It looks like we've got some kind of mismatch on our cluster, similar to 
the recent CG problem report?

Nothing is reported by slurmctld; but following any job, the errant 
compute nodes report:

[2013-07-23T17:10:17.148] error: Malformed RPC of type 5016 received
[2013-07-23T17:10:17.148] error: slurm_receive_msg_and_forward: Header 
lengths are longer than data received
[2013-07-23T17:10:17.152] error: service_connection: slurm_receive_msg: 
Header lengths are longer than data received
[2013-07-23T17:10:17.158] error: service_connection: slurm_receive_msg: 
Header lengths are longer than data received
[2013-07-23T17:11:13.131] [294.0] Abandoning IO 60 secs after job 
shutdown initiated

(Lots of the "Malformed RPC 5016" messages appear per job.)

Any thoughts?
Andy

-- 
Andy Riebs
Hewlett-Packard Company
High Performance Computing
+1 404 648 9024
My opinions are not necessarily those of HP

Reply via email to