RT for crt.cs.wisc.edu
RT for crt.cs.wisc.edu
| # | Thu Oct 02 11:09:16 2008 | burt@fnal.gov - Ticket created | [Reply] | |||||||||
I just cut the GIP 1.0.4 release: http://home.fnal.gov/~burt/gip_1_0_4.tgz Alain, could you put it in a test cache please so that Jeff and Suchandra can install and test? - B --- Changelog (since 1.0.2) == 1.0.4 == * Bugfix for GlueVOInfoLocalID, which needs to include GlueSEUniqueID.== 1.0.3 == * Sets reasonable defaults for GlueCEPolicyMaxObtainableWallClockTime, GlueCEPolicyMaxObtainableCPUTime, GlueCEStateWorstResponseTime |
||||||||||||
| # | Fri Oct 03 13:14:50 2008 | roy - Taken | ||
| # | Fri Oct 03 13:15:11 2008 | roy - Priority changed from (no value) to '3' | ||
| # | Fri Oct 03 13:15:12 2008 | roy - Fix scheduled CUR added | ||
| # | Fri Oct 03 13:16:56 2008 | roy - Cc sthapa@ci.uchicago.edu added | ||
| # | Fri Oct 03 13:16:57 2008 | roy - Cc rjporter@lbl.gov added | ||
| # | Fri Oct 03 13:18:16 2008 | roy - Correspondence added | [Reply] | |
|
On Thu Oct 02 11:09:16 2008, burt@fnal.gov wrote: > I just cut the GIP 1.0.4 release: > > http://home.fnal.gov/~burt/gip_1_0_4.tgz > > Alain, could you put it in a test cache please so that Jeff and > Suchandra can install and test? > > - B Hi everyone, We're busy finalizing an update for VDT 1.10.1 which will be released on Monday. We'll get to this shortly thereafter (perhaps Tuesday?) and let you know. It looks like this is a pretty straightforward bug fix, so it's likely to move quickly into a release, right? Thanks, -alain ----------------------------------------------------------------- Alain Roy vdt-support@opensciencegrid.org VDT Support http://vdt.cs.wisc.edu/support.html |
||||
| # | Fri Oct 03 13:18:17 2008 | RT_System - Status changed from 'new' to 'open' | ||
| # | Fri Oct 03 14:09:27 2008 | burt@fnal.gov - Correspondence added | [Reply] | |||||||||||
Alain Roy via RT wrote: > We're busy finalizing an update for VDT 1.10.1 which will be released on > Monday. We'll get to this shortly thereafter (perhaps Tuesday?) and let > you know. > > It looks like this is a pretty straightforward bug fix, so it's likely > to move quickly into a release, right? I wish. The last release included in the VDT is 1.0.1, so this rolls up the changes from 1.0.2 and 1.0.3 as well. 1.0.2 had some big changes to the logging subsystem that I'd like tested, and I don't think Jeff and Suchandra tested 1.0.2. - B |
||||||||||||||
| # | Fri Oct 03 14:24:46 2008 | roy - Correspondence added | [Reply] | |||||||||
>> It looks like this is a pretty straightforward bug fix, so it's >>> likely >> to move quickly into a release, right? > I wish. > > The last release included in the VDT is 1.0.1, so this rolls up the > changes from 1.0.2 and 1.0.3 as well. 1.0.2 had some big changes to > the > logging subsystem that I'd like tested, and I don't think Jeff and > Suchandra tested 1.0.2. Ah, gotcha. OK, we'll get a test cache out early next week for you to try out. -alain |
||||||||||||
| # | Tue Oct 07 10:51:12 2008 | roy - Given to kronenfe | ||
| # | Tue Oct 07 15:52:55 2008 | kronenfe - Correspondence added | [Reply] | |
|
Hi Burt, I'm preparing the test cache for the new GIP version. The configure_gip perl script which is included in the tarball has syntax errors. This causes an install time failure. For your reference, here's the error. You can also get this broken installation by doing the following: pacman -get http://vdt.cs.wisc.edu/test-cache/gip:Generic-Information-Provider [root@vdt-rhap5-amd64 test]# perl -c vdt/setup/configure_gip Backticks found where operator expected at vdt/setup/configure_gip line 1783, near "my $contents = `" (Might be a runaway multi-line `` string starting on line 1469) (Missing semicolon on previous line?) syntax error at vdt/setup/configure_gip line 1783, near "my $contents = `" Global symbol "$queue" requires explicit package name at vdt/setup/configure_gip line 1469. Global symbol "$queue" requires explicit package name at vdt/setup/configure_gip line 1469. Global symbol "$queue" requires explicit package name at vdt/setup/configure_gip line 1469. Global symbol "$queue" requires explicit package name at vdt/setup/configure_gip line 1469. Global symbol "$queue" requires explicit package name at vdt/setup/configure_gip line 1469. Global symbol "$queue" requires explicit package name at vdt/setup/configure_gip line 1469. Global symbol "$queue" requires explicit package name at vdt/setup/configure_gip line 1469. Global symbol "$queue" requires explicit package name at vdt/setup/configure_gip line 1469. Global symbol "$queue" requires explicit package name at vdt/setup/configure_gip line 1469. Global symbol "$queue" requires explicit package name at vdt/setup/configure_gip line 1469. Global symbol "$queue" requires explicit package name at vdt/setup/configure_gip line 1469. Global symbol "$queue" requires explicit package name at vdt/setup/configure_gip line 1469. syntax error at vdt/setup/configure_gip line 1783, near "create -c " vdt/setup/configure_gip has too many errors. |
||||
| # | Wed Oct 08 09:19:27 2008 | burt@fnal.gov - Correspondence added | [Reply] | |||||||||||
Thanks, Scot. You're right -- I'm not sure how we were succesfully testing GIP 1.0.2 and above. I'll release a 1.0.5 with this fixed for you shortly. - B |
||||||||||||||
| # | Wed Oct 08 10:50:08 2008 | burt@fnal.gov - Correspondence added | [Reply] | |||||||||||
Scot Kronenfeld via RT wrote: > Hi Burt, > I'm preparing the test cache for the new GIP version. The configure_gip > perl script which is included in the tarball has syntax errors. This > causes an install time failure. > > For your reference, here's the error. You can also get this broken > installation by doing the following: > pacman -get > http://vdt.cs.wisc.edu/test-cache/gip:Generic-Information-Provider > > > [root@vdt-rhap5-amd64 test]# perl -c vdt/setup/configure_gip > Backticks found where operator expected at vdt/setup/configure_gip line > 1783, near "my $contents = `" > (Might be a runaway multi-line `` string starting on line 1469) Scot, I've released GIP 1.0.5 with this syntax error fixed. http://home.fnal.gov/~burt/gip_1_0_5.tgz - B |
||||||||||||||
| # | Wed Oct 08 11:40:00 2008 | kronenfe - Correspondence added | [Reply] | |||||||||
Burt, Thanks, that fixed the problem. I've setup a test cache with GIP 1.0.5. To fetch it 1) Fetch whatever globus package you need pacman -get http://vdt.cs.wisc.edu/test-cache/gip:Some-Globus-Package 2) Then fetch the GIP pacman -get http://vdt.cs.wisc.edu/test-cache/gip:Generic-Information-Provider You can't get the new version of the GIP in any existing install that you have, because the software would be coming from two different pacman caches (this would cause problems for a lot of the VDT infrastructure). Let me know if you have any further questions. -scot |
||||||||||||
| # | Fri Oct 10 11:08:28 2008 | kronenfe - Subject changed from 'GIP 1.0.4 release for testing' to 'GIP 1.0.5 release for testing' | ||
| # | Tue Oct 14 20:59:41 2008 | rjporter@lbl.gov - Correspondence added | [Reply] | |||||||||||
Hi Scot, I installed the 'OSG-CE' package from the gip-test cache which includes gip 1.0.5. The installation seemed to go fine except for a couple of things: 1.0 When I run 'vdt-version' I see: Software Status -------- ------ Apache HTTPD 2.2.9 OK ....snip.... - Generic Information Provider 1.0.5 UPDATE AVAILABLE Is there a reason for the "UPDATE AVAILABLE"? 2.0 I got six identical text dumps during the install that I've not seen before. It may be due to the test-cache or installing OSG-CE without the ce.pacman wrapper from the OSG cache. I"ll show the text dump below. Suchandra, Is there a svn version of the configure scripts I should grab for these tests instead of the OSG-1.0 version? thanks, Jeff Text dump: export ANT_HOME="/opt/osg/gip-test/ant" export DISPLAY="localhost:11.0" export GLOBUS_ERROR_VERBOSE="true" export GLOBUS_LOCATION="/opt/osg/gip-test/globus" export GLOBUS_MYSQL_PATH="/opt/osg/gip-test/mysql" export GLOBUS_OPTIONS="-Xmx512M" export GPT_LOCATION="/opt/osg/gip-test/gpt" export G_BROKEN_FILENAMES="1" export HISTSIZE="1000" export HOME="/root" export HOSTNAME="osp1.lbl.gov" export INPUTRC="/etc/inputrc" export JAVA_HOME="/opt/osg/gip-test/jdk1.5" export KDEDIR="/usr" export LANG="en_US.UTF-8" export LD_LIBRARY_PATH="/opt/osg/gip-test/jdk1.5/jre/lib/i386:/opt/osg/gip-test/jdk1.5/jre/lib/i386/server:/opt/osg/gip-test/jdk1.5/jre/lib/i386/client:/opt/osg/gip-test/mysql/lib/mysql:/opt/osg/gip-test erkeley-db/lib:/opt/osg/gip-test/expat/lib" export LESSOPEN="|/usr/bin/lesspipe.sh %s" export LOGNAME="root" export LS_COLORS="no=00:fi=00:di=00;34:ln=00;36:pi=40;33:so=00;35:bd=40;33;01:cd=40;33;01:or=01;05;37;41:mi=01;05;37;41:ex=00;32:*.cmd=00;32:*.exe=00;32:*.com=00;32:*.btm=00;32:*.bat=00;32:*.sh=00;32:*.c =00;32:*.tar=00;31:*.tgz=00;31:*.arj=00;31:*.taz=00;31:*.lzh=00;31:*.zip=00;31:*.z=00;31:*.Z=00;31:*.gz=00;31:*.bz2=00;31:*.bz=00;31:*.tz=00;31:*.rpm=00;31:*.cpio=00;31:*.jpg=00;35:*.gif=00;35:*.bmp=00;3 *.xbm=00;35:*.xpm=00;35:*.png=00;35:*.tif=00;35:" export MAIL="/var/spool/mail/root" export MANPATH="/opt/osg/gip-test/vdt/man:/opt/osg/gip-test/perl/man:/opt/osg/gip-test/expat/man:/opt/osg/gip-test/logrotate/man:/opt/osg/gip-test/wget/man:/opt/osg/gip-test/mysql/man:/opt/osg/gip-test/j 1.5/man" export MYSQL_UNIX_PORT="/opt/osg/gip-test/vdt-app-data/mysql/var/mysql.sock" export OLDPWD export OLD_VDT_LOCATION="/opt/osg/1.0-ce" export PACMAN_INSTALLATION="/opt/osg/gip-test" export PACMAN_LOCATION="/opt/osg/pacman" export PAC_ANCHOR="/opt/osg/gip-test" export PATH="/opt/osg/gip-test/ant/bin:/opt/osg/gip-test/jdk1.5/bin:/opt/osg/gip-test/mysql/bin:/opt/osg/gip-test/wget/bin:/opt/osg/gip-test/logrotate/sbin:/opt/osg/gip-test/gpt/sbin:/opt/osg/gip-test/vd bootstrap/vdt/bin:/opt/osg/gip-test/vdt-bootstrap/vdt/sbin:/opt/osg/pacman/bin:/opt/osg/gip-test/vdt/sbin:/opt/osg/gip-test/vdt/bin:/opt/osg/pacman/bin:/usr/kerberos/sbin:/usr/kerberos/bin:/usr/local/sbi /usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin:/usr/X11R6/bin:/root/bin" export PEGASUS_HOME="/opt/osg/gip-test/pegasus" export PERL5LIB="/opt/osg/gip-test/vdt/lib" export PWD="/opt/osg/gip-test" export SHELL="/bin/bash" export SHLVL="2" export SSH_ASKPASS="/usr/libexec/openssh/gnome-ssh-askpass" export TERM="xterm" export TMP="/opt/osg/gip-test/o..pacman..o/tmp" export USER="root" export VDTSETUP_ENABLE_GATEKEEPER export VDTSETUP_ENABLE_GLOBUS_ROTATE="y" export VDTSETUP_ENABLE_GRIDFTP="y" export VDT_INSTALL_LOG="vdt-install.log" export VDT_LOCATION="/opt/osg/gip-test" export VDT_POSTINSTALL_README="/opt/osg/gip-test/post-install/README" export X509_CERT_DIR="/opt/osg/gip-test/globus/TRUSTED_CA" export XAUTHORITY="/root/.xauthA2Yoz0" export y /opt/osg/gip-test/vdt-questions.sh: line 19: VDTSETUP_ENABLE_APACHE: command not found /opt/osg/gip-test/vdt-questions.sh: line 27: VDTSETUP_ENABLE_APACHE: command not found /opt/osg/gip-test/vdt-questions.sh: line 30: VDTSETUP_ENABLE_TOMCAT: command not found /opt/osg/gip-test/vdt-questions.sh: line 31: VDTSETUP_ENABLE_MYSQL: command not found /opt/osg/gip-test/vdt-questions.sh: line 32: VDTSETUP_ENABLE_ROTATE: command not found Scot Kronenfeld via RT wrote: > Burt, > Thanks, that fixed the problem. I've setup a test cache with GIP 1.0.5. > > To fetch it > > 1) Fetch whatever globus package you need > pacman -get http://vdt.cs.wisc.edu/test-cache/gip:Some-Globus-Package > > 2) Then fetch the GIP > pacman -get http://vdt.cs.wisc.edu/test-cache/gip:Generic-Information-Provider > > > You can't get the new version of the GIP in any existing install that > you have, because the software would be coming from two different > pacman caches (this would cause problems for a lot of the VDT > infrastructure). > > Let me know if you have any further questions. > > -scot > > > |
||||||||||||||
| # | Tue Oct 14 23:42:34 2008 | kronenfe - Correspondence added | [Reply] | |||||||||
> Generic Information Provider 1.0.5 UPDATE > AVAILABLE > > Is there a reason for the "UPDATE AVAILABLE"? Good catch, but it is not a concern while in the test cache. It's comparing to the 1.10.99 cache which thinks that the GIP version is 1.0.4. Since the versions are not the same, it thinks there is an update available. This won't be an issue when we move it into the production cache, and it won't affect anything you are testing. > 2.0 I got six identical text dumps during the install that I've not > seen before. It may be due to the test-cache or installing OSG-CE > without the ce.pacman wrapper from the OSG cache. I"ll show the text > dump below. When you say text dump, you mean it's printing to the screen? I installed the Globus and Generic-Information-Provider package, and didn't see these messages. I'll try the OSG-CE package tomorrow to see if I have that problem. -scot |
||||||||||||
| # | Tue Oct 14 23:45:40 2008 | kronenfe - Correspondence added | [Reply] | |||||||||
> When you say text dump, you mean it's printing to the screen? I > installed the Globus and Generic-Information-Provider package, and > didn't see these messages. I'll try the OSG-CE package tomorrow to > see if I have that problem. Actually, my OSG-CE install finished quicker than I anticipated. I did not see any unusual text output to the screen. Here is the command I ran: pacman -get http://vdt.cs.wisc.edu/test-cache/gip:OSG-CE How did you install it? |
||||||||||||
| # | Wed Oct 15 00:36:40 2008 | rjporter@lbl.gov - Correspondence added | [Reply] | |||||||||||
Yep the same way. [root@osp1 gip-test]# pacman -get http://vdt.cs.wisc.edu/test-cache/gip:OSG-CE Except I had set OLD_VDT_LOCATION to my current CE installation. Thinking this might be a difference I just now did another install without setting that variable. This time I was asked a series of questions that were pre-answered (e.g. do you want service X to run automatically). And this time I didn't get the screen dump text. So I think this is likely ~normal behavior - or at least not related to the new GIP. I might double check in the morning with an OSG-CE install from the standard cache but I'll probably work on configuring the gip and testing. Suchandra, any update on which configure-osg.py and configure_modules to use? thanks, Jeff Scot Kronenfeld via RT wrote: >> When you say text dump, you mean it's printing to the screen? I >>> installed the Globus and Generic-Information-Provider package, and >> didn't see these messages. I'll try the OSG-CE package tomorrow to >> see if I have that problem. >> > Actually, my OSG-CE install finished quicker than I anticipated. I > did not see any unusual text output to the screen. Here is the > command I ran: > > pacman -get http://vdt.cs.wisc.edu/test-cache/gip:OSG-CE > > How did you install it? > > > |
||||||||||||||
| # | Wed Oct 15 09:04:28 2008 | kronenfe - Correspondence added | [Reply] | |||||||||
Thanks for letting us know. I'm going to set OLD_VDT_LOCATION and try to figure out where the text is coming from. On Wed, Oct 15, 2008 at 12:37 AM, rjporter@lbl.gov via RT <vdt-support@opensciencegrid.org> wrote: > Yep the same way. > > [root@osp1 gip-test]# pacman -get > http://vdt.cs.wisc.edu/test-cache/gip:OSG-CE > > Except I had set OLD_VDT_LOCATION to my current CE installation. > Thinking this might be a difference I just now did another install > without setting that variable. This time I was asked a series of > questions that were pre-answered (e.g. do you want service X to run > automatically). And this time I didn't get the screen dump text. So I > think this is likely ~normal behavior - or at least not related to the > new GIP. I might double check in the morning with an OSG-CE install > from the standard cache but I'll probably work on configuring the gip > and testing. |
||||||||||||
| # | Wed Oct 15 15:05:01 2008 | kronenfe - Correspondence added | [Reply] | |
|
The vdt-questions.sh file is not being generated correctly and is causing extra output to the screen. I'm going to do some maintenance on this cache to test a change. Unless I hear an objection, the gip test cache will be unavailable for about 15 minutes starting around 3:20 central. This means that you will not be able to do any installs for this period. |
||||
| # | Mon Oct 20 14:43:05 2008 | kronenfe - Correspondence added | [Reply] | |
|
On Tue Oct 14 20:59:41 2008, rjporter@lbl.gov wrote: > 2.0 I got six identical text dumps during the install that I've not > seen before. It may be due to the test-cache or installing OSG-CE > without the ce.pacman wrapper from the OSG cache. I"ll show the text > dump below. Hi Jeff, I've fixed a bug in our questions system that was generating this. If you do another install, the vdt-questions.sh file will be generated correctly, and you won't see any text dumps. However, I think your previous installation is probably fine to use. The only effect the vdt-questions.sh file will have is whether or not services are enabled/disabled at installation time. Thanks for bringing this to our attention. -scot |
||||
| # | Wed Oct 22 17:23:38 2008 | rjporter@lbl.gov - Correspondence added | [Reply] | |||||||||||
Hi Scot & all, The cache worked fine this time. I've been testing the new gip on LBNL_VTB - please check the data. I made some spot checks against the Changelog that Burt sent and they look good. I did want to report a couple of oddities - noting that I installed the CE piece-by-piece so that I could re-point to the test cache and use still the OSG cache. 1. The "auto-vdt-1.0.0" includes the file, "OSGv1.0.0-release-notes" but the script "install.sh" that comes with the package tries to copy "OSGv1.0.0-release.notes". Thus, the release notes are not copied to the VDT directory and you get an error message (that's hidden by the pacman install). 2. The first time I ran configure-osg.py, it claimed "OSG_LOCATION=/opt/osg/ngip-test: No such file or directory" However, that directory is correct and it didn't complain the 2nd time I ran the script. I don't think it caused any problems and may have been an oddity related to the way I did the install. thanks, Jeff Scot Kronenfeld via RT wrote: > On Tue Oct 14 20:59:41 2008, rjporter@lbl.gov wrote: > >> 2.0 I got six identical text dumps during the install that I've not >>> seen before. It may be due to the test-cache or installing OSG-CE >> without the ce.pacman wrapper from the OSG cache. I"ll show the text >> dump below. >> > Hi Jeff, > I've fixed a bug in our questions system that was generating this. If > you do another install, the vdt-questions.sh file will be generated > correctly, and you won't see any text dumps. > > However, I think your previous installation is probably fine to use. > The only effect the vdt-questions.sh file will have is whether or not > services are enabled/disabled at installation time. > > Thanks for bringing this to our attention. > > -scot > > |
||||||||||||||
| # | Thu Oct 23 22:27:47 2008 | kronenfe - Correspondence added | [Reply] | |||||||||
> I did want to report a couple of oddities - noting that I installed the > CE piece-by-piece so that I could re-point to the test cache and use > still the OSG cache. Hi Jeff, I added a ce.pacman file to our test cache so that if you have to install again for any reason, you can just do this: pacman -get http://vdt.cs.wisc.edu/test-cache/gip:ce Note that this is not kept in sync with the OSG's ce.pacman, so if they change their version (which does not happen frequently), I would have to regenerate mine. The two issues you reported are both related to the OSG cache, so I will let the appropriate people comment on them. Thanks for testing, scot |
||||||||||||
| # | Tue Oct 28 13:58:32 2008 | kronenfe - Correspondence added | [Reply] | |
|
On Wed Oct 22 17:23:38 2008, rjporter@lbl.gov wrote: > I've been testing the new gip on LBNL_VTB - please check the data. I > made some spot checks against the Changelog that Burt sent and they look > good. Is the new version of GIP ready to be released by the VDT? If so, is there a timeframe it should be released within, or can it wait until the next lettered release (which would probably be within 1-2 weeks, depending on what else comes up). Thanks, scot |
||||
| # | Mon Nov 10 10:08:17 2008 | kronenfe - Correspondence added | [Reply] | |
|
Hi Burt, Is GIP 1.0.5 ready to be released in the VDT? We have it on hold until we hear back from you. Thanks, scot |
||||
| # | Mon Nov 10 11:15:15 2008 | rjporter@lbl.gov - Correspondence added | [Reply] | |||||||||||
Hi Scot & Burt, I haven't heard from Suchandra on whether he has tested it. We've changed hardware on my ITB site that has Condor deployed so I can run a new test on that system. I should be able to do that today. Otherwise, I found no show stoppers. Jeff Scot Kronenfeld via RT wrote: > Hi Burt, > Is GIP 1.0.5 ready to be released in the VDT? We have it on hold until > we hear back from you. > > Thanks, > scot > > > |
||||||||||||||
| # | Mon Nov 10 11:30:28 2008 | sthapa@ci.uchicago.edu - Correspondence added | [Reply] | |||||||||||
I'm currently getting the condor resource on a cluster here up. I'll run through the testing today or tomorrow morning. Suchandra On Nov 10, 2008, at 11:15 AM, rjporter@lbl.gov via RT wrote: > Hi Scot & Burt, > > I haven't heard from Suchandra on whether he has tested it. We've > changed hardware on my ITB site that has Condor deployed so I can > run a > new test on that system. I should be able to do that today. > Otherwise, > I found no show stoppers. > > Jeff > > > Scot Kronenfeld via RT wrote: >> Hi Burt, >>> Is GIP 1.0.5 ready to be released in the VDT? We have it on hold >> until >> we hear back from you. >> >> Thanks, >> scot >> >> >> > > -- > View ticket at <http://crt.cs.wisc.edu/Ticket/Display.html?user=guest&pass=guest&id=4094 > > > VDT Support: vdt-support@opensciencegrid.org--- Suchandra Thapa sthapa@ci.uchicago.edu Computation Institute Research Institutes Bldg, Ste 421 5640 S Ellis Avenue Chicago, IL 60637 USA |
||||||||||||||
| # | Mon Nov 10 11:48:15 2008 | roy - Correspondence added | [Reply] | |||||||||
Suchandra is on this ticket to. Suchandra, can you weigh in? Thanks, -alain On Nov 10, 2008, at 11:15 AM, rjporter@lbl.gov via RT wrote: > http://crt.cs.wisc.edu/Ticket/Display.html?id=4094 > > Hi Scot & Burt, > > I haven't heard from Suchandra on whether he has tested it. We've > changed hardware on my ITB site that has Condor deployed so I can > run a > new test on that system. I should be able to do that today. > Otherwise, > I found no show stoppers. > > Jeff > > > Scot Kronenfeld via RT wrote: >> Hi Burt, >> Is GIP 1.0.5 ready to be released in the VDT? We have it on hold >> until >> we hear back from you. >> >> Thanks, >> scot >> >> >> |
||||||||||||
| # | Mon Nov 10 11:55:26 2008 | sthapa@ci.uchicago.edu - Correspondence added | [Reply] | |||||||||||
Hi Alain, I'm currently setting up a condor site to do the testing. I should have testing done by tomorrow. Suchandra On Nov 10, 2008, at 11:48 AM, Alain Roy via RT wrote: > Suchandra is on this ticket to. Suchandra, can you weigh in? > > Thanks, > -alain > > On Nov 10, 2008, at 11:15 AM, rjporter@lbl.gov via RT wrote: > >> http://crt.cs.wisc.edu/Ticket/Display.html?id=4094 >>> >> Hi Scot & Burt, >> >> I haven't heard from Suchandra on whether he has tested it. We've >> changed hardware on my ITB site that has Condor deployed so I can >> run a >> new test on that system. I should be able to do that today. >> Otherwise, >> I found no show stoppers. >> >> Jeff >> >> >> Scot Kronenfeld via RT wrote: >>> Hi Burt, >>> Is GIP 1.0.5 ready to be released in the VDT? We have it on hold >>> until >>> we hear back from you. >>> >>> Thanks, >>> scot >>> >>> >>> > > -- > View ticket at <http://crt.cs.wisc.edu/Ticket/Display.html?user=guest&pass=guest&id=4094 > > > VDT Support: vdt-support@opensciencegrid.org--- Suchandra Thapa sthapa@ci.uchicago.edu Computation Institute Research Institutes Bldg, Ste 421 5640 S Ellis Avenue Chicago, IL 60637 USA |
||||||||||||||
| # | Wed Nov 12 15:12:51 2008 | kronenfe - Correspondence added | [Reply] | |
|
Hi Suchandra, Have you had a chance to do the testing on this? Thanks, scot |
||||
| # | Wed Nov 12 15:42:09 2008 | rjporter@lbl.gov - Correspondence added | [Reply] | |||||||||||
Hi Scot, I did do the install on my condor site and, though I didn't have any noticeable problems during the stall, it does not appear to be reporting correctly. I'm a bit cautious since this is a debian-4 system which may have some hidden traps (e.g. /bin/env doesn't exist while /usr/bin/env does). The site is LBNL_DSD_ITB. It is reporting on the CEmon monitoring pages: https://osg-ress-1.fnal.gov:8443/ReSS/ReSS-int-ClassadValidity.html https://osg-ress-1.fnal.gov:8443/ReSS/ReSS-int-History.html But not on the CEMon-bdii status page: http://is-itb.grid.iu.edu/cgi-bin/status.cgi I also cannot find it's info via an ldapbrowser or ldapsearch. I sent an email in for clarification about this (regarding my VTB site, LBNL_VTB) to the gip-team late on Monday. I figured I would sit with them at the site-admin meeting tomorrow if I didn't hear anythiing sooner. thanks, Jeff Scot Kronenfeld via RT wrote: > Hi Suchandra, > Have you had a chance to do the testing on this? > > Thanks, > scot > > |
||||||||||||||
| # | Wed Nov 12 15:56:22 2008 | kronenfe - Correspondence added | [Reply] | |||||||||
Ok, thanks for the info. Please keep us posted. The new GIP is on hold until I hear otherwise. |
||||||||||||
| # | Wed Nov 12 16:04:24 2008 | burt@fnal.gov - Correspondence added | [Reply] | |||||||||||
Scot Kronenfeld via RT wrote: > Ok, thanks for the info. Please keep us posted. The new GIP is on > hold until I hear otherwise. Thanks, Scot. I think this is a problem actually at the BDII/collector level, but various folks are en route to the site admin workshop, so we won't have accurate data to test against for a bit. - B |
||||||||||||||
| # | Thu Nov 13 00:16:22 2008 | roy - Correspondence added | [Reply] | |||||||||
On Nov 12, 2008, at 2:04 PM, burt@fnal.gov via RT wrote: > http://crt.cs.wisc.edu/Ticket/Display.html?id=4094 > > Scot Kronenfeld via RT wrote: >> Ok, thanks for the info. Please keep us posted. The new GIP is on >>> hold until I hear otherwise. > Thanks, Scot. I think this is a problem actually at the BDII/ > collector > level, but various folks are en route to the site admin workshop, so > we > won't have accurate data to test against for a bit. Thanks Burt. Let's keep an eye on this so we can release it when we feel it's ready. -alain ----------------------------------------------------------------- Alain Roy vdt-support@opensciencegrid.org VDT Support http://vdt.cs.wisc.edu/support.html |
||||||||||||
| # | Mon Nov 17 09:48:06 2008 | roy - Queue changed from vdt-support to vdt-internal | ||
| # | Thu Dec 04 12:49:59 2008 | roy - Correspondence added | [Reply] | |
|
Hi, As best I can tell, the GIP 1.0.5 (1.0.6?) upgrade has been in limbo for a few weeks. We're ready to roll with it if people want to release it, but we haven't gotten the all-clear from you. I think the ball is in your court. What should we do? Thanks, -alain ----------------------------------------------------------------- Alain Roy vdt-support@opensciencegrid.org VDT Support http://vdt.cs.wisc.edu/support.html On Thu Nov 13 00:16:22 2008, roy wrote: > On Nov 12, 2008, at 2:04 PM, burt@fnal.gov via RT wrote: > > http://crt.cs.wisc.edu/Ticket/Display.html?id=4094 > > > > > Scot Kronenfeld via RT wrote: > >> Ok, thanks for the info. Please keep us posted. The new GIP is on > >> >> hold until I hear otherwise. > > Thanks, Scot. I think this is a problem actually at the BDII/ > > collector > > level, but various folks are en route to the site admin workshop, so > > we > > won't have accurate data to test against for a bit. > Thanks Burt. Let's keep an eye on this so we can release it when we > feel it's ready. > > -alain > > ----------------------------------------------------------------- > Alain Roy vdt-support@opensciencegrid.org > VDT Support http://vdt.cs.wisc.edu/support.html |
||||
| # | Thu Dec 04 12:57:01 2008 | sthapa@ci.uchicago.edu - Correspondence added | [Reply] | |||||||||
Hi Alain, Spoke with Brian and since the logging changes are the only major change, I've confirmed that it's working. It looks good from my side. Suchandra On Dec 4, 2008, at 12:50 PM, Alain Roy via RT wrote: > Hi, > > As best I can tell, the GIP 1.0.5 (1.0.6?) upgrade has been in limbo > for > a few weeks. We're ready to roll with it if people want to release it, > but we haven't gotten the all-clear from you. > > I think the ball is in your court. What should we do? > > Thanks, > -alain > > ----------------------------------------------------------------- > Alain Roy vdt-support@opensciencegrid.org > VDT Support http://vdt.cs.wisc.edu/support.html > > On Thu Nov 13 00:16:22 2008, roy wrote: >> On Nov 12, 2008, at 2:04 PM, burt@fnal.gov via RT wrote: >>>> http://crt.cs.wisc.edu/Ticket/Display.html?id=4094 >>>>> >>> Scot Kronenfeld via RT wrote: >>>> Ok, thanks for the info. Please keep us posted. The new GIP is on >>>>>>> hold until I hear otherwise. >>> Thanks, Scot. I think this is a problem actually at the BDII/ >>> collector >>> level, but various folks are en route to the site admin workshop, so >>> we >>> won't have accurate data to test against for a bit. >> Thanks Burt. Let's keep an eye on this so we can release it when we >> feel it's ready. >> >> -alain >> >> ----------------------------------------------------------------- >> Alain Roy vdt-support@opensciencegrid.org >> VDT Support http://vdt.cs.wisc.edu/support.html > > > > -- > View ticket at <http://crt.cs.wisc.edu/Ticket/Display.html?user=guest&pass=guest&id=4094 > > --- Suchandra Thapa sthapa@ci.uchicago.edu Computation Institute Research Institutes Bldg, Ste 421 5640 S Ellis Avenue Chicago, IL 60637 USA |
||||||||||||
| # | Thu Dec 04 13:19:17 2008 | rjporter@lbl.gov - Correspondence added | [Reply] | |||||||||||
Hi Suchandra, Did you test the install from the gip-test cache? Looking at the new gip validation page, my pbs site is reporting fine but my condor site shows CEInfoTotalCPUs is 0 for the condor job managers. It is reporting correctly to CEMon from that validation page. I was thinking this must be pilot error on my part or some unforseen problem with my site that was rebuilt just prior to this install. However, I see that one of Burt's test sites appears to be showing the same behaviour. Burt have you seen this? I have meant to track this down but so far haven't been able to. thanks, Jeff Suchandra Thapa via RT wrote: > Hi Alain, > > Spoke with Brian and since the logging changes are the only major > change, I've confirmed that it's working. It looks good from my side. > > Suchandra > > On Dec 4, 2008, at 12:50 PM, Alain Roy via RT wrote: > > >> Hi, >>> >> As best I can tell, the GIP 1.0.5 (1.0.6?) upgrade has been in limbo >> for >> a few weeks. We're ready to roll with it if people want to release it, >> but we haven't gotten the all-clear from you. >> >> I think the ball is in your court. What should we do? >> >> Thanks, >> -alain >> >> ----------------------------------------------------------------- >> Alain Roy vdt-support@opensciencegrid.org >> VDT Support http://vdt.cs.wisc.edu/support.html >> >> On Thu Nov 13 00:16:22 2008, roy wrote: >> >>> On Nov 12, 2008, at 2:04 PM, burt@fnal.gov via RT wrote: >>>>> >>>> http://crt.cs.wisc.edu/Ticket/Display.html?id=4094 >>> Thanks Burt. Let's keep an eye on this so we can release it when we>>>> >>>> Scot Kronenfeld via RT wrote: >>>> >>>>> Ok, thanks for the info. Please keep us posted. The new GIP is on >>>> Thanks, Scot. I think this is a problem actually at the BDII/>>>>> hold until I hear otherwise. >>>>> >>>> collector >>>> level, but various folks are en route to the site admin workshop, so >>>> we >>>> won't have accurate data to test against for a bit. >>>> >>> feel it's ready. >>> >>> -alain >>> >>> ----------------------------------------------------------------- >>> Alain Roy vdt-support@opensciencegrid.org >>> VDT Support http://vdt.cs.wisc.edu/support.html >>> >> >> -- >> View ticket at <http://crt.cs.wisc.edu/Ticket/Display.html?user=guest&pass=guest&id=4094 >> > --- > Suchandra Thapa > sthapa@ci.uchicago.edu > Computation Institute > Research Institutes Bldg, Ste 421 > 5640 S Ellis Avenue > Chicago, IL 60637 USA > > > > > |
||||||||||||||
| # | Fri Dec 05 12:00:49 2008 | burt@fnal.gov - Correspondence added | [Reply] | |||||||||||
Alain Roy via RT wrote: > Hi, > > As best I can tell, the GIP 1.0.5 (1.0.6?) upgrade has been in limbo for > a few weeks. We're ready to roll with it if people want to release it, > but we haven't gotten the all-clear from you. > > I think the ball is in your court. What should we do? The ball is still in our court. I want to make sure Suchandra and Jeff's output is ok -- there's some things in Jeff's I don't understand yet. Soon, soon. - B |
||||||||||||||
| # | Tue Dec 09 10:04:50 2008 | burt@fnal.gov - Correspondence added | [Reply] | |||||||||||
Alain Roy via RT wrote: > Hi, > > As best I can tell, the GIP 1.0.5 (1.0.6?) upgrade has been in limbo for > a few weeks. We're ready to roll with it if people want to release it, > but we haven't gotten the all-clear from you. > > I think the ball is in your court. What should we do? Alain, I've just cut the GIP 1.0.6 release (GIP 1.0.5 plus a few fixes that we feel confident don't need wider testing). http://home.fnal.gov/~burt/gip_1_0_6.tgz Thanks for indulging our extra long testing period! I think we'll have a suitable GIP 1.1 release for you shortly. - B |
||||||||||||||
| # | Tue Dec 09 11:58:32 2008 | kronenfe - Correspondence added | [Reply] | |||||||||||
> I've just cut the GIP 1.0.6 release (GIP 1.0.5 plus a few fixes that we feel > confident don't need wider testing). GIP 1.0.6 is now in the VDT's gip test cache. If you already have an installation from that test cache (http://vdt.cs.wisc.edu/test-cache/gip), you should be able to update it using pacman -update Generic-Information-Provider If you have any problems, let me know. -scot |
||||||||||||||
| # | Tue Dec 09 12:08:43 2008 | kronenfe - Correspondence added | [Reply] | |||||||||||
Hi Burt, The configure_gip script that is packaged with GIP 1.0.6 (and earlier) fails if $VDT_LOCATION/globus does not exist (in the initialize subroutine). However, I think the only place in the code outside the initialize subroutine that is looking in $VDT_LOCATION/globus is the fix_for_lsf_queues subroutine. This is a problem because if someone tries to install the Generic--Information-Provider package standalone, without previously installing a globus component , the installation fails because $VDT_LOCATION/globus does not exist and the configuration script errors out. This issue was reported to us by someone trying to install CEMon-Server standalone, which depends on Generic-Information-Provider, but does not depend on any Globus components. You may already have fixed this in the latest GIP. If not, is it possible to fix the configure_gip script to remove this error condition? Thanks, scot |
||||||||||||||
| # | Tue Dec 09 12:23:21 2008 | burt@fnal.gov - Correspondence added | [Reply] | |||||||||||
Scot Kronenfeld via RT wrote: > Hi Burt, > The configure_gip script that is packaged with GIP 1.0.6 (and earlier) > fails if $VDT_LOCATION/globus does not exist (in the initialize > subroutine). However, I think the only place in the code outside the > initialize subroutine that is looking in $VDT_LOCATION/globus is the > fix_for_lsf_queues subroutine. > > This is a problem because if someone tries to install the > Generic--Information-Provider package standalone, without previously > installing a globus component , the installation fails because > $VDT_LOCATION/globus does not exist and the configuration script > errors out. This issue was reported to us by someone trying to > install CEMon-Server standalone, which depends on > Generic-Information-Provider, but does not depend on any Globus > components. > > You may already have fixed this in the latest GIP. If not, is it > possible to fix the configure_gip script to remove this error > condition? Hi Scot, I don't think we had this issue previously reported to us. We'll have a look. - B |
||||||||||||||
| # | Thu Dec 18 11:26:20 2008 | burt@fnal.gov - Correspondence added | [Reply] | |||||||||||
Scot Kronenfeld via RT wrote: > Hi Burt, > The configure_gip script that is packaged with GIP 1.0.6 (and earlier) > fails if $VDT_LOCATION/globus does not exist (in the initialize > subroutine). However, I think the only place in the code outside the > initialize subroutine that is looking in $VDT_LOCATION/globus is the > fix_for_lsf_queues subroutine. > > This is a problem because if someone tries to install the > Generic--Information-Provider package standalone, without previously > installing a globus component , the installation fails because > $VDT_LOCATION/globus does not exist and the configuration script > errors out. This issue was reported to us by someone trying to > install CEMon-Server standalone, which depends on > Generic-Information-Provider, but does not depend on any Globus > components. > > You may already have fixed this in the latest GIP. If not, is it > possible to fix the configure_gip script to remove this error > condition? Scot, I'll push a GIP 1.0.7 with this fixed out to you shortly. I just moved the globus location logic inside the fix_for_lsf_queues subroutine. - B |
||||||||||||||
| # | Thu Dec 18 11:33:26 2008 | burt@fnal.gov - Correspondence added | [Reply] | |||||||||||
Scot Kronenfeld via RT wrote: > Hi Burt, > The configure_gip script that is packaged with GIP 1.0.6 (and earlier) > fails if $VDT_LOCATION/globus does not exist (in the initialize > subroutine). However, I think the only place in the code outside the > initialize subroutine that is looking in $VDT_LOCATION/globus is the > fix_for_lsf_queues subroutine. > > This is a problem because if someone tries to install the > Generic--Information-Provider package standalone, without previously > installing a globus component , the installation fails because > $VDT_LOCATION/globus does not exist and the configuration script > errors out. This issue was reported to us by someone trying to > install CEMon-Server standalone, which depends on > Generic-Information-Provider, but does not depend on any Globus > components. > > You may already have fixed this in the latest GIP. If not, is it > possible to fix the configure_gip script to remove this error > condition? Scot, I've cut GIP 1.0.7 with this fix. http://home.fnal.gov/~burt/gip/gip_1_0_7.tgz We don't need any further testing in the test cache before putting it in the next VDT release (although you may want to ask the user running CEMon-Server standalone to give it a try). - B |
||||||||||||||
| # | Thu Dec 18 11:50:11 2008 | kronenfe - Correspondence added | [Reply] | |||||||||
Ok, we will prepare to release GIP 1.0.7 in the next lettered release to 1.10.1. Thanks Burt! |
||||||||||||
| # | Thu Dec 18 11:55:49 2008 | rjporter@lbl.gov - Correspondence added | [Reply] | |||||||||||
Hi Scot & Burt, Is this different from what is going into the 'last big update of 2008'? I know we have a different test cache for that and I believe full VTB testing is waiting on some items - perhaps from GIP. thanks, Jeff Scot Kronenfeld via RT wrote: > Ok, we will prepare to release GIP 1.0.7 in the next lettered release to 1.10.1. > > Thanks Burt! > > > |
||||||||||||||
| # | Thu Dec 18 11:57:55 2008 | burt@fnal.gov - Correspondence added | [Reply] | |||||||||||
rjporter@lbl.gov via RT wrote: > Hi Scot & Burt, > > Is this different from what is going into the 'last big update of > 2008'? I know we have a different test cache for that and I believe > full VTB testing is waiting on some items - perhaps from GIP. Not appreciably. The 1.0.6 and 1.0.7 have minor changes compared to what we tested on your site with 1.0.5. - b |
||||||||||||||
| # | Thu Dec 18 14:19:39 2008 | kronenfe - Subject changed from 'GIP 1.0.5 release for testing' to 'GIP 1.0.7 ready for release' | ||
| # | Fri Dec 19 12:55:13 2008 | kronenfe - Status changed from 'open' to 'resolved' | ||
| # | Fri Dec 19 13:14:31 2008 | kronenfe - Fixed in 1.10.1r added | ||
Time to display: 6.252436
»|« RT 3.8.2 Copyright 1996-2008 Best Practical Solutions, LLC.
