"V-16-1-10651 Cluster UUID received from snapshot is not matching with one existing on server101" appears when starting VCS, even though the value of cluster UUID on the configuration file /etc/vx/.uuids/clusuuid is same on all the nodes
Article ID: 100042626
Updated On:
Description
Error Message
Below is the error message seen from the engine log when starting the cluster on node server101 while the cluster is already running fine on node server102
2018/03/26 14:25:29 VCS INFO V-16-1-10466 End of snapshot received from node: 1. snapped_membership: 0x3 current_membership: 0x3 current_jeopardy_membership: 0x0
2018/03/26 14:25:29 VCS ERROR V-16-1-10651 Cluster UUID received from snapshot is not matching with one existing on server101. VCS Stopping. Manually restart VCS after configuring correct CID in the cluster.
The cluster UDID in the configuration file /etc/vx/.uuids/clusuuid is reflecting the same for both of the nodes. But, VCS still fails to start on node server101.
root@server102#/opt/VRTSvcs/bin/uuidconfig.pl -clus -display -use_llthost
Finding existing UUID information ...
server101 .... exists.
server102 .... exists.
Done
root@server102#
[root@server102 ]# cat /etc/vx/.uuids/clusuuid
{bdaf4e70-1dd1-11y2-915f-9909xd1d9b48}
[root@server101 ]# cat /etc/vx/.uuids/clusuuid
{bdaf4e70-1dd1-11y2-915f-9909xd1d9b48}
When checking the CID (Cluster ID) attribute for the cluster on node server102, using the haclus command, the Cluster UDID is showing a different value than it is in the /etc/vx/.uuids/clusuuid file.
[root@server102 ]#haclus -display | grep -i cid
CID {2a5fdfcc-1dd2-11b2-82x2-f748f851by37}
[root@server102 ]#
The CID provides a universally unique identification for a cluster. VCS populates this attribute once the engine passes an hacf-generated snapshot to it. This happens when VCS is about to go to a RUNNING state from the LOCAL_BUILD state.
The cluster UDID value on the configuration file may have changed due to manual intervention.
Resolution
Copy the Cluster UDID which was returned by the haclus command output to /etc/vx/.uuids/clusuuid on both of the nodes and restart the cluster on the problematic node (in this example, server101).
[root@server102 ]#haclus -display | grep -i cid
CID {2a5fdfcc-1dd2-11b2-82x2-f748f851by37}
[root@server102 ]#
Issue/Introduction
Additional Information
