we’re running a 3-node XtraDB-Cluster and observed the following issue: if one node looses connectivity to the rest of the cluster (using iptables DROP for testing), then inserting enough to a table so that an SST becomes necessary and try to re-join the firewalled node, it can’t do an SST stating “WSREP: You have configured ‘xtrabackup-v2’ state snapshot transfer method which cannot be performed on a running server. Wsrep provider won’t be able to fall back to it if other means of state transfer are unavailable. In that case you will need to restart the server.”
We don’t think this is expected beharviour?
Please see attached logs.
Here the node gets connectivity again, tries to rejoin the cluster, and fails:
And this is the log of the donor Node where the previously offline node gains connectivity again:
Versions are as follows (Debian wheezy, packages from Percona):
Do you have any idea why this is not working? Thanks in advance! Btw, logs were generated with wsrep_debug, wsrep_log_conflicts and general_log set to ON. If you need even more verbosity, let me know!
Best regards and thanks in advance,