Long delays when joining a cluster or rebooting a node in the active cluster

Products

• CA Privileged Access Manager (PAM)

Issue/Introduction

We had a previous case open for a problem where any PAM node reboot took a very long time after upgrade to 4.1. We tested a solution successfully on standalone nodes, and the solution is included in a hotfix we got on top of 4.1.2. But we observe that even with this hotfix applied, it takes very long, about 20 minutes, for a node to be available again if we reboot it while it is an active member of a cluster. We also see long delays when we have a new node join the cluster, which does not involve a reboot.

Environment

Release : 4.1-4.1.4

Cause

Various startup scripts could be held up, sometimes for a few minutes, while logging to the common console log.

Resolution

This problem is expected to be fixed in 4.1.5+ and 4.2+.