The fencing script "fence_vmware_soap" is very slow on fencing a clustered vm node

Solution Verified - Updated -

Environment

  • Red Hat Enterprise Linux Server 5.7 (with the High Availability Add on)
  • Red Hat Enterprise Linux Server 6.1 (with the High Availability Add on)

Issue

  • The fencing script fence_vmware_soap is very slow on fencing a clustered vm node.
  • fence_vmware_soap takes too long when trying to fence or get status of other node in vmware guest cluster setup.

Resolution

Root Cause

  • The fence_vmware_soap script requested unnecessary data from vSphere thus slowing the fencing process.

Diagnostic Steps

  • Use the time tool to measure the fencing time for a system and notice the high user time because it is python processing the vSphere server list and status.
$ time fence_vmware_soap -z  -l <user> -p <password> -a <host> -o status -n <system> -v
Status: ON

real    4m23.936s
user    4m14.208s
sys     0m0.508s

This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.

Comments

Log in to comment