Hello,
this upcoming Monday we’ll be processing an OCP 4 (Openshift) cluster upgrade for the latest set of security and bug fixes to land. You can expect intermittent failures trying to reach to major GNOME services while the systems reboot after the upgrade has been applied.
In the near future with our move to Fastly Load Balancers we’ll be able to reduce services downtime by disabling the affected node while it performs maintenance. Please see Andrea Veri's Blog · Me, myself and I for more information around this specific current architecture limitation.
Thanks!