808 | Cluster application stopped | Critical | Outstanding | Application [NC] on node [TEST-vSZ] stopped.
These messages you are getting because for a system IO intensive scenario on SZ, the check process scripts may not be able to respond in 10 seconds, then led to process checking failed. Solution will be to extend the process check timeout to align with the check alive timeout (10) minutes.
We have a reported issue for this. And the issue will be fixed in 18.104.22.168.0 670+.
No mention of a patch to get us by until this new version.