[torqueusers] Signalling on multi node jobs.
Roy.Dragseth at cc.uit.no
Mon Sep 19 14:39:26 MDT 2005
On the mpiexec list we have been discussing how to get suspend/resume work
with mpiexec. I thought that if you send a signal using qsig or whatever it
gets forwarded to all nodes in a job, but that does not seem to be the case.
Only the mother superior receives the signal, is this the intended behaviour?
This has implications on how mpiexec is supposed to handle signals. If the
above behaviour is the right one then it must forward the signals. If not,
then it can ignore them as they will be sent to all nodes anyway.
The Computer Center, University of Tromsø, N-9037 TROMSØ, Norway.
phone:+47 77 64 41 07, fax:+47 77 64 41 00
Roy Dragseth, High Performance Computing System Administrator
Direct call: +47 77 64 62 56. email: royd at cc.uit.no
More information about the torqueusers