Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
Tags

[it's a mystery] Extension not working? Hit unreadable memory at side=1

A topic by Carson K. created Apr 07, 2019 Views: 342 Replies: 3
Viewing posts 1 to 4
(1 edit)

I followed this tutorial exactly as is: https://medium.com/@meseta/releasing-gamemaker-games-automating-bug-reports-like-a-pro-part-1-b2f388215e0e

And when I encounter a crash, I see normal GMS2 behavior. When adding "crash_error_init();" to my game start it made it so once the game crashed it just closed the game instantly, without running the sentry code (Yes I tried putting a message box or something else in there and they don't trigger either so I don't think it's a problem with GMSentry). When trying to set both fatal and normal modes to _queue, I still see the same results.

If it helps, here's what I'm seeing in console: https://pastebin.com/pLZ7TUgh

The first 7 lines seem like catch_error errors, and then the error I occurred printed out 4 times.

Not sure if I'm doing something wrong or if there's an issue with the extension but I'm posting here in hopes of getting some help :)

Developer (1 edit)

Did you flag errors to be caught as per documentation? e.g.

catch_error_init();
catch_error_set_normal(catch_error_normal_queue);
catch_error_set_fatal(catch_error_fatal_queue);

Edit: You seem to be. What runtime version, and could you assemble a small sample project that reproduces this? Don't think I've seen "malformed variable" errors outside of literal memory corruption

(1 edit)

IDE v2.2.2.413 Runtime v2.2.2.326

It seems to work fine in a new project. I think it's my own game (or this crash in specific) that's causing problems.
After getting var s and checking if s != "", I have a simple line that checks for a key press and just does "yeet = yote" (yote is not a real variable), and the game didn't crash and instead ran the sentry code.

The crash that I encountered in the original post was in a different room called from a different object, does that have anything to do with it? The object that contains the catch_error code is persistent, so it's not like it's going away, but the crash occurred after entering a different room and was caused by a completely different object so I'm not quite sure what the problem is :/

EDIT: It seems to work fine now actually, it's just the one bug in particular that just crashes the program instantly (I'm referring to the one in the OP and pastebin) 

Developer

I can't get that specific error to pop up no matter what so I can only assume that it is a combination of factors causing GM to end up in an unusual state after something goes wrong - cases like this will get rarer as GMS2 moves toward built-in exception support (recently announced), which will leave the extension to catch anything that the game code didn't.