We had three Ceph OSDs crash overnight. They went down & out. I was able to start them back up and the VMs would reboot, but the same OSDs would crash a few minutes later. This morning, I destroyed the three affected OSDs (they were stopped and marked as out). The Ceph rebalanced, but any VM that I start is getting stuck at the boot screen with the Proxmox logo. If the VM is using SeaBIOS, I receive a message that says Booting from Hard Disk, but it hangs there as well.
When I try to view the VM Disk in the pool through any of the nodes, I receive a Communication failure (0) error.
When I try to view the VM Disk in the pool through any of the nodes, I receive a Communication failure (0) error.
Last edited: