RT for crt.cs.wisc.edu
RT for crt.cs.wisc.edu
configure_gip.tar.gz
osg-info-dynamic-pbs.tar.gz
Comments about this user:
No comment entered about this user This user's 10 highest priority tickets:
|
|
| # | Mon Sep 10 14:30:35 2007 | timm@fnal.gov - Ticket created | [Reply] | |||||||||
There is a bug in the configure_gip script, both in the version as shipped with vdt-1.6.1 and vdt-1.8.0, and presumably it has always been there. The problem happens when you have a user ACL on a queue, thus: (here's output from qstat -Q -f) Queue: d0farm queue_type = Execution max_queuable = 1000 total_jobs = 1 state_count = Transit:0 Queued:0 Held:0 Waiting:0 Running:1 Exiting:0 max_running = 800 acl_host_enable = True acl_hosts = d0cabosg2.fnal.gov acl_user_enable = True acl_users = dzeroana,dzeropro,samgrid resources_max.mem = 950mb resources_max.pmem = 950mb resources_assigned.mem = 0b resources_assigned.nodect = 1 enabled = True started = True ------------- These three users (dzeropro, samgrid, and dzeroana) are all part of the dzero VO. What the configurator should do is make a GlueCEAccessBaseControlRule of VO:dzero, by means of the osg-user-vo-map.txt. What it does, instead is to make GlueCEAccessControlBaseRule: VO:dzeroana GlueCEAccessControlBaseRule: VO:dzeropro GlueCEAccessControlBaseRule: VO:samgrid These user id's are not VO's and can't be used as such in ReSS or BDII or anything. We need to get this fixed. (This also exposes the meta-question of what happens if a site wants to support only certain groups or certain roles of a VO, but I digress.) Thanks Steve Timm -- ------------------------------------------------------------------ Steven C. Timm, Ph.D (630) 840-8525 timm@fnal.gov http://home.fnal.gov/~timm/ Fermilab Computing Division, Scientific Computing Facilities, Grid Facilities Department, FermiGrid Services Group, Assistant Group Leader. |
||||||||||||
| # | Tue Sep 18 10:48:18 2007 | roy - Taken | ||
| # | Tue Sep 18 10:48:45 2007 | roy - Cc law@cac.cornell.edu added | ||
| # | Tue Sep 18 10:58:47 2007 | roy - Correspondence added | [Reply] | |
|
Sorry for the slow response. I've added Will Law to the ticket, so he can help, because he is the current maintainer of configure_gip. Will--I do'nt understand the underlying problem here. Do you understand it, and how to fix it? Thanks, -alain > There is a bug in the configure_gip script, both in the version > as shipped with vdt-1.6.1 and vdt-1.8.0, and presumably > it has always been there. > > The problem happens when you have a user ACL on a queue, thus: > (here's output from qstat -Q -f) > > > Queue: d0farm > queue_type = Execution > max_queuable = 1000 > total_jobs = 1 > state_count = Transit:0 Queued:0 Held:0 Waiting:0 Running:1 Exiting:0 > max_running = 800 > acl_host_enable = True > acl_hosts = d0cabosg2.fnal.gov > acl_user_enable = True > acl_users = dzeroana,dzeropro,samgrid > resources_max.mem = 950mb > resources_max.pmem = 950mb > resources_assigned.mem = 0b > resources_assigned.nodect = 1 > enabled = True > started = True > > > ------------- > These three users (dzeropro, samgrid, and dzeroana) are > all part of the dzero VO. What the configurator should do is > make a GlueCEAccessBaseControlRule of VO:dzero, by means of > the osg-user-vo-map.txt. > > What it does, instead is to make > > GlueCEAccessControlBaseRule: VO:dzeroana > GlueCEAccessControlBaseRule: VO:dzeropro > GlueCEAccessControlBaseRule: VO:samgrid > > > These user id's are not VO's and can't be used as such in ReSS or BDII > or anything. We need to get this fixed. > > (This also exposes the meta-question of what happens if a site wants to > support only certain groups or certain roles of a VO, but I digress.) > > Thanks > > Steve Timm > > > > > |
||||
| # | Tue Sep 18 10:58:48 2007 | RT_System - Status changed from 'new' to 'open' | ||
| # | Tue Sep 18 14:23:24 2007 | law@cac.cornell.edu - Correspondence added | [Reply] | |||||||||||
Hi Alain - I am out of the office right now because my Mom became extremely ill suddenly on Friday. I'm not completely sure when I will be back. Ruth Pordes and Rob Gardner are working on getting some sort of help for the moment - I am not sure about the status. I am not sure about the source of the issue exactly right now, I'll admit I am a bit rattled and exhausted. Will On 9/18/07, Alain Roy via RT <vdt-support@opensciencegrid.org> wrote: > > Sorry for the slow response. > > I've added Will Law to the ticket, so he can help, because he is the > current maintainer of configure_gip. > > Will--I do'nt understand the underlying problem here. Do you understand > it, and how to fix it? > > Thanks, > -alain > > > There is a bug in the configure_gip script, both in the version > Exiting:0> > as shipped with vdt-1.6.1 and vdt-1.8.0, and presumably > > it has always been there. > > > > The problem happens when you have a user ACL on a queue, thus: > > (here's output from qstat -Q -f) > > > > > > Queue: d0farm > > queue_type = Execution > > max_queuable = 1000 > > total_jobs = 1 > > state_count = Transit:0 Queued:0 Held:0 Waiting:0 Running:1 > > max_running = 800 >> > acl_host_enable = True > > acl_hosts = d0cabosg2.fnal.gov > > acl_user_enable = True > > acl_users = dzeroana,dzeropro,samgrid > > resources_max.mem = 950mb > > resources_max.pmem = 950mb > > resources_assigned.mem = 0b > > resources_assigned.nodect = 1 > > enabled = True > > started = True > > > > > > ------------- > > These three users (dzeropro, samgrid, and dzeroana) are > > all part of the dzero VO. What the configurator should do is > > make a GlueCEAccessBaseControlRule of VO:dzero, by means of > > the osg-user-vo-map.txt. > > > > What it does, instead is to make > > > > GlueCEAccessControlBaseRule: VO:dzeroana > > GlueCEAccessControlBaseRule: VO:dzeropro > > GlueCEAccessControlBaseRule: VO:samgrid > > > > > > These user id's are not VO's and can't be used as such in ReSS or BDII > > or anything. We need to get this fixed. > > > > (This also exposes the meta-question of what happens if a site wants to > > support only certain groups or certain roles of a VO, but I digress.) > > > > Thanks > > > > Steve Timm > > > > > > > > > > > > -- > View ticket at < > http://vdt.cs.wisc.edu/rt/Ticket/Display.html?user=guest&pass=guest&id=2959 > > > VDT Support, vdt-support@ivdgl.org> |
||||||||||||||
| # | Tue Sep 18 14:33:38 2007 | timm@fnal.gov - Correspondence added | [Reply] | |||||||||||
The source of the issue is the following: Queue X in PBS is declared to be accessible to only three users: The GIP advertises it as being accessible to three VO's with the names of those users, even though those user names do not correspond to any VO name that is known or unknown. The GIP should parse the user ACL and look up the user-to-vo correspondence in osg-user-vo-map.txt. In this case, it would have found out that those three users all correspond to DZERO VO and advertised it accordingly. This particular section, PBS configuration--I seem to find a new bug in the logic every time I try to do something new. Former maintainers of the code didn't have much expertise in PBS nor did they have a complicated PBS cluster with which to test the code. Steve Timm On Tue, 18 Sep 2007, William Law wrote: > Hi Alain - > > I am out of the office right now because my Mom became extremely ill > suddenly on Friday. I'm not completely sure when I will be back. Ruth > Pordes and Rob Gardner are working on getting some sort of help for the > moment - I am not sure about the status. > > I am not sure about the source of the issue exactly right now, I'll admit I > am a bit rattled and exhausted. > > Will > > On 9/18/07, Alain Roy via RT <vdt-support@opensciencegrid.org> wrote: >> >>> Sorry for the slow response. >> >> I've added Will Law to the ticket, so he can help, because he is the >> current maintainer of configure_gip. >> >> Will--I do'nt understand the underlying problem here. Do you understand >> it, and how to fix it? >> >> Thanks, >> -alain >> >>> There is a bug in the configure_gip script, both in the version >> Exiting:0>>> as shipped with vdt-1.6.1 and vdt-1.8.0, and presumably >>> it has always been there. >>> >>> The problem happens when you have a user ACL on a queue, thus: >>> (here's output from qstat -Q -f) >>> >>> >>> Queue: d0farm >>> queue_type = Execution >>> max_queuable = 1000 >>> total_jobs = 1 >>> state_count = Transit:0 Queued:0 Held:0 Waiting:0 Running:1 >>> max_running = 800 >>>>> acl_host_enable = True >>> acl_hosts = d0cabosg2.fnal.gov >>> acl_user_enable = True >>> acl_users = dzeroana,dzeropro,samgrid >>> resources_max.mem = 950mb >>> resources_max.pmem = 950mb >>> resources_assigned.mem = 0b >>> resources_assigned.nodect = 1 >>> enabled = True >>> started = True >>> >>> >>> ------------- >>> These three users (dzeropro, samgrid, and dzeroana) are >>> all part of the dzero VO. What the configurator should do is >>> make a GlueCEAccessBaseControlRule of VO:dzero, by means of >>> the osg-user-vo-map.txt. >>> >>> What it does, instead is to make >>> >>> GlueCEAccessControlBaseRule: VO:dzeroana >>> GlueCEAccessControlBaseRule: VO:dzeropro >>> GlueCEAccessControlBaseRule: VO:samgrid >>> >>> >>> These user id's are not VO's and can't be used as such in ReSS or BDII >>> or anything. We need to get this fixed. >>> >>> (This also exposes the meta-question of what happens if a site wants to >>> support only certain groups or certain roles of a VO, but I digress.) >>> >>> Thanks >>> >>> Steve Timm >>> >>> >>> >>> >>> >> >> -- >> View ticket at < >> http://vdt.cs.wisc.edu/rt/Ticket/Display.html?user=guest&pass=guest&id=2959 >>> >> VDT Support, vdt-support@ivdgl.org>> -- ------------------------------------------------------------------ Steven C. Timm, Ph.D (630) 840-8525 timm@fnal.gov http://home.fnal.gov/~timm/ Fermilab Computing Division, Scientific Computing Facilities, Grid Facilities Department, FermiGrid Services Group, Assistant Group Leader. |
||||||||||||||
| # | Tue Sep 18 15:33:25 2007 | roy - Correspondence added | [Reply] | |||||||||
> I am out of the office right now because my Mom became extremely ill > suddenly on Friday. I'm not completely sure when I will be back. > Ruth > Pordes and Rob Gardner are working on getting some sort of help > for the > moment - I am not sure about the status. Will-- Family health issues are WAY more important than the VDT. Don't think about this issue any more. If I can't solve it, we'll solve it with your help in a future update after VDT 1.8.1 is released. I'm truly sorry to hear about your mother's illness. I hope she recovers--best of luck! -alain |
||||||||||||
| # | Tue Sep 18 15:33:25 2007 | roy - Correspondence added | [Reply] | |||||||||
> The source of the issue is the following: > Queue X in PBS is declared to be accessible to only three users: > The GIP advertises it as being accessible to three VO's with the names > of those users, even though those user names do not correspond to any > VO name that is known or unknown. > The GIP should parse the user ACL and look up the user-to-vo > correspondence in osg-user-vo-map.txt. In this case, it would have > found > out that those three users all correspond to DZERO VO and > advertised it > accordingly. > > This particular section, PBS configuration--I seem to find a new bug > in the logic every time I try to do something new. Former > maintainers of > the code didn't have much expertise in PBS nor did they have a > complicated > PBS cluster with which to test the code. Steve-- I also have no experience with the PBS, nor this code in particular. Do you have the ability and time to look at it and suggest a change that we can put into the VDT? Thanks, -alain |
||||||||||||
| # | Thu Oct 11 14:56:06 2007 | roy - Priority changed from (no value) to '3' | ||
| # | Thu Oct 11 14:56:07 2007 | roy - Fix scheduled CUR added | ||
| # | Thu Oct 11 14:56:52 2007 | roy - Cc anand.iowa@gmail.com added | ||
| # | Thu Oct 11 14:57:59 2007 | roy - Correspondence added | [Reply] | |
|
Hi Anand, Are you able to help with this VDT ticket that is about the GIP? Will Law is unable to help right now due to personal issues. You can see the full text of the VDT ticket here: http://vdt.cs.wisc.edu/rt/index.html?user=guest&pass=guest&q=2959 Thanks, -alain |
||||
| # | Thu Oct 11 15:36:39 2007 | apadmana-old - Correspondence added | [Reply] | |||||||||||
Hi Alain, I will be happy to help address this ticket. I will let you know as soon as I am able to fix it. Thanks Anand Alain Roy via RT wrote: > Hi Anand, > > Are you able to help with this VDT ticket that is about the GIP? Will > Law is unable to help right now due to personal issues. > > You can see the full text of the VDT ticket here: > http://vdt.cs.wisc.edu/rt/index.html?user=guest&pass=guest&q=2959 > > Thanks, > -alain > > > |
||||||||||||||
| # | Thu Oct 11 16:37:56 2007 | roy - Correspondence added | [Reply] | |||||||||
Great, thanks Anand! Steve--when we have a fix, it will go into the next update of VDT 1.8.1. -alain |
||||||||||||
| # | Fri Oct 12 16:48:39 2007 | apadmana-old - Correspondence added | [Reply] | |||||||||||
Hi Steve/Alain, I did find the source of the problem and updated it, but I don't have a complicated environment like FNAL to test it fully. So Steve if you could test it and let me know if this works, we can finalize the changes. Anand Alain Roy via RT wrote: > Great, thanks Anand! > > Steve--when we have a fix, it will go into the next update of VDT 1.8.1. > > -alain > > > Message body not shown because it is too large or is not plain text. |
||||||||||||||
| # | Sun Oct 14 20:36:06 2007 | roy - Correspondence added | [Reply] | |||||||||
Steve-- Let us know how your testing goes. If it is successful, we'll include it in an update to VDT 1.8.1. Thanks, -alain On Oct 12, 2007, at 4:48 PM, Anand Padmanabhan via RT wrote: > http://vdt.cs.wisc.edu/rt/Ticket/Display.html?id=2959 > > Hi Steve/Alain, > > I did find the source of the problem and updated it, but I don't > have a > complicated environment like FNAL to test it fully. So Steve if you > could test it and let me know if this works, we can finalize the > changes. > > Anand > > Alain Roy via RT wrote: >> Great, thanks Anand! >>> >> Steve--when we have a fix, it will go into the next update of VDT >> 1.8.1. >> >> -alain >> >> >> > <configure_gip.tar.gz> |
||||||||||||
| # | Mon Oct 15 10:39:00 2007 | timm@fnal.gov - Correspondence added | [Reply] | |||||||||||
Anand and all-- I have looked at the new configure_gip and it appears that the new code will do the right thing. Unfortunately this appears to be a glue schema v1.3 gip and the only PBS system of this complexity I have to test it out against is currently vdt-1.6.1/OSG 0.6.0 glue 1.2 based. So I may not be able to test it out right away. What I could do is to simply patch the code from the new configure_gip into the queue and acl parsing section of the old one and see if that particular problem is solved. I will let you know when this is done. Steve Timm On Fri, 12 Oct 2007, Anand Padmanabhan wrote: > Hi Steve/Alain, > > I did find the source of the problem and updated it, but I don't have a > complicated environment like FNAL to test it fully. So Steve if you could > test it and let me know if this works, we can finalize the changes. > > Anand > > Alain Roy via RT wrote: >> Great, thanks Anand! >>> >> Steve--when we have a fix, it will go into the next update of VDT 1.8.1. >> >> -alain >> >> >> > -- ------------------------------------------------------------------ Steven C. Timm, Ph.D (630) 840-8525 timm@fnal.gov http://home.fnal.gov/~timm/ Fermilab Computing Division, Scientific Computing Facilities, Grid Facilities Department, FermiGrid Services Group, Assistant Group Leader. |
||||||||||||||
| # | Mon Oct 15 11:04:05 2007 | timm@fnal.gov - Correspondence added | [Reply] | |||||||||||
I patched the modified subroutine add_to_pbs_queues into my production configure_gip vdt 1.6.1, along with the patched call of the extra parameter, and it now reports the correct GlueCEAccessControlBaseRule I'm confident this will do the fix in vdt 1.8.1 as well. Thanks for the fix. Steve Timm On Mon, 15 Oct 2007, Steven Timm wrote: > Anand and all-- > I have looked at the new configure_gip and it appears that the > new code will do the right thing. Unfortunately this appears > to be a glue schema v1.3 gip and the only PBS system of this > complexity I have to test it out against is currently vdt-1.6.1/OSG 0.6.0 > glue 1.2 based. So I may not be able to test it out right away. > What I could do is to simply patch the code from the new configure_gip > into the queue and acl parsing section of the old one and see if that > particular problem is solved. I will let you know when this is done. > > Steve Timm > > > On Fri, 12 Oct 2007, Anand Padmanabhan wrote: > >> Hi Steve/Alain, >>> >> I did find the source of the problem and updated it, but I don't have a >> complicated environment like FNAL to test it fully. So Steve if you could >> test it and let me know if this works, we can finalize the changes. >> >> Anand >> >> Alain Roy via RT wrote: >>> Great, thanks Anand! >> >>> >>> Steve--when we have a fix, it will go into the next update of VDT 1.8.1. >>> >>> -alain >>> >>> >>> >> > -- ------------------------------------------------------------------ Steven C. Timm, Ph.D (630) 840-8525 timm@fnal.gov http://home.fnal.gov/~timm/ Fermilab Computing Division, Scientific Computing Facilities, Grid Facilities Department, FermiGrid Services Group, Assistant Group Leader. |
||||||||||||||
| # | Mon Oct 15 11:12:48 2007 | apadmana-old - Correspondence added | [Reply] | |||||||||||
Steve, The patch should be easily portable to GIP in vdt-1.6.1. All you would need to do is: 1. Replace the add_to_pbs_queues() function with the one from new configure_gip 2. Add an extra parameter when this function is called (in fix_for_pbs_queues()): Line ~713: change add_to_pbs_queues(\@pbs_groups,\@tmp_groups); to add_to_pbs_queues(\@pbs_groups,\@tmp_groups,0); Line ~716: change add_to_pbs_queues(\@pbs_users,\@tmp_groups); to add_to_pbs_queues(\@pbs_users,\@tmp_groups,1); Let me know if you run into problems or have any questions. Thanks, Anand Steven Timm wrote: > Anand and all-- > I have looked at the new configure_gip and it appears that the > new code will do the right thing. Unfortunately this appears > to be a glue schema v1.3 gip and the only PBS system of this > complexity I have to test it out against is currently vdt-1.6.1/OSG 0.6.0 > glue 1.2 based. So I may not be able to test it out right away. > What I could do is to simply patch the code from the new configure_gip > into the queue and acl parsing section of the old one and see if that > particular problem is solved. I will let you know when this is done. > > Steve Timm > > > On Fri, 12 Oct 2007, Anand Padmanabhan wrote: > >> Hi Steve/Alain, > >> >> I did find the source of the problem and updated it, but I don't have >> a complicated environment like FNAL to test it fully. So Steve if you >> could test it and let me know if this works, we can finalize the changes. >> >> Anand >> >> Alain Roy via RT wrote: >>> Great, thanks Anand! >>>>> >>> Steve--when we have a fix, it will go into the next update of VDT 1.8.1. >>> >>> -alain >>> >>> >>> >> |
||||||||||||||
| # | Mon Oct 15 11:32:49 2007 | timm@fnal.gov - Correspondence added | [Reply] | |||||||||||
The steps below are exactly what I did--just before I got this E-mail. Thanks Steve Timm ------------------------------------------------------------------ Steven C. Timm, Ph.D (630) 840-8525 timm@fnal.gov http://home.fnal.gov/~timm/ Fermilab Computing Division, Scientific Computing Facilities, Grid Facilities Department, FermiGrid Services Group, Assistant Group Leader. On Mon, 15 Oct 2007, Anand Padmanabhan via RT wrote: > Steve, > > The patch should be easily portable to GIP in vdt-1.6.1. All you would > need to do is: > 1. Replace the add_to_pbs_queues() function with the one from new > configure_gip > 2. Add an extra parameter when this function is called (in > fix_for_pbs_queues()): > Line ~713: change add_to_pbs_queues(\@pbs_groups,\@tmp_groups); to > add_to_pbs_queues(\@pbs_groups,\@tmp_groups,0); > Line ~716: change add_to_pbs_queues(\@pbs_users,\@tmp_groups); to > add_to_pbs_queues(\@pbs_users,\@tmp_groups,1); > > Let me know if you run into problems or have any questions. > > Thanks, > Anand > Steven Timm wrote: >> Anand and all-- >>> I have looked at the new configure_gip and it appears that the >> new code will do the right thing. Unfortunately this appears >> to be a glue schema v1.3 gip and the only PBS system of this >> complexity I have to test it out against is currently vdt-1.6.1/OSG 0.6.0 >> glue 1.2 based. So I may not be able to test it out right away. >> What I could do is to simply patch the code from the new configure_gip >> into the queue and acl parsing section of the old one and see if that >> particular problem is solved. I will let you know when this is done. >> >> Steve Timm >> >> >> On Fri, 12 Oct 2007, Anand Padmanabhan wrote: >> >>> Hi Steve/Alain, >>>>> >>> I did find the source of the problem and updated it, but I don't have >>> a complicated environment like FNAL to test it fully. So Steve if you >>> could test it and let me know if this works, we can finalize the changes. >>> >>> Anand >>> >>> Alain Roy via RT wrote: >>>> Great, thanks Anand! >>>>>>> >>>> Steve--when we have a fix, it will go into the next update of VDT 1.8.1. >>>> >>>> -alain >>>> >>>> >>>> >>> > > -- > View ticket at <http://vdt.cs.wisc.edu/rt/Ticket/Display.html?user=guest&pass=guest&id=2959> > VDT Support, vdt-support@ivdgl.org > |
||||||||||||||
| # | Mon Oct 15 12:17:51 2007 | roy - Correspondence added | [Reply] | |||||||||
So is the conclusion that I should take the new configure_gip as-is from Anand and put it into VDT 1.8.1b? Thanks, -alain |
||||||||||||
| # | Mon Oct 15 12:17:55 2007 | apadmana-old - Correspondence added | [Reply] | |||||||||||
Thanks Steve! Anand Steven Timm wrote: > I patched the modified subroutine add_to_pbs_queues > into my production configure_gip vdt 1.6.1, along with the patched > call of the extra parameter, and it now reports the correct > GlueCEAccessControlBaseRule > I'm confident this will do the fix in vdt 1.8.1 as well. > Thanks for the fix. > > Steve Timm > > > On Mon, 15 Oct 2007, Steven Timm wrote: > >> Anand and all-- > >> I have looked at the new configure_gip and it appears that the >> new code will do the right thing. Unfortunately this appears >> to be a glue schema v1.3 gip and the only PBS system of this >> complexity I have to test it out against is currently vdt-1.6.1/OSG 0.6.0 >> glue 1.2 based. So I may not be able to test it out right away. >> What I could do is to simply patch the code from the new configure_gip >> into the queue and acl parsing section of the old one and see if that >> particular problem is solved. I will let you know when this is done. >> >> Steve Timm >> >> >> On Fri, 12 Oct 2007, Anand Padmanabhan wrote: >> >>> Hi Steve/Alain, >>>>> >>> I did find the source of the problem and updated it, but I don't have >>> a complicated environment like FNAL to test it fully. So Steve if you >>> could test it and let me know if this works, we can finalize the >>> changes. >>> >>> Anand >>> >>> Alain Roy via RT wrote: >>>> Great, thanks Anand! >>>>>>> >>>> Steve--when we have a fix, it will go into the next update of VDT >>>> 1.8.1. >>>> >>>> -alain >>>> >>>> >>>> >>> >> |
||||||||||||||
| # | Mon Oct 15 12:23:59 2007 | timm@fnal.gov - Correspondence added | [Reply] | |||||||||||
That is certainly my recommendation, yes. Steve On Mon, 15 Oct 2007, Alain Roy via RT wrote: > So is the conclusion that I should take the new configure_gip as-is > from Anand and put it into VDT 1.8.1b? > > Thanks, > -alain > > > -- ------------------------------------------------------------------ Steven C. Timm, Ph.D (630) 840-8525 timm@fnal.gov http://home.fnal.gov/~timm/ Fermilab Computing Division, Scientific Computing Facilities, Grid Facilities Department, FermiGrid Services Group, Assistant Group Leader. |
||||||||||||||
| # | Mon Oct 15 12:46:51 2007 | roy - Comments added | [Reply] | |||||||
Commit comment: Updated from Anand to fix bug #2959. Changed files: U vdt/branches/vdt-1.8.1/Configure-GIP/vdt/setup/configure_gip To generate a diff: svn diff -c 6761 file:///p/vdt/workspace/svn |
||||||||||
| # | Mon Oct 15 12:59:00 2007 | apadmana-old - Correspondence added | [Reply] | |||||||||||
Yes. Also Alain could you update dynamic pbs (as attached, goes into lcg/libexec). I found it was missing a newline which was causing GlueCEPolicyAssignedJobSlots to repeated twice under VOView that might cause issue with interoperability. Thanks, Anand Steven Timm wrote: > > That is certainly my recommendation, yes. > > Steve > > > On Mon, 15 Oct 2007, Alain Roy via RT wrote: > >> So is the conclusion that I should take the new configure_gip as-is >>> from Anand and put it into VDT 1.8.1b? >> >> Thanks, >> -alain >> >> >> Message body not shown because it is too large or is not plain text. |
||||||||||||||
| # | Mon Oct 15 13:12:11 2007 | roy - Comments added | [Reply] | |||||||
Commit comment: Update from Anand: add a newline that was missing. Changed files: U vdt/branches/vdt-1.8.1/Generic-Information-Provider/lcg/libexec/osg-info-dynamic-pbs To generate a diff: svn diff -c 6762 file:///p/vdt/workspace/svn |
||||||||||
| # | Mon Oct 15 13:20:56 2007 | roy - Correspondence added | [Reply] | |
|
Thanks Anand and Steve! I've taken both updates (one to configure_gip and one to osg-info-dynamic-pbs), and they will be in VDT 1.8.1b. I'm going to resolve this ticket now, but feel free to send more email if further problems are discovered. Thanks, -alain |
||||
| # | Mon Oct 15 13:21:08 2007 | roy - Status changed from 'open' to 'resolved' | ||
Time to display: 3.411071
»|« RT 3.8.2 Copyright 1996-2008 Best Practical Solutions, LLC.
