Upd:Problems with Infiniband and GPFS
LRZ aktuell
publish at lrz.de
Sa Sep 29 15:05:07 CEST 2012
Changes to this message: GPFS unavailable again
2nd Update
Unfortunately, during Saturday morning GPFS has become unavailable
again on most nodes of the system.
-----------------------------------------------------------------------
Dear users of SuperMUC,
the reason for the problems with the GPFS filesystem on SuperMUC seem
to be defective Infiniband hardware component(s), which broke as a
consequence of the power failure on Sunday, September 23rd.
Unfortunately, the particular hardware components could not yet be
identified and we cannot reliably estimate how long the search will
take.
IBM is still working on the problem.
We apologize for the inconveniences caused by this failure. We will
inform you through an update to this announcement as soon as the
problem is fixed.
This information is also available on our web server
http://www.lrz-muenchen.de/services/compute/supermuc/aktuell/ali4424/
Markus Mueller
Mehr Informationen über die Mailingliste aktuell