[Snic-users] Triolith and Gamma: Running jobs lost

[Mats Kronberg]

Update: Triolith and Gamma are now available.

As far as we can tell, no data has been damaged or lost. Most jobs
that were running on Triolith and Gamma at 07:30 this morning failed.

It's clear what happened but not why. The vendor is still
investigating. We have applied a software update that might help,
that's about all we can do for now.

If you have any questions regarding this outage, please contact
support at nsc.liu.se.

FYI: I'm not at all happy with the number of outages caused by the
storage system recently. We're doing what we can to stabilize things,
but almost every time there has been an outage the cause have been a
different one. E.g the August 12th outage was caused by a server
hardware failure, and today it was a previously unseen (at least by
NSC) software problem.


-- 
Mats Kronberg, NSC Support <support at nsc.liu.se>