What were you trying to do that didn't work?

After disabling standby mode on a node, the node is continuously in “pending” state. Entries from pacemaker.log show this:

Apr 09 00:03:45.608 svdb2psx01 pacemaker-schedulerd[32520] (unpack_transient_attributes)        info: svdb2psx04 is in standby mode

Apr 09 00:03:45.608 svdb2psx01 pacemaker-schedulerd[32520] (determine_online_status_fencing)    info: svdb2psx04 is not ready to run resources

…

Apr 09 00:03:45.704 svdb2psx01 pacemaker-schedulerd[32520] (unpack_transient_attributes)        info: svdb2psx04 is in standby mode

Apr 09 00:03:45.704 svdb2psx01 pacemaker-schedulerd[32520] (determine_online_status_fencing)    info: svdb2psx04 is not ready to run resources

Apr 09 00:03:45.704 svdb2psx01 pacemaker-schedulerd[32520] (determine_online_status)    info: svdb2psx04 is pending

What is the impact of this issue to you?

Since the host never becomes “started”, resources are not able to run on this host

Please provide the package NVR for which the bug is seen: 2.1.9-1

How reproducible is this bug?: This bug is not easily reproducible.

Steps to reproduce

• /usr/sbin/crm_standby -v on -N <hostname>
• Run “systemctl disable pacemaker” on host
• Run “systemctl disable corosync” on host
• Run “systemctl stop pacemaker” on host
• Run “systemctl stop corosync” on host
• Run “systemctl enable corosync pacemaker --now” on host
• /usr/sbin/crm_standby -v off -N <hostname>

Expected results

Host is no longer in standby mode and returns to “started” state

Actual results

Host is no longer in standby mode and is stuck in “pending” state