Advice: QRIScloud NeCTAR Volume Storage Hazard: 2018-06-08 onwards

Follow

Starting on Friday 8th June, we will be replacing servers in the QRIScloud NeCTAR Ceph storage cluster with servers from the decommissioned QRIScloud GCI Ceph cluster.  This work will take a number of weeks, and should be transparent to users.  The storage cluster will maintain its current "2 copy" replication policy at all times.

We do not anticipate any problems from this procedure.  However, we advise all users of QRIScloud NeCTAR volume storage to ensure that their backup procedures are working. If you do not take regular backups, we recommend that you take a backup before the hazard window starts.

UPDATE 2018-06-08 - For staff availability reasons, the work has been rescheduled to start on Monday 11th July.

UPDATE 2018-07-31 - 14:00 - We are currently experiencing some significant performance problems with the Ceph cluster, caused (we think) by a network topology issue.  The symptoms are that some file system requests are taking an inordinately long time.  This is leading to file system lockups, and problems attaching and detaching volumes.  These problems started last Friday.

QRIScloud operations staff are working to diagnose and resolve the issue, but we do not have an ETA at this stage.  We apologize for the inconvenience this is causing.

Have more questions? Submit a request

Comments

Powered by Zendesk