[Beowulf] mpich2 complain about nodes that i dont use

Mark Hahn hahn at physics.mcmaster.ca
Fri Sep 30 19:53:04 PDT 2005


> I indeed was surprised when I saw that (signal 9) with one of our codes
> as well. In that case it turned out to be code that needed a larger
> stacksize than was permitted under the current settings (ulimit, etc.).

right you are!  I expected memory-limit problems to be signaled by SIGBUS
or perhaps SEGV, but plain as day, the OOM killer in the kernel uses KILL.
I guess the reason is that only STOP and KILL cannot be caught/ignored.

thanks!
-mark




More information about the Beowulf mailing list