Hi
We have a dev environment of SQL 2012 that has 18GB assigned to it. The server has 48GB total and has 2 other instances. For one particular instance, let's call it E1, have a database restored to it every day from production. DB is about 1TB with a couple of other smaller ones. The load on his DB is very light from the web application, only about 2 or 3 users in it testing specific data issues. We're seeing either the users complaining about the following error (it's same error in event logs, sometimes it is thrown back to user)
"There is insufficient system memory in resource pool 'internal' to run this query'"
At this point a restart is usually required and things start working. From looking at dbcc MemoryStatus, one thing we have noticed is that the value for Pages Allocated for 'CACHESTORE_BROKERTO (node 0)' seems to be grow/be at a high value. I'm not exactly sure what this value means.
When looking at sys.converation_endpoints there are only around 200 entries...I’ve seen some articles that talk about leaks but I don't see that here given the number of rows.
After a restart of SQL, I clear the conversation_endpoints and then check memory status and the Pages Allocated to cachestore_brokerto continues to grow. I don't know if this is the source of the problem but it's the one value that seems to stand out.
One could argue that the RAM memory allocation is low, but the usage on this db is also very low. If there is RAM pressure wouldn't SQL utilize pagefile?
I'll post output from memorystatus in next post.....
SQL version
Microsoft SQL Server 2012 - 11.0.5058.0 (X64) May 14 2014 18:34:29 Copyright (c) Microsoft Corporation Enterprise Edition (64-bit) on Windows NT 6.1 <X64> (Build 7601: Service Pack 1)
Anyone got any thoughts or suggestions on how to investigate this further?
Thanks
Amit