reported Random "RemoveLeaf" CTD playing in MP

[Dread091]

Hi!

A friend and I had a CTD in our server when we flight the Mi-24.

If it could be useful: dedicated server runs a Pretense dynamic mission + Splash Damage script, SRS.

 

Zip files are from my side. dcs.log from my friend.

I see that the dump information are pretty similar but I'm not experienced enough to find the issue.

My friend believe that the issus come from SRS and uninstalled it. The problem hasn't reappeared but I don't think it's related.

 

I have the impression that it is linked to an effect that is launched by the mission.

Thanks in advance.

dcs.log-20240418-190556.zip dcs.log-20240418-145812.zip dcs.log

[Flappie]

You have a bunch of mods and some of them cause errors. Do the crashes happen when you disable them?

2024-04-07 14:20:21.015 ALERT   Scripting (Main): loadfile: can't open 'C:\Users\franc\Saved Games\DCS\Mods/tech/SH-70B Seahawk Folded/Database/db_Seahawk_objects.lua'
2024-04-07 14:20:21.020 ALERT   Scripting (Main): loadfile: [string "C:\Users\franc\Saved Games\DCS\Mods/tech/Virpil/VPC_RAF_Car.lua"]:6: '=' expected near '-'
2024-04-07 14:20:21.021 INFO    Scripting (Main): OBSOLETE mount_vfs_sound_path() used in C:\Users\franc\Saved Games\DCS\Mods/aircraft/A-4E-C/entry.lua
2024-04-07 14:20:21.059 ERROR   Scripting (Main): [string "CoreMods/aircraft/AircraftWeaponPack/AS_Missiles.lua"]:25: attempt to call global 'ContainerWithSubshapes' (a nil value)

 

2024-04-07 14:20:22.316 INFO    SOUND (Main): Added sound path: C:\Users\franc\Saved Games\DCS\Mods/aircraft/A-4E-C/Sounds
2024-04-07 14:20:22.355 INFO    SOUNDER (Main): Loaded 4 sounders.
2024-04-07 14:20:22.392 INFO    SOUND (Main): Added sound path: C:\Users\franc\Saved Games\DCS\Mods/aircraft/Hercules ver 6.8.2/Sounds
2024-04-07 14:20:22.412 INFO    SOUNDER (Main): Loaded 1 sounders.
2024-04-07 14:20:22.426 INFO    SOUND (Main): Added sound path: C:\Users\franc\Saved Games\DCS\Mods/aircraft/Su-30_EFM_V2.7.54b/Sounds
2024-04-07 14:20:22.459 INFO    SOUNDER (Main): Loaded 7 sounders.
2024-04-07 14:20:22.505 INFO    SOUND (Main): Added sound path: C:\Users\franc\Saved Games\DCS\Mods/aircraft/UH-60L/Sounds
2024-04-07 14:20:22.527 INFO    SOUNDER (Main): Loaded 6 sounders.

 

[Dread091]

Thanks for your answer.

Worth a try to disable some mods on the server.

Is it also necessary to remove them from the Saved Games folder (client side)?

Edited April 19, 2024 by Dread091

[Flappie]

Yes, they should be disabled on the client side so DCS does not load them at all.

[Dread091]

I think we disable all mods and same error yesterday.

This time I was in an F-14. CTD 10 seconds after I select slot.

 

 

dcs.20240425-193638.log

Edited April 26, 2024 by Dread091

[Flappie]

Some shaders break IC which means some files were altered:

2024-04-25 18:08:07.743 WARNING SECURITYCONTROL (15248): IC fail: /shaders/enlight/cirrus.fx @ bazar/shaders
2024-04-25 18:08:07.834 WARNING SECURITYCONTROL (15248): IC fail: /shaders/particlesystem2/lightning.fx @ bazar/shaders
...
2024-04-25 18:08:08.530 WARNING SECURITYCONTROL (15248): IC fail: config/effects/clouds.lua
...
2024-04-25 18:08:09.092 WARNING SECURITYCONTROL (15248): IC fail: config/effects/particlesystem2/lightning.lua
...
2024-04-25 18:10:25.753 WARNING SECURITYCONTROL (25928): IC fail: /textures/grass_n_4 @ bazar/textures/buildings_sum.zip

Try a DCS cleanup and repair (don't forget the cleanup step, and choose the "slow" repair mode).

Edited April 26, 2024 by Flappie

[Dread091]

Repair done but just had a new CTD. Seems the same error.

dcs.log

[Flappie]

Do you use T.A.R.G.E.T. scripts? See this thread (same symptoms):

Another user had the same error, and many others. In this case, the RAM was faulty:

 

[Dread091]

No I don't have any T.A.R.G.E.T. scripts.

I run a memory test from Windows 11 app and everything seems OK.

Edited April 27, 2024 by Dread091

[Dread091]

And here the log from my friend.

I see that we have pretty much the same pattern.

First this line:

2024-04-27 14:49:47.605 WARNING EFFECTS2 (Main): OParticleSystem: preset 'smoke' not found in 'speedSmoke.lua', using default

And few minutes later crash with these lines:

2024-04-27 14:51:20.582 INFO    EDCORE (14832): 0x00000000000272a6 (edCore): b2DynamicTree::RemoveLeaf + 0xC6
2024-04-27 14:51:20.583 INFO    EDCORE (14832): 0x0000000000026c19 (edCore): b2DynamicTree::MoveProxy + 0xC9
2024-04-27 14:51:20.585 INFO    EDCORE (14832): 0x000000000003e2aa (Effects): Effects::PipeSmokeProviderProxy::setSmokeDensity + 0x650A
2024-04-27 14:51:20.585 INFO    EDCORE (14832): 0x000000000003edee (Effects): Effects::PipeSmokeProviderProxy::setSmokeDensity + 0x704E
2024-04-27 14:51:20.585 INFO    EDCORE (14832): 0x0000000000161dcf (Visualizer): smSceneManager::DestroySceneManager + 0x1147F
2024-04-27 14:51:20.585 INFO    EDCORE (14832): 0x00000000001671b8 (Visualizer): smSceneManager::regLua + 0x4FC8
2024-04-27 14:51:20.586 INFO    EDCORE (14832): 0x000000000014e96c (Visualizer): smCamera_Implement::getClipRegion + 0x19CAC
2024-04-27 14:51:20.586 INFO    EDCORE (14832): 0x000000000013a08c (Visualizer): smCamera_Implement::getClipRegion + 0x53CC
2024-04-27 14:51:20.586 INFO    EDCORE (14832): 0x0000000000151b72 (Visualizer): smSceneManager::DestroySceneManager + 0x1222
2024-04-27 14:51:20.586 INFO    EDCORE (14832): 0x00000000000339c1 (edCore): ed::thread::_get_current_thread_id + 0x71
2024-04-27 14:51:20.586 INFO    EDCORE (14832): 0x0000000000029333 (ucrtbase): _recalloc + 0xA3
2024-04-27 14:51:20.586 INFO    EDCORE (14832): 0x000000000001257d (KERNEL32): BaseThreadInitThunk + 0x1D

 

dcs.log

Edited April 27, 2024 by Dread091

[Flappie]

Windows RAM utility is known to miss RAM issues. If you want to be sure, you should use MemTest86.

 

But since you say your friend is having the same issue, RAM may not be the issue here (unless you are synchronized and unlucky people).

Does it only happen with the Pretense mission?

[Dread091]

We're attempting a free flight mission without any units (with SRS only), and so far, there have been no crashes. Nevertheless, in my opinion, conducting a test without AI units, explosions, or missile/bomb shots doesn't seem representative of how we typically use DCS.

I understand that DCS is a complex environment and I am not trying to discredit ED. I know you want to rule out all possible variables (I work in IT, I will do the same)
Maybe Pretense is the problem but if we don't have crashes outside of Pretense, it would be too simple to say that the issue comes from there.

I'd like to mention that there are approximately ten pilots flying on Pretense, and two of us have experienced recurring random crashes (possibly a third—I'm currently recovering the log).

 

The randomness of this crash makes debugging difficult. It can occur in the first minutes after selecting a slot or after 30 minutes - an hour of flight and sometimes even a day without meeting it.

 

Could enabling the full log tell us more?

 

[Maverick87Shaka]

1 hour ago, Dread091 said:

Could enabling the full log tell us more?

 

Did you perform the full Memtest on your system as suggested? Is not worth try to find something else if you didn't check this.

[Dread091]

[Flappie]

I asked about Pretense because there was a very annoying crash case that was mostly appearing in Pretense missions. It would happen when respawning. This got fixed.

But maybe there's more to it. My goal is to reproduce the crash on my computer. So if you tell me that it only happens on Pretense, I'll go there and try.

[Dread091]

As far as I remember, yes always happend on Pretense.

I've just flight for 2 hours this evening on an empty mission with 3 friends and everything's OK.

Edited April 29, 2024 by Dread091

[Dread091]

Oh I forgot to mention that we are on Pretense Syria.

We are trying a mission with some more units but not Pretense yet.

 

Would it be useful for you if you have access to our server when we retried Pretense?

[Flappie]

10 hours ago, Dread091 said:

Would it be useful for you if you have access to our server when we retried Pretense?

Yes, it would. I'd be happy to accompany you in a Mi-24P. This way I'd have more chances to reproduce the crash, flying where you guys fly. I'm GMT+1 by the way.