[torquedev] [Bug 98] Allocation of incorrect pointer in src/scheduler.cc/samples/fifo/job_info.c: update_job_comment causes random crash.

bugzilla-daemon at supercluster.org bugzilla-daemon at supercluster.org
Wed Nov 10 02:29:29 MST 2010


--- Comment #7 from Simon Toth <SimonT at mail.muni.cz> 2010-11-10 02:29:28 MST ---
(In reply to comment #5)
> The empirical evidence is that pointer to writable memory is required by
> pbs_alterjob() as without my patch pbs_sched would run for a few minutes before
> dying with a SEGV in sprintf(). (This is with around 2500 jobs in the queue by
> the way.) With this patch it's rock solid.
> Simon, agreed, strdup() is probably the way to go.

This is weird. Our scheduler is based of FIFO (and the specified code is
actually identical). Still it doesn't crash (and we did extensive stress
testing - hundreds of tests, each with 5000 jobs).

Could you post the stack trace of one of the crashes?

Configure bugmail: http://www.clusterresources.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

More information about the torquedev mailing list