We are running on Episerver 6 R2, IIS 7, Microsoft server 2008 R2 and SQL server.
We have installed a new Episerver site which has 2 front application servers (for visitors) and one application server for editors. Now and then the hole site is 'freezing' and the only way to get it up and running is to kill the application pool. When searching for a solution we have shut down 1 of two fronts so we know which front which will have all requests. When the site freezse the strange part is that if I am loging in on current front server cpu is on 0-1% usage, w3wp-process in on 0-1% of cpu usage, RAM is on 1,5Gb (we are haveing 8Gb as max) so it looks like the front server is idleing... but I can se that requests are stacking up in the working process so something is freezing the application. As far as I know DB server is idleing to. If there would be 'bad code' my quess is that cpu usage should be much higher as it is know it is on 0-1% usage
I wonder if anyone have had the same problem and solved it or if of anyone recognize this symptoms and could point me in right direction how to trace and find the bottleneck.
Thanks in advance!
Any error messages in the event viewer - application log or anywhere else?
The low cpu activity indicates that the threads are blocked (waiting for something). I would recommend to take a memory dump of the process e.g. using adplus. Then I would look at the stack dump to see what the different threas are doing. If they are blocked you should be able to see what call that is blocking.