[mvapich-discuss] FATAL event 3

[Matthew Koop]

> Do you know what may lead to the following error message?
>
> [1241] Abort: [1241] Got FATAL event 3
>  at line 2451 in file viacheck.c
> [1] Abort: [rhea50:1] Got completion with error, code=10, dest rank=1025
>  at line 382 in file viacheck.c
>
> This error only occurs for large-scale runs with a couple thousand tasks.

Adam,

Does this occur only for one particular application or is this a problem
that shows up in various runs? It's a QP access error, so we'd need some
more information to determine what the problem could be.

What version of MVAPICH is this error showing up for?

Matt