IIS 5 & IIS 6
W3SVC event id 1012
Last post Jan 12, 2009 12:18 PM by Rovastar
Jan 12, 2009 10:15 AM|AlanR54NC|LINK
Strange problem occurring sporadically over the last few weeks. Web farm of 10 IIS 6 servers (2K3 sp2, current patch level). Applications are a collection of classic asp, .net1.1 thru 3.5. Each server is configured identically and runs about 15 app pools.
The app pools are configured to recycle each night. (staggered times between 2AM and 3AM). random servers are throwing continous errors where the app pools are not starting during the expected times. The error is not specific to an app pool, and in fact appears
to be true for any app pool accessed on the affected server. An IISReset corrects the problem. Debug Diag dumps taken during the 'event' have revealed no insights (at least to internal reviews).
Points of note: Not all servers have exhibited the problem. It does seem to closely mirror the staggerd app pool resets, but there have been 2-3 occurances where the events did not follow that pattern. Once a server starts exhibiting the issue it seems to
be more prone to reoccurances. Server reboots seems to clear up the re-occurance issue. It is still a little early to determine if it has truely 'fixed' the problem. Any thoughts, experiences, etc ?
Jan 12, 2009 12:18 PM|Rovastar|LINK
What is the description of the error from the event log?
Anything in the httperr log?
Something like: A process serving application pool 'DefaultAppPool' exceeded time limits during start up.
If so I would look at changing the properties of your app pool. In the health tab there are a few there. e.g. Work process startup time might need increasing.
Really you have to experiment when you have a machine with the problem and not quickly do a iisreset quick fix.
Turn on log event on recycling for all instances to see if that sheds any light. http://support.microsoft.com/kb/332088