-
-
Notifications
You must be signed in to change notification settings - Fork 413
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Discord Rich crash #1349
Comments
It's no longer instantly crashing for me on either of the offsets, but according to crash stats some users still crash on the offset that peaked yesterday (below is from r20496, client.dll @ 0017e4c0): Dumptrace
It's not easy to explain this, but since I have knownledge on nightly build distribution rates (and the speed these crashes appear after lifting the nightly channel) i can safely say this:
But now, just 3 minutes after briefly lifting the nightly auto-update channel, someone had the exact same crash.. we won't be collecting further stats though, because the lift was reverted for this reason (it could become worse again in a matter of hours, if the crash risk again is elevated) but mostly due to another crash I posted in #development - if neccesary, we can raise nightly and collect stats after said disconnect crash is resolved by LopSided - but this "3 minutes after nightly push" thing probably means you can speculate it's not as isolated as that "single victim on r20488", and it would grow if we kept updating nightly users. In that case, the crash risk is still elevated. So it could be like this:
I mean, maybe you just need to give it a deeper thought: why can this happen.. even should account for the 'odd' case, no one should really crash like this. I read your comment in #dev that it's a point something isn't initialized yet.. so then think: when can the same scenario apply? I'm just afraid that Discord Rich integration will be the same type of bottleneck to MTA stability as CEF has been for a long time.. I am also seeing various other crashes on Discord module and core.dll (similar offsets), they are for later and I don't have time to debug and report them all. But i am afraid of stability yeah. Maybe needs a general safety/checks in code review. |
Describe the bug
There are 2 new crashes caused by Discord Rich implementation (or recent changes, i suspect a4f447a since i reproduced it twice on this build - MTA can barely be ran and is very unstable)
To reproduce
Launch MTA, and idle in the server browser (i crashed on offset 0017e67d) or join a server (i crashed on offset 0017e4c0 after a short while)
Version
Client: 1.5.7-r20492
Additional context
Dumptrace:
Crash 1 (offset 0017e67d)
Crash 2 (offset 0017e4c0)
Dumps: https://upload.mtasa.com/u/916651105/dumps.zip_ (can only be accessed by MTA team, which isn't a problem anyways since contributors won't have release symbols either.. assuming that @0x416c69 will be working on this issue, you will really need to reproduce it.. or judge by the stack trace i provided. I reproduced it constantly.. should be pretty easy, the latest nightly is really, really unstable..)
Note: discussion of this issue in #development started after this message: https://discordapp.com/channels/278474088903606273/366384007535001612/696624285678043136
The text was updated successfully, but these errors were encountered: