-02 at 21:17 -0700, James Peach wrote: >>> >>> platform_defs.h contains this little gem: >>> >>> #if defined(HAVE_MALLOC_H) >>> #include >>> #else >>> /* #undef ptrdiff_t */ >>> #define ptrdiff_t long >>> #endif >>> >> >> Oh, you're not using my git tree - I fixed that one months >> back. Your platform_defs.h should have a ptrdiff section >> along the lines... >> >> /* Check if ptrdiff_t type is available */ >> #define HAVE_PTRDIFF_T 1 >> #ifndef HAVE_PTRDIFF_T >> #define ptrdiff_t long >> #endif >> >> Your earlier snippet rings bells in my distant memory as to >> what the original pcp header looked like. > > Yep, this is fixed in yr git tree. I guess I had a onld version of PCP > header installed into /usr/include :( > >> >> >> >> WRT, the qmake/xcode issue, I may be able to get hold of a >> 10.5 system tonight - I'll see if I can reproduce and will >> look into it a bit more. >> >> cheers. >> >> -- >> Nathan >> > > From kimbrr@sgi.com Thu Apr 10 23:06:56 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 10 Apr 2008 23:07:20 -0700 (PDT) Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with SMTP id m3B66qk8024033 for ; Thu, 10 Apr 2008 23:06:55 -0700 Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.54.149]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id QAA26307; Fri, 11 Apr 2008 16:07:31 +1000 Received: from snort.melbourne.sgi.com (localhost [127.0.0.1]) by snort.melbourne.sgi.com (SGI-8.12.5/8.12.5) with ESMTP id m3B67UsT126722901; Fri, 11 Apr 2008 16:07:30 +1000 (AEST) Received: from localhost (kimbrr@localhost) by snort.melbourne.sgi.com (SGI-8.12.5/8.12.5/Submit) with ESMTP id m3B67Tuj129063712; Fri, 11 Apr 2008 16:07:30 +1000 (AEST) X-Authentication-Warning: snort.melbourne.sgi.com: kimbrr owned process doing -bs Date: Fri, 11 Apr 2008 16:07:29 +1000 From: Michael Newton X-X-Sender: kimbrr@snort.melbourne.sgi.com To: pcp@oss.sgi.com cc: Tim Serong Subject: Don't delete PMCD logs Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-archive-position: 1588 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: kimbrr@sgi.com Precedence: bulk X-list: pcp Every time pmcd restarts, it starts a new pmcd.log, the old one becomes pmcd.log.prev, and the old pmcd.log.prev is lost. Many PMDAs do likewise This sometimes makes it incredibly difficult to diagnose problems, because if you happen to need to restart PCP once or twice while triaging a bug, you lose all your logs. Worse, if you've obtained access to a machine well after a problem has occurred, the logs might have all been rotated away before you even got there. We propose doing away with the ".prev" logfiles altogether, and just continually appending to the main log files. Some "base OS" functionality can then be used to rotate the log files as necessary, eg: on SLES, ship a logrotate config file for PCP, which by default keeps log files around for a long time (weeks/months) and let the logrotate package take care of log rotation. What thoughts? Dr.Michael("Kimba")Newton kimbrr@sgi.com From nscott@aconex.com Thu Apr 10 23:23:34 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 10 Apr 2008 23:23:54 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3B6NXCo025906 for ; Thu, 10 Apr 2008 23:23:34 -0700 X-ASG-Debug-ID: 1207895050-040001050000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 0AFDA77D11F for ; Thu, 10 Apr 2008 23:24:10 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id QfjBb6OBePBYtLmO for ; Thu, 10 Apr 2008 23:24:10 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id B966D53475E; Fri, 11 Apr 2008 16:24:06 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id fL9mALVEtaPDNqZZ; Fri, 11 Apr 2008 16:24:06 +1000 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id A3BCD92C2AC; Fri, 11 Apr 2008 16:24:06 +1000 (EST) X-ASG-Orig-Subj: Re: Don't delete PMCD logs Subject: Re: Don't delete PMCD logs From: Nathan Scott Reply-To: nscott@aconex.com To: Michael Newton Cc: pcp@oss.sgi.com, Tim Serong In-Reply-To: References: Content-Type: text/plain Organization: Aconex Date: Fri, 11 Apr 2008 16:24:06 +1000 Message-Id: <1207895046.21048.250.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207895052 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.47456 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1589 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp On Fri, 2008-04-11 at 16:07 +1000, Michael Newton wrote: > Every time pmcd restarts, it starts a new pmcd.log, the old one > becomes pmcd.log.prev, and the old pmcd.log.prev is lost. > Many PMDAs do likewise > > This sometimes makes it incredibly difficult to diagnose problems, because > if you happen to need to restart PCP once or twice while triaging a bug, > you lose all your logs. Worse, if you've obtained access to a machine > well after a problem has occurred, the logs might have all been rotated > away before you even got there. > > We propose doing away with the ".prev" logfiles altogether, and just > continually appending to the main log files. Some "base OS" functionality > can then be used to rotate the log files as necessary, eg: on SLES, ship a > logrotate config file for PCP, which by default keeps log files around for > a long time (weeks/months) and let the logrotate package take care of log > rotation. > > What thoughts? > Not all platforms have logrotate setup (like Cygwin and possibly Mac) - so as long as this is optional (maybe via a runtime tests on -f /etc/logrotate.d/pcp and also -x `which logrotate`) then this seems OK to me. cheers. -- Nathan From nscott@aconex.com Fri Apr 11 00:02:39 2008 Received: with ECARTIS (v1.0.0; list pcp); Fri, 11 Apr 2008 00:03:00 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3B72cjR032591 for ; Fri, 11 Apr 2008 00:02:39 -0700 X-ASG-Debug-ID: 1207897395-4b2f014d0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 0A24C99D831 for ; Fri, 11 Apr 2008 00:03:15 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id eENqZgEH9KFh7wDJ for ; Fri, 11 Apr 2008 00:03:15 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 6BE12534883; Fri, 11 Apr 2008 17:03:13 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id ioz3Ej2xweMHRgb3; Fri, 11 Apr 2008 17:03:13 +1000 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id 59CD492C0E0; Fri, 11 Apr 2008 17:03:13 +1000 (EST) X-ASG-Orig-Subj: PATH setting for PCP daemons and PMDAs Subject: PATH setting for PCP daemons and PMDAs From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Cc: jamespeach@mac.com Content-Type: text/plain Organization: Aconex Date: Fri, 11 Apr 2008 17:03:12 +1000 Message-Id: <1207897392.21048.274.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207897397 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.47457 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1590 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Hi all, The snippit of code below from pcp.env fills me with fear and loathing. I know James is not a big fan either. :) The explicit setting of LD_LIBRARY_PATH on Solaris gives me some security concerns (this is used by the PCP start and cron scripts, so is often executed as root) - it's better to do that stuff through the /etc/ld.so.conf.d/ equivalent in Solaris, not here. The Solaris-specific part then specifies /usr/local/bin, which is respecified later - a bit odd. Also, theres no way to extend the PATH seen by PMDAs. I hit a situation yesterday where it would've been really useful to have a configuration hook there. Finally, every time we run PCP scripts, we potentially have to go off and stat an ever-increasing number of directories, many being there for other platforms, which just isn't a clean design. So, I'm thinking: - the Solaris piece should be completely removed - the platform-specific part could be replaced by a pcp.conf variable $PCP_PLATFORM_PATHS, which sysadmins can add to if they have "special" PATH needs for custom PMDAs, like me! This is the current code, read it and weep: # This needs to be the union of directories in which PCP tools may expect # to find executable commands ... # PATH="" if [ "$PCP_PLATFORM" = solaris ] then # GNU tools and libs hide in /usr/local ... make sure we find these # early enough in the searching # PATH=/usr/local/bin: LD_LIBRARY_PATH=/usr/local/lib export LD_LIBRARY_PATH fi PATH=$PATH/usr/sbin:/sbin:/bin:/usr/bin:/etc:${PCP_BIN_DIR}: ${PCP_BINADM_DIR}:${PCP_SHARE_DIR}/bin:${PCP_SHARE_DIR}/lib \ # the more obscure ones are conditional # for dir in /usr/bsd /usr/etc /usr/bin/X11 /usr/local/bin \ /opt/sfw/bin /usr/ccs/bin /cygdrive/c/WINDOWS/system32 \ /usr/contrib/bin /opt/local/bin do [ -d $dir ] || continue echo "$PATH" | egrep ":$dir(:|\$)" >/dev/null && continue PATH=$PATH:$dir done export PATH New version: PATH=/usr/sbin:/sbin:/bin:/usr/bin:/etc:${PCP_BIN_DIR}: ${PCP_BINADM_DIR}:${PCP_SHARE_DIR}/bin:${PCP_SHARE_DIR}/lib: ${PCP_PLATFORM_PATHS} And PCP_PLATFORM_PATHS can be set to an explicit value inside the configure script (so theres a bit of extra code in there to be fair, but no runtime code). Finally, can someone with access to the original rlog verify and fill in the blanks in this platform<->path mapping: /usr/bsd - FreeBSD /usr/etc - IRIX /usr/bin/X11 - UNIX platforms /usr/local/bin - UNIX platforms /opt/sfw/bin - ? /usr/ccs/bin - ? /cygdrive/c/WINDOWS/system32 - Windows/Cygwin /usr/contrib/bin - ? /opt/local/bin - Mac OS X cheers. -- Nathan From jamespeach@mac.com Fri Apr 11 09:53:16 2008 Received: with ECARTIS (v1.0.0; list pcp); Fri, 11 Apr 2008 09:53:25 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3BGrCEL025685 for ; Fri, 11 Apr 2008 09:53:16 -0700 X-ASG-Debug-ID: 1207932825-0c8400ad0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from smtpoutm.mac.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 009811081631 for ; Fri, 11 Apr 2008 09:53:45 -0700 (PDT) Received: from smtpoutm.mac.com (smtpoutm.mac.com [17.148.16.70]) by cuda.sgi.com with ESMTP id YJ4JZtnfpzG0UHLU for ; Fri, 11 Apr 2008 09:53:45 -0700 (PDT) Received: from mac.com (asmtp005-s [10.150.69.68]) by smtpoutm.mac.com (Xserve/smtpout007/MantshX 4.0) with ESMTP id m3BGrh7u001835; Fri, 11 Apr 2008 09:53:43 -0700 (PDT) Received: from [17.151.107.134] ([17.151.107.134]) (authenticated bits=0) by mac.com (Xserve/asmtp005/MantshX 4.0) with ESMTP id m3BGrcfi001742 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Fri, 11 Apr 2008 09:53:40 -0700 (PDT) Cc: pcp@oss.sgi.com Message-Id: <9D97D5D7-E318-478E-A2D9-73DF3191ED18@mac.com> From: James Peach To: nscott@aconex.com In-Reply-To: <1207897392.21048.274.camel@edge.scott.net.au> Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v919.2) X-ASG-Orig-Subj: Re: PATH setting for PCP daemons and PMDAs Subject: Re: PATH setting for PCP daemons and PMDAs Date: Fri, 11 Apr 2008 09:53:35 -0700 References: <1207897392.21048.274.camel@edge.scott.net.au> X-Mailer: Apple Mail (2.919.2) X-Barracuda-Connect: smtpoutm.mac.com[17.148.16.70] X-Barracuda-Start-Time: 1207932831 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.47497 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1591 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: jamespeach@mac.com Precedence: bulk X-list: pcp On 11/04/2008, at 12:03 AM, Nathan Scott wrote: > Hi all, > > The snippit of code below from pcp.env fills me with fear > and loathing. I know James is not a big fan either. :) > > The explicit setting of LD_LIBRARY_PATH on Solaris gives > me some security concerns (this is used by the PCP start > and cron scripts, so is often executed as root) - it's > better to do that stuff through the /etc/ld.so.conf.d/ > equivalent in Solaris, not here. The Solaris-specific > part then specifies /usr/local/bin, which is respecified > later - a bit odd. > > Also, theres no way to extend the PATH seen by PMDAs. I > hit a situation yesterday where it would've been really > useful to have a configuration hook there. > > Finally, every time we run PCP scripts, we potentially > have to go off and stat an ever-increasing number of > directories, many being there for other platforms, which > just isn't a clean design. > > So, I'm thinking: > - the Solaris piece should be completely removed yes > > - the platform-specific part could be replaced by a > pcp.conf variable $PCP_PLATFORM_PATHS, which sysadmins > can add to if they have "special" PATH needs for custom > PMDAs, like me! Sounds OK. IMHO PCP should (by default) only look in the minimum set of paths ... /usr/bin and friends > > > > This is the current code, read it and weep: > > # This needs to be the union of directories in which PCP tools may > expect > # to find executable commands ... > # > PATH="" > if [ "$PCP_PLATFORM" = solaris ] > then > # GNU tools and libs hide in /usr/local ... make sure we find these > # early enough in the searching > # > PATH=/usr/local/bin: > LD_LIBRARY_PATH=/usr/local/lib > export LD_LIBRARY_PATH > fi > PATH=$PATH/usr/sbin:/sbin:/bin:/usr/bin:/etc:${PCP_BIN_DIR}: > ${PCP_BINADM_DIR}:${PCP_SHARE_DIR}/bin:${PCP_SHARE_DIR}/lib \ > # the more obscure ones are conditional > # > for dir in /usr/bsd /usr/etc /usr/bin/X11 /usr/local/bin \ > /opt/sfw/bin /usr/ccs/bin /cygdrive/c/WINDOWS/system32 \ > /usr/contrib/bin /opt/local/bin > do > [ -d $dir ] || continue > echo "$PATH" | egrep ":$dir(:|\$)" >/dev/null && continue > PATH=$PATH:$dir > done > export PATH > > > New version: > PATH=/usr/sbin:/sbin:/bin:/usr/bin:/etc:${PCP_BIN_DIR}: > ${PCP_BINADM_DIR}:${PCP_SHARE_DIR}/bin:${PCP_SHARE_DIR}/lib: > ${PCP_PLATFORM_PATHS} > > And PCP_PLATFORM_PATHS can be set to an explicit value inside > the configure script (so theres a bit of extra code in there > to be fair, but no runtime code). > > Finally, can someone with access to the original rlog verify > and fill in the blanks in this platform<->path mapping: > > /usr/bsd - FreeBSD > /usr/etc - IRIX > /usr/bin/X11 - UNIX platforms > /usr/local/bin - UNIX platforms > /opt/sfw/bin - ? Sun Freeware > > /usr/ccs/bin - ? solaris BSD compat subsystem? > > /cygdrive/c/WINDOWS/system32 - Windows/Cygwin > /usr/contrib/bin - ? > /opt/local/bin - Mac OS X MacPorts - There is the fink project, which sold we /sw/bin > > > cheers. > > -- > Nathan > From nscott@aconex.com Fri Apr 11 13:02:43 2008 Received: with ECARTIS (v1.0.0; list pcp); Fri, 11 Apr 2008 13:02:52 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3BK2gcI015763 for ; Fri, 11 Apr 2008 13:02:43 -0700 X-ASG-Debug-ID: 1207944198-630900d40000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 497E71291286 for ; Fri, 11 Apr 2008 13:03:19 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id ilvvabDUR0Nau7PM for ; Fri, 11 Apr 2008 13:03:19 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 9F376538DC8; Sat, 12 Apr 2008 06:03:17 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id 4DxHrxOgKC9gAziT; Sat, 12 Apr 2008 06:03:17 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id 8CAB992C2AD; Sat, 12 Apr 2008 06:03:17 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Sat, 12 Apr 2008 06:03:17 +1000 (EST) Message-ID: <59757.192.168.3.1.1207944197.squirrel@mail.aconex.com> In-Reply-To: <9D97D5D7-E318-478E-A2D9-73DF3191ED18@mac.com> References: <1207897392.21048.274.camel@edge.scott.net.au> <9D97D5D7-E318-478E-A2D9-73DF3191ED18@mac.com> Date: Sat, 12 Apr 2008 06:03:17 +1000 (EST) X-ASG-Orig-Subj: Re: PATH setting for PCP daemons and PMDAs Subject: Re: PATH setting for PCP daemons and PMDAs From: nscott@aconex.com To: "James Peach" Cc: pcp@oss.sgi.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207944201 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.47 X-Barracuda-Spam-Status: No, SCORE=-1.47 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.47509 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name X-archive-position: 1592 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp > On 11/04/2008, at 12:03 AM, Nathan Scott wrote: >> >> - the platform-specific part could be replaced by a >> pcp.conf variable $PCP_PLATFORM_PATHS, which sysadmins >> can add to if they have "special" PATH needs for custom >> PMDAs, like me! > > Sounds OK. IMHO PCP should (by default) only look in the minimum set > of paths ... /usr/bin and friends Its occurred to me since we last discussed this why it is as it is - the open source PCP build and runtime has a heavy dependency on many GNU tools (like gmake, esp). These are often not on the usual PATH, and are squirrelled away someplace special - except on Linux and Mac OS X and Cygwin, where they are the default tools. For those platforms, and now that we changed the configure AWK test, which looks in all manner of places, I think we can safely remove some of the PATH extensions, for those platforms. >> /usr/bsd - FreeBSD >> /usr/etc - IRIX >> /usr/bin/X11 - UNIX platforms >> /usr/local/bin - UNIX platforms >> /opt/sfw/bin - ? > > Sun Freeware > Taa. >> >> /usr/ccs/bin - ? > > > solaris BSD compat subsystem? > Yep - google found it, and also mail saying its been deprecated for awhile (in favour of /usr/bin, etc). >> >> /cygdrive/c/WINDOWS/system32 - Windows/Cygwin >> /usr/contrib/bin - ? >> /opt/local/bin - Mac OS X > > MacPorts - There is the fink project, which sold we /sw/bin > I think I'm finally convinced we can drop these two. :) The one remaining oddity I see is that "/etc" is part of the default PCP path ... anyone know why that might be? Was that a need from IRIX? cheers. -- Nathan From nscott@aconex.com Fri Apr 11 13:36:31 2008 Received: with ECARTIS (v1.0.0; list pcp); Fri, 11 Apr 2008 13:36:40 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3BKaUj6022679 for ; Fri, 11 Apr 2008 13:36:31 -0700 X-ASG-Debug-ID: 1207946227-622d01f50000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id E1067129130E for ; Fri, 11 Apr 2008 13:37:08 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id c8BXbEyjOcBULjJx for ; Fri, 11 Apr 2008 13:37:08 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 21106538EFE for ; Sat, 12 Apr 2008 06:37:07 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id oAWUebXZ7lAfY7tN for ; Sat, 12 Apr 2008 06:37:07 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id 13CB592C0E0 for ; Sat, 12 Apr 2008 06:37:07 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Sat, 12 Apr 2008 06:37:07 +1000 (EST) Message-ID: <60169.192.168.3.1.1207946227.squirrel@mail.aconex.com> Date: Sat, 12 Apr 2008 06:37:07 +1000 (EST) X-ASG-Orig-Subj: pcp updates Subject: pcp updates From: nscott@aconex.com To: pcp@oss.sgi.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1207946228 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -0.97 X-Barracuda-Spam-Status: No, SCORE=-0.97 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=BSF_RULE7568M, NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.47513 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name 0.50 BSF_RULE7568M Custom Rule 7568M X-archive-position: 1593 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/pcp.git configure.in | 12 +++++ src/include/pcp.conf.in | 1 src/include/pcp.env | 50 --------------------- src/libpcp/src/GNUmakefile | 2 src/pmcd/rc_pcp | 105 --------------------------------------------- src/xconfirm/GNUmakefile | 2 src/xconfirm/README | 6 -- src/xconfirm/xconfirm | 2 8 files changed, 18 insertions(+), 162 deletions(-) commit d389a511601e608c8911893eac0261c444621097 Author: Nathan Scott Date: Sat Apr 12 06:33:54 2008 +1000 Rework PATH handling in all PCP shell scripts (via pcp.env) to address some issues disussed on the mailing list... > The explicit setting of LD_LIBRARY_PATH on Solaris gives > me some security concerns (this is used by the PCP start > and cron scripts, so is often executed as root) - it's > better to do that stuff through the /etc/ld.so.conf.d/ > equivalent in Solaris, not here. The Solaris-specific > part then specifies /usr/local/bin, which is respecified > later - a bit odd. > > Also, theres no way to extend the PATH seen by PMDAs. I > hit a situation yesterday where it would've been really > useful to have a configuration hook there. > > Finally, every time we run PCP scripts, we potentially > have to go off and stat an ever-increasing number of > directories, many being there for other platforms, which > just isn't a clean design. commit c6fd458bde44f0e694fc4a44fb7438c29b871c9f Author: Nathan Scott Date: Sat Apr 12 06:20:51 2008 +1000 Remove reference to removed README file from makefile. commit 251b0ce15fc7c350145192e5e8ed70416847f8e2 Author: Nathan Scott Date: Sat Apr 12 06:16:05 2008 +1000 Make the xconfirm script (the default PCP_XCONFIRM_PROG) run kmconfirm, if it finds it. This is always a better choice if found, since its a modern interface and is also cross-platform. Remove odd README describing xconfirm as a generic tool SUBDIR. commit 1bffa41e085da9daa4c27a7c5afad619d7c44b84 Author: Nathan Scott Date: Fri Apr 11 16:11:54 2008 +1000 Remove several large IRIX-specific shell snippets from the pcp rc script. commit a0062586e468c5c19232451f7ce0c2509065366c Author: Nathan Scott Date: Fri Apr 11 16:04:01 2008 +1000 Remove funky TRIX shell snippet, as per yesterdays meeting re IRIX. commit 444a0688f30498466d478f0d6e9a9b7dc6f092e6 Author: Nathan Scott Date: Fri Apr 11 11:14:03 2008 +1000 Use DSOSUFFIX macro when creating dsotbl.h for local context mode on MacOSX. From nscott@aconex.com Sat Apr 12 22:41:48 2008 Received: with ECARTIS (v1.0.0; list pcp); Sat, 12 Apr 2008 22:42:03 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3D5fii2025945 for ; Sat, 12 Apr 2008 22:41:48 -0700 X-ASG-Debug-ID: 1208065341-724401c60000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 0F12D9BD741 for ; Sat, 12 Apr 2008 22:42:21 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id QvMrKDEMSvApVX66 for ; Sat, 12 Apr 2008 22:42:21 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id E494253F74D for ; Sun, 13 Apr 2008 15:23:31 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id oFiWzQy5JoCzUGe3 for ; Sun, 13 Apr 2008 15:23:31 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id D446D92C0DE for ; Sun, 13 Apr 2008 15:23:31 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Sun, 13 Apr 2008 15:23:31 +1000 (EST) Message-ID: <56968.192.168.3.1.1208064211.squirrel@mail.aconex.com> Date: Sun, 13 Apr 2008 15:23:31 +1000 (EST) X-ASG-Orig-Subj: kmchart updates Subject: kmchart updates From: nscott@aconex.com To: pcp@oss.sgi.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1208065344 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.47 X-Barracuda-Spam-Status: No, SCORE=-1.47 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.47644 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name X-archive-position: 1594 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/kmchart.git README | 4 ---- build/mac/GNUmakefile | 11 ++++++----- images/kmquery.icns |binary src/chart/GNUmakefile | 38 +++++++++++++++++++++++++++++++------- src/chart/kmchart.cpp | 4 +++- src/include/builddefs.in | 4 ++-- src/query/GNUmakefile | 44 +++++++++++++++++++++++++++++++++++++++----- src/query/kmconfirm.sh | 4 ---- src/query/kmconfirm.sh.IN | 2 ++ src/query/kmmessage.sh | 4 ---- src/query/kmmessage.sh.IN | 2 ++ src/query/kmquery.pro | 1 + src/query/kmquery.sh.IN | 2 ++ src/time/GNUmakefile | 38 +++++++++++++++++++++++++++++++++----- src/time/kmtime.sh.IN | 2 ++ 15 files changed, 123 insertions(+), 37 deletions(-) commit eb84ca9b83ea381852801c829ee39eaa32165512 Merge: 408ae96... 69a6f9f... Author: Nathan Scott Date: Sun Apr 13 15:19:00 2008 +1000 Merge branch 'master' of git://oss.sgi.com:8090/nathans/kmchart commit 69a6f9f23ae30016894542d68c0378e9d96fb40b Author: Nathan Scott Date: Sun Apr 13 15:13:54 2008 +1000 Install kmchart and helper binaries as "real" MacOSX Application binaries. commit 408ae96137e5154a41cef34b28621156067e4538 Author: Nathan Scott Date: Thu Apr 10 11:18:57 2008 +1000 Fix position of the date label for images generated from the command line. From nscott@aconex.com Sun Apr 13 22:38:35 2008 Received: with ECARTIS (v1.0.0; list pcp); Sun, 13 Apr 2008 22:38:49 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3E5cXjU022368 for ; Sun, 13 Apr 2008 22:38:35 -0700 X-ASG-Debug-ID: 1208151551-52aa014e0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 8B672560E1 for ; Sun, 13 Apr 2008 22:39:12 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id 1ljSkoApfxWwnXQB for ; Sun, 13 Apr 2008 22:39:12 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 4CB6F547AAF for ; Mon, 14 Apr 2008 15:39:10 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id LbIsMrYfILUNtUeb for ; Mon, 14 Apr 2008 15:39:10 +1000 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id 3F40392C0E3 for ; Mon, 14 Apr 2008 15:39:10 +1000 (EST) X-ASG-Orig-Subj: kmchart updates Subject: kmchart updates From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Content-Type: text/plain Organization: Aconex Date: Mon, 14 Apr 2008 15:39:09 +1000 Message-Id: <1208151549.21048.285.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1208151552 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.47738 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1595 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/kmchart.git debian/control | 1 man/man1/kmsnap.1 | 286 +++++++++++++++++++++++++++++++ src/GNUmakefile | 2 src/chart/chart.cpp | 7 src/chart/chart.h | 1 src/chart/view.cpp | 2 src/snap/GNUmakefile | 27 ++ src/snap/Snap | 30 +++ src/snap/control | 71 +++++++ src/snap/crontab.IN | 6 src/snap/kmsnap.sh | 457 ++++++++++++++++++++++++++++++++++++++++++++++++++ src/snap/summary.html | 34 +++ 12 files changed, 923 insertions(+), 1 deletion(-) commit f686712302eac5f2f6794442d44977b2425c6256 Author: Nathan Scott Date: Mon Apr 14 15:36:34 2008 +1000 Deleted plots should not appear in saved kmchart views. commit f08780d6f9575d97d006721aa1281a30fcb147e2 Author: Nathan Scott Date: Mon Apr 14 15:33:27 2008 +1000 Add kmsnap support, for automated generation of kmchart images. From mort@sgi.com Mon Apr 14 11:25:43 2008 Received: with ECARTIS (v1.0.0; list pcp); Mon, 14 Apr 2008 11:25:57 -0700 (PDT) Received: from relay.sgi.com (relay1.corp.sgi.com [192.26.58.214]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3EIPgG0028142 for ; Mon, 14 Apr 2008 11:25:42 -0700 Received: from attica.americas.sgi.com (attica.americas.sgi.com [128.162.236.44]) by relay1.corp.sgi.com (Postfix) with ESMTP id 4F5998F8065 for ; Mon, 14 Apr 2008 11:26:19 -0700 (PDT) Received: by attica.americas.sgi.com (Postfix, from userid 44682) id 145D11EC678; Mon, 14 Apr 2008 13:26:19 -0500 (CDT) Date: Mon, 14 Apr 2008 13:26:18 -0500 From: Martin Hicks To: pcp@sgi.com Subject: disabling syslog in pmieconf Message-ID: <20080414182618.GA30732@alcatraz.americas.sgi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.9i X-archive-position: 1596 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: mort@sgi.com Precedence: bulk X-list: pcp Hi, Just trying to figure out how to create pmieconf rules. I've figured most of it out, but I'd like to disable the syslog action for my new rule by default. Here's a really simple example: #pmieconf-rules 1 # --- DO NOT MODIFY THIS FILE --- see pmieconf(4) rule tempo.simple default = "$string$" predicate = "some_inst ( kernel.all.load > 0.1 )" enabled = no version = 1 help = "A dummy pmieconf file"; string rule default = "A simple pmieconf rule. No real purpose" display = no modify = no; string delta default = "4 sec"; shell simple_action default = "echo $action_expand$ > /tmp/pmieout" display = no modify = no; syslog syslog_action enabled = no default = ""; When I enable this rule I'd *like* it to only have my $simple_action$ and disable the syslog_action. What's the right way to do that? The above produces this pmie: // 1 tempo.simple delta = 4 sec; tempo.simple = some_inst ( kernel.all.load > 0.1 ) -> shell 10 min "echo %v@%h > /tmp/pmieout" & syslog 10 min "A simple pmieconf rule. No real purpose" " %v@%h"; thanks mh From nscott@aconex.com Tue Apr 15 00:46:22 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 15 Apr 2008 00:46:36 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3F7kMnv025852 for ; Tue, 15 Apr 2008 00:46:22 -0700 X-ASG-Debug-ID: 1208245618-063701030000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 0AB4B9E6F13 for ; Tue, 15 Apr 2008 00:46:59 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id XQviW7IM4yfO0RwM for ; Tue, 15 Apr 2008 00:46:59 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 8C520555990 for ; Tue, 15 Apr 2008 17:46:49 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id ubEk8uzXCLBFO7gN for ; Tue, 15 Apr 2008 17:46:49 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id C16AE92C2D4 for ; Tue, 15 Apr 2008 17:46:49 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Tue, 15 Apr 2008 17:46:49 +1000 (EST) Message-ID: <43211.192.168.3.1.1208245609.squirrel@mail.aconex.com> Date: Tue, 15 Apr 2008 17:46:49 +1000 (EST) X-ASG-Orig-Subj: pcp updates Subject: pcp updates From: nscott@aconex.com To: pcp@oss.sgi.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1208245621 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -0.97 X-Barracuda-Spam-Status: No, SCORE=-0.97 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=BSF_RULE7568M, NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.47836 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name 0.50 BSF_RULE7568M Custom Rule 7568M X-archive-position: 1597 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/pcp.git VERSION.pcp | 2 - configure.in | 54 ++++++++++++++++++++++++----------------- debian/changelog | 9 ++++++ src/include/pcp.conf.in | 6 ++++ src/include/pcp.env | 11 ++++---- src/pmcd/rc_pcp | 2 - src/pmdas/news/news.agent | 2 - src/pmie/pmie_check.sh | 2 - src/pmlogctl/pmlogger_check.sh | 2 - src/pmlogctl/pmnewlog.sh | 2 - src/pmproxy/rc_pmproxy | 2 - 11 files changed, 60 insertions(+), 34 deletions(-) commit 00b014f419cdc60b44e42c9abc2b44037b10f417 Author: Nathan Scott Date: Tue Apr 15 17:41:53 2008 +1000 Bump build version number. commit 7c9c955f9848c35ce0457904e1af4cfa8b0b2d02 Author: Nathan Scott Date: Tue Apr 15 17:41:21 2008 +1000 Add configuration magic for knsnap control file - needs to be here as this is where the platform-specific control file locations are decided, and I really don't want to create /etc/kmchart.conf. And since precendent is set for layered products adding config info to pcp.conf by PCP_MPI_DIRS, I'm sure noone will complain. ;-) commit 59133b4c7418337430f22bb7629422de2974907f Author: Nathan Scott Date: Tue Apr 15 17:32:52 2008 +1000 Add AIX path setting, mvoe platform-PATH setting up with other platform macros. commit e113e6f689faceaf460d05eda7efd5f6a0c60909 Author: Nathan Scott Date: Tue Apr 15 17:30:30 2008 +1000 Its poor form to override user-specified CFLAGS, even for Interix. commit cfdbfdff5c4151fb699bac590ddb1a05b2c911fd Author: Nathan Scott Date: Tue Apr 15 17:12:54 2008 +1000 Remove remaining vistages of config.done, which is no longer used anywhere. commit 61290eeffd753cdef958b03c49cbf2a291c0c802 Author: Nathan Scott Date: Tue Apr 15 17:12:22 2008 +1000 Fix _get_pids_by_name for Cygwin. Here we have a ps binary that does not report a TIME column (only STIME, which is the process start time, which only for awhile produces output in the time string format HH:MM:SS, then switches to a "Mar 14" style format). Happily, it turns out the procps.exe binary produces output in the regular form we're after - this patch introduces $PCP_PS_PROG so that we can use this other ps. commit 28b7ebee690240ab0bbb2a87f3b4e4639284ff82 Author: Nathan Scott Date: Tue Apr 15 16:50:45 2008 +1000 Fix _get_pids_by_name for Mac OS X. Here we have a ps binary that can return a time string of the form HH:MM.SS instead of the expected HH:MM:SS used everywhere else. From mort@sgi.com Tue Apr 15 04:58:38 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 15 Apr 2008 04:58:49 -0700 (PDT) Received: from relay.sgi.com (relay1.corp.sgi.com [192.26.58.214]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3FBwbq5028625 for ; Tue, 15 Apr 2008 04:58:38 -0700 Received: from attica.americas.sgi.com (attica.americas.sgi.com [128.162.236.44]) by relay1.corp.sgi.com (Postfix) with ESMTP id 7719F8F808B; Tue, 15 Apr 2008 04:59:15 -0700 (PDT) Received: by attica.americas.sgi.com (Postfix, from userid 44682) id F313F1EC678; Tue, 15 Apr 2008 06:59:14 -0500 (CDT) Date: Tue, 15 Apr 2008 06:59:14 -0500 From: Martin Hicks To: Ken McDonell Cc: pcp@sgi.com Subject: Re: disabling syslog in pmieconf Message-ID: <20080415115914.GG30732@alcatraz.americas.sgi.com> References: <20080414182618.GA30732@alcatraz.americas.sgi.com> <1208254893.26640.0.camel@bozo.localdomain> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1208254893.26640.0.camel@bozo.localdomain> User-Agent: Mutt/1.5.9i X-archive-position: 1598 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: mort@sgi.com Precedence: bulk X-list: pcp On Tue, Apr 15, 2008 at 08:21:33PM +1000, Ken McDonell wrote: > pmieconf is one of the pcp pieces that never made it to open source ... > if sgi would like to address that issue, I'd be happy to respond ... > 8^)> Well that further highlights my newness to PCP and explains why the documentation seems pretty crufty and incomplete :) mh > > > On Mon, 2008-04-14 at 13:26 -0500, Martin Hicks wrote: > > Hi, > > > > Just trying to figure out how to create pmieconf rules. I've figured > > most of it out, but I'd like to disable the syslog action for my new > > rule by default. Here's a really simple example: > > > > #pmieconf-rules 1 > > # --- DO NOT MODIFY THIS FILE --- see pmieconf(4) > > > > rule tempo.simple > > default = "$string$" > > predicate = "some_inst ( kernel.all.load > 0.1 )" > > enabled = no > > version = 1 > > help = "A dummy pmieconf file"; > > > > string rule > > default = "A simple pmieconf rule. No real purpose" > > display = no > > modify = no; > > > > string delta > > default = "4 sec"; > > > > shell simple_action > > default = "echo $action_expand$ > /tmp/pmieout" > > display = no > > modify = no; > > > > syslog syslog_action > > enabled = no > > default = ""; > > > > > > When I enable this rule I'd *like* it to only have my $simple_action$ > > and disable the syslog_action. What's the right way to do that? The > > above produces this pmie: > > > > // 1 tempo.simple > > delta = 4 sec; > > tempo.simple = > > some_inst ( kernel.all.load > 0.1 ) -> shell 10 min "echo %v@%h > /tmp/pmieout" > > & syslog 10 min "A simple pmieconf rule. No real purpose" " %v@%h"; > > > > thanks > > mh > > > > From makc@melbourne.sgi.com Tue Apr 15 11:46:56 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 15 Apr 2008 11:47:07 -0700 (PDT) Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with SMTP id m3FIkqOH006336 for ; Tue, 15 Apr 2008 11:46:54 -0700 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 EAA23075; Wed, 16 Apr 2008 04:47:27 +1000 Received: from kuku.melbourne.sgi.com (localhost [127.0.0.1]) by kuku.melbourne.sgi.com (SGI-8.12.11.20060308/8.12.11) with ESMTP id m3FIlPHQ053995; Wed, 16 Apr 2008 04:47:25 +1000 (EST) Received: (from makc@localhost) by kuku.melbourne.sgi.com (SGI-8.12.11.20060308/8.12.11/Submit) id m3FIlNmV053981; Wed, 16 Apr 2008 04:47:23 +1000 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <18436.63546.849993.626865@kuku.melbourne.sgi.com> Date: Wed, 16 Apr 2008 04:47:22 +1000 From: Max Matveev To: nscott@aconex.com Cc: pcp@oss.sgi.com Subject: Re: PATH setting for PCP daemons and PMDAs In-Reply-To: <1207897392.21048.274.camel@edge.scott.net.au> References: <1207897392.21048.274.camel@edge.scott.net.au> <9D97D5D7-E318-478E-A2D9-73DF3191ED18@mac.com> <59757.192.168.3.1.1207944197.squirrel@mail.aconex.com> X-Mailer: VM 7.07 under 21.4 (patch 15) "Security Through Obscurity" XEmacs Lucid X-archive-position: 1599 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: makc@sgi.com Precedence: bulk X-list: pcp On Fri, 11 Apr 2008 17:03:12 +1000, Nathan Scott wrote: nscott> /usr/bsd - FreeBSD Also on Irix for bsd versions of some utilities. nscott> /usr/ccs/bin - ? Sun compilers use to be there. On Sat, 12 Apr 2008 06:03:17 +1000 (EST), nscott wrote: nscott> The one remaining oddity I see is that "/etc" is part of the nscott> default PCP path ... anyone know why that might be? Was that nscott> a need from IRIX? Not just Irix - lots of System V derviations have binaries in /etc. Also IIRC pcp.env was an executable once. max From nscott@aconex.com Thu Apr 17 18:23:15 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 17 Apr 2008 18:23:28 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3I1NDSm013194 for ; Thu, 17 Apr 2008 18:23:15 -0700 X-ASG-Debug-ID: 1208481832-145901a20000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 06752897303 for ; Thu, 17 Apr 2008 18:23:53 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id dcjwTAJ2k2ORfj4y for ; Thu, 17 Apr 2008 18:23:53 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id DDB4B571D94 for ; Fri, 18 Apr 2008 11:23:47 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id rOav67rlPMHR4dZT for ; Fri, 18 Apr 2008 11:23:47 +1000 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id BA10792C0E3 for ; Fri, 18 Apr 2008 11:23:47 +1000 (EST) X-ASG-Orig-Subj: pcp updates Subject: pcp updates From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Content-Type: text/plain Organization: Aconex Date: Fri, 18 Apr 2008 11:23:47 +1000 Message-Id: <1208481827.23166.4.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.22.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1208481834 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.52 X-Barracuda-Spam-Status: No, SCORE=-1.52 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=BSF_RULE7568M X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.48081 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.50 BSF_RULE7568M Custom Rule 7568M X-archive-position: 1600 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/pcp.git VERSION.pcp | 2 +- debian/changelog | 6 ++++++ debian/control | 2 ++ src/libpcp_pmda/src/cache.c | 2 +- 4 files changed, 10 insertions(+), 2 deletions(-) commit 43d1416029a3e32b44262073b44b63b6c27b97af Author: Nathan Scott Date: Fri Apr 18 11:20:43 2008 +1000 Add Debian pkg conflict on pgpool2, which also has an /etc/pcp.conf. commit e6f8ac177cf4832f0f7977d2aca6282346118197 Author: Nathan Scott Date: Fri Apr 18 11:19:34 2008 +1000 Remove spurious use of a static local in libpcp_pmda cache.c code. From nscott@aconex.com Thu Apr 17 20:48:05 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 17 Apr 2008 20:48:14 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3I3m4Yc024166 for ; Thu, 17 Apr 2008 20:48:05 -0700 X-ASG-Debug-ID: 1208490522-6de702f70000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 75120A3BB44 for ; Thu, 17 Apr 2008 20:48:42 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id JY3tphTC0TC6ht9g for ; Thu, 17 Apr 2008 20:48:42 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 5DA0D572FBB for ; Fri, 18 Apr 2008 13:48:40 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id Z5q3QUykuz5zMH3p for ; Fri, 18 Apr 2008 13:48:40 +1000 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id 3F9CB92C0E3 for ; Fri, 18 Apr 2008 13:48:40 +1000 (EST) X-ASG-Orig-Subj: Warnings galore Subject: Warnings galore From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Content-Type: text/plain; charset=UTF-8 Organization: Aconex Date: Fri, 18 Apr 2008 13:48:39 +1000 Message-Id: <1208490519.23166.11.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.22.1 Content-Transfer-Encoding: 8bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1208490524 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.48091 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1601 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Just upgraded gcc to ... 13:47 nathans@edge /source/git/pcp 7> gcc --version gcc (GCC) 4.2.3 (Debian 4.2.3-3) and, wheee, tons of shiny new warnings to work through, if anyone's feeling brave... connect.c: In function ‘__pmConnectPMCD’: connect.c:453: warning: ‘sts’ is used uninitialized in this function connect.c:325: warning: ‘fd’ may be used uninitialized in this function p_instance.c: In function ‘__pmDecodeInstance’: p_instance.c:216: warning: ‘rp’ may be used uninitialized in this function p_result.c: In function ‘__pmDecodeResult’: p_result.c:224: warning: ‘i’ may be used uninitialized in this function p_pmns.c: In function ‘__pmDecodeNameList’: p_pmns.c:304: warning: ‘status’ may be used uninitialized in this function pmns.c: In function ‘load’: pmns.c:2159: warning: ‘np’ may be used uninitialized in this function pmns.c:2159: note: ‘np’ was declared here pmns.c: In function ‘pmGetChildrenStatus’: pmns.c:3084: warning: ‘newname’ may be used uninitialized in this function pmns.c: In function ‘pmTraversePMNS’: pmns.c:3565: warning: ‘namelist’ may be used uninitialized in this function units.c: In function ‘pmUnitsStr’: units.c:122: warning: ‘spacestr’ may be used uninitialized in this function units.c:123: warning: ‘timestr’ may be used uninitialized in this function units.c:124: warning: ‘countstr’ may be used uninitialized in this function util.c: In function ‘pmflush’: util.c:899: warning: ‘eptr’ may be used uninitialized in this function logutil.c: In function ‘__pmGetArchiveEnd’: logutil.c:1947: warning: ‘physend’ may be used uninitialized in this function logutil.c:1937: warning: ‘save’ may be used uninitialized in this function interp.c: In function ‘__pmLogFetchInterp’: interp.c:499: warning: ‘pcp’ may be used uninitialized in this function rtime.c: In function ‘pmParseTimeWindow’: rtime.c:614: warning: ‘delta’ may be used uninitialized in this function spec.c: In function ‘pmParseMetricSpec’: spec.c:93: warning: ‘a_end’ may be used uninitialized in this function spec.c:97: warning: ‘i_end’ may be used uninitialized in this function optfetch.c: In function ‘__pmOptFetchAdd’: optfetch.c:369: warning: ‘pmp’ may be used uninitialized in this function callback.c: In function ‘pmdaInstance’: callback.c:220: warning: ‘idp’ may be used uninitialized in this function pmnsmerge.c: In function ‘addpmns’: pmnsmerge.c:122: warning: ‘lastp’ may be used uninitialized in this function pmnsdel.c: In function ‘delpmns’: pmnsdel.c:52: warning: ‘lastp’ may be used uninitialized in this function pminfo.c: In function ‘report’: pminfo.c:144: warning: ‘vsp’ may be used uninitialized in this function pmnscomp.c: In function ‘main’: pmnscomp.c:235: warning: ‘startsum’ may be used uninitialized in this function config.c: In function ‘CreateAgent’: config.c:1532: warning: ‘argv’ may be used uninitialized in this function config.c: In function ‘ContactAgents’: config.c:1956: warning: ‘sts’ may be used uninitialized in this function dopdus.c: In function ‘DoInstance’: dopdus.c:303: warning: ‘fdfail’ may be used uninitialized in this function dopdus.c: In function ‘DoDesc’: dopdus.c:224: warning: ‘fdfail’ may be used uninitialized in this function pmie.c: In function ‘load’: pmie.c:206: warning: ‘sts’ may be used uninitialized in this function symbol.c: In function ‘symIntern’: symbol.c:134: warning: ‘scoop’ may be used uninitialized in this function dstruct.c: In function ‘instExpr’: dstruct.c:713: warning: ‘u.scaleSpace’ may be used uninitialized in this function dstruct.c:713: warning: ‘u.scaleTime’ may be used uninitialized in this function dstruct.c:713: warning: ‘u.scaleCount’ may be used uninitialized in this function dstruct.c:713: warning: ‘u.pad’ may be used uninitialized in this function show.c: In function ‘lookupHostInst’: show.c:155: warning: ‘dbg_dump’ may be used uninitialized in this function andor.c: In function ‘cndAnd_n_n’: andor.c:266: warning: ‘ip2’ may be used uninitialized in this function pmprobe.c: In function ‘main’: pmprobe.c:99: warning: ‘host’ may be used uninitialized in this function pmprobe.c:89: warning: ‘fetch_sts’ may be used uninitialized in this function pmval.c: In function ‘main’: pmval.c:564: warning: ‘tmp_mode’ is used uninitialized in this function pmval.c:1515: note: ‘tmp_mode’ was declared here pmval.c:1545: warning: ‘idx2’ may be used uninitialized in this function pmlogger.c: In function ‘do_dialog’: pmlogger.c:325: warning: ‘msgf’ may be used uninitialized in this function callback.c: In function ‘log_callback’: callback.c:336: warning: ‘pdu_last_pmid’ may be used uninitialized in this function callback.c:335: warning: ‘pdu_first_pmid’ may be used uninitialized in this function :1438: warning: ‘yyunput’ defined but not used gram.y: In function ‘dometric’: gram.y:397: warning: ‘sts’ may be used uninitialized in this function pmlogger.o: In function `do_dialog': /source/build/pcp-2.7.4/src/pmlogger/pmlogger.c:412: warning: the use of `tmpnam' is dangerous, better use `mkstemp' pmdumplog.c: In function ‘main’: pmdumplog.c:363: warning: ‘rawfile’ may be used uninitialized in this function pmdumplog.c:225: warning: ‘log_size’ may be used uninitialized in this function pmdumplog.c:225: note: ‘log_size’ was declared here pmdumplog.c:224: warning: ‘meta_size’ may be used uninitialized in this function pmdumplog.c:224: note: ‘meta_size’ was declared here metriclist.c: In function ‘searchmlist’: metriclist.c:198: warning: ‘vlistp’ may be used uninitialized in this function :1163: warning: ‘yyunput’ defined but not used pmstore.c: In function ‘main’: pmstore.c:290: warning: ‘host’ may be used uninitialized in this function pmstore.c:301: warning: ‘aggr_len’ may be used uninitialized in this function pmlogsummary.c: In function ‘main’: pmlogsummary.c:661: warning: ‘rate’ may be used uninitialized in this function pmlogsummary.c:661: note: ‘rate’ was declared here pmclient.c: In function ‘main’: pmclient.c:261: warning: ‘host’ may be used uninitialized in this function pmkstat.c: In function ‘main’: pmkstat.c:99: warning: ‘host’ may be used uninitialized in this function pmkstat.c:112: warning: ‘lastdkwrite’ may be used uninitialized in this function pmkstat.c:111: warning: ‘lastdkread’ may be used uninitialized in this function pmkstat.c:107: warning: ‘prev’ may be used uninitialized in this function pmlc.c: In function ‘main’: pmlc.c:80: warning: ‘sts’ may be used uninitialized in this function actions.c: In function ‘Status’: actions.c:601: warning: ‘size’ may be used uninitialized in this function actions.c:600: warning: ‘vol’ may be used uninitialized in this function actions.c:599: warning: ‘state’ may be used uninitialized in this function actions.c:598: warning: ‘hostname’ may be used uninitialized in this function actions.c:597: warning: ‘timenow’ may be used uninitialized in this function actions.c:596: warning: ‘last’ may be used uninitialized in this function actions.c:595: warning: ‘start’ may be used uninitialized in this function actions.c:594: warning: ‘tzlogger’ may be used uninitialized in this function actions.c: In function ‘ShowLoggers’: actions.c:214: warning: ‘pport’ may be used uninitialized in this function :1487: warning: ‘yyunput’ defined but not used dso.c: In function ‘dodso’: dso.c:215: warning: ‘desc_list’ may be used uninitialized in this function dso.c:213: warning: ‘sts’ may be used uninitialized in this function pmda.c: In function ‘fillResult’: pmda.c:504: warning: ‘nbyte’ may be used uninitialized in this function pmda.c: In function ‘dopmda’: pmda.c:268: warning: ‘desc_list’ may be used uninitialized in this function :1632: warning: ‘yyunput’ defined but not used pmstat.c: In function ‘cntDiff’: pmstat.c:103: warning: ‘diff’ may be used uninitialized in this function pmcd.c: In function ‘pmcd_instance’: pmcd.c:649: warning: ‘nports’ may be used uninitialized in this function pmcd.c:651: warning: ‘pmiecount’ may be used uninitialized in this function pmcd.c:648: warning: ‘getname’ may be used uninitialized in this function pmcd.c: In function ‘pmcd_fetch’: pmcd.c:915: warning: ‘dp’ may be used uninitialized in this function weblog.c: In function ‘web_fetch’: weblog.c:2530: warning: ‘m_type’ may be used uninitialized in this function weblog.c:2529: warning: ‘m_offset’ may be used uninitialized in this function txrecord.c: In function ‘main’: txrecord.c:44: warning: ‘sp’ may be used uninitialized in this function telnet.c: In function ‘grab_cisco’: telnet.c:338: warning: ‘skip’ may be used uninitialized in this function probe.c: In function ‘main’: probe.c:118: warning: ‘name’ may be used uninitialized in this function probe.c:118: note: ‘name’ was declared here probe.c:117: warning: ‘ctype’ may be used uninitialized in this function probe.c:117: note: ‘ctype’ was declared here probe.c:113: warning: ‘defer’ may be used uninitialized in this function probe.c:113: note: ‘defer’ was declared here jstat.c: In function ‘jstat_indom_check’: jstat.c:410: warning: ‘sts’ is used uninitialized in this function jstat.c: In function ‘jstat_refresh’: jstat.c:593: warning: ‘error’ is used uninitialized in this function linuxlnk.c: In function ‘SetBaudCOM’: linuxlnk.c:380: warning: ‘baud’ may be used uninitialized in this function telnet-probe.c: In function ‘main’: telnet-probe.c:35: warning: ‘port’ may be used uninitialized in this function -- Nathan From nscott@aconex.com Tue Apr 22 17:29:33 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 22 Apr 2008 17:29:44 -0700 (PDT) Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3N0TUMh031489 for ; Tue, 22 Apr 2008 17:29:33 -0700 X-ASG-Debug-ID: 1208910610-06a100680000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id A66F915B8E40 for ; Tue, 22 Apr 2008 17:30:11 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id uEAxcOrhcBkdA04m for ; Tue, 22 Apr 2008 17:30:11 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 456C75931AB for ; Wed, 23 Apr 2008 10:30:05 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id ic34PG9byGbB1Qh3 for ; Wed, 23 Apr 2008 10:30:05 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id 06E0A92C0E3 for ; Wed, 23 Apr 2008 10:30:09 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Wed, 23 Apr 2008 10:30:09 +1000 (EST) Message-ID: <50513.192.168.3.1.1208910609.squirrel@mail.aconex.com> Date: Wed, 23 Apr 2008 10:30:09 +1000 (EST) X-ASG-Orig-Subj: kmchart updates Subject: kmchart updates From: nscott@aconex.com To: pcp@oss.sgi.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1208910612 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -0.97 X-Barracuda-Spam-Status: No, SCORE=-0.97 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=BSF_RULE7568M, NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.48554 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name 0.50 BSF_RULE7568M Custom Rule 7568M X-archive-position: 1602 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/kmchart.git debian/control | 2 doc/CHANGES | 4 qa/009 | 2 qa/010 | 4 qa/010.out | 10 - qa/011 | 4 qa/common.sh | 37 +++++++ src/chart/chart.cpp | 6 - src/chart/kmchart.cpp | 48 ++++----- src/chart/kmchart.h | 12 +- src/chart/kmchart.pro | 7 - src/chart/kmchart.ui | 222 -------------------------------------------- src/chart/statusbar.cpp | 104 ++++++++++++++++++++ src/chart/statusbar.h | 61 ++++++++++++ src/chart/tabwidget.cpp | 6 + src/chart/tabwidget.h | 2 src/chart/timeaxis.cpp | 14 +- src/dumptext/kmdumptext.cpp | 8 - src/query/kmmessage.sh.IN | 2 19 files changed, 279 insertions(+), 276 deletions(-) commit 25fc5e1a5adff3e8a409d632fd34b0f0514f4a18 Merge: 867eedf... 6783ebf... Author: Nathan Scott Date: Wed Apr 23 10:26:08 2008 +1000 Merge branch 'master' of git://oss.sgi.com:8090/nathans/kmchart commit 867eedfb0587f2a13f71df6579bfb782b94ef363 Author: Nathan Scott Date: Wed Apr 23 10:24:07 2008 +1000 Rework the way the unified time axis is constructed. We're now using a subclass of QStatusBar, and the time axis and button are children of this class. This is a first step in a planned complete redesign of QLayout management in kmchart, for which we need to be starting from a more conventional QApplication layout (i.e. with a status bar and with one real "central" widget, the tab widget). The change has also had several beneficial side-effects, mainly from just looking at this code once more: - the time axis and date/label values now sit atop a toolbutton widget, which means we can take action when pressed() and also it usually inherits the nice shaded gradient look from the theme. - resolved the time button size issue, such that the icon is no longer scaled down by two pixels, resulting in a fuzzy font. - fixed the value-label-clearing timeout, which regressed with the kmsnap changes. commit 0e0bbba31d71092a1df7d0bdb18818763557520d Author: Nathan Scott Date: Tue Apr 22 09:31:37 2008 +1000 Fix a typo in the whatsthis text for the VCR state button. commit b1b3358e8522c11f9cd4796c02fc2d7475fe2e9c Author: Nathan Scott Date: Wed Apr 16 08:15:15 2008 +1000 Fix -g/-G mismatches in dumptext QA tests. commit 7c4fff619fd3764f322689656dd35ac8a74af345 Author: Nathan Scott Date: Wed Apr 16 08:09:19 2008 +1000 Fix issues found in dumptext QA test 010. commit 6783ebfb439ac829333a39a299320e160ee7706b Author: Nathan Scott Date: Mon Apr 14 16:01:56 2008 +1000 Fix typo in checked in script wrapper. commit 8558fd29c66de356220a67a055369efe15f66250 Author: Nathan Scott Date: Wed Mar 12 09:28:18 2008 +1100 Update Debian package "Priority" field to match reality. From nscott@aconex.com Tue Apr 22 18:01:47 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 22 Apr 2008 18:01:59 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3N11jsE000878 for ; Tue, 22 Apr 2008 18:01:47 -0700 X-ASG-Debug-ID: 1208912541-36a903a40000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 503B6D8968 for ; Tue, 22 Apr 2008 18:02:22 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id iug1PEFBZGxPLllL for ; Tue, 22 Apr 2008 18:02:22 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id BA5CC5937FC; Wed, 23 Apr 2008 11:01:59 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id eyIzEuiQBsuFgQM7; Wed, 23 Apr 2008 11:01:59 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id 7A1E992C2EE; Wed, 23 Apr 2008 11:02:03 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Wed, 23 Apr 2008 11:02:03 +1000 (EST) Message-ID: <52405.192.168.3.1.1208912523.squirrel@mail.aconex.com> Date: Wed, 23 Apr 2008 11:02:03 +1000 (EST) X-ASG-Orig-Subj: pcpweb updates Subject: pcpweb updates From: nscott@aconex.com To: kimbrr@sgi.com, markgw@sgi.com, jkwaoz@sgi.com Cc: pcp@oss.sgi.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1208912547 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0207 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.47 X-Barracuda-Spam-Status: No, SCORE=-1.47 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.48556 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name X-archive-position: 1603 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp [ SGI guys, could someone update the web pages please? You're many revisions behind here now ... thanks. Alternatively, I'd be happy to do the web page updates directly from the git tree, if you've not got time to update them? Its just some trivial cron scripting from my end.] Changes committed to git://oss.sgi.com:8090/nathans/pcpweb.git index.html | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 71885e1e91b216332595855698f1584a7b83aa4f Author: Nathan Scott Date: Wed Apr 23 10:53:25 2008 +1000 Update documented snapshot development version of pcp and kmchart. From nscott@aconex.com Wed Apr 23 03:54:36 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 23 Apr 2008 03:54:50 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3NAsYjG015239 for ; Wed, 23 Apr 2008 03:54:36 -0700 X-ASG-Debug-ID: 1208948112-7b3b035b0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id DECAE7874B0 for ; Wed, 23 Apr 2008 03:55:13 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id usEF8q17gUXOZdBP for ; Wed, 23 Apr 2008 03:55:13 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id DD48559971C for ; Wed, 23 Apr 2008 20:55:03 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id aQDsT0XwCLnT292L for ; Wed, 23 Apr 2008 20:55:03 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id 0AA8192C2EE for ; Wed, 23 Apr 2008 20:55:08 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Wed, 23 Apr 2008 20:55:08 +1000 (EST) Message-ID: <52887.192.168.3.1.1208948108.squirrel@mail.aconex.com> Date: Wed, 23 Apr 2008 20:55:08 +1000 (EST) X-ASG-Orig-Subj: kmchart updates Subject: kmchart updates From: nscott@aconex.com To: pcp@oss.sgi.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1208948115 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -0.97 X-Barracuda-Spam-Status: No, SCORE=-0.97 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=BSF_RULE7568M, NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.48596 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name 0.50 BSF_RULE7568M Custom Rule 7568M X-archive-position: 1604 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/kmchart.git VERSION | 4 debian/changelog | 6 doc/CHANGES | 27 man/man1/kmsnap.1 | 2 src/chart/chart.cpp | 35 src/chart/chart.h | 1 src/chart/exportdialog.cpp | 15 src/chart/kmchart.cpp | 14 src/chart/kmchart.pro | 2 src/chart/tab.cpp | 1 src/libqwt/qwt_legend_item.cpp | 11 src/libqwt/qwt_plot_curve.cpp | 29 src/libqwt/qwt_plot_curve.h | 3 src/libqwt3d/gl2ps.c | 5986 ++++++++++++++++++++++++++++++++++++++++ src/libqwt3d/gl2ps.h | 199 + src/libqwt3d/libqwt3d.pro | 8 src/libqwt3d/qwt3d_io_gl2ps.cpp | 2 src/snap/kmsnap.sh | 6 src/time/kmtimearch.cpp | 5 src/time/kmtimearch.ui | 18 src/time/kmtimelive.cpp | 5 src/time/kmtimelive.ui | 24 22 files changed, 6336 insertions(+), 67 deletions(-) commit 00bc76d4c4c88ce57ca100f3a14314957437382c Author: Nathan Scott Date: Wed Apr 23 20:42:38 2008 +1000 Bump version number to 1.1. commit 9f3bdf1813220cc4fbf4555d047c979759ad7fa5 Author: Nathan Scott Date: Wed Apr 23 20:27:20 2008 +1000 It turns out that Qt can be built without gif support, and often is on Linux distros due to the age old gif patent issue. Its a bad format to default to - so, we now default to png, which is always available on all platforms. commit 912327790321a79962572bf2bea0dcf470a95b00 Author: Nathan Scott Date: Wed Apr 23 19:41:48 2008 +1000 Fix all legend issues (both on-screen and in generated images) by completely removing the fixLegendPen() workaround and fixing the underlying issue properly by allowing a different width Pen to be used for rendering each legend item. Also set the legend font at an opportune time inside kmchart, so libqwt is sure to have seen it prior to the times it uses it. Added some cosmetic improvements as well - the plot legend box is now displayed with a thin outline, and the Win-push-button look for each legend is reversed (not depressed by default). commit eb4e472a64ce6aacd7d7c29d00aafd6326c44af6 Author: Nathan Scott Date: Wed Apr 23 17:24:28 2008 +1000 Fix shell issue in logical OR test in kmsnap. commit 1f77dd51c8dfacdf866c06de0480a0acbaff08ac Author: Nathan Scott Date: Wed Apr 23 12:56:36 2008 +1000 Resolve initial window sizing differences on Mac OS X. Because Qt automatically "removes" the toolbar from the application, and uses it as per "normal" Mac applications, across the top of the display, we need to account for that size change in the initial window size, but only on the Mac platform of course. commit 272bb439d0eca6a500c8dbaae8c9f186c86c0117 Author: Nathan Scott Date: Wed Apr 23 12:52:51 2008 +1000 Add a vertical spacer at the base of the kmtime windows, to help with layout issues on both Mac and Linux (first part of proper fix for Mac in this area). commit 96acb3db2b9939d32cbfb0df9e0aecc962783523 Author: Nathan Scott Date: Wed Apr 23 12:04:57 2008 +1000 Resolve linkage issue when trying qwt3d library. From kenj@internode.on.net Wed Apr 23 15:15:43 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 23 Apr 2008 15:15:56 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3NMFeph014060 for ; Wed, 23 Apr 2008 15:15:43 -0700 X-ASG-Debug-ID: 1208988980-267c002f0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from ipmail01.adl6.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 591F0DE53F for ; Wed, 23 Apr 2008 15:16:21 -0700 (PDT) Received: from ipmail01.adl6.internode.on.net (ipmail01.adl6.internode.on.net [203.16.214.146]) by cuda.sgi.com with ESMTP id stUXir5Bbfq86bFf for ; Wed, 23 Apr 2008 15:16:21 -0700 (PDT) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AkwBAIJSD0g7p65s/2dsb2JhbAAIriU X-IronPort-AV: E=Sophos;i="4.25,701,1199626200"; d="scan'208";a="88075291" Received: from ppp59-167-174-108.lns1.mel4.internode.on.net (HELO [192.168.1.64]) ([59.167.174.108]) by ipmail01.adl6.internode.on.net with ESMTP; 24 Apr 2008 07:46:18 +0930 X-ASG-Orig-Subj: Re: Warnings galore Subject: Re: Warnings galore From: Ken McDonell To: nscott@aconex.com Cc: pcp@oss.sgi.com In-Reply-To: <1208490519.23166.11.camel@edge.scott.net.au> References: <1208490519.23166.11.camel@edge.scott.net.au> Content-Type: text/plain; charset=UTF-8 Date: Wed, 23 Apr 2008 22:16:11 +0000 Message-Id: <1208988971.27113.16.camel@bozo.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 (2.12.3-3.fc8) Content-Transfer-Encoding: 8bit X-Barracuda-Connect: ipmail01.adl6.internode.on.net[203.16.214.146] X-Barracuda-Start-Time: 1208988982 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0026 1.0000 -2.0042 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.00 X-Barracuda-Spam-Status: No, SCORE=-2.00 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.48638 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1605 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: kenj@internode.on.net Precedence: bulk X-list: pcp I've looked at the first one. I can PROVE mathematically, that sts at line 453 of connect.c MUST be initialized in all cases. This appears to be a case of some "still young enough to know everything" gcc zealot who has cocked the whole thing up. My recommendation is to not upgrade gcc, and leave the PCP source alone. On Fri, 2008-04-18 at 13:48 +1000, Nathan Scott wrote: > Just upgraded gcc to ... > > 13:47 nathans@edge /source/git/pcp 7> gcc --version > gcc (GCC) 4.2.3 (Debian 4.2.3-3) > > and, wheee, tons of shiny new warnings to work through, if > anyone's feeling brave... > > connect.c: In function ‘__pmConnectPMCD’: > connect.c:453: warning: ‘sts’ is used uninitialized in this function > connect.c:325: warning: ‘fd’ may be used uninitialized in this function > p_instance.c: In function ‘__pmDecodeInstance’: > p_instance.c:216: warning: ‘rp’ may be used uninitialized in this > function > p_result.c: In function ‘__pmDecodeResult’: > p_result.c:224: warning: ‘i’ may be used uninitialized in this function > p_pmns.c: In function ‘__pmDecodeNameList’: > p_pmns.c:304: warning: ‘status’ may be used uninitialized in this > function > pmns.c: In function ‘load’: > pmns.c:2159: warning: ‘np’ may be used uninitialized in this function > pmns.c:2159: note: ‘np’ was declared here > pmns.c: In function ‘pmGetChildrenStatus’: > pmns.c:3084: warning: ‘newname’ may be used uninitialized in this > function > pmns.c: In function ‘pmTraversePMNS’: > pmns.c:3565: warning: ‘namelist’ may be used uninitialized in this > function > units.c: In function ‘pmUnitsStr’: > units.c:122: warning: ‘spacestr’ may be used uninitialized in this > function > units.c:123: warning: ‘timestr’ may be used uninitialized in this > function > units.c:124: warning: ‘countstr’ may be used uninitialized in this > function > util.c: In function ‘pmflush’: > util.c:899: warning: ‘eptr’ may be used uninitialized in this function > logutil.c: In function ‘__pmGetArchiveEnd’: > logutil.c:1947: warning: ‘physend’ may be used uninitialized in this > function > logutil.c:1937: warning: ‘save’ may be used uninitialized in this > function > interp.c: In function ‘__pmLogFetchInterp’: > interp.c:499: warning: ‘pcp’ may be used uninitialized in this function > rtime.c: In function ‘pmParseTimeWindow’: > rtime.c:614: warning: ‘delta’ may be used uninitialized in this function > spec.c: In function ‘pmParseMetricSpec’: > spec.c:93: warning: ‘a_end’ may be used uninitialized in this function > spec.c:97: warning: ‘i_end’ may be used uninitialized in this function > optfetch.c: In function ‘__pmOptFetchAdd’: > optfetch.c:369: warning: ‘pmp’ may be used uninitialized in this > function > callback.c: In function ‘pmdaInstance’: > callback.c:220: warning: ‘idp’ may be used uninitialized in this > function > pmnsmerge.c: In function ‘addpmns’: > pmnsmerge.c:122: warning: ‘lastp’ may be used uninitialized in this > function > pmnsdel.c: In function ‘delpmns’: > pmnsdel.c:52: warning: ‘lastp’ may be used uninitialized in this > function > pminfo.c: In function ‘report’: > pminfo.c:144: warning: ‘vsp’ may be used uninitialized in this function > pmnscomp.c: In function ‘main’: > pmnscomp.c:235: warning: ‘startsum’ may be used uninitialized in this > function > config.c: In function ‘CreateAgent’: > config.c:1532: warning: ‘argv’ may be used uninitialized in this > function > config.c: In function ‘ContactAgents’: > config.c:1956: warning: ‘sts’ may be used uninitialized in this function > dopdus.c: In function ‘DoInstance’: > dopdus.c:303: warning: ‘fdfail’ may be used uninitialized in this > function > dopdus.c: In function ‘DoDesc’: > dopdus.c:224: warning: ‘fdfail’ may be used uninitialized in this > function > pmie.c: In function ‘load’: > pmie.c:206: warning: ‘sts’ may be used uninitialized in this function > symbol.c: In function ‘symIntern’: > symbol.c:134: warning: ‘scoop’ may be used uninitialized in this > function > dstruct.c: In function ‘instExpr’: > dstruct.c:713: warning: ‘u.scaleSpace’ may be used uninitialized in this > function > dstruct.c:713: warning: ‘u.scaleTime’ may be used uninitialized in this > function > dstruct.c:713: warning: ‘u.scaleCount’ may be used uninitialized in this > function > dstruct.c:713: warning: ‘u.pad’ may be used uninitialized in this > function > show.c: In function ‘lookupHostInst’: > show.c:155: warning: ‘dbg_dump’ may be used uninitialized in this > function > andor.c: In function ‘cndAnd_n_n’: > andor.c:266: warning: ‘ip2’ may be used uninitialized in this function > pmprobe.c: In function ‘main’: > pmprobe.c:99: warning: ‘host’ may be used uninitialized in this function > pmprobe.c:89: warning: ‘fetch_sts’ may be used uninitialized in this > function > pmval.c: In function ‘main’: > pmval.c:564: warning: ‘tmp_mode’ is used uninitialized in this function > pmval.c:1515: note: ‘tmp_mode’ was declared here > pmval.c:1545: warning: ‘idx2’ may be used uninitialized in this function > pmlogger.c: In function ‘do_dialog’: > pmlogger.c:325: warning: ‘msgf’ may be used uninitialized in this > function > callback.c: In function ‘log_callback’: > callback.c:336: warning: ‘pdu_last_pmid’ may be used uninitialized in > this function > callback.c:335: warning: ‘pdu_first_pmid’ may be used uninitialized in > this function > :1438: warning: ‘yyunput’ defined but not used > gram.y: In function ‘dometric’: > gram.y:397: warning: ‘sts’ may be used uninitialized in this function > pmlogger.o: In function `do_dialog': > /source/build/pcp-2.7.4/src/pmlogger/pmlogger.c:412: warning: the use of > `tmpnam' is dangerous, better use `mkstemp' > pmdumplog.c: In function ‘main’: > pmdumplog.c:363: warning: ‘rawfile’ may be used uninitialized in this > function > pmdumplog.c:225: warning: ‘log_size’ may be used uninitialized in this > function > pmdumplog.c:225: note: ‘log_size’ was declared here > pmdumplog.c:224: warning: ‘meta_size’ may be used uninitialized in this > function > pmdumplog.c:224: note: ‘meta_size’ was declared here > metriclist.c: In function ‘searchmlist’: > metriclist.c:198: warning: ‘vlistp’ may be used uninitialized in this > function > :1163: warning: ‘yyunput’ defined but not used > pmstore.c: In function ‘main’: > pmstore.c:290: warning: ‘host’ may be used uninitialized in this > function > pmstore.c:301: warning: ‘aggr_len’ may be used uninitialized in this > function > pmlogsummary.c: In function ‘main’: > pmlogsummary.c:661: warning: ‘rate’ may be used uninitialized in this > function > pmlogsummary.c:661: note: ‘rate’ was declared here > pmclient.c: In function ‘main’: > pmclient.c:261: warning: ‘host’ may be used uninitialized in this > function > pmkstat.c: In function ‘main’: > pmkstat.c:99: warning: ‘host’ may be used uninitialized in this function > pmkstat.c:112: warning: ‘lastdkwrite’ may be used uninitialized in this > function > pmkstat.c:111: warning: ‘lastdkread’ may be used uninitialized in this > function > pmkstat.c:107: warning: ‘prev’ may be used uninitialized in this > function > pmlc.c: In function ‘main’: > pmlc.c:80: warning: ‘sts’ may be used uninitialized in this function > actions.c: In function ‘Status’: > actions.c:601: warning: ‘size’ may be used uninitialized in this > function > actions.c:600: warning: ‘vol’ may be used uninitialized in this function > actions.c:599: warning: ‘state’ may be used uninitialized in this > function > actions.c:598: warning: ‘hostname’ may be used uninitialized in this > function > actions.c:597: warning: ‘timenow’ may be used uninitialized in this > function > actions.c:596: warning: ‘last’ may be used uninitialized in this > function > actions.c:595: warning: ‘start’ may be used uninitialized in this > function > actions.c:594: warning: ‘tzlogger’ may be used uninitialized in this > function > actions.c: In function ‘ShowLoggers’: > actions.c:214: warning: ‘pport’ may be used uninitialized in this > function > :1487: warning: ‘yyunput’ defined but not used > dso.c: In function ‘dodso’: > dso.c:215: warning: ‘desc_list’ may be used uninitialized in this > function > dso.c:213: warning: ‘sts’ may be used uninitialized in this function > pmda.c: In function ‘fillResult’: > pmda.c:504: warning: ‘nbyte’ may be used uninitialized in this function > pmda.c: In function ‘dopmda’: > pmda.c:268: warning: ‘desc_list’ may be used uninitialized in this > function > :1632: warning: ‘yyunput’ defined but not used > pmstat.c: In function ‘cntDiff’: > pmstat.c:103: warning: ‘diff’ may be used uninitialized in this function > pmcd.c: In function ‘pmcd_instance’: > pmcd.c:649: warning: ‘nports’ may be used uninitialized in this function > pmcd.c:651: warning: ‘pmiecount’ may be used uninitialized in this > function > pmcd.c:648: warning: ‘getname’ may be used uninitialized in this > function > pmcd.c: In function ‘pmcd_fetch’: > pmcd.c:915: warning: ‘dp’ may be used uninitialized in this function > weblog.c: In function ‘web_fetch’: > weblog.c:2530: warning: ‘m_type’ may be used uninitialized in this > function > weblog.c:2529: warning: ‘m_offset’ may be used uninitialized in this > function > txrecord.c: In function ‘main’: > txrecord.c:44: warning: ‘sp’ may be used uninitialized in this function > telnet.c: In function ‘grab_cisco’: > telnet.c:338: warning: ‘skip’ may be used uninitialized in this function > probe.c: In function ‘main’: > probe.c:118: warning: ‘name’ may be used uninitialized in this function > probe.c:118: note: ‘name’ was declared here > probe.c:117: warning: ‘ctype’ may be used uninitialized in this function > probe.c:117: note: ‘ctype’ was declared here > probe.c:113: warning: ‘defer’ may be used uninitialized in this function > probe.c:113: note: ‘defer’ was declared here > jstat.c: In function ‘jstat_indom_check’: > jstat.c:410: warning: ‘sts’ is used uninitialized in this function > jstat.c: In function ‘jstat_refresh’: > jstat.c:593: warning: ‘error’ is used uninitialized in this function > linuxlnk.c: In function ‘SetBaudCOM’: > linuxlnk.c:380: warning: ‘baud’ may be used uninitialized in this > function > telnet-probe.c: In function ‘main’: > telnet-probe.c:35: warning: ‘port’ may be used uninitialized in this > function > > From kjw@pocketeer.rightsock.com Wed Apr 23 16:51:05 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 23 Apr 2008 16:51:18 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3NNp0OX025871 for ; Wed, 23 Apr 2008 16:51:05 -0700 X-ASG-Debug-ID: 1208994702-267102690000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from mordred.punk.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 39E80D30B1 for ; Wed, 23 Apr 2008 16:51:42 -0700 (PDT) Received: from mordred.punk.net (noodle.rightsock.com [69.36.241.36]) by cuda.sgi.com with ESMTP id dlXxA5VSv0xyTyVi for ; Wed, 23 Apr 2008 16:51:42 -0700 (PDT) Received: from pocketeer.rightsock.com (localhost [127.0.0.1]) by mordred.punk.net (8.12.10/8.12.10) with ESMTP id m3NNpPDj043544; Wed, 23 Apr 2008 16:51:25 -0700 (PDT) (envelope-from kjw@pocketeer.rightsock.com) Received: (from kjw@localhost) by pocketeer.rightsock.com (8.13.5.20060308/8.13.5/Submit) id m3NNrak4010039; Wed, 23 Apr 2008 16:53:36 -0700 Date: Wed, 23 Apr 2008 16:53:36 -0700 From: Kevin Wang To: Ken McDonell Cc: pcp@oss.sgi.com X-ASG-Orig-Subj: Re: Warnings galore Subject: Re: Warnings galore Message-ID: <20080423235335.GA9349@rightsock.com> References: <1208490519.23166.11.camel@edge.scott.net.au> <1208988971.27113.16.camel@bozo.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1208988971.27113.16.camel@bozo.localdomain> User-Agent: Mutt/1.5.11 X-Barracuda-Connect: noodle.rightsock.com[69.36.241.36] X-Barracuda-Start-Time: 1208994703 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.48648 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1606 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: kjw@pocketeer.rightsock.com Precedence: bulk X-list: pcp From Ken McDonell > This appears to be a case of some "still young enough to know > everything" gcc zealot who has cocked the whole thing up. My > recommendation is to not upgrade gcc, and leave the PCP source alone. \wave Hi Ken! Long time no see. (kjw@sgi.com some 8 yrs ago?) Glad to see you're still alive and kicking! - Kevin From nscott@aconex.com Wed Apr 23 17:16:33 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 23 Apr 2008 17:16:47 -0700 (PDT) Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3O0GVkf027736 for ; Wed, 23 Apr 2008 17:16:33 -0700 X-ASG-Debug-ID: 1208996229-3c4603d70000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 9FDFE15C64BB for ; Wed, 23 Apr 2008 17:17:10 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id 3988tCvkdwG5yHOo for ; Wed, 23 Apr 2008 17:17:10 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id A51AA59D500 for ; Thu, 24 Apr 2008 10:17:03 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id oYENYj41rQlFbmEs for ; Thu, 24 Apr 2008 10:17:03 +1000 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id 50E8192C2AB for ; Thu, 24 Apr 2008 10:17:08 +1000 (EST) X-ASG-Orig-Subj: kmchart updates Subject: kmchart updates From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Content-Type: text/plain Organization: Aconex Date: Thu, 24 Apr 2008 10:17:07 +1000 Message-Id: <1208996228.23166.80.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.22.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1208996233 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.48648 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1607 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/kmchart.git README | 40 +++++++++++++++++++++++++++++++++++++--- VERSION | 4 ++-- debian/changelog | 10 ++++++++-- doc/CHANGES | 7 +++++++ src/chart/GNUmakefile | 2 +- src/chart/kmchart.cpp | 5 ----- src/libqwt3d/GNUmakefile | 2 +- src/query/GNUmakefile | 2 +- 8 files changed, 57 insertions(+), 15 deletions(-) commit 16a18a803196ee99b5f904d4dd5de5a77624dbb7 Author: Nathan Scott Date: Thu Apr 24 10:10:13 2008 +1000 Bump point release number again. commit c896505420cd68fbf6d683818b8fdeb31574e3d8 Author: Nathan Scott Date: Thu Apr 24 10:08:00 2008 +1000 Fix two more makefiles that were not constructing the SOURCE macro correctly. commit e64146efb11cc8b8eb6405efbdc25e04f8946ba1 Author: Nathan Scott Date: Thu Apr 24 09:44:09 2008 +1000 Bump release number to 1.1.2 for couple more minor tweaks. commit f902239bd8fd6aef75a8e80ca8d56abff9f87c09 Author: Nathan Scott Date: Thu Apr 24 09:39:17 2008 +1000 Fix makefile issue with SOURCES macro not picking up all files. commit 621516699e881bcc4a11eb6ca349ef99aea9b4f8 Author: Nathan Scott Date: Thu Apr 24 08:29:43 2008 +1000 Update README, setting out the roadmap for kmchart-2.0. commit 9b3a353787b11f2c756c07cfa755a6c3a5daa175 Author: Nathan Scott Date: Wed Apr 23 22:40:20 2008 +1000 Revert change forcing kmchart main window size to be fixed on Mac, kmtime only. From nscott@aconex.com Thu Apr 24 21:34:03 2008 Received: with ECARTIS (v1.0.0; list pcp); Thu, 24 Apr 2008 21:34:20 -0700 (PDT) Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3P4Xx88028578 for ; Thu, 24 Apr 2008 21:34:02 -0700 X-ASG-Debug-ID: 1209098079-4cd203660000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 07B0015D65EF for ; Thu, 24 Apr 2008 21:34:40 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id XUdFB6Y6qhhPFrgR for ; Thu, 24 Apr 2008 21:34:40 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 082CB5A677F for ; Fri, 25 Apr 2008 14:34:31 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id 6Ca0EhvR6WetKDlm for ; Fri, 25 Apr 2008 14:34:31 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id AEBDB92C0E3; Fri, 25 Apr 2008 14:34:37 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Fri, 25 Apr 2008 14:34:37 +1000 (EST) Message-ID: <34675.192.168.3.1.1209098077.squirrel@mail.aconex.com> Date: Fri, 25 Apr 2008 14:34:37 +1000 (EST) X-ASG-Orig-Subj: pcp updates Subject: pcp updates From: nscott@aconex.com To: pcp@oss.sgi.com Cc: mward@aconex.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1209098081 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.47 X-Barracuda-Spam-Status: No, SCORE=-1.47 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.48762 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name X-archive-position: 1608 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/pcp.git src/pmcd/rc-proc.sh | 2 +- src/pmcd/src/pmcd.c | 3 ++- src/pmdas/jstat/jstat.c | 9 ++++----- 3 files changed, 7 insertions(+), 7 deletions(-) commit bc917fa42e08fb26b28ab8a9540fdd2a917a16b7 Author: Nathan Scott Date: Fri Apr 25 14:30:46 2008 +1000 Tighten the pmcd-is-already-running error message, when bind fails. I saw EPERM here from Cygwin recently, when pmcd was not running at all, which resulted in a misleading diagnostic - we now guard on an EADDRINUSE. The message was also changed to pmcd "may" be running instead of "is" running, as it may be something else bound to the requested port. commit e563365fbeed09554fa354d1044a6551de0d7d03 Author: Nathan Scott Date: Fri Apr 25 14:26:29 2008 +1000 Fix two invalid uses of status/error variables in the jstat PMDA. commit d76f9691a0ce999c29c16a19d9ac83bb0a770647 Author: Mikel Ward Date: Fri Apr 25 14:21:44 2008 +1000 /etc/init.d/pmie status fails if there's more than one PMIE running. The fix seems to be at line 372 of /usr/share/pcp/lib/rc-proc.sh. < [ $_n -eq 1 ] && return 0 > [ $_n -ge 1 ] && return 0 RC_CHECKPROC claims to confirm to the LSB definition of rc_checkproc, but LSB doesn't define an rc_checkproc or a checkproc. http://refspecs.linux-foundation.org/LSB_3.2.0/LSB-Core-generic/LSB-Core-generic/iniscrptfunc.html The closest thing there is is checkproc on SUSE, which says exit code 0 means "process is running" without making it explicit whether than means one or many. http://www.freebsd.org/cgi/man.cgi?query=checkproc&apropos=0&sektion=0&manpath=SuSE+Linux%2Fi386+8.2&format=html From nscott@aconex.com Sun Apr 27 18:35:31 2008 Received: with ECARTIS (v1.0.0; list pcp); Sun, 27 Apr 2008 18:35:39 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3S1ZTKO014937 for ; Sun, 27 Apr 2008 18:35:31 -0700 X-ASG-Debug-ID: 1209346570-2ca5001b0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 5840EF84C3 for ; Sun, 27 Apr 2008 18:36:10 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id xxfxeN1cq6FRyCvk for ; Sun, 27 Apr 2008 18:36:10 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 66D565B1AC5 for ; Mon, 28 Apr 2008 11:36:01 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id CSui1CFvaONf9ZLc for ; Mon, 28 Apr 2008 11:36:01 +1000 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id A262D92C302 for ; Mon, 28 Apr 2008 11:36:09 +1000 (EST) X-ASG-Orig-Subj: pcp updates Subject: pcp updates From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Content-Type: text/plain Organization: Aconex Date: Mon, 28 Apr 2008 11:36:09 +1000 Message-Id: <1209346569.23166.124.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.22.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1209346572 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.49038 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1609 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/kmchart.git VERSION | 2 +- debian/changelog | 6 ++++++ doc/CHANGES | 5 +++++ src/chart/chart.cpp | 7 +++++++ src/chart/chart.h | 1 + src/chart/chartdialog.cpp | 2 +- src/chart/exportdialog.cpp | 2 +- src/chart/kmchart.cpp | 12 ++++++++++++ src/chart/kmchart.h | 1 + src/chart/statusbar.cpp | 18 ++++++++++++++---- src/chart/statusbar.h | 4 ++++ 11 files changed, 53 insertions(+), 7 deletions(-) commit 9990828392a1007aa9313bb39e8fb220aa04dfba Author: Nathan Scott Date: Mon Apr 28 11:31:21 2008 +1000 Bump version number to 1.1.4. commit 35e11e7b64539a7fae346e7cb86c243273ab3f50 Author: Nathan Scott Date: Mon Apr 28 11:27:09 2008 +1000 Fix an issue where plot colors were getting lost on chart editing. commit 273b6d37c7731068d711cf4d6bad32bf7ae3fa29 Author: Nathan Scott Date: Mon Apr 28 10:14:44 2008 +1000 Fix format string comparison for output image files. commit e36d3ccb701f06d0a15abe1b5c8559f1af2616f4 Author: Nathan Scott Date: Mon Apr 28 08:37:36 2008 +1000 Fix a problem with the way the time axis is now implemented within a QStatusBar - this means a vertical toolbar doesn't extend all the way to the base of the window, but rather it ends above the status bar, which meant the unified time axis was extending beyond the width of each chart. From nscott@aconex.com Sun Apr 27 21:42:02 2008 Received: with ECARTIS (v1.0.0; list pcp); Sun, 27 Apr 2008 21:42:10 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3S4g0SG001163 for ; Sun, 27 Apr 2008 21:42:02 -0700 X-ASG-Debug-ID: 1209357762-786e02790000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id C9DC91213992 for ; Sun, 27 Apr 2008 21:42:42 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id YQP3m5cEOUyEenWS for ; Sun, 27 Apr 2008 21:42:42 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 8E34E5B39B9 for ; Mon, 28 Apr 2008 14:42:32 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id NY4JGNopUgAZepvl for ; Mon, 28 Apr 2008 14:42:32 +1000 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id E958292C304 for ; Mon, 28 Apr 2008 14:42:40 +1000 (EST) X-ASG-Orig-Subj: kmchart updates Subject: kmchart updates From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Content-Type: text/plain Organization: Aconex Date: Mon, 28 Apr 2008 14:42:40 +1000 Message-Id: <1209357760.23166.140.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.22.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1209357763 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.49051 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1610 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/kmchart.git VERSION | 2 +- debian/changelog | 4 ++-- doc/CHANGES | 3 +++ src/chart/kmchart.cpp | 9 +++------ src/chart/kmchart.h | 1 + src/chart/statusbar.cpp | 6 +++--- 6 files changed, 13 insertions(+), 12 deletions(-) commit 01af239a6fa33faf87f386990bd0af9e56d2f99e Author: Nathan Scott Date: Mon Apr 28 14:33:16 2008 +1000 Bump version number to 1.1.5. commit daacfee4028c7676b261565160b19e6763186f74 Author: Nathan Scott Date: Mon Apr 28 14:32:05 2008 +1000 Revise the earlier for time axis right-alignment when vertical toolbar is in use - it turns out that Qt4.3 behaves differently (and wrongly) to 4.2 in the way the height/width of the toolbar are reported - this reworks things in such a way that will work for all versions (dimensions are extracted after initial widget creation, where we know they are correct, stored and used later for on-the-fly layout changing). From mort@sgi.com Tue Apr 29 09:19:02 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 29 Apr 2008 09:19:14 -0700 (PDT) Received: from relay.sgi.com (relay1.corp.sgi.com [192.26.58.214]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3TGJ1PB022541 for ; Tue, 29 Apr 2008 09:19:01 -0700 Received: from attica.americas.sgi.com (attica.americas.sgi.com [128.162.236.44]) by relay1.corp.sgi.com (Postfix) with ESMTP id CE19A8F80AB for ; Tue, 29 Apr 2008 09:19:42 -0700 (PDT) Received: by attica.americas.sgi.com (Postfix, from userid 44682) id 8F6941EC678; Tue, 29 Apr 2008 11:19:42 -0500 (CDT) Date: Tue, 29 Apr 2008 11:19:42 -0500 From: Martin Hicks To: pcp Subject: Args run together in pmie actions Message-ID: <20080429161941.GF6076@alcatraz.americas.sgi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.9i X-archive-position: 1611 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: mort@sgi.com Precedence: bulk X-list: pcp Hi, I found this problem in the SGI version of the pcp rpm, and just duplicated it with the git tree that I checked out a few minutes ago. Seems like a bug to me... When multiple instances are true for a particular pmie rule, the action is done only once, with the strings all concatenated together. For example: mort@laplace:~/src$ pmie -V some_inst kernel.all.load > 0 -> syslog "there is a load %i"; expr_1: true The resulting syslog message is: Apr 29 12:11:59 laplace pcp-pmie[8568]: there is a load 1 minutethere is a load 5 minutethere is a load 15 minute What I think I *should* see is: Apr 29 12:11:59 laplace pcp-pmie[8568]: there is a load 1 minute Apr 29 12:11:59 laplace pcp-pmie[8568]: there is a load 5 minute Apr 29 12:11:59 laplace pcp-pmie[8568]: there is a load 15 minute This behavior is the same if the action is: -> shell "my_command %i"; The result being: my_command 1 minutemy_command 5 minutemy_command 15 minute being executed...and failing. A bug or a feature that I don't understand? thanks mh From mort@sgi.com Tue Apr 29 14:16:44 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 29 Apr 2008 14:16:58 -0700 (PDT) Received: from relay.sgi.com (relay1.corp.sgi.com [192.26.58.214]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3TLGiqQ027505 for ; Tue, 29 Apr 2008 14:16:44 -0700 Received: from attica.americas.sgi.com (attica.americas.sgi.com [128.162.236.44]) by relay1.corp.sgi.com (Postfix) with ESMTP id 894CB8F808E; Tue, 29 Apr 2008 14:17:25 -0700 (PDT) Received: by attica.americas.sgi.com (Postfix, from userid 44682) id B68941EC678; Tue, 29 Apr 2008 16:17:24 -0500 (CDT) Date: Tue, 29 Apr 2008 16:17:24 -0500 From: Martin Hicks To: Ken McDonell Cc: pcp Subject: Re: Args run together in pmie actions Message-ID: <20080429211724.GN6076@alcatraz.americas.sgi.com> References: <20080429161941.GF6076@alcatraz.americas.sgi.com> <1209502039.31660.9.camel@bozo.localdomain> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1209502039.31660.9.camel@bozo.localdomain> User-Agent: Mutt/1.5.9i X-archive-position: 1612 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: mort@sgi.com Precedence: bulk X-list: pcp On Tue, Apr 29, 2008 at 08:47:19PM +0000, Ken McDonell wrote: > This is not a bug, it is designed to work that way ... 8^)> > > The %i %v and %h substitution occurs in buidling the string and the > action is fired once. > > If you rewrite the rule thusly, all will be well ... > > ... -> syslog "there is a load" " [%i] %v"; Thanks guys. So "shell" actions just have to iterate through multiple args and figure out how many instances are true... That's no big deal, just not the way I thought it worked. Thanks mh PS. Some braindamaged SGI mailserver is rewriting my e-mails to the pcp list and removing the "oss" part of the domain. Hopefully they'll fix this soon, and hopefully we'll all stop sending Paul Pedersen e-mail. From nscott@aconex.com Tue Apr 29 14:50:27 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 29 Apr 2008 14:50:35 -0700 (PDT) Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3TLoPri030469 for ; Tue, 29 Apr 2008 14:50:26 -0700 X-ASG-Debug-ID: 1209505866-7efe00570000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 8E4DF160858F for ; Tue, 29 Apr 2008 14:51:06 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id faHxoL2zEiAhZxMP for ; Tue, 29 Apr 2008 14:51:06 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 3DDA85C29B2 for ; Wed, 30 Apr 2008 07:50:52 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id x1P5lBQRnzHb0VpS for ; Wed, 30 Apr 2008 07:50:52 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id 1E4C092C302 for ; Wed, 30 Apr 2008 07:51:02 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Wed, 30 Apr 2008 07:51:02 +1000 (EST) Message-ID: <36023.192.168.3.1.1209505862.squirrel@mail.aconex.com> Date: Wed, 30 Apr 2008 07:51:02 +1000 (EST) X-ASG-Orig-Subj: pcp updates Subject: pcp updates From: nscott@aconex.com To: pcp@oss.sgi.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1209505869 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.47 X-Barracuda-Spam-Status: No, SCORE=-1.47 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.49206 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name X-archive-position: 1613 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/pcp.git VERSION.pcp | 2 +- debian/changelog | 7 +++++++ src/pmdas/linux/help | 10 +++++++--- src/pmdas/linux/pmda.c | 8 ++++---- src/pmdas/linux/proc_net_dev.c | 4 ++++ 5 files changed, 23 insertions(+), 8 deletions(-) commit dcd62fce655fa40c8f341d03c97030fb74b07481 Author: Nathan Scott Date: Wed Apr 30 07:47:17 2008 +1000 Bump build version number to 20080430. commit e4ec4631e2d8072d3d815396cf993fdf2ee5f0a4 Author: Ken McDonell Date: Wed Apr 30 07:43:19 2008 +1000 Fix issues in the network interface speed/baudrate metrics. network.interface.speed is scaled from Megabits/second to Megabytes/second using division by 8 instead of 10, units of the two metrics were not right, and the conversion from the speed/baudrate exposed by the ethtool ioctl interface into these units was also incorrect. From mort@sgi.com Tue Apr 29 15:34:29 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 29 Apr 2008 15:34:43 -0700 (PDT) Received: from relay.sgi.com (netops-testserver-3.corp.sgi.com [192.26.57.72]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3TMYRtk002372 for ; Tue, 29 Apr 2008 15:34:29 -0700 Received: from attica.americas.sgi.com (attica.americas.sgi.com [128.162.236.44]) by netops-testserver-3.corp.sgi.com (Postfix) with ESMTP id 67D949089B; Tue, 29 Apr 2008 15:35:08 -0700 (PDT) Received: by attica.americas.sgi.com (Postfix, from userid 44682) id F398E1EC678; Tue, 29 Apr 2008 17:35:07 -0500 (CDT) Date: Tue, 29 Apr 2008 17:35:07 -0500 From: Martin Hicks To: nscott@aconex.com Cc: Ken McDonell , pcp Subject: Re: Args run together in pmie actions Message-ID: <20080429223507.GO6076@alcatraz.americas.sgi.com> References: <20080429161941.GF6076@alcatraz.americas.sgi.com> <1209502039.31660.9.camel@bozo.localdomain> <20080429211724.GN6076@alcatraz.americas.sgi.com> <35796.192.168.3.1.1209504742.squirrel@mail.aconex.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <35796.192.168.3.1.1209504742.squirrel@mail.aconex.com> User-Agent: Mutt/1.5.9i X-archive-position: 1614 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: mort@sgi.com Precedence: bulk X-list: pcp On Wed, Apr 30, 2008 at 07:32:22AM +1000, nscott@aconex.com wrote: > > > > On Tue, Apr 29, 2008 at 08:47:19PM +0000, Ken McDonell wrote: > >> This is not a bug, it is designed to work that way ... 8^)> > >> > >> The %i %v and %h substitution occurs in buidling the string and the > >> action is fired once. > >> > >> If you rewrite the rule thusly, all will be well ... > >> > >> ... -> syslog "there is a load" " [%i] %v"; > > > > Thanks guys. So "shell" actions just have to iterate through multiple > > args and figure out how many instances are true... > > Only the "true" instances are passed to the shell action. Right. Bad description on my part. mh From nscott@aconex.com Tue Apr 29 15:54:50 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 29 Apr 2008 15:54:58 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3TMsmdx003955 for ; Tue, 29 Apr 2008 15:54:50 -0700 X-ASG-Debug-ID: 1209509728-6c5b01cc0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 4195710DE8EB for ; Tue, 29 Apr 2008 15:55:28 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id ftW2lb3hzeVqR4Br for ; Tue, 29 Apr 2008 15:55:28 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id D70A75C2D4C for ; Wed, 30 Apr 2008 08:55:13 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id EAnILvRjHHIGiRpi for ; Wed, 30 Apr 2008 08:55:13 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id C4FA992C0E3 for ; Wed, 30 Apr 2008 08:55:23 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Wed, 30 Apr 2008 08:55:23 +1000 (EST) Message-ID: <37814.192.168.3.1.1209509723.squirrel@mail.aconex.com> Date: Wed, 30 Apr 2008 08:55:23 +1000 (EST) X-ASG-Orig-Subj: pcpweb updates Subject: pcpweb updates From: nscott@aconex.com To: pcp@oss.sgi.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1209509732 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.47 X-Barracuda-Spam-Status: No, SCORE=-1.47 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.49211 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name X-archive-position: 1615 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/pcpweb.git contribute.html | 19 ++++++++----------- index.html | 6 +++--- 2 files changed, 11 insertions(+), 14 deletions(-) commit 6148b6d44358258444ba6d238575ca934ff512a0 Author: Nathan Scott Date: Wed Apr 30 08:53:28 2008 +1000 Updates to the how-to-contribute page, crossing off completed items, revising others. commit 1a2f5c383db0514fcc6a1640c615d41301bb037a Author: Nathan Scott Date: Tue Apr 29 21:28:17 2008 +1000 Bump kmchart minor version number once more. From markgw@sgi.com Tue Apr 29 16:40:11 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 29 Apr 2008 16:40:24 -0700 (PDT) Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with SMTP id m3TNe6sx011884 for ; Tue, 29 Apr 2008 16:40:08 -0700 Received: from [134.14.55.13] (dhcp13.melbourne.sgi.com [134.14.55.13]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id JAA02568; Wed, 30 Apr 2008 09:40:43 +1000 Message-ID: <4817B1FB.6080607@sgi.com> Date: Wed, 30 Apr 2008 09:40:43 +1000 From: Mark Goodwin Reply-To: markgw@sgi.com Organization: SGI Engineering User-Agent: Thunderbird 1.5.0.14 (Windows/20071210) MIME-Version: 1.0 To: Ken McDonell CC: nscott@aconex.com, pcp@oss.sgi.com Subject: Re: Warnings galore References: <1208490519.23166.11.camel@edge.scott.net.au> <1208988971.27113.16.camel@bozo.localdomain> In-Reply-To: <1208988971.27113.16.camel@bozo.localdomain> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-archive-position: 1616 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: markgw@sgi.com Precedence: bulk X-list: pcp Ken McDonell wrote: > I've looked at the first one. Anyone looked at the rest of them? They can't all be bogus?? Cheers > > I can PROVE mathematically, that sts at line 453 of connect.c MUST be > initialized in all cases. > > This appears to be a case of some "still young enough to know > everything" gcc zealot who has cocked the whole thing up. My > recommendation is to not upgrade gcc, and leave the PCP source alone. > > On Fri, 2008-04-18 at 13:48 +1000, Nathan Scott wrote: >> Just upgraded gcc to ... >> >> 13:47 nathans@edge /source/git/pcp 7> gcc --version >> gcc (GCC) 4.2.3 (Debian 4.2.3-3) >> >> and, wheee, tons of shiny new warnings to work through, if >> anyone's feeling brave... >> >> connect.c: In function ‘__pmConnectPMCD’: >> connect.c:453: warning: ‘sts’ is used uninitialized in this function >> connect.c:325: warning: ‘fd’ may be used uninitialized in this function >> p_instance.c: In function ‘__pmDecodeInstance’: >> p_instance.c:216: warning: ‘rp’ may be used uninitialized in this >> function >> p_result.c: In function ‘__pmDecodeResult’: >> p_result.c:224: warning: ‘i’ may be used uninitialized in this function >> p_pmns.c: In function ‘__pmDecodeNameList’: >> p_pmns.c:304: warning: ‘status’ may be used uninitialized in this >> function >> pmns.c: In function ‘load’: >> pmns.c:2159: warning: ‘np’ may be used uninitialized in this function >> pmns.c:2159: note: ‘np’ was declared here >> pmns.c: In function ‘pmGetChildrenStatus’: >> pmns.c:3084: warning: ‘newname’ may be used uninitialized in this >> function >> pmns.c: In function ‘pmTraversePMNS’: >> pmns.c:3565: warning: ‘namelist’ may be used uninitialized in this >> function >> units.c: In function ‘pmUnitsStr’: >> units.c:122: warning: ‘spacestr’ may be used uninitialized in this >> function >> units.c:123: warning: ‘timestr’ may be used uninitialized in this >> function >> units.c:124: warning: ‘countstr’ may be used uninitialized in this >> function >> util.c: In function ‘pmflush’: >> util.c:899: warning: ‘eptr’ may be used uninitialized in this function >> logutil.c: In function ‘__pmGetArchiveEnd’: >> logutil.c:1947: warning: ‘physend’ may be used uninitialized in this >> function >> logutil.c:1937: warning: ‘save’ may be used uninitialized in this >> function >> interp.c: In function ‘__pmLogFetchInterp’: >> interp.c:499: warning: ‘pcp’ may be used uninitialized in this function >> rtime.c: In function ‘pmParseTimeWindow’: >> rtime.c:614: warning: ‘delta’ may be used uninitialized in this function >> spec.c: In function ‘pmParseMetricSpec’: >> spec.c:93: warning: ‘a_end’ may be used uninitialized in this function >> spec.c:97: warning: ‘i_end’ may be used uninitialized in this function >> optfetch.c: In function ‘__pmOptFetchAdd’: >> optfetch.c:369: warning: ‘pmp’ may be used uninitialized in this >> function >> callback.c: In function ‘pmdaInstance’: >> callback.c:220: warning: ‘idp’ may be used uninitialized in this >> function >> pmnsmerge.c: In function ‘addpmns’: >> pmnsmerge.c:122: warning: ‘lastp’ may be used uninitialized in this >> function >> pmnsdel.c: In function ‘delpmns’: >> pmnsdel.c:52: warning: ‘lastp’ may be used uninitialized in this >> function >> pminfo.c: In function ‘report’: >> pminfo.c:144: warning: ‘vsp’ may be used uninitialized in this function >> pmnscomp.c: In function ‘main’: >> pmnscomp.c:235: warning: ‘startsum’ may be used uninitialized in this >> function >> config.c: In function ‘CreateAgent’: >> config.c:1532: warning: ‘argv’ may be used uninitialized in this >> function >> config.c: In function ‘ContactAgents’: >> config.c:1956: warning: ‘sts’ may be used uninitialized in this function >> dopdus.c: In function ‘DoInstance’: >> dopdus.c:303: warning: ‘fdfail’ may be used uninitialized in this >> function >> dopdus.c: In function ‘DoDesc’: >> dopdus.c:224: warning: ‘fdfail’ may be used uninitialized in this >> function >> pmie.c: In function ‘load’: >> pmie.c:206: warning: ‘sts’ may be used uninitialized in this function >> symbol.c: In function ‘symIntern’: >> symbol.c:134: warning: ‘scoop’ may be used uninitialized in this >> function >> dstruct.c: In function ‘instExpr’: >> dstruct.c:713: warning: ‘u.scaleSpace’ may be used uninitialized in this >> function >> dstruct.c:713: warning: ‘u.scaleTime’ may be used uninitialized in this >> function >> dstruct.c:713: warning: ‘u.scaleCount’ may be used uninitialized in this >> function >> dstruct.c:713: warning: ‘u.pad’ may be used uninitialized in this >> function >> show.c: In function ‘lookupHostInst’: >> show.c:155: warning: ‘dbg_dump’ may be used uninitialized in this >> function >> andor.c: In function ‘cndAnd_n_n’: >> andor.c:266: warning: ‘ip2’ may be used uninitialized in this function >> pmprobe.c: In function ‘main’: >> pmprobe.c:99: warning: ‘host’ may be used uninitialized in this function >> pmprobe.c:89: warning: ‘fetch_sts’ may be used uninitialized in this >> function >> pmval.c: In function ‘main’: >> pmval.c:564: warning: ‘tmp_mode’ is used uninitialized in this function >> pmval.c:1515: note: ‘tmp_mode’ was declared here >> pmval.c:1545: warning: ‘idx2’ may be used uninitialized in this function >> pmlogger.c: In function ‘do_dialog’: >> pmlogger.c:325: warning: ‘msgf’ may be used uninitialized in this >> function >> callback.c: In function ‘log_callback’: >> callback.c:336: warning: ‘pdu_last_pmid’ may be used uninitialized in >> this function >> callback.c:335: warning: ‘pdu_first_pmid’ may be used uninitialized in >> this function >> :1438: warning: ‘yyunput’ defined but not used >> gram.y: In function ‘dometric’: >> gram.y:397: warning: ‘sts’ may be used uninitialized in this function >> pmlogger.o: In function `do_dialog': >> /source/build/pcp-2.7.4/src/pmlogger/pmlogger.c:412: warning: the use of >> `tmpnam' is dangerous, better use `mkstemp' >> pmdumplog.c: In function ‘main’: >> pmdumplog.c:363: warning: ‘rawfile’ may be used uninitialized in this >> function >> pmdumplog.c:225: warning: ‘log_size’ may be used uninitialized in this >> function >> pmdumplog.c:225: note: ‘log_size’ was declared here >> pmdumplog.c:224: warning: ‘meta_size’ may be used uninitialized in this >> function >> pmdumplog.c:224: note: ‘meta_size’ was declared here >> metriclist.c: In function ‘searchmlist’: >> metriclist.c:198: warning: ‘vlistp’ may be used uninitialized in this >> function >> :1163: warning: ‘yyunput’ defined but not used >> pmstore.c: In function ‘main’: >> pmstore.c:290: warning: ‘host’ may be used uninitialized in this >> function >> pmstore.c:301: warning: ‘aggr_len’ may be used uninitialized in this >> function >> pmlogsummary.c: In function ‘main’: >> pmlogsummary.c:661: warning: ‘rate’ may be used uninitialized in this >> function >> pmlogsummary.c:661: note: ‘rate’ was declared here >> pmclient.c: In function ‘main’: >> pmclient.c:261: warning: ‘host’ may be used uninitialized in this >> function >> pmkstat.c: In function ‘main’: >> pmkstat.c:99: warning: ‘host’ may be used uninitialized in this function >> pmkstat.c:112: warning: ‘lastdkwrite’ may be used uninitialized in this >> function >> pmkstat.c:111: warning: ‘lastdkread’ may be used uninitialized in this >> function >> pmkstat.c:107: warning: ‘prev’ may be used uninitialized in this >> function >> pmlc.c: In function ‘main’: >> pmlc.c:80: warning: ‘sts’ may be used uninitialized in this function >> actions.c: In function ‘Status’: >> actions.c:601: warning: ‘size’ may be used uninitialized in this >> function >> actions.c:600: warning: ‘vol’ may be used uninitialized in this function >> actions.c:599: warning: ‘state’ may be used uninitialized in this >> function >> actions.c:598: warning: ‘hostname’ may be used uninitialized in this >> function >> actions.c:597: warning: ‘timenow’ may be used uninitialized in this >> function >> actions.c:596: warning: ‘last’ may be used uninitialized in this >> function >> actions.c:595: warning: ‘start’ may be used uninitialized in this >> function >> actions.c:594: warning: ‘tzlogger’ may be used uninitialized in this >> function >> actions.c: In function ‘ShowLoggers’: >> actions.c:214: warning: ‘pport’ may be used uninitialized in this >> function >> :1487: warning: ‘yyunput’ defined but not used >> dso.c: In function ‘dodso’: >> dso.c:215: warning: ‘desc_list’ may be used uninitialized in this >> function >> dso.c:213: warning: ‘sts’ may be used uninitialized in this function >> pmda.c: In function ‘fillResult’: >> pmda.c:504: warning: ‘nbyte’ may be used uninitialized in this function >> pmda.c: In function ‘dopmda’: >> pmda.c:268: warning: ‘desc_list’ may be used uninitialized in this >> function >> :1632: warning: ‘yyunput’ defined but not used >> pmstat.c: In function ‘cntDiff’: >> pmstat.c:103: warning: ‘diff’ may be used uninitialized in this function >> pmcd.c: In function ‘pmcd_instance’: >> pmcd.c:649: warning: ‘nports’ may be used uninitialized in this function >> pmcd.c:651: warning: ‘pmiecount’ may be used uninitialized in this >> function >> pmcd.c:648: warning: ‘getname’ may be used uninitialized in this >> function >> pmcd.c: In function ‘pmcd_fetch’: >> pmcd.c:915: warning: ‘dp’ may be used uninitialized in this function >> weblog.c: In function ‘web_fetch’: >> weblog.c:2530: warning: ‘m_type’ may be used uninitialized in this >> function >> weblog.c:2529: warning: ‘m_offset’ may be used uninitialized in this >> function >> txrecord.c: In function ‘main’: >> txrecord.c:44: warning: ‘sp’ may be used uninitialized in this function >> telnet.c: In function ‘grab_cisco’: >> telnet.c:338: warning: ‘skip’ may be used uninitialized in this function >> probe.c: In function ‘main’: >> probe.c:118: warning: ‘name’ may be used uninitialized in this function >> probe.c:118: note: ‘name’ was declared here >> probe.c:117: warning: ‘ctype’ may be used uninitialized in this function >> probe.c:117: note: ‘ctype’ was declared here >> probe.c:113: warning: ‘defer’ may be used uninitialized in this function >> probe.c:113: note: ‘defer’ was declared here >> jstat.c: In function ‘jstat_indom_check’: >> jstat.c:410: warning: ‘sts’ is used uninitialized in this function >> jstat.c: In function ‘jstat_refresh’: >> jstat.c:593: warning: ‘error’ is used uninitialized in this function >> linuxlnk.c: In function ‘SetBaudCOM’: >> linuxlnk.c:380: warning: ‘baud’ may be used uninitialized in this >> function >> telnet-probe.c: In function ‘main’: >> telnet-probe.c:35: warning: ‘port’ may be used uninitialized in this >> function >> >> > > -- Mark Goodwin markgw@sgi.com Engineering Manager for XFS and PCP Phone: +61-3-99631937 SGI Australian Software Group Cell: +61-4-18969583 ------------------------------------------------------------- From nscott@aconex.com Tue Apr 29 16:45:17 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 29 Apr 2008 16:45:31 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3TNjHXp012144 for ; Tue, 29 Apr 2008 16:45:17 -0700 X-ASG-Debug-ID: 1209512760-625d03760000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 3859710DEDF8 for ; Tue, 29 Apr 2008 16:46:00 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id tDN9kLBJV7QM423x for ; Tue, 29 Apr 2008 16:46:00 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 36E0B5C31ED; Wed, 30 Apr 2008 09:45:49 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id ztv62DjUjotSJKW4; Wed, 30 Apr 2008 09:45:49 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id 27A2392C0E3; Wed, 30 Apr 2008 09:45:59 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Wed, 30 Apr 2008 09:45:59 +1000 (EST) Message-ID: <40134.192.168.3.1.1209512759.squirrel@mail.aconex.com> In-Reply-To: <4817B1FB.6080607@sgi.com> References: <1208490519.23166.11.camel@edge.scott.net.au> <1208988971.27113.16.camel@bozo.localdomain> <4817B1FB.6080607@sgi.com> Date: Wed, 30 Apr 2008 09:45:59 +1000 (EST) X-ASG-Orig-Subj: Re: Warnings galore Subject: Re: Warnings galore From: nscott@aconex.com To: markgw@sgi.com Cc: "Ken McDonell" , pcp@oss.sgi.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1209512761 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0003 1.0000 -2.0194 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.47 X-Barracuda-Spam-Status: No, SCORE=-1.47 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.49211 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name X-archive-position: 1617 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp > > > Ken McDonell wrote: >> I've looked at the first one. > > Anyone looked at the rest of them? They can't all be bogus?? I fixed the two jstat PMDA ones toward the bottom - they were legit (different warning message though - "is used uninitialised" versus "may be used uninitialised"). I've not looked at any of the others. cheers. -- Nathan From nscott@aconex.com Tue Apr 29 17:08:02 2008 Received: with ECARTIS (v1.0.0; list pcp); Tue, 29 Apr 2008 17:08:10 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3U07u1S013316 for ; Tue, 29 Apr 2008 17:08:02 -0700 X-ASG-Debug-ID: 1209514119-6c57035b0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 97F5010DEAEC for ; Tue, 29 Apr 2008 17:08:39 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id LqCXiQGdMwxwFgwC for ; Tue, 29 Apr 2008 17:08:39 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id D29735C2A58 for ; Wed, 30 Apr 2008 10:08:28 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id qFsjzn5ukkwKb44V for ; Wed, 30 Apr 2008 10:08:28 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id D1F2392C0E3 for ; Wed, 30 Apr 2008 10:08:38 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Wed, 30 Apr 2008 10:08:38 +1000 (EST) Message-ID: <40997.192.168.3.1.1209514118.squirrel@mail.aconex.com> Date: Wed, 30 Apr 2008 10:08:38 +1000 (EST) X-ASG-Orig-Subj: pmParseMetricSpec(3) problems Subject: pmParseMetricSpec(3) problems From: nscott@aconex.com To: pcp@oss.sgi.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1209514120 X-Barracuda-Bayes: INNOCENT GLOBAL 0.2115 1.0000 -0.7674 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -0.22 X-Barracuda-Spam-Status: No, SCORE=-0.22 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.49211 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name X-archive-position: 1618 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Hi all, I've mentioned the first of these issues to Mark before, but just wanted to get the word out to a wider audience, in case someone else can fix this for me. :) I've come across two issues in pmParseMetricSpec(3). The first is that it does not do the right thing for (archive) file names containing a colon (:) - these it treats as a hostname (up to the first colon). This is not that uncommon since archives might well be named containing HH:MM notation for hours/minutes, as kmchart once did for its recordings. Second, I've just recently come across the fact that theres no way to specify use of the local context through this interface - it handles only hostnames and archive filenames in its parsing, and the API "int isarch" parameter makes resolving this quite tricky. I'm not sure how best to fix this one. It may be that we need a new interface here, which allows all context types to be passed in? Or could we change the "int isarch" field to be the context "int type" and also make localhost:/metric[] to mean local context? (the latter bit seems quite dodgey to me - but I don't see a better way). cheers. -- Nathan From kenj@internode.on.net Wed Apr 30 14:19:36 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 30 Apr 2008 14:19:50 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3ULJSuB025602 for ; Wed, 30 Apr 2008 14:19:36 -0700 X-ASG-Debug-ID: 1209590410-5ab802c70000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from ipmail01.adl6.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 9B853112151 for ; Wed, 30 Apr 2008 14:20:10 -0700 (PDT) Received: from ipmail01.adl6.internode.on.net (ipmail01.adl6.internode.on.net [203.16.214.146]) by cuda.sgi.com with ESMTP id BlCHlEWZi4apA3zs for ; Wed, 30 Apr 2008 14:20:10 -0700 (PDT) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AqEBALt9GEg7p65s/2dsb2JhbAAIrg0 X-IronPort-AV: E=Sophos;i="4.25,730,1199626200"; d="scan'208";a="93018769" Received: from ppp59-167-174-108.lns1.mel4.internode.on.net (HELO [192.168.1.65]) ([59.167.174.108]) by ipmail01.adl6.internode.on.net with ESMTP; 01 May 2008 06:50:08 +0930 X-ASG-Orig-Subj: Re: pmParseMetricSpec(3) problems Subject: Re: pmParseMetricSpec(3) problems From: Ken McDonell To: nscott@aconex.com Cc: pcp@oss.sgi.com In-Reply-To: <40997.192.168.3.1.1209514118.squirrel@mail.aconex.com> References: <40997.192.168.3.1.1209514118.squirrel@mail.aconex.com> Content-Type: text/plain Date: Thu, 01 May 2008 07:20:07 +1000 Message-Id: <1209590407.2870.19.camel@localhost.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 (2.12.3-4.fc8) Content-Transfer-Encoding: 7bit X-Barracuda-Connect: ipmail01.adl6.internode.on.net[203.16.214.146] X-Barracuda-Start-Time: 1209590412 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.49298 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1619 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: kenj@internode.on.net Precedence: bulk X-list: pcp I think these are both solvable. >From a spec point of view (see PCPIntro(1)) a) eating multiple colons in the archive name is a no brainer b) if neither host: nor archive/ is present the metric spec is still valid, so disk.all.total and disk.dev.total[sda1] and disk.dev.total[mydisk,yourdisk theirdisk] are all valid and refer to the local pmcd's view of the metrics world. >From an implementation point of view, a) is simple. b) requires a bit more research. On Wed, 2008-04-30 at 10:08 +1000, nscott@aconex.com wrote: > Hi all, > > I've mentioned the first of these issues to Mark before, but just wanted > to get > the word out to a wider audience, in case someone else can fix this for > me. :) > > I've come across two issues in pmParseMetricSpec(3). The first is that it > does > not do the right thing for (archive) file names containing a colon (:) - > these it > treats as a hostname (up to the first colon). This is not that uncommon > since > archives might well be named containing HH:MM notation for hours/minutes, > as kmchart once did for its recordings. > > Second, I've just recently come across the fact that theres no way to specify > use of the local context through this interface - it handles only hostnames > and archive filenames in its parsing, and the API "int isarch" parameter > makes > resolving this quite tricky. > > I'm not sure how best to fix this one. It may be that we need a new > interface > here, which allows all context types to be passed in? Or could we change the > "int isarch" field to be the context "int type" and also make > localhost:/metric[] > to mean local context? (the latter bit seems quite dodgey to me - but I > don't > see a better way). > > cheers. > > -- > Nathan > > From nscott@aconex.com Wed Apr 30 14:38:53 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 30 Apr 2008 14:39:07 -0700 (PDT) Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m3ULcjWH026784 for ; Wed, 30 Apr 2008 14:38:53 -0700 X-ASG-Debug-ID: 1209591567-029602ce0000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 283241616D6A for ; Wed, 30 Apr 2008 14:39:28 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id Hxi1A2Cd4k9F0Xd6 for ; Wed, 30 Apr 2008 14:39:28 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 08F7D5CBEB3; Thu, 1 May 2008 07:39:14 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id EYR79DtBw6N76n7i; Thu, 01 May 2008 07:39:14 +1000 (EST) Received: from mail.aconex.com (castle.yarra.acx [192.168.3.3]) by postoffice.aconex.com (Postfix) with ESMTP id D2B8292C2F0; Thu, 1 May 2008 07:39:25 +1000 (EST) Received: from 192.168.3.1 (proxying for 58.107.42.33) (SquirrelMail authenticated user nscott) by mail.aconex.com with HTTP; Thu, 1 May 2008 07:39:25 +1000 (EST) Message-ID: <54508.192.168.3.1.1209591565.squirrel@mail.aconex.com> In-Reply-To: <1209590407.2870.19.camel@localhost.localdomain> References: <40997.192.168.3.1.1209514118.squirrel@mail.aconex.com> <1209590407.2870.19.camel@localhost.localdomain> Date: Thu, 1 May 2008 07:39:25 +1000 (EST) X-ASG-Orig-Subj: Re: pmParseMetricSpec(3) problems Subject: Re: pmParseMetricSpec(3) problems From: nscott@aconex.com To: "Ken McDonell" Cc: pcp@oss.sgi.com User-Agent: SquirrelMail/1.4.8-4.el4.centos MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1209591570 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.47 X-Barracuda-Spam-Status: No, SCORE=-1.47 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=NO_REAL_NAME X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.49298 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.55 NO_REAL_NAME From: does not include a real name X-archive-position: 1620 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp > I think these are both solvable. > >>From a spec point of view (see PCPIntro(1)) > a) eating multiple colons in the archive name is a no brainer *nod*. Simple Matter Of Programming there. > b) if neither host: nor archive/ is present the metric spec is still > valid, so disk.all.total and disk.dev.total[sda1] and > disk.dev.total[mydisk,yourdisk theirdisk] are all valid and refer to the > local pmcd's view of the metrics world. Thats pretty much how it is today - except I think that if there is no archive/host specified at all, then the passed in value "isarch" is used in the result structure passed back out of pmParseMetricSpec. But, my issue was local contexts (not pmcd) - I don't see any way for a pmMetricSpec to specify this third kind of context today, and not sure what the best approach to take to implement that is. > On Wed, 2008-04-30 at 10:08 +1000, nscott@aconex.com wrote: >> ... >> Second, I've just recently come across the fact that theres no way to >> specify >> use of the local context through this interface - it handles only >> hostnames >> and archive filenames in its parsing, and the API "int isarch" parameter >> makes >> resolving this quite tricky. >> >> I'm not sure how best to fix this one. It may be that we need a new >> interface >> here, which allows all context types to be passed in? Or could we >> change the >> "int isarch" field to be the context "int type" and also make >> localhost:/metric[] >> to mean local context? (the latter bit seems quite dodgey to me - but I >> don't >> see a better way). cheers. -- Nathan From markgw@sgi.com Wed Apr 30 16:18:22 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 30 Apr 2008 16:18:41 -0700 (PDT) Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with SMTP id m3UNIJbK001590 for ; Wed, 30 Apr 2008 16:18:21 -0700 Received: from [134.14.55.13] (dhcp13.melbourne.sgi.com [134.14.55.13]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id JAA04146; Thu, 1 May 2008 09:18:57 +1000 Message-ID: <4818FE5B.5000105@sgi.com> Date: Thu, 01 May 2008 09:18:51 +1000 From: Mark Goodwin Reply-To: markgw@sgi.com Organization: SGI Engineering User-Agent: Thunderbird 1.5.0.14 (Windows/20071210) MIME-Version: 1.0 To: nscott@aconex.com CC: Ken McDonell , pcp@oss.sgi.com Subject: Re: pmParseMetricSpec(3) problems References: <40997.192.168.3.1.1209514118.squirrel@mail.aconex.com> <1209590407.2870.19.camel@localhost.localdomain> <54508.192.168.3.1.1209591565.squirrel@mail.aconex.com> In-Reply-To: <54508.192.168.3.1.1209591565.squirrel@mail.aconex.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-archive-position: 1621 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: markgw@sgi.com Precedence: bulk X-list: pcp nscott@aconex.com wrote: >> I think these are both solvable. >> >> >From a spec point of view (see PCPIntro(1)) >> a) eating multiple colons in the archive name is a no brainer > > *nod*. Simple Matter Of Programming there. > >> b) if neither host: nor archive/ is present the metric spec is still >> valid, so disk.all.total and disk.dev.total[sda1] and >> disk.dev.total[mydisk,yourdisk theirdisk] are all valid and refer to the >> local pmcd's view of the metrics world. > > Thats pretty much how it is today - except I think that if there is no > archive/host specified at all, then the passed in value "isarch" is used > in the result structure passed back out of pmParseMetricSpec. > > But, my issue was local contexts (not pmcd) - I don't see any way for a > pmMetricSpec to specify this third kind of context today, and not sure > what the best approach to take to implement that is. perhaps a reserved word, e.g. LOCALCONTEXT: Do we also want to try and entertain a chain of proxies, separated by ':'? I think you mentioned you have no need for more than one, and our need has gone away entirely now. Cheers > >> On Wed, 2008-04-30 at 10:08 +1000, nscott@aconex.com wrote: >>> ... >>> Second, I've just recently come across the fact that theres no way to >>> specify >>> use of the local context through this interface - it handles only >>> hostnames >>> and archive filenames in its parsing, and the API "int isarch" parameter >>> makes >>> resolving this quite tricky. >>> >>> I'm not sure how best to fix this one. It may be that we need a new >>> interface >>> here, which allows all context types to be passed in? Or could we >>> change the >>> "int isarch" field to be the context "int type" and also make >>> localhost:/metric[] >>> to mean local context? (the latter bit seems quite dodgey to me - but I >>> don't >>> see a better way). > > cheers. > > -- > Nathan > > > -- Mark Goodwin markgw@sgi.com Engineering Manager for XFS and PCP Phone: +61-3-99631937 SGI Australian Software Group Cell: +61-4-18969583 ------------------------------------------------------------- From nscott@aconex.com Wed Apr 30 17:03:15 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 30 Apr 2008 17:03:23 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m41037wM009295 for ; Wed, 30 Apr 2008 17:03:15 -0700 X-ASG-Debug-ID: 1209600231-202703370000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 222B81136BB for ; Wed, 30 Apr 2008 17:03:51 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id 8rJnr2ZSQiRBdbOE for ; Wed, 30 Apr 2008 17:03:51 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id 3DAAA5CC947 for ; Thu, 1 May 2008 10:03:39 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id Ove9Z6pyNYFxIt0H for ; Thu, 01 May 2008 10:03:39 +1000 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id 2631F92C302; Thu, 1 May 2008 10:03:50 +1000 (EST) X-ASG-Orig-Subj: Re: pmParseMetricSpec(3) problems Subject: Re: pmParseMetricSpec(3) problems From: Nathan Scott Reply-To: nscott@aconex.com To: Ken McDonell Cc: pcp@oss.sgi.com In-Reply-To: <1209594900.2885.13.camel@localhost.localdomain> References: <40997.192.168.3.1.1209514118.squirrel@mail.aconex.com> <1209590407.2870.19.camel@localhost.localdomain> <54508.192.168.3.1.1209591565.squirrel@mail.aconex.com> <1209594900.2885.13.camel@localhost.localdomain> Content-Type: text/plain Organization: Aconex Date: Thu, 01 May 2008 10:03:49 +1000 Message-Id: <1209600229.25294.44.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.22.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1209600232 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.49308 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1622 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp On Thu, 2008-05-01 at 08:34 +1000, Ken McDonell wrote: > Oops, sorry for missing the whole point! > > Since the original rationale for PM_CONTEXT_LOCAL no longer exists (let > me know if anyone is interested in the history lesson), and it was > probably a bad idea even at the time, much less now, I'm curious why > this one is of interest. > > Killing off PM_CONTEXT_LOCAL would solve your problem ... 8^)> > > Alternately, choose a character that cannot be part of a valid hostname > (e.g. @) and declare that @:metric means PM_CONTEXT_LOCAL ... in which > case "host" would be specified in the string, and isarch would not be > used as an input parameter, and extend isarch in the return result to be > -1 or 2 for PM_CONTEXT_LOCAL. > > In the open source code base, pmParseMetricSpec() is only called from: > src/libpcp_pmc/src/Metric.c++ > src/pmdumptext/pmdumptext.c++ > src/pmval/pmval.c > so we're probably not looking a big risk even if the API semantics are > bent a little to accommodate PM_CONTEXT_LOCAL. I'm in favour of the latter, even having had the history lesson, for the reasons I mentioned earlier - mainly the option for (even) less CPUburn when running kmchart, the ability to run it without a network (heh, ok, thats just my occassionally wierd setup), and most importantly as a way to reduce the barrier-to-entry for people trying out kmchart (no need for pmcd running, and everything can be run as a regular user with no additional priveledges). As discussed, the @:[inst] syntax sounds good to me. cheers. -- Nathan From nscott@aconex.com Wed Apr 30 17:07:17 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 30 Apr 2008 17:07:30 -0700 (PDT) Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m4107A85009445 for ; Wed, 30 Apr 2008 17:07:17 -0700 X-ASG-Debug-ID: 1209600472-4a1302c10000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id A2F541618432 for ; Wed, 30 Apr 2008 17:07:52 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id 2ZkCD4YNtU1ggMEx for ; Wed, 30 Apr 2008 17:07:52 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id D01835CC6BD; Thu, 1 May 2008 10:07:40 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id AqTlcD5lLvPVBMYR; Thu, 01 May 2008 10:07:40 +1000 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id B794492C2F0; Thu, 1 May 2008 10:07:51 +1000 (EST) X-ASG-Orig-Subj: Re: pmParseMetricSpec(3) problems Subject: Re: pmParseMetricSpec(3) problems From: Nathan Scott Reply-To: nscott@aconex.com To: markgw@sgi.com Cc: Ken McDonell , pcp@oss.sgi.com In-Reply-To: <4818FE5B.5000105@sgi.com> References: <40997.192.168.3.1.1209514118.squirrel@mail.aconex.com> <1209590407.2870.19.camel@localhost.localdomain> <54508.192.168.3.1.1209591565.squirrel@mail.aconex.com> <4818FE5B.5000105@sgi.com> Content-Type: text/plain Organization: Aconex Date: Thu, 01 May 2008 10:07:51 +1000 Message-Id: <1209600471.25294.49.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.22.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1209600474 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -2.02 X-Barracuda-Spam-Status: No, SCORE=-2.02 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests= X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.49308 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- X-archive-position: 1623 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp On Thu, 2008-05-01 at 09:18 +1000, Mark Goodwin wrote: > > nscott@aconex.com wrote: > > But, my issue was local contexts (not pmcd) - I don't see any way for a > > pmMetricSpec to specify this third kind of context today, and not sure > > what the best approach to take to implement that is. > > perhaps a reserved word, e.g. LOCALCONTEXT: That's a valid hostname though. Ken's suggestion of a single "@" character sounds good to me - no hostname will have that, and its unambiguous even with the extended pmproxy syntax. > Do we also want to try and entertain a chain of proxies, separated by ':'? > I think you mentioned you have no need for more than one, and our > need has gone away entirely now. That concept is still feasible - with multiple "@"s though, not multiple ":"s. cheers. -- Nathan From nscott@aconex.com Wed Apr 30 17:58:56 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 30 Apr 2008 17:59:04 -0700 (PDT) Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m410wlCM013074 for ; Wed, 30 Apr 2008 17:58:56 -0700 X-ASG-Debug-ID: 1209603563-573703610000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 0B0ED1618123 for ; Wed, 30 Apr 2008 17:59:23 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id WUAVCgiTJ16INTbZ for ; Wed, 30 Apr 2008 17:59:23 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id EECD95CCEDF for ; Thu, 1 May 2008 10:59:10 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id SIHoWxvuPhw5Oadm for ; Thu, 01 May 2008 10:59:10 +1000 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id D7BB992C302 for ; Thu, 1 May 2008 10:59:21 +1000 (EST) X-ASG-Orig-Subj: pcp updates Subject: pcp updates From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Content-Type: text/plain Organization: Aconex Date: Thu, 01 May 2008 10:59:21 +1000 Message-Id: <1209603561.25294.61.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.22.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1209603564 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.52 X-Barracuda-Spam-Status: No, SCORE=-1.52 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=BSF_RULE7568M X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.49312 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.50 BSF_RULE7568M Custom Rule 7568M X-archive-position: 1624 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/pcp.git src/pmdas/jstat/jstat.c | 1 - src/pmdas/pmcd/src/pmcd.c | 41 ++++++++++++++++++++++++++++------------- src/pmdas/windows/data.c | 4 ++-- 3 files changed, 30 insertions(+), 16 deletions(-) commit ecebb7e60135f46c672f57dd6892be0c97c378dc Merge: 3ea795f... dcd62fc... Author: Nathan Scott Date: Thu May 1 10:57:47 2008 +1000 Merge branch 'nathans' of git://oss.sgi.com:8090/nathans/pcp into nathans commit 3ea795fbd385ae71c3a19e9b6856d4c0272e60b5 Author: Nathan Scott Date: Thu May 1 10:56:17 2008 +1000 Fix memory leaks in the pmdapmcd code dealing with the pmie indom. commit 644ccf2f8a59de32e2b7616007cb7eda2c73df04 Author: Nathan Scott Date: Thu May 1 10:54:19 2008 +1000 Remove an unused variable in the jstat PMDA. commit 8769b45dced83c9df0cd1d29fa4ef4861d4e4806 Author: Nathan Scott Date: Mon Apr 28 16:14:05 2008 +1000 Turns out that the sqlserver cache_mgr cache_object_count is signed. From nscott@aconex.com Wed Apr 30 19:22:18 2008 Received: with ECARTIS (v1.0.0; list pcp); Wed, 30 Apr 2008 19:22:27 -0700 (PDT) Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m412MCFe018271 for ; Wed, 30 Apr 2008 19:22:18 -0700 X-ASG-Debug-ID: 1209608571-737a01d50000-S8gJnT X-Barracuda-URL: http://cuda.sgi.com:80/cgi-bin/mark.cgi Received: from postoffice2.aconex.com (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 8D52F161895F for ; Wed, 30 Apr 2008 19:22:52 -0700 (PDT) Received: from postoffice2.aconex.com (prod.aconex.com [203.89.192.138]) by cuda.sgi.com with ESMTP id 8um6hO9PMtSy9BKX for ; Wed, 30 Apr 2008 19:22:52 -0700 (PDT) Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam Firewall) with ESMTP id EB1CC5CDC38 for ; Thu, 1 May 2008 12:17:25 +1000 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.3.1]) by postoffice2.aconex.com with ESMTP id chJDryABPtDDHNAx for ; Thu, 01 May 2008 12:17:25 +1000 (EST) Received: from [192.168.5.76] (unknown [203.89.192.141]) by postoffice.aconex.com (Postfix) with ESMTP id DD7A692C2F0 for ; Thu, 1 May 2008 12:17:36 +1000 (EST) X-ASG-Orig-Subj: kmchart updates (1.2 release) Subject: kmchart updates (1.2 release) From: Nathan Scott Reply-To: nscott@aconex.com To: pcp@oss.sgi.com Content-Type: text/plain Organization: Aconex Date: Thu, 01 May 2008 12:17:36 +1000 Message-Id: <1209608256.25294.107.camel@edge.scott.net.au> Mime-Version: 1.0 X-Mailer: Evolution 2.22.1 Content-Transfer-Encoding: 7bit X-Barracuda-Connect: prod.aconex.com[203.89.192.138] X-Barracuda-Start-Time: 1209608576 X-Barracuda-Bayes: INNOCENT GLOBAL 0.0000 1.0000 -2.0210 X-Barracuda-Virus-Scanned: by cuda.sgi.com at sgi.com X-Barracuda-Spam-Score: -1.52 X-Barracuda-Spam-Status: No, SCORE=-1.52 using per-user scores of TAG_LEVEL=2.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=2.1 tests=BSF_RULE7568M X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.49316 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.50 BSF_RULE7568M Custom Rule 7568M X-archive-position: 1625 X-ecartis-version: Ecartis v1.0.0 Sender: pcp-bounce@oss.sgi.com Errors-to: pcp-bounce@oss.sgi.com X-original-sender: nscott@aconex.com Precedence: bulk X-list: pcp Changes committed to git://oss.sgi.com:8090/nathans/kmchart.git README | 23 +++++++++---------- VERSION | 4 +-- debian/changelog | 6 +++++ doc/CHANGES | 7 ++++++ src/chart/main.cpp | 24 ++++++++++++++++---- src/chart/namespace.cpp | 56 ++++++++++++++++++++++++++++++++++++++++++------ src/chart/tab.cpp | 7 +++++- 7 files changed, 101 insertions(+), 26 deletions(-) commit 48b3f7b22064d9c3b0f44d25ef79d7c4d81392c7 Author: Nathan Scott Date: Thu May 1 12:13:06 2008 +1000 Bump stable version to 1.2. commit 1f46d09456bb6041dbe31c8d48d98978df800758 Author: Ken McDonell Date: Thu May 1 12:04:55 2008 +1000 Fix the metric selection tree widget code, where expanding parts of the namespace would randomly do some children and not others. The Qt routine sortChildren() does not work as we maintain our own pointers into the tree items via my.back ... if sortChildren() is used, our expansion picking does not work later on. So, we now sort the PMNS children lexicographically by name before adding them into the QTreeWidget ... only tricky part is that we need to sort both offspring[] AND status[]. commit 84d32d4e66c3ea1dd7f9b7eaae09e3ac364aade5 Author: Nathan Scott Date: Thu May 1 09:06:19 2008 +1000 Update README after recent fixes. commit 61811d23aa17a96625c280ec9a9f13716df2385b Author: Nathan Scott Date: Thu May 1 09:01:33 2008 +1000 Fix timezone reporting issue with the timestamp in the value label. commit 5fd335e21c30d8270fea309e52dbfcf4fd37f1ef Author: Nathan Scott Date: Thu May 1 08:49:57 2008 +1000 Explicitly update the time axis state button on Live mode Stop event. commit 7f4f2a0b9ef57bcf00915f8e6828a04c6a4ad025 Author: Nathan Scott Date: Thu May 1 08:35:27 2008 +1000 Fix archive offset to ensure new start position is within archive bounds.