Pacemaker not able to promote node in a 3 node (2 node sync +1 config replica)

[snehasish3005]

Hello, This is my PCS Status.

pcs status
Cluster name: agcluster
Cluster Summary:

• Stack: corosync (Pacemaker is running)
• Current DC: primarynode (version 2.1.6-6fdc9deea29) - partition with quorum
• Last updated: Sat Feb 14 00:46:28 2026 on primarynode
• Last change: Fri Feb 13 23:47:34 2026 by hacluster via crmd on witnessnode
• 3 nodes configured
• 8 resource instances configured

Node List:

• Online: [ primarynode secondarynode witnessnode ]

Full List of Resources:

• Clone Set: ag_vetter-clone [ag_vetter] (promotable):
  • Promoted: [ primarynode ]
  • Unpromoted: [ secondarynode ]
• ag_vetter_vip (ocf:heartbeat:IPaddr2): Started primarynode
• hz_fip_assign_vetter (systemd:hz-fip-assign-vetter): Started primarynode
• Clone Set: ag_dekra-clone [ag_dekra] (promotable):
  • Promoted: [ primarynode ]
  • Unpromoted: [ secondarynode ]
• ag_dekra_vip (ocf:heartbeat:IPaddr2): Started primarynode
• hz_fip_assign_deckra (systemd:hz-fip-assign): Started primarynode

Daemon Status:
corosync: active/enabled
pacemaker: active/enabled
pcsd: active/enabled

When I try to simulate a failover using pcs standby node, I get an error saying that Not enough replicas to promote SecondaryNode.
Need 2 but have 1.
But I can see that there are 2 replicas. I sync + 1 configuration replica.

Can you please help here ?

I am using SQL Server 2025 in Linux for the HA set up.

[oalbrigt]

This is a PostgreSQL agent, so seems like you posted in the wrong repository.

MS SQL agent issues can be reported here: https://github.com/microsoft/mssql-server-ha/