Peter (and clix), when you say "crash" do you mean you find the cf process no longer running? Or just not responsive? That's an important distinction.
If it's truly down, unexpectedly, one place you can look for logs is the cfusion/bin folder. If the jvm crashed, you may find hs_err*.log files. (If you have multiple instances of CF, see the instancename/bin folder instead).
Look also for any other logs in the cf logs folder, and pay attention specifically to the time it crashes or becomes unresponsive, as well as to when cf starts as there maybe a client there.
Finally, if indeed cf is still running, but "unresponsive", it would be very helpful to have a cf monitor showing what's running. It may be that in fact many requests are hung, and once there are more than the "max simultaneous requests" value in the cf admin requests tuning page, cf queues any subsequent requests (for 60 seconds by default).
As for monitoring, if you are on cf2018 or above, cf now includes the PMT (which many find challenging to get running for various reasons that can often be resolved), or a great 3rd party option is fusionreactor (which has 14 day free trial, but requires a cf restart to take effect). These tools will show WHAT (if any) requests are running, and a whole lot more that shines light into the cf black box.
I can help with installing/using these, as well as resolving this challenge you both face, if you prefer that over back-and-forth here. See my signature.
Or write back with a response and I or others will carry on trying to help here.
/Charlie (troubleshooter, carehart.org)
5
Replies
AdChoices