This is a 3-node cluster with gfs2 filesystem resources configured using dlm and clvmd. Stonith is enabled. dlm and gfs2 resources are set to fence on failures. Pacemaker version on the cluster - 2.1.4-5.el8 (CentOS 8 Stream)
Pacemaker default fence actions are set to power off the node instead of a reboot by setting "pcmk_reboot_action=off" on IPMI fence devices. Also, cluster wide stonith default action is set to poweroff the node instead of a reboot # pcs property config |grep stonith stonith-action: off stonith-enabled: true But still the node is rebooting as dlm_controld resource is sending a reboot action as part of fencing. Below are the logs obtained from syslog to confirm the same. dlm_controld[73542]: 813117 fence request 3 pid 1695739 nodedown time 1690006193 fence_all dlm_stonith pacemaker-fenced[15706]: notice: Client stonith-api.1695739 wants to fence (reboot) 3 using any device I need the node to be powered off from fencing operations rather than a reboot. Disabling fencing on dlm resources is not an option. Is there any other way to solve this and make dlm issue a poweroff action instead of a reboot as part of fencing. Regards P V K Sai Siddhartha The information contained in this transmission may be confidential. Any disclosure, copying, or further distribution of confidential information is not permitted unless such privilege is explicitly granted in writing by Quantum. Quantum reserves the right to have electronic communications, including email and attachments, sent across its networks filtered through security software programs and retain such messages in order to comply with applicable data security and retention requirements. Quantum is not responsible for the proper and complete transmission of the substance of this communication or for any delay in its receipt.
_______________________________________________ Manage your subscription: https://lists.clusterlabs.org/mailman/listinfo/users ClusterLabs home: https://www.clusterlabs.org/
