This issue occurs specifically with a MySQL cluster using Galera.
If one of the virtual machines is hosted on a slow datastore, when being updated, the node may take significantly longer to sync with the other nodes in the cluster.
The synchronization progress can be monitored using the wsrep_ist_receive_status
variable, which provides details about the Incremental State Transfer (IST) process. To view this status, run the following query:
SHOW STATUS LIKE 'wsrep%';"
If wsrep_ist_receive_status takes few minutes to move forward, you may be experience this issue.
TAS Mysql or VMware Tanzu for MySQL on Cloud Foundry clusters.
A slow datastore can significantly delay the synchronization process of the updating node with the other nodes in the cluster.
If the slowness can't be solved and the infrastructure supports it, the recommended solution is to move the virtual machine's disk to a performant datastore. This can be achieved using tools like Storage vMotion in VMware vCenter.