Good morning community!
I have a three-node HA cluster running on a three-node Gluster shared storage. At different times a day a scheduled backup job starts doing a backup of some VM's onto an external NFS storage.
Last night the third node with some VM's that were currently doing a backup, got a faulty power supply and the server crashed. All VM's on that node were migrated to the other nodes. Everything good so far. But all the VM's that were doing a backup till the power supply got faulty, didn't powered up on the other nodes, because the conf files of the VM's were still locked. So I needed to manually unlock all the files and then start the VM's. This resulted in multiple hours of downtime, till I even noticed that some VM's didn't powered up.
Is there a bug? Is there some quick fix for such case?
Thanks in advance and greetings from Germany!
Andy
I have a three-node HA cluster running on a three-node Gluster shared storage. At different times a day a scheduled backup job starts doing a backup of some VM's onto an external NFS storage.
Last night the third node with some VM's that were currently doing a backup, got a faulty power supply and the server crashed. All VM's on that node were migrated to the other nodes. Everything good so far. But all the VM's that were doing a backup till the power supply got faulty, didn't powered up on the other nodes, because the conf files of the VM's were still locked. So I needed to manually unlock all the files and then start the VM's. This resulted in multiple hours of downtime, till I even noticed that some VM's didn't powered up.
Is there a bug? Is there some quick fix for such case?
Thanks in advance and greetings from Germany!
Andy