IIS 5 & IIS 6
Webservice Application Crashes
Last post Jul 18, 2008 03:12 PM by chuck.boyer
Jul 18, 2008 08:20 AM|chuck.boyer|LINK
I have a problem with a webservices application (.NET 1.1 running on IIS 6.0). After approximately 9 hours it crashes and we receive HTTP/1.1 500 ( The specified network name is no longer available.
The only event in the event log that I can find that corresponds to the same timeframe is the following:
Event Type: Information
Event Source: W3SVC
Event Category: None
Event ID: 1074
Time: 9:15:03 PM
A worker process with process id of '10148' serving application pool '[mysite.tld]' has requested a recycle because the worker process reached its allowed processing time limit.
For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.
Does anyone have any suggestions on where to start looking for the problem? I have downloaded the debugdiag tool, but I'm a bit unsure of how I should configure the rules to capture the crash.
Jul 18, 2008 10:22 AM|lshields|LINK
Welcome to the IIS Technet Boards.
The message you are seeing in the Event logs points to some settings on the application pool for mysite.tld. Do you have any of the options set for Recycling this application pool? You can set these pools to recycle based on memory consumption, processor
consumption and also time.
Do you have any of these settings? If so, can you post them?
Jul 18, 2008 11:04 AM|Rovastar|LINK
Additionally you can log when the app pools recycle. For some servers you have done this however you have not enable this for all events that can occur (you have only enabled: AppPoolRecycleTime,
It may be useful to have this for all events that occur.
RecycleISAPIUnhealthly is useful to have.
To turn them all on (for AppPool: DefaultAppPool):
cscript adsutil.vbs Set w3svc/AppPools/DefaultAppPool/LogEventOnRecycle
Jul 18, 2008 03:12 PM|chuck.boyer|LINK
Thanks for the tips guys. Recycle worker process is set to every 720 minutes. All other options are unchecked.
I've enabled logging on the events suggested, so I'll keep an eye on things.
I would think though that the auto recycling of the worker process shouldn't cause the application to crash.