-
Bug
-
Resolution: Unresolved
-
Major
-
None
-
RH436 - RHEL 8.3 1
-
None
-
False
-
-
False
-
-
-
en-US (English)
Please fill in the following information:
URL: | https://rol.redhat.com/rol/app/courses/rh436-8.3/pages/ch07s03 |
Reporter RHNID: | shasingh01 |
Section Title: | Configuring a Two-node Cluster |
Issue description
Description: Create the fence device for the second node without the pcmk_delay_base option. With this configuration, the node1 machine always wins the fence race because the cluster causes the other node wait for 30 seconds before allowing it to fence node1, giving node1 an opportunity to fence node2 first.
for the above paragraph, it should be "With this configuration, the node2 machine always wins the fence race because the cluster causes node1 (with the delay) to wait for 30 seconds before allowing it to fence node2, giving node2 an opportunity to fence node1 first."
That configuration does not prevent the cluster from fencing the node1 machine. If the cluster communication network is up, and the node1 machine fails for another reason, then the cluster can still fence the node1 machine, but with a 30-second delay.
and for this above paragraph, node1 must be replaced by node2.
Steps to reproduce:
Workaround:
Expected result: