Closed 0sorkon closed 6 months ago
Hello, thanks for the report, although this is more a DRBD issue than related to LINSTOR.
The first block is "just a warning", that should have been fixed in a more recent version of DRBD. The other stack traces is something from libvirtd
, so neither DRBD nor LINSTOR - at least nothing directly. It could be the case that those "quite noisy" DRBD warnings could have contributed somehow to this libvirtd behavior, that is hard to tell.
I will however close this issue now, since this is not an issue with LINSTOR (i.e. wrong project). Please try to upgrade DRBD and if this issue still persists, please reopen a new issue in the DRBD project
There is a linstore pool on top of a thin lvm on ssd and hdd discs (different storage pools). Replication factor 2. It was necessary to take one of the three nodes out of operation while preserving resources. When trying to evacuate a node as described in the documentation, a message appears in the logs:
This message is repeated several times and then the following message:
In the end, there is no other option but to reboot the node. The same situation has happened several times with other pool members. Only
linstore node restore
and deactivation of resources to stop replicationlinstor r deact
helped to stabilise the situation, because, in my opinion, it was the active replication of resources that caused this problem. Could you please tell me what could be the cause of this problem and how to properly shut down the server with correct migration of resources that are deployed on it?