1

I am very new to Windows Server, so sorry for dummy questions. I have a Windows Server 2008 R2 Enterprise. There are few AppPools. The problem is, that DefaultAppPool stops responding after 40 users online (depending on loadimpact tests).

Strange thing is that when I do a load test on websites configured on TestPool or Test2Pool (different domains, different websites and databases than DefaultAppPool), DefaultAppPool website hangs, while TestPool website is ok.

This load test has no impact on server's cpu or memory usage. I have tried monitoring HTTP Service Request Queues and W3SVC_W3WP variables with Performance monitor, but with no signs of strange things. Also checked Connection limits - it's default value.

How could I track this issue?

1 Answer 1

0

Any directory marked as an Application in IIS can run in any App Pool. Virtual directories and physical directories just run in their parent app/site/whatever's above them in the hierarchy.

As DefaultAppPool is the catch-all App Pool, I'm guessing that at some point, a portion of one of those other websites was marked as an App, and that app was set to run in DefaultAppPool.

So while TestPool.com/ runs in TestPool, maybe TestPool/AppA is set to run in DefaultAppPool?

While the GUI creates an App Pool per site, and tries to keep things organized, scripts need to manage that sort of detail themselves, which can lead to a counter-intuitive-seeming situation like this.

Otherwise, it seems likely there's a component which calls back into DefaultAppPool when the other sites run - like an authentication proxy or another web service module - because App Pools are generally isolated from each other, and only ever affect each other through shared dependencies.

You can usually right-click an App Pool and choose View Applications in order to work out whether some little apps from the other sites have invaded a particular App Pool.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .