[Bioclusters] mpiBLAST memory leak?

Iddo Friedberg idoerg at gmail.com
Thu Nov 15 13:32:58 EST 2007

I am running mpiBLAST 1.4.0 using mpich and SGE on a Rocks (4.2.1) cluster.
I have been getting what seems to be consistent memory leaks that crashed
nodes after about 5-6 hours of running. Memory consumption steadily
increases until the node from which the job is running crashes.

I tried setting P4_GLOBMEMSIZE to under sysctl.shmmax. Still no good.
/var/log or the SGE logfiles do not contain information. Any ideas?




I. Friedberg

"The only problem with troubleshooting is that
sometimes trouble shoots back."

More information about the Bioclusters mailing list