4

So I set up our SQL servers to notify us of alerts at severity 17 or higher (as advised by various SQL Server articles). I've done this to each node of a SQL Server Always On Availability Cluster, with three multi-subnet nodes.

Over the weekend we've received almost a hundred notifications about a single "informational only" message from one of the secondary nodes. The other nodes aren't doing this and the alert doesn't seem to be causing any actual problems. The alert is:

DESCRIPTION:    Skipping the default startup of database '<database>' because the database belongs to an availability group (Group ID: <...>). 
The database will be started by the availability group. 
This is an informational message only. No user action is required.

Possible culprit: Our network has a few issues, the node could be disconnecting and reconnecting or timing out.

I suppose my questions are, is there any action I should take to resolve this alert? Why is it only affecting one of my nodes? If it really is informational and no action needs to be taken, can I configure my alerts to ignore these alerts?

Upon a comment, I looked at the logs and it appears that the server has been running all that time but that it disconnected from the always on server around that time:

AlwaysOn Availability Groups: Local Windows Server Failover Cluster node is no longer online.  
The Availability replica is going offline because the local ...  
The Waiting for local Windows Server Failover Clustering service to start ...

(Then it goes through the informational messages regarding starting up each database) 
Local Windows Server Failover Clustering service started.
  • Chances are you're having connectivity issues on your secondary server. Even though the notifications are informational, you might want to troubleshoot the connectivity issues. If you ever really need a fail-over to happen and your secondary is having network issues... – Reaces Dec 15 '14 at 10:14
  • 1
    There's a "problem" in SQL 2012 AAG's that takes secondaries offline when they lose connection to the primary (brentozar.com/archive/2015/02/alwayson-availability-groups-faq). I reckon that might be the cause here. What should you do about it? Probably nothing if it's just a network dropout. – Greg Nov 23 '17 at 1:09
0

I would suggest you this script.
The @message_id = 35202 and 35206, it was extremely useful for me when I had configured an AwaysOn with 4 Nodes between 2 different data centers.
How many replicas do you have? Just be aware of Quorum Votes.
This script helps you to find out.

SELECT  member_name
      , member_state_desc
      , number_of_quorum_votes
 FROM   sys.dm_hadr_cluster_members;
1

Same kind of message appeared in our environment as well.

What I understand is, as the DB was in AAG, and when the node (secondary AAG group) goes under restart, the DB start-up will be taken care of by AlwaysOn instead of the regular SQL Server 3-phase process!

If the message appears without any secondary node restart, we may have to consider troubleshooting, but in our condition we can ignore that as the message states it is just informational.

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.