4

I have SQL Server 2014 running on Windows Server 2012 R2.

There are three nodes, one is remote DataCenter. So two nodes have voting. Failover Cluster is using file share witness on remote server.

If primary server is down, AG automatic Failover works but when SQL Server is down AG automatic failover does not work. AG waits on Resolving state.

Also:

  • I stopped service from configuration manager, AG Failover is success.
  • Stopped service from services.msc, AG Failover is failed.
  • Killed service from Proccess List, AG Failover is failed.

Voting;

Server         Assigned Vote       Current Vote
Node 1              1                   1
Node 2              1                   1
Node 3              0                   0

I disabled firewall for all network, nothing changed, problem continuous.

Can someone help me to solve problem?

Critical Error

The role of this availability replica is unhealthy.
The replica does not have either the primary or secondary role.

Failover is from node1 to node2. Both are set to synchronous and automatic failover. Nodes are not in a paused state. I have installed all nodes as stand alone instances.

|improve this question
  • 1.) are all 3 nodes configured to be synchronous and with automatic failover? 2.) when node 1 fails, does automatic failover work to node 2? 3.) when node 2 fails, does automatic failover work to node 3 or node1? 4.) why does node 3 not have a vote in the quorum configuration? 5.) are you changing settings for the group inside of SQL or from cluster admin? 6.) does your file share witness have a vote in the quorum configuration? – Chris Lumnah May 16 '17 at 0:42
7

It could be the issue described in INF: AlwaysOn – The secondary database doesn’t come automatically when the primary instance of SQL Server goes down by Arvindh Kalidasan - Support Engineer, Microsoft GTSC.

In this blog we would discuss about behavior of AlwaysOn availability group where the secondary database doesn't come automatically when the primary instance goes down. The secondary database goes into Resolving state. On the failover cluster manager the resource appears in fail state.

[...] we found that if we stop SQL Service manually on primary replica, it fails over to second node only once. Any further attempts of stopping SQL Service (to test auto failover) would not cause failover.

The workaround posted there is:

[...] to have the value set to a higher number for "Maximum Failures in the specified period".

  1. Maximum Failures in the specified period: set to 60
  2. Period (Hours): set to 1
|improve this answer
  • FYI the msdn link is not working, I am not sure what is the issue with blogs.msdn link many of them are not working. If it is migrated to new platform they should add redirection – Shanky Mar 22 at 8:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.