hanode-relation-changed Job for corosync.service failed because a timeout was exceeded

Recently, various timeout problems (eg: [1]) have occurred in CI system. I did some analysis for one CI sosreport [2], I found that these timeout issues were ultimately caused by the timeout issue of hacluster_octavia

$ grep -r 'timeout' 4d46f4e0-bc78-4702-92bf-c7e9f723e51c/octavia_3/var/log/juju/unit-hacluster-octavia-3.log |tail -n1
2021-11-17 11:09:53 WARNING hanode-relation-changed Job for corosync.service failed because a timeout was exceeded.

Other context about this log pls refer - https://pastebin.ubuntu.com/p/J44tRQcHGP/

and I didn't find any more info about corosync from this CI sosreport.

[1] https://review.opendev.org/c/openstack/charm-octavia/+/787700/
[2] https://openstack-ci-reports.ubuntu.com/artifacts/5ee/787700/29/check/bionic-ussuri-ha-ovn/5ee01ba/log/juju-crashdump-4d46f4e0-bc78-4702-92bf-c7e9f723e51c.tar.xz