I switched our dev and qa clusters to Galera, and they seem to work fine, but I see this error in the logs:

Feb 27 11:24:08 us01odc-dev1-ctrl1 mysqld[361982]: 2020-02-27 11:24:08 484069 [ERROR] InnoDB: WSREP: referenced FK check fail: Lock wait index `PRIMARY` table `neutron`.`provisioningblocks`

I googled around and people say that it is a harmless error, but it seems to be time-correlated with creation of VMs that aren’t pingable.

https://jira.mariadb.org/browse/MDEV-18562

Is this a harmless error? Do I need to look elsewhere for the reason why some VMs fail to return pings?