is_managed and cluster resource maintenance (HEARTBEAT)
This document (7000553) is provided subject to the disclaimer at the end of this document.
Environment
Situation
From the documentation:
/usr/share/heartbeat/crm.dtd * is_managed_default (boolean, default=TRUE) Unless the resource's definition says otherwise, * TRUE : resources will be started, stopped, monitored and moved as necessary/required * FALSE : resources will not started if stopped, stopped if started nor have any recurring actions scheduled.it might seem viable to do this by just setting the attribute "is_managed" to "false" on the affected resource.
This is NOT so!
If there is a monitor operation running on the resource, which is the case in most enviroments, then the resource will be in status "unmanaged" but the monitor will still be done. So as soon as the resource gives anything back to the monitor that triggers the failure, the cluster will act accordingly.
Resolution
There is two solutions to set the cluster to maintenance:
1. set the global attribute "is_managed" to false (via GUI or CLI):
this will set all resources to "is_managed" "false". Any resource can now be modified. But during this period of time the cluster will NOT check any resource.
2. set the resource attribute "is_managed" to "false" and set the monitor operation of this resource "on_fail" to "ignore".
After the maintenance all settings have to be reverted to make the cluster functioning again.
Disclaimer
This Support Knowledgebase provides a valuable tool for SUSE customers and parties interested in our products and solutions to acquire information, ideas and learn from one another. Materials are provided for informational, personal or non-commercial use within your organization and are presented "AS IS" WITHOUT WARRANTY OF ANY KIND.
- Document ID:7000553
- Creation Date: 04-Jun-2008
- Modified Date:10-Mar-2021
-
- SUSE Linux Enterprise Server
For questions or concerns with the SUSE Knowledgebase please contact: tidfeedback[at]suse.com