Special K

Copy, thanks. Please keep us posted, if it happens on any other mission.

Does this happen on a specific server (if yes, which one) or is any of you guys hosting the mission and if yes, which of you? Does it happen on different maps or just one? Have you tried the SC, too, or is it just the Stennis?

TEMP folder was deleted, too. Still crashes.

What you can do in addition is running the attached cmd. That is checking the registry and the most common files in Windows (I created that cmd for convenience, you can run the commands by yourself if you look into it). After it ran, you should do a reboot. So maybe do that in one of your maintenance windows. Has to be run as Administrator. It does not do any harm, just standard windows commands. check_integrity.cmd

Ok, got it. I don't see any issues in your log. Try the realtek thing. D3W had similar issues, iirc that fixed it for them.

If users have issues with disconnects, etc. I'd recommend that they check three main things: a) RAM If they have below 32 GB of RAM they should make sure that they have a pagefile of at least 16 GB. The preload radius should not exceed 30.000 b) HDD vs SSD You all know it already, but it is true for the pagefile, too. Especially with low RAM systems, people should check that their game and pagefile is on an SSD. c) Antivirus Solutions They shall make sure that they have excluded at least the game directory, best case even Saves Games\DCS(.openbeta) from their AV read scan. My bad, rechecked the version. Sorry for that.

Do you guys have any realtek software on that server? If yes, I assume you don't need it, so please deinstall it. I've seen a similar issue with another server recently.

I've had a look at the user logfiles. a) dcs.log That user has lots of mods that are broken. Besides that, they don't have much RAM and a pagefile that is likely not large enough. I expect them to have paging issues. b) dcs(1).log That user has an even smaller pagfile. I would expect them to get even out of memory errors, if they play for a longer time. Might be the same issue here - disconnects due to insufficient hardware. c) dcs(2).log That one is interesting, and I would ask you for the user name. That is no official DCS version but a closed beta. It had known issues. And it is no log to provide in public. => my bad, version was correct, all good

Honestly the report of that Nvidia user can be anything. If they are the single needle in the haystack, it's more likely that it is something completely different. I've seen lots of complaints in the past month and that is the only user with an Nvidia card claiming that issue. Don't get me wrong, I fully agree that this issue is very painful. I don't agree though that a software vendor necessarily has to work around driver bugs, if this is any. That's still an assumption as it at least seems to happen 99% with AMD cards (if we count that user). But all of that doesn't help. The logs do not say anything at that point. Absolutely no issue. That is even more pointing into a driver direction as DCS does not seem to get any error back from the DX API. So to get things forward - does anyone know any way to enable some debug mode in the driver? Debug logging etc? With that, we could try to track it down, if some guys could enable it and send their logs after the issue happens.

Do you have a log of any Nvidia user reporting the same issue? I have not seen any.

As DCS is using Direct X and it is only happening on AMD cards, it's most likely their implementation of Direct X that is wrong at some point. Unfortunately, ED can't fix issues in the AMD graphics drivers. So if they - by chance - find out, what exactly is causing the issue, there might be a workaround. But that is unclear. Most likely we will see another AMD patch in the near future and the issue vanishes with that.

Would you guys please update your AMD Radeon driver to the latest one, which should be the following and try again? I have positive reports that the issue is vanished with this update. Radeon Software Adrenalin 21.8.2 Driver Version 21.20.25.01 for Windows® 10 (Windows Driver Store Version 27.20.22025.1006).

Seems to be the same thing:

Not sure, if that is related, but we see the following errors in the logs. As it was an eject, I'm pretty sure it was not a static in this case. 2021-04-30 09:32:55.377 INFO Scripting: event:type=eject,target=,t=30085.501,initiatorMissionID=1004162,targetMissionID=0, 2021-04-30 09:32:55.377 INFO Scripting: event:type=kill,weapon=AIM-120C-5,target=F-4E,t=30085.501,initiatorMissionID=1004132,targetMissionID=1004162, 2021-04-30 09:32:55.377 ERROR Lua::Config: Call error world.onEvent:[string "-- world events hook..."]:461: Unit doesn't exist stack traceback: [C]: ? [C]: in function 'getName' [string "-- world events hook..."]:461: in function <[string "-- world events hook..."]:440>. 2021-04-30 09:32:56.367 INFO Scripting: event:type=eject,target=,t=30086.501,initiatorMissionID=1004162,targetMissionID=0, 2021-04-30 09:32:56.367 ERROR Lua::Config: Call error world.onEvent:[string "-- world events hook..."]:461: Unit doesn't exist stack traceback: [C]: ? [C]: in function 'getName' [string "-- world events hook..."]:461: in function <[string "-- world events hook..."]:440>. 2021-04-30 09:32:57.820 INFO Scripting: event:weapon=SM-2,t=30087.948,type=shot,initiatorMissionID=3,

philstyle reported an issue lately this evening, when people that try to join got a client crash. Nobody was ably to join anymore. We have the crash logs from 2 ppl, both of them have the exact same trace in it, so here is just one: 2021-04-26 21:16:24.962 INFO EDCORE: # -------------- 20210426-211625 -------------- 2021-04-26 21:16:24.963 INFO EDCORE: DCS/2.7.0.5118 (x86_64; Windows NT 10.0.19041) 2021-04-26 21:16:24.964 INFO EDCORE: C:\Steam\steamapps\common\DCSWorld\bin\DCS.exe 2021-04-26 21:16:24.965 INFO EDCORE: # C0000005 ACCESS_VIOLATION at 07F65FBB 00:00000000 2021-04-26 21:16:24.969 INFO EDCORE: SymInit: Symbol-SearchPath: '.;C:\Steam\steamapps\common\DCSWorld;C:\Steam\steamapps\common\DCSWorld\bin;C:\WINDOWS;C:\WINDOWS\system32;SRV*C:\websymbols*https://msdl.microsoft.com/download/symbols;', symOptions: 530, UserName: 'user0' 2021-04-26 21:16:24.970 INFO EDCORE: OS-Version: 10.0.19041 () 0x300-0x1 2021-04-26 21:16:25.701 INFO EDCORE: 0x00000000007B5FBB (DCS): SW + 0x4C560B 2021-04-26 21:16:25.702 INFO EDCORE: 0x00000000008EF8AD (DCS): SW + 0x5FEEFD 2021-04-26 21:16:25.702 INFO EDCORE: 0x000000000093508C (DCS): SW + 0x6446DC 2021-04-26 21:16:25.702 INFO EDCORE: 0x0000000000939D88 (DCS): SW + 0x6493D8 2021-04-26 21:16:25.702 INFO EDCORE: 0x00000000000A6CE1 (WorldGeneral): slot_decompose_ex + 0xCF1 2021-04-26 21:16:25.703 INFO EDCORE: 0x00000000000AADE7 (WorldGeneral): WorldManager::notifyBirth + 0x37 2021-04-26 21:16:25.703 INFO EDCORE: 0x00000000000CD786 (Weapons): wBomb::netCreate + 0x2B6 2021-04-26 21:16:25.703 INFO EDCORE: 0x00000000000DD274 (Weapons): weaponInitRandomSeed + 0x1494 2021-04-26 21:16:25.703 INFO EDCORE: 0x0000000000100743 (edCore): Mail::Reader::stream + 0x1473 2021-04-26 21:16:25.703 INFO EDCORE: 0x00000000009A4D4F (DCS): SW + 0x6B439F 2021-04-26 21:16:25.703 INFO EDCORE: 0x00000000009A5011 (DCS): SW + 0x6B4661 2021-04-26 21:16:25.703 INFO EDCORE: 0x00000000009D2720 (DCS): SW + 0x6E1D70 2021-04-26 21:16:25.703 INFO EDCORE: 0x00000000009B96E9 (DCS): SW + 0x6C8D39 2021-04-26 21:16:25.703 INFO EDCORE: 0x000000000098F7C2 (DCS): SW + 0x69EE12 2021-04-26 21:16:25.704 INFO EDCORE: 0x00000000007FF2C0 (DCS): SW + 0x50E910 2021-04-26 21:16:25.704 INFO EDCORE: 0x00000000007D02A4 (DCS): SW + 0x4DF8F4 2021-04-26 21:16:25.704 INFO EDCORE: 0x00000000007D067D (DCS): SW + 0x4DFCCD 2021-04-26 21:16:25.704 INFO EDCORE: 0x0000000001C2ACAB (DCS): AmdPowerXpressRequestHighPerformance + 0xE3FCA7 2021-04-26 21:16:25.704 INFO EDCORE: 0x0000000000A067DE (DCS): SW + 0x715E2E 2021-04-26 21:16:25.704 INFO EDCORE: 0x0000000000017034 (KERNEL32): BaseThreadInitThunk + 0x14 2021-04-26 21:16:25.704 INFO EDCORE: 0x0000000000052651 (ntdll): RtlUserThreadStart + 0x21 2021-04-26 21:16:26.218 INFO EDCORE: Minidump created. --------- Server side looks good so far (times are a bit off): 2021-04-26 21:27:17.020 INFO NET: accepting connection from 188.194.226.232:53311 2021-04-26 21:27:17.033 INFO NET: adding client[7] 2021-04-26 21:27:17.385 INFO NET: added client[7] name=No.54 Philstyle addr=188.194.226.232:53311 2021-04-26 21:27:17.392 INFO LuaNET: SLMOD INFO: connected client 7 2021-04-26 21:27:17.392 INFO LuaNET: SLMOD INFO: num clients: 23 2021-04-26 21:27:18.472 INFO NET: client[7] UDP port changed from 53311 to 53087 ----- As the user count dropped to about 10, people were able to connect again.