What I like most about Galera Cluster is the multi master setup, it’s just a lot easier to manage compared to the master slave approach we used before. We are currently running it with 3 DB nodes and it handles failover pretty well, so even if one node goes down the others keep working. We were using a similar setup earlier on Red Hat Enterprise Linux and now on Rocky, and it’s been pretty consistent. Replication is fast and data stays in sync across nodes without much delay, which is really important for us. There were some issues at one point where a node kept failing repeatedly, and it took some time to figure out what was going wrong. Debugging that wasn’t very straightforward. But once it was fixed, things have been stable. Overall it’s been working well for our setup.