Error 976 Severity: 14, State: 1. Always on Clustering

Question

This has been happening in both the replicas (secondaries) at the same time ....Nothing shows on the primary replica.

Error: 976, Severity: 14, State: 1.

The target database is participating in an availability group and is currently not accessible for queries. Either data movement is suspended or the availability replica is not enabled for read access.

To allow read-only access to this and other databases in the availability group, enable read access to one or more secondary availability replicas in the group.

For more information, see the ALTER AVAILABILITY GROUP statement in SQL Server Books Online

Any ideas on where to start or what might be the problem?

Answer 1

You are attempting to connect to databases on secondary replicas when the availabilty group replica is not configured to allow connections on anything other than the primary replica.

The following script will tell you the state of the secondary replicas, and whether they will accept read connections:

SELECT ag.name,
       replica_server_name,
       secondary_role_allow_connections_desc
FROM sys.availability_replicas ar
    JOIN sys.availability_groups ag
        ON ag.group_id = ar.group_id;

The result of each replica with be:

• NO - no connections allowed to the database (default behavior)
• READ_ONLY - connections allowed when the ReadIntent flag is passed in a connection string
• ALL - all connections to the databases are permitted.

To change the setting for a replica you can either use the GUI, or a script:

ALTER AVAILABILITY GROUP [AGName]
MODIFY REPLICA ON 'ReplicaName'
WITH (SECONDARY_ROLE (ALLOW_CONNECTIONS = ALL, READ_ONLY_ROUTING_URL = 'TCP://ReplicaName:Port'))

For more details see the Microsoft documentation.

Answer 2

I'm having the same issue. It is due to scheduled transaction log backups. These have been setup using standard SQL maintenance plans. I have it configured to take the backup on the primary only. The SQL generated by the transaction log backup subtask contains the logic to prevent the backup from taking place on the secondary and it works HOWEVER, these connection errors are still logged in SQL on the secondaries which saturates the SQL log...