Hi all,<br><br>TORQUE 3.0.5 is now available for general use.<br><br>There were no new features added to TORQUE 3.0.5. Below is the content of the CHANGELOG for 3.0.5.<br><br>TORQUE 3.0.x is recommended for sites using NUMA systems similar to SGI's Ultra-violet. The same functionality is now also available in TORQUE 4.0.1 which we recommend for everyone.<br>
<br>Regards<br><br>Ken Nielson<br>Adaptive Computing<br><br>3.0.5 Change log<br><br> b - fix for writing too much data when job_script is saved to job log.<br> b - fix for where pbs_mom would not automatically set gpu mode.<br>
b - fix for alligning qstat -r output when configured with -DTXT.<br> e - Change size of transfer block used on job rerun from 4k to 64k.<br> b - With nvidia gpus, TORQUE was losing the directive of what nodes it should<br>
run the job on from Moab. Corrected.<br> e - add the $PBS_WALLTIME variable to jobs, thanks to a patch from Mark Roberts<br> n - change moab_array_compatible server parameter so it defaults to true<br> b - fix so comma separated file list can be used with qsub -W stagein/stageout.<br>
Matches qsub documentation again. (backported from 4.0.1)<br> e - change to allow pbs_mom to run if configured with --enable-nvidia-gpus but<br> installed on a node without Nvidia gpus.<br> b - fix for where pbsnodes displays outdated gpu_status information.<br>
b - Fixed a problem where starting pbs_mom with the default -p option would delete<br> running jobs instead of trying to recover them if cpusets were enabled. TRQ-828<br> Fixes Bugzilla 174.<br> b - Added a check for a NULL job pointer in modify_job. This fixes TRQ-818<br>
b - fix problem with '+ and segfault when using multiple node gpu requests.<br> (backported from 4.0.1)<br> b - fix problem with mom segfault when using 8 or more gpus on mom node.<br> (backported from 4.0.1)<br>
b - Fixed a bug for NUMA cpusets where Nodes were being underallocated for the job.<br> TRQ-845<br> b - Fixed a problem with cpusets where if a boot cpuset is present under /dev/cpuset<br> torque would only write out 1024 cpus. This was changed from a hardcoded<br>
value to be the maximum size found in /dev/cpuset/cpus. TRQ-840<br> b - Fix so child pbs_mom does not remain running after qdel on slow starting job.<br> TRQ-860. (backported from 4.0.1)<br> b - Fix so pbs_mom won't segfault after a qdel is done for a job that is still<br>
running the prologue. TRQ-832. (backported from 4.0.1)<br> b - Fixed yet another case where procct would get promoted to Moab causing jobs<br> to never be run. Added remove_procct to ROUTE_PERM_FAILURE case in<br>
default_router.<br> b - Fix so qrun jobid[] does not cause pbs_server segfault. TRQ-865. (backported from 4.0.2)<br><br><br>