AlliedModders

AlliedModders (https://forums.alliedmods.net/index.php)
-   Extensions (https://forums.alliedmods.net/forumdisplay.php?f=134)
-   -   Accelerator - Crash Reporting That Doesn't Suck (https://forums.alliedmods.net/showthread.php?t=277703)

cravenge 07-31-2022 03:44

Re: Accelerator - Crash Reporting That Doesn't Suck
 
Quote:

Originally Posted by hefiwhfcds2 (Post 2785245)
does anyone make sig for l4d2 with linux?

The extension already has a Linux signature for every game. What else are you looking for?

hefiwhfcds2 08-01-2022 04:53

Re: Accelerator - Crash Reporting That Doesn't Suck
 
Quote:

Originally Posted by cravenge (Post 2785263)
The extension already has a Linux signature for every game. What else are you looking for?

In The past, I could view The Last 500 Lines of Console Output using The accelerator,i don't see it now

8guawong 10-22-2022 02:41

Re: Accelerator - Crash Reporting That Doesn't Suck
 
after the latest csgo (windows) update i get the following

Code:

L 10/22/2022 - 14:39:28: [CRASH] Accelerator failed to upload crash dump: The requested URL returned error: 500 (22)
2022-10-22 14:39:28: basic_code_modules.cc:73: ERROR: Module steamclient.dll could not be stored
2022-10-22 14:39:28: basic_code_modules.cc:73: ERROR: Module imagehlp.dll could not be stored
2022-10-22 14:39:28: basic_code_modules.cc:73: ERROR: Module VERSION.dll could not be stored
2022-10-22 14:39:28: basic_code_modules.cc:73: ERROR: Module vstdlib_s.dll could not be stored
2022-10-22 14:39:28: basic_code_modules.cc:73: ERROR: Module tier0_s.dll could not be stored
2022-10-22 14:39:28: basic_code_modules.cc:73: ERROR: Module Secur32.dll could not be stored
2022-10-22 14:39:28: basic_code_modules.cc:73: ERROR: Module SSPICLI.DLL could not be stored

is this accelerator related?

cravenge 10-22-2022 04:49

Re: Accelerator - Crash Reporting That Doesn't Suck
 
Looks like the website is down at the moment.

xiaoli 12-09-2022 11:27

Re: Accelerator - Crash Reporting That Doesn't Suck
 
would it be to possible to make stack trace information more detailed when crash is triggered by watchdog?
for example by having accelerator display last stacktrace before process jumped to watchdog, as it'd be helpful if we could see what function server got stuck at before watchdog got triggered :bee:

https://i.imgur.com/MW0UVla.png

asherkin 12-09-2022 12:19

Re: Accelerator - Crash Reporting That Doesn't Suck
 
If it can be recovered it’ll be displayed further down the stack trace, but generally it can not be due to how the watchdog is implemented.

eyal282 12-26-2022 17:30

Re: Accelerator - Crash Reporting That Doesn't Suck
 
https://crash.limetech.org/6z7xpj2lvyyu

I can't say I like calling spawning the 2049th entity the same as "This crash was caused by a plugin trying to spawn an incorrectly configured entity."

Wasted a lot of time believing I'm improperly messing with dispatchkeyvalue.

AdRiAnIlloO 12-27-2022 11:00

Re: Accelerator - Crash Reporting That Doesn't Suck
 
Quote:

Originally Posted by xiaoli (Post 2794847)
would it be to possible to make stack trace information more detailed when crash is triggered by watchdog?
for example by having accelerator display last stacktrace before process jumped to watchdog, as it'd be helpful if we could see what function server got stuck at before watchdog got triggered :bee:

https://i.imgur.com/MW0UVla.png

I'm fairly sure your crash could be related to my patched CollisionHook extension, per the update I posted on the thread. I got the exact same stack trace after running my patched version for +1 week without restarting server. See here.

A1mDev 01-15-2023 16:05

Re: Accelerator - Crash Reporting That Doesn't Suck
 
It is not related to your extension, because i built an empty extension and got a crash too. Apparently the compiler is to blame, it creates debug symbols that the accelerator extension cannot read. I don't even know what to do in this case? Downgrade compiler? Specify flag '-gdwarf-4'? ... Of course if you use the new compiler, extension accelerator needs to be updated anyway.

Note1: The compiler option '-gdwarf-4' solves this problem of course, but I don't think it's a good idea.
Note2: gdb version 9.2 also fails to read some debug symbols (Dwarf Error: DW_FORM_strx1 found in non-DWO CU).

bakugo 03-07-2023 05:31

Re: Accelerator - Crash Reporting That Doesn't Suck
 
Any plans to fix the crash on dump upload on certain distros, or is this unmaintained?


All times are GMT -4. The time now is 22:52.

Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.