Network outage at VUB main campus#
Updated on 14/07/2022
16:30 The network outage in VUB main campus is resolved. Access to both the Hydra HPC cluster and the Pixiu object storage has been reestablished.
The job scheduler of Hydra is back online and jobs in the queue are being processed as usual. Jobs already running before the outage might have failed due to the disconnect of the VSC_HOME and VSC_DATA storage, users that had running jobs should check the output for errors and re-submit any failed jobs. Jobs that failed to start during the outage have been re-queued automatically. Jobs of non-VUB users in Hydra will be held in queue until Monday, July 17th.
On Friday, July 14th at around 11:30 (CEST), the main VUB campus suffered a generalised network outage that is also affecting the data center hosting the Tier-2 HPC cluster Hydra and the Pixiu object storage.
These are the main consequences for our users:
loss of access to the login nodes of Hydra
loss of access to the notebook platform at notebooks.hpc.vub.be
loss of access to the Pixiu object storage
many running jobs in Hydra will be severely impacted due to the loss of connection of the cluster with the storage serving the home and data directories of the users
All partitions in Hydra are currently down and the job scheduler has been put on pause until the situation resolves.
The network outage at VUB campus is still ongoing. We will keep this news post up to date with any new information.