Hello everybody . I wanted to report anomalous behavior regarding replicas on a cluster of a VM. The VM is replicating on 3 nodes on a ZFS share named "zshare". I realized that if I take snapshots from that VM and restore the snapshot after a short time, it happens that the cluster reports an error in the replication, even as if the disk has I / O problems. But this is incorrect as this VM has been moved several times to other nodes. The fact is that the error only occurs when you take snapshots. Here is the error:
Furthermore, the snapshot bug still persists which is not deleted on replication to other nodes, ie if I delete a snapshot on the VM and this snasphot is replicated it is not deleted. the following bug
By deleting the replication job and deleting the replicated images and creating the job from scratch, everything starts working again
May 02 19:34:03 pve4 pvesr[20526]: send/receive failed, cleaning up snapshot(s)..
May 02 19:34:03 pve4 pvesr[20526]: 103-1: got unexpected replication job error - command 'set -o pipefail && pvesm export zshare:vm-103-disk-0 zfs - -with-snapshots 1 -snapshot __replicate_103-1_1619976840__ | /usr/bin/ssh -e none -o 'BatchMode=yes' -o 'HostKeyAlias=pve3' root@192.168.2.37 -- pvesm import zshare:vm-103-disk-0 zfs - -with-snapshots 1 -allow-rename 0' failed: exit code 255
May 02 19:34:03 pve4 systemd[1]: pvesr.service: Succeeded.
May 02 19:34:03 pve4 systemd[1]: Started Proxmox VE replication runner.
Furthermore, the snapshot bug still persists which is not deleted on replication to other nodes, ie if I delete a snapshot on the VM and this snasphot is replicated it is not deleted. the following bug
By deleting the replication job and deleting the replicated images and creating the job from scratch, everything starts working again
Attachments
Last edited:
