Data Index stops working and proto files are lost on Infinispan running on OpenShift

Sometimes, the PV attached to the Infinispan instance deployed by the Infinispan Operator gets corrupted / unread for some reason, breaking the Data Index.

To get around it, we need to recreate the Infinispan instance and delete the Data Index pod manually. The problem is that all the data processed so far get lost.

We need to investigate the reason behind this breaks and understand how Infinispan is treating their data on OpenShift, like storage recommendations, deployment best practices and so on.