Network problem 15 Feb 2017

From ScientificComputing
Revision as of 12:49, 16 February 2017 by Sfux (talk | contribs)

Jump to: navigation, search

Due to a misconfiguration of the cluster's internal network, the compute and login nodes lost network connectivity. This caused all running jobs to fail without being able to write back their results.

The network has been repaired and all nodes have been brought back into operation.

We have reopened the login nodes, but LSF queues remain closed until we can assure that the cluster is running well.

For updates on the status of the problem, please check the system status and this page.


We are sorry for the inconvenience.

Update 16.02.2017

LSF queues are progressively being opened to bring the system back into production.