SQL 2012 Cluster CheckServiceAlive spams application event log - by jrsqlguy

Status : 

  Duplicate<br /><br />
		This item appears to be a duplicate of another existing Connect or internal item.<br /><br />
		A more detailed explanation for the resolution of this particular item may have been provided in the comments section.

Sign in
to vote
ID 737820 Comments
Status Closed Workarounds
Type Bug Repros 13
Opened 4/19/2012 10:55:28 PM
Access Restriction Public


When I cluster sql 2012, about every 15 seconds or so, the following is posted into the application event log

[sqagtres] CheckServiceAlive: returning TRUE (success)

[sqagtres] LooksAlive request.

Is this by design or debug code not disabled? 
Sign in to post a comment.
Posted by Jason Fay on 3/10/2013 at 10:08 AM
I wasn't hit with this (SQL 2012 RTM, SQL 2008 R2 SP2, SQL 2008 SP3) all on a cluster, UNTIL last night some Windows Updates were applied. Then all of a sudden I started getting the Event ID 53 spam - FOR ALL INSTANCES, including the SQL 2008 instance.    Upgrading the 2008 R2 instances to CU3 did not resolve. Upgrading the SQL 2012 instance to SP1 did.

List of Windows updates applied that triggered this:

Will spend some time this week trying to figure out which one.
Posted by middon on 9/6/2012 at 11:32 AM
I opened a ticket with MS support and they were able to reproduce the issue on SQL 2008 R2 SP2. There is no current hotfix but it may "tentatively" be fixed and released in the coming CU3 around 10/15.
Posted by Dave_Mc on 8/10/2012 at 9:06 AM
Just wondering -- since Microsoft has already marked this problem "Closed," and that was before anyone had reported it on 2008 R2 SP2, does that mean that no one from Microsoft will be looking at this item again? I'm thinking that we might need to resubmit it as a new bug in order to get it associated with SQL 2008 R2.

Anyone there know the answer..?
Posted by Dave_Mc on 8/9/2012 at 3:02 PM
I meant to add that you need to stop and start the SQL Agent service for the change in polling intervals to take effect.
Posted by Dave_Mc on 8/9/2012 at 3:00 PM
We're also hitting this problem consistently after installing SP2 for SQL Server 2008 R2. We were getting 2 entries in the application event log every 5 seconds.

At least in our case, it's for the SQL Agent service. I don't consider that to be as critical (or generally as troublesome) as the SQL RDBMS service itself, so I reduced the polling interval in order to bring down the rate of messages somewhat. This is done on the Advanced Policies tab of the Properties window for the SQL Agent resource in Failover Cluster Manager. I'm not sure what I think about this for a system that is in active use, but I'm willing to try it in our DBA test cluster. I went from "Basic resource health check interval" of 5 seconds to 40 seconds, and from "Thorough resource health check interval" of 1 minute to 2 minutes.

Has anyone found a better work-around? It seems odd that Microsoft has apparently not even acknowledged it as a problem with SQL 2008 R2 SP2.

Posted by middon on 8/3/2012 at 12:14 PM
We are also seeing the same issue after installing SP2 for SQL Server 2008 R2. Will there be a fix for 2008 R2 SP2?
Posted by Ajay.G on 7/12/2012 at 10:23 PM
after deploying sp2 on sql server 2008 R2 we are also getting same.
Posted by Microsoft on 6/13/2012 at 2:27 PM

Just wanted to update you that we are targeting to fix this issue in Cumulative update #3 for SQL Server 2012. In addition the fix will be available in SP1.

Thank you
Alex Grach[MSFT]
Posted by _dfinn on 6/7/2012 at 11:32 AM
Please add something to stop this, it's amazing how much app log spam this creates once you start building multiple instances across a cluster.
Posted by Maciej Chojnacki on 5/15/2012 at 11:23 PM
Any chance to got that fixed?
Posted by Sethu Srinivasan on 4/24/2012 at 12:43 PM
Thanks for reporting this issue. We shall take a look at this issue and reply back with our findings

Sethu Srinivasan [MSFT]
Posted by johanljunggren on 4/24/2012 at 7:19 AM
I got the same issue on an SQL 2012 Cluster.