I have a 3 node Galera cluster running MariaDB on Ubuntu 18. I ran a few batch files to add data to the database, then shut down each node as this cluster is being relocated to a different subnet at a new location. After trying to bring the cluster back up, each node fails. Examining /var/lib/mysql/grastate.dat for ALL nodes show seqno = -1, safe_to_bootstrap = 0. I ran mysqld --wsrep-recover and ALL nodes report WSREP: Recovered position :100344, which is expected since the nodes were idle for sometime before shutdown. The question is how to bring up the cluster again, I keep getting "Failed to start MariaDB 10.1.38 database server" when checking status. I set safe_to_bootstrap: 1 but that did not work.


The working fix for this was executing sudo galera_new_clusteron any node. I had the other pair shutdown when executing this, then started each in turn and the cluster was back up.

  • Better would have been to shutdown-move-restart each one, one at a time. – Rick James Apr 29 '19 at 0:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.