I think it is a CPU spike that triggered the CPU Limit-based recycle.
I think you need to decide whether your metric is correct (i.e. you *should*
be getting the event log warnings and should be figuring out why your ASP
app would spike the CPU), or the metric is incorrect and should be adjusted.
If the server can be debugged, I suggest looking up the AppPool property
OrphanActionExe, which allows you to configure a command to run when IIS is
going to recycle/abandon an application pool. In your case, you want to
configure it to run a command like adplus or iisstate to take a snapshot of
the process so that you can see *what* is spiking the CPU. I suspect it
will be something in your ASP page, and this can help narrow down which page
and what sort of action it was doing.
I remember you can use %1, %2, etc for w3wp PID, w3wp AppPoolId, .etc, but
not certain of which (I believe %1 = PID, %2 = failure reasonn, %3 =
AppPoolId), and it does not appear to be in documentation. I'll open a
documentation bug on it, but in the meantime, you can quickly experiment and
figure out which.
This posting is provided "AS IS" with no warranties, and confers no rights.
3.5 GB physical ram.
4 x 2GHz Xeon CPU
I think it might be do to high CPU for a few seconds.
Thanks for your help