rhel7: "pcs resource move" should not lead to hard failover failures
Issue
For example, let's say the resource res1 is on nodex, the administrator once execute the "pcs resource move res1", hence a cli-ban-res1/nodex is set in configuration, and then for whatever reason, he forgets to execute the "pcs resource clear res1" to remove the cli-ban line.
If, later, there is hard failover case which leads pacemaker to re-start all resources on nodex, but as the cli-ban-res1/nodex is always here for res1, the failover fails. We can't have this risk of failover failing just because an administrator has not executed a "pcs resource clear".
Environment
- Red Hat Entreprise Linux (RHEL) 7
- pcs
- pacemaker
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.