[Nebula-users] Nebula job starts stopped while troubleshooting
Jonas Stare
jonst at nsc.liu.se
Thu Oct 3 17:20:13 CEST 2019
We found the error and nebula is available again.
A big thank you to Peter W who found it for us before it became a huge problem.
If your job started between 15 and 17 it might be a good idea to check that it ran correctly.
Sorry for any inconvenience.
best regards
Jonas Stare
On 2019-10-03 15:36, Kent Engström wrote:
> Torbjörn Lönnemark <ketl at nsc.liu.se> writes:
>> The downtime has now concluded, and it's possible to queue jobs again.
>
> We have got reports about problems running MPI jobs, and it seems to be
> related to the number of ranks allocated per node.
>
> We have started to troubleshoot this and have stopped new jobs from
> starting until we know more. We will update your later.
>
More information about the Nebula-users
mailing list