On Sat, Mar 12, 2011 at 1:02 PM, Rayson Ho <[email protected]> wrote: > On Sat, Mar 12, 2011 at 8:11 AM, Lane Schwartz <[email protected]> wrote: >>>> I do know that more scheduling info can be printed out. There's >>>> currently a job waiting in the queue that asked for mem_free=40960M, > > Why 40960M?? You want 40G for that job?? Is that a typo??
No, not a typo. We've got some big jobs. :) >>> and one source of Rayson's patch for it that I have to hand is >>> https://arc.liv.ac.uk/trac/SGE/changeset/3551/sge >> >> I'm definitely running 64-bit hosts. Do you know if the bug you're >> referring to is fixed in the current open source release? > > The problem you are encountering is related to scheduling, but the fix > I have is a workaround for accounting. I don't believe they are > related. > > And it is in Open Grid Scheduler (OGS) since last year, and was > included in "OGS 6.2 update 5 patch 1" that was released last month: > > http://gridscheduler.sourceforge.net/ OK, thanks for the info. I do have a testbed grid running 6.2u5p1. The main grid, though, where I'm encountering this issue is running 6.2u6. If I had reason to believe that the problem would go away if we switched from 6.2u6 to 6.2u5p1, we could probably do so. But if not then we'll probably stay with 6.2u6 for the moment. Cheers, Lane _______________________________________________ users mailing list [email protected] https://gridengine.org/mailman/listinfo/users
