I'm sorry for not asking this first and if you've already done this.
Am I understanding correctly that you're running M-G 2.x?
If so, do you have a local dev environment or dev environment that you can this app up on? One that you're just using? Or in this case, since M-G won't load maybe that doesn't matter since no one else can use dev.
Personally, I'd get that set up on my box. Make sure the exact same error is occuring. Then I'd just drop M-G 3.2 in there and check. Knock on wood, that does the trick.
I get not wanting to upgrade everything, but if you're running into this now with M-G 2 even if you go into the M-G code and fix the issue you're likely to run into similar problems. That's a more than huge jump. M-G 3, not v2, v3, came out back in 2008. The same with ColdFusion 8.
Oh, and that call chain I see ColdSpring. What version are you using? Have you tried the last version of it? Did you turn on debugging in it and/or check the ColdSpring logs to see what error(s) are happening when M-G tries to load?
That said ----
Just to ask the obvious, what error(s) were in the CF logs? One common problem is that something else is erroring out elsewhere but in a way that doesn't stop processing the request. That object not having the method has that sort of smell ( among others ).
Have you turned on debugging in M-G 2? This looks like it's where everything for the framework is being spun up ( instantiated ). That may not work because of that. If it doesn't, you should be able to fiddle around and throw in a line of code there, something like arguments.events.getLog() or .writeLog(),, if I remember correctly ( IIRC ).
I say this because to me it look like the object got created but not in a way that those methods were added. Something went there. Depending on the built in debug in M-G an exactly what it gives you, you may need to stick in your own temp debugging before that line 170. write to a file the object and list out it's functions, too. IIRC you need to write to a file and not use cfdump to screen because CF did, or once did, have some issues with cfdump reflecting methods that have been added to an object after it's been instantiated.
If you don't find anything there, look through what's being fired and make sure to do similar logging in that chain anywhere that it could be instantiating an object or appending methods to an object.
Gues --> My gutt feeling is that since M-G 3 was written for CF8 there's a bad chance that something in there is doing something that CF2016 no longer likes. And at that it sounds like you're running M-G which was _Probably_ written for CF6 or maybe CF7? Anyone know? Can we FedEx Joe Rhinehart a beer to get an answer?
Well it could be that there is now ( as of CF9 that is ) a local scope. I'd still kinda wonder why that didn't get fixed in M-G2, but, it could happen. Or maybe something with the local scope
In short, yuck. But, eh, what else do you do? CF8 was long ago unsupported. And it has to run on an unsupported version of Java. And moving to another framework's an even larger problem.