When the backend checks qstat on pbspro it seems that sometimes a job is
(temporarily) not found. We should add at least a few retries on that:
2012-08-27 16:17:34+0800 [-] ssh attempting remote command: module load
blast&&cat > /dev/null && qstat -x -f '1932269.epic' | sed -ne
'1h;1!H;${;g;s/\n\t//g;p;}'
2012-08-27 16:17:35+0800 [-] 192.168.1.96 - - [27/Aug/2012:16:17:35 +0800] "GET
/fs/copy?priority=<snip> "-" "YabiGevent/0.1"
2012-08-27 16:17:35+0800 [-] Error! We atempted to qstat the job
<1932269.epic>, the call was successful, but we got no data at all. The job
just VANISHED! We are marking this job as errored
2012-08-27 16:17:35+0800 [-] Cannot call RemoteInfo call for job 1932269.epic
Original issue reported on code.google.com by amacgregor on 28 Aug 2012 at 6:44
Original issue reported on code.google.com by
amacgregoron 28 Aug 2012 at 6:44