Basic Pacemaker cluster without any SAP software or resources installed -- detection failed

[pirat013]

We have a tiny cluster on AWS with stonith and VIP resource installed only.

# crm_mon -1
Stack: corosync
Current DC: node2 (version 2.0.1+20190417.13d370ca9-3.24.1-2.0.1+20190417.13d370ca9) - partition with quorum
Last updated: Wed Mar 30 12:49:22 2022
Last change: Thu Mar 24 15:27:49 2022 by root via crm_resource on node1

2 nodes configured
2 resources configured

Online: [ node1 node2 ]

Active resources:

 res_AWS_IP	(ocf::suse:aws-vpc-move-ip):	Started node1
 res_AWS_STONITH	(stonith:external/ec2):	Started node2

The agent journal look like this:

# journalctl -f -u trento-agent
-- Logs begin at Wed 2022-03-30 11:50:21 UTC. --
Mar 30 12:12:51 node2 trento[2340]: time="2022-03-30 12:12:51" level=info msg="Identifying if the VM is running in a cloud environment..."
Mar 30 12:12:51 node2 trento[2340]: time="2022-03-30 12:12:51" level=info msg="VM is running on aws"
Mar 30 12:12:51 node2 trento[2340]: time="2022-03-30 12:12:51" level=info msg="Identifying the SUSE subscription details..."
Mar 30 12:12:53 node2 trento[2340]: time="2022-03-30 12:12:53" level=info msg="Subscription (7 entries) discovered"
Mar 30 12:12:53 node2 trento[2340]: time="2022-03-30 12:12:53" level=info msg="Discovery tick output: No HA cluster discovered on this host\n\nNo SAP system discovered on this host\n\nCloud provider aws discovered\n\nSubscription (7 entries) discovered\n\nHost with name: node2 successfully discovered"
...

[schlosstom]

We found the reason why the cluster was not detected.

After adding:

cluster_name: testcluster

in the corosync.conf the cluster has been detected by trento.

However the "cluster_name" option is not mentioned in any Best Practice papers. This means if customer following our papers it could most likely happen that this option will never be set.