The Balena HPC service will be offline for maintenance from 14th - 18th of January to complete a few essential tasks.
ClusterVision (our supplier) and the ARC team will be working on the system during this period.
Some of the maintenance tasks that will be taking place are:
- BeeGFS upgrade
- BMC upgrade on some of the Skylake nodes
- Fix multi-path failing issues on headnodes
- Repair data objects on BeeGFS
- Master node failover test
During this maintenance there will be no access to Balena and workload scheduling will be paused to prevent any jobs from running, this will resume once the system is returned to production.
If you have any concerns please contact arc-support@bath.ac.uk.
--
Team HPC
Respond