Thanks for the follow-up. This might take some time to get back feedback on this issue due to various reasons. Once I have further details or findings I'll post an update.
The system experiencing the issue is a production system and it's in the process of being migrated to a more powerful host running bare kvm/libvirt. The system would also have a mirrored set for the ZFS pool so the frequency of the issue occurring might change or disappear. It's also worth noting this might be happening under high IO load in particular (intensive write operations). There might also be a misconfiguration, however such condition should be handled properly. The issue happened randomly and sometimes it took as little as a few hours but sometimes it was running for about a month... The primary DB host is still on 22.04 with 5.15 kernel and seems to be stable in this respect. The only notable difference between the two, prior to the upgrade to 24.04 on the troubled system, is the ZFS drive for MariaDB on the primary being mirrored. It does seem to have something to do with either the kernel or driver/module (maybe some sort of regression or race condition?) as the system was functioning well for a long time before the upgrade of 22.04 (5.15.0-118-generic) to 24.04 (6.8.x). Having said that, we'll keep monitoring this and I'll try to find some time to attempt to isolate and reproduce the issue. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/2098375 Title: ZFS pool randomly encounters error condition and hangs To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2098375/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
