Why does a node fence other(s) when it is first joining a High Availability cluster in RHEL 4, 5, or 6?

Solution Verified - Updated -

Issue

  • Why do the nodes in my Red Hat Enterprise Linux cluster fence each other before they have even joined the cluster?
  • Nodes are fencing each other on startup, one node that is starting cman boots another during boot.
  • Cluster nodes in two node cluster repeatedly fence each other at boot time in a loop. How to troubleshoot if the cluster formation fails at initial stage like this?
  • Cluster server appeared hung, when the node 1 is rebooted manually, node 2 is fenced. What's the root cause, RCA of this behaviour?
  • When powering off/on one of the nodes of a RHEL cluster, the rebooted node fenced off the surviving node during bootup, resulting in reservation conflict for the surviving node and it lost all cluster services.
  • Not able to start cluster services, every time nodes are getting fenced. When try to start cluster services and also during manual reboot.
  • RedHat cluster causes the active node to reboot, after fencing off the passive node that crashed?
  • In active-active GFS2 cluster, if one server come up, second goes down automatically.
  • In two nodes cluster when one node starts, it fences another node.
  • post_join_delay , FENCE_MEMBER_DELAY

Environment

  • Red Hat Cluster Suite (RHCS) 4
  • Red Hat Enterprise Linux (RHEL) 5 or 6 with the High Availability Add On

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.

Current Customers and Partners

Log in for full access

Log In

New to Red Hat?

Learn more about Red Hat subscriptions

Using a Red Hat product through a public cloud?

How to access this content