[mpich-discuss] Assertion failed

Darius Buntinas buntinas at mcs.anl.gov
Fri Dec 5 17:44:08 CST 2008

I guess what I'm asking is whether the execution of one mpi program
affects the execution of the second.  It seemed like you said that when
you run the java program you eventually hit the assertion after many
messages, and then if you run another program after that, you hit the
assertion immediately.  Am I understanding you correctly?


On 12/05/2008 05:16 PM, Xavier Olive wrote:
> Le 6 déc. 08 à 01:50, Darius Buntinas a écrit :
>> Hmm, are saying that if you run one mpi program that does a lot of sends
>> and hits the assert, then run a second simple program that, e.g., just
>> does a send and receive, that the second program fails too?
>> If the failure persists between executions of two separate mpi jobs,
>> that may indicate a problem with the network (card/driver/etc).
>> Are you using a special network (e.g., InfiniBand, Myrinet)?
> The first test was in local, on my desktop computer.
> The second test was on a supercomputer w/ Infiniband.
> I will try on Monday to recompile with nemesis option
> Thanks

More information about the mpich-discuss mailing list