Am 12.08.2011 um 12:01 schrieb Dave Love: > I wrote: > >> The problem is simply that the shepherd files appear >> to be empty, though I suspect they're not, at the time >> they're read. The rest of the mail is (now) OK. >> I'll debug it eventually... > > I'd forgotten that the hostfile is one of the shepherd files and does > get printed. > > To avoid confusion, here's a sample message from an 8.0.0a installation, > which behaves the same as a 6.2u5-ish one:
I'm still on 6.2u5 and I get more. The shepherd trace is recorded in the spool directory (active_jobs) of the node before. Is this empty there too? > Job 67 caused action: none > User = root > Queue = al...@lv3fn.nw-grid.ac.uk > Start Time = <unknown> > End Time = <unknown> > failed before writing exit_status:shepherd exited with exit status 19: > before writing exit_status > Shepherd trace: > > Shepherd pe_hostfile: > lv3fn.nw-grid.ac.uk 1 al...@lv3fn.nw-grid.ac.uk UNDEFINED > _______________________________________________ > users mailing list > users@gridengine.org > https://gridengine.org/mailman/listinfo/users _______________________________________________ users mailing list users@gridengine.org https://gridengine.org/mailman/listinfo/users
