Hi, Using tag-based HA LVM. If NodeA loses access to the disks, it's naturally unable to strip the tags out. NodeB will not recover the service because the tags of NodeA are in the VG. Even though NodeB can fully communicate with the NodeA, quorum is established, etc. Question 1) couldn't I somehow set up NodeB to fence NodeA and strip the tags? This is what it does if it thinks NodeA is offline, anyway. Question 2) How can I recover most gracefully from this? I'm stuck in a situation where NodeB has access to disks but refuses to activate them due to the tags of NodeA being there. I can manually remove the tag, or I can fence NodeA myself and then NodeB will happily declare "I can claim this volume group" and survive. The former is a little scary in terms of potential human error, while the latter is awful since other services might be successfully still running. Any better options out there? Thanks in advance. -- Linux-cluster mailing list Linux-cluster@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/linux-cluster
