Coldfusion 2018 - CF Administrator 500 error

[tevansresdat]

I've recently inherited a Coldfusion 2018 instance. I was provided a clone of the production server which I have running as a VM on an external drive. It's also up on AWS so that others on my team can access it remotely. I can access the CF adminitrator app just fine from the local VM at the typical address (http://localhost:8500/CFIDE/administrator/index.cfm). However, on the AWS VM (which is theoretically identical), I get the default CF 500 error screen at that same address. Not seeing any errors logged, so I'm not sure what is causing it. Has anyone else run into a similar issue? I'm at a loss for how to troubleshoot without any detailed errors.

[Charlie Arehart]

You say default 500 error. Can you be more specific, sharing the details? And have you confirmed the 8500 port is what cf is listening on? If you don't know how, tell us if it's a windows or Linux vm.

Finally, if it is listening on 8500, can you try 127.0.0.1? If that works, we can see why once you can login. 

[tevansresdat]

Thanks for the follow up Charlie. By default 500 error screen I just mean it's showing the default template:

It is running on Windows and Coldfusion appears to be listening on that port. From netstat -a -b:

Getting the same thing using 127.0.0.1, unfortunately.

Let me know if any other info would be useful.

[Charlie Arehart]

If you're getting that error screen, it will log the error to the cf application log. 

But you said there was no error in the logs. Sure about that? 

[tevansresdat]

I've checked both the standard log directory (<cf_root>\cfusion\logs) and the log directory specified in <cf_root>\cfusion\lib\neo-logging.xml and I can't find anything new in the cf application log (or any other logs in those directories for that matter).

Another bit of backstory that may be part of the issue: the Coldfusion instance is on a different drive from the log files. When the VM was initially setup on AWS, the drive letters were not the same as the original production server this was cloned from. I discovered this after the Coldfusion service would not start. I have since remapped the drives correctly, restarted the VM, and the Coldfusion service is now running. I'm wondering though if that may still be causing issues. Maybe the cause of the 500 error and why it's seemingly not being logged - which would be a real double whammy. And maybe outside the scope of this forum, but I would appreciate any other suggestions before I dive down that rabbit hole.

[Charlie Arehart]

Is ANYTHING being logged to the cfusion/logs files? Like when cf is restarted, if nothing else?

I will say that given the tweaks you've had to deal with, we may quickly run out of runway trying to sort things out. 

In that case, and if no one else helps, I'll say this may be a case where eyes on the server with directly may be best. I offer such help, as may others, if it comes to that. Until then, we'll keep on going here. 

[tevansresdat · Accepted solution]

Your last reply got me thinking. No, it wasn't logging anything, even when restarting. Turns out, when the VM got pushed to AWS, it appears to have blown away the directory permissions for the Coldfusion user account. After re-adding those, everything appears to be working now. I appreciate you taking the time to work this through with me. Be safe and stay healthy!