We have a Galera cluster using rh-mariadb101 with 3 nodes. One node failed, and will not rejoin the cluster. There is nothing in the error logs to indicate why it won't rejoin; just the service starting, beginning replication from another node, and then shutting down. We expect the node to complete replication and rejoin the cluster.
We suspect a file system corruption because
df -Thshows only 1G in use where the other two nodes show over 50G in use, although xfs_repair doesn't appear to have done anything and the directories all look right.
- Red Hat Enterprise Linux 7.7 (RHEL)
- Red Hat Software Collections (RHSC)
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.