Sitecore Eventqueue table growing rapidly after upgrading to 6.6 causing sql timeout

Question

We are seeing slow performance and 408 errors on website Investigation revealed that sitecore eventqueue table is growing rapidly(1million rows per hour), which is eventually causing sql queries taking forever to run(please see below query) eating up all resources and stopping sql server connectivity, which caused 408 errors

SELECT [EventType], [InstanceType], [InstanceData], [InstanceName], [UserName], [Stamp], [Created] FROM [EventQueue] WHERE ([InstanceName] <> @targetInstance AND [RaiseGlobally] = 1 OR [InstanceName] = @targetInstance AND [RaiseLocally] = 1) AND [Stamp] >= CAST(@fromTimestamp AS TIMESTAMP) ORDER BY [Stamp]

clearing eventqueue table followed by app pool recycle brings back the website to normal. But we endup doing this everyday to keep website running, which is not correct.

Further investigation revealed that there are lots of RuntimeRolesUpdatedRemoteEvent events in this table which apparently not used in code anywhere.

Sitecore recently been upgraded to version Sitecore.NET 6.6.0 (rev. 130529)

Can events written in eventqueue table be controlled? Is there anyway to stop RuntimeRolesUpdatedRemoteEvent events going in eventqueue table?

Answer 1

Sitecore found a bug which was causing to write multiple event in eventqueue table at authentication, and afterward when user browsing the site every time authentication is checked (i.e. user clicked anywhere in website) more rows added in this table. Sitecore have provided a patch which has fixed this issue.