RHEL5 cluster locks up after a node leaves the cluster when there are more than 100 GFS/GFS2 file systems mounted

Solution Verified - Updated -

Issue

  • clusters with more than 100 GFS/GFS2 file systems fail to recover when a node leaves the cluster.  The cluster appears to be "locked up", with cluster/lvm operations stalling indefinitely

  • cluster with 100+ GFS2 filesystems across clustered LVM.  The cluster will randomly start "hanging" (indefinitely waiting) on various LVM commands, a rebooted node is unable to mount any of the volumes and "hangs" at the lvm monitor stage. When this happens, other nodes report the cluster is quorate but their lvm commands fail to return as well.

Environment

  • Red Hat Enterprise Linux 5.6
  • Greater than 100 GFS or GFS2 file systems mounted

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