Frequent crash on deploy in DZ mode

[stephenangelico]

Most times when I get into a Danger Zone match, I can play through the warmup perfectly fine, then choose a deployment position, and then once deployment starts and I'd expect to spawn, the game crashes to desktop. Earlier today (Melbourne time) it happened about every second or third time, but now it happens almost every single time. This was the case both with the 12/7/2018 update and the Late 12/7/2018 update. I have been mainly solo queueing since the latter update, but the crashes have happened on a duos match. This occurs both on my machine and my brother's.

My PC:
Intel i5-4460
32GB DDR3
GTX 1060 6GB
Ubuntu MATE 18.04

Other PC:
Intel i7-6700K
48GB DDR4
GTX 960
Debian Stretch

Edit 20181212: As hypothesized by @Rosuav and tested by many, setting Shader Detail to Low or Medium seems to consistently avoid the issue. The current theory is a bad texture only visible up high around the Radio area.

Edit 20190107: There appears to be another map bug affecting the south area of the map, in the area of Industry/Crane/Silos/Shipping. This causes a crash even on the lowest Shader Detail setting. No workaround has yet been found.

[blogdron]

Driver versions? Steam beta or not?

[stephenangelico]

Driver versions? Steam beta or not?

On my system, 390.87 from ppa:graphics-drivers. I know there's a stable 410.87 and a beta 415.13 but until recently they were in dependency hell. I'll look at upgrading some time unless it's really a problem. But my brother uses version 384.130 from the Debian non-free repository, so it's not a common driver issue between us.

I am on the Steam beta, as is my brother. I just now tested the stable Steam client and got the same crash in-game.

[blogdron]

@stephenangelico thank wait official responce)) next time take info from here System information from steam (Steam -> Help -> System Information) and paste here https://gist.github.com/ this greatly simplifies finding and fixing the problem. 😸 🐛

@kisak-valve, hello bro :D I saw the same thing on the forum https://www.linux.org.ru/forum/games/14657652?cid=14659389

[figuracorta]

same happens to me on macOS since yesterday's #4 update

[mmihon]

@kisak-valve, hello bro :D I saw the same thing on the forum https://www.linux.org.ru/forum/games/14657652?cid=14659389

I don't know english, sorry

green - work ok, red game crashes to desktop https://www.dropbox.com/s/3iezqzjjux6ttil/20181208162056_1.jpg?dl=0

[blogdron]

@mmihon (Приложи ссылкой инфу от сюда (Steam -> Help -> System Information) сюда https://gist.github.com/) Add info from (Steam -> Help -> System Information) to https://gist.github.com/ and paste link here

[mmihon]

@mmihon (Приложи ссылкой инфу от сюда (Steam -> Help -> System Information) сюда https://gist.github.com/) Add info from (Steam -> Help -> System Information) to https://gist.github.com/ and paste link here

https://gist.github.com/mmihon/13b1d61b7f64327814316374d331ca6f

[vanitasvitae]

Same here :/ Really annoying when queing with a mate that is then left alone.
Gist

[ethric]

+1

Here's my system information

I can also confirm @mmihon's observation: some deployment locations work, other's don't. Seems to be fully reproducible.

edit: It did work with the initial version, though. So, it was introduced with one of the day 1 patches.

[d10sfan]

Getting the same thing, for a while in the initial patch, was able to run just fine.

[Unaccounted4]

I too am experiencing this.

Changing setting.mem_level to 0 in ~/.local/share/Steam/userdata//730/local/cfg/video.txt made the game stop crashing. It seems to have introduced some herky jerks even if the FPS readout says it's holding over 100.

Specs: https://gist.github.com/Unaccounted4/96a17b0a97f78356dbebaea9fbe108b7

[DarkArc]

Same thing, I seemed to be fine on danger zone release. Perhaps some of the bug fixes for windows users caused this?

Processor: Ryzen 1800X
GPU: AMD RX 580

[DarkArc]

Though about using gdb to give a back trace... Not sure if having a debugger attached would cause VAC to trigger though; so playing it safe.

[DarkArc]

Anecdotally, I seem to get the crash most frequent around this deployment.

dmesg:

[ 2291.258892] csgo_linux64[11111]: segfault at fffffffffffffff8 ip 00007f0128e1a5d4 sp 00007ffc280eea20 error 7 in materialsystem_client.so[7f0128d93000+17a000]
[ 3765.595069] csgo_linux64[11641]: segfault at fffffffffffffff8 ip 00007f28dc53deba sp 00007ffe93f82100 error 7 in materialsystem_client.so[7f28dc48a000+17a000]
[ 4337.309186] csgo_linux64[12535]: segfault at fffffffffffffff8 ip 00007f69e1c9a05d sp 00007ffeef07f3b0 error 7 in materialsystem_client.so[7f69e1be6000+17a000]

[blogdron]

I assume that this is due to the preloading of the map data, try NOT to CHOOSE the landing site and trust in the automatic distribution of the landing on the map.

Perhaps this will work until the error is corrected. )))))))

[blogdron]

its real Danger Zone 😄

P.S. I do not have this error! Everything works well

[ZortacDev]

Got the same error, crashes when not choosing a starting location, too.

dmesg:

[15945.223171] csgo_linux64[15424]: segfault at fffffffffffffff8 ip 00007f2bf14fe008 sp 00007ffcff53f3e0 error 7 in materialsystem_client.so[7f2bf1459000+17a000]
[16635.789110] csgo_linux64[26158]: segfault at fffffffffffffff8 ip 00007f26bac46a57 sp 00007fff8f041080 error 7 in materialsystem_client.so[7f26bab9f000+17a000]

System information:
https://gist.github.com/ZortacDev/3eb3aa519ad5e21555da4eb2ce5cca32

[rage311]

I'm having the exact same issue as @stephenangelico. It SEEMS to work perfectly for the first game after a reboot, but once a crash occurs, it seems to crash every time after that. I'll do some more testing to see if I can figure out a pattern. Every crash under Linux has been just before deployment and after selecting a drop location.

It's accompanied by the segfault message in dmesg that others have mentioned:
[ 1853.877907] csgo_linux64[4468]: segfault at fffffffffffffff8 ip 00007fe280141eba sp 00007ffe2a644110 error 7 in materialsystem_client.so[7fe28008e000+17a000]

System information: https://gist.github.com/rage311/5c8494e6f942a34407d742eaee06ee4d

Edit for some high level system info:
Arch Linux 64-bit 4.19.4
Nvidia GTX 1070 Ti with driver 415.22
Intel i7-4790k @ stock clocks
16GB DDR3

[ThrosturX]

Just want to chime in, Gas Station always crashes for me. 3 attempts so far.

[ethric]

@rage311 At least for me it's not related to rebooting, as that's (of course ;)) the first thing I tried. It seems to be more related to chosen drop location.

Picnic always works for me, btw.

edit:

next game, Picnic was taken, tried "Hatch" -> crash

dmesg also says:
csgo_linux64[4817]: segfault at fffffffffffffff8 ip 00007f37b896e488 sp 00007ffd405938d0 error 7 in materialsystem_client.so[7f37b88cd000+17a000]

[thomasfinstad]

I too am experiencing this.

Changing setting.mem_level to 0 in ~/.local/share/Steam/userdata//730/local/cfg/video.txt made the game stop crashing. It seems to have introduced some herky jerks even if the FPS readout says it's holding over 100.

Specs: https://gist.github.com/Unaccounted4/96a17b0a97f78356dbebaea9fbe108b7

I tried the same, no help, same crash as all others on deploy.

[29026.976332] csgo_linux64[21789]: segfault at fffffffffffffff8 ip 00007f0bc4704eba sp 00007ffe8f2e4910 error 7 in materialsystem_client.so[7f0bc4651000+17a000]

Spec: https://gist.github.com/penetal/fcf26c338f476c061dfde42818165b35

[Rosuav]

For me, the crash is every 2-4 games, so it's far from consistent - but still plenty annoying. I've successfully landed at Radio in duos, if that helps.

[Rosuav]

There are also occasional crashes as warmup starts, so it's definitely not based entirely on where you drop. (Valve, it'd be great if a crashed member of a squad could rejoin during warmup.)

[winny-]

Same issue here. Crash just after the deploy screen disappears.

System information

dmesg:

[344429.439011] csgo_linux64[4541]: segfault at fffffffffffffff8 ip 00007f15e0dae2f8 sp 00007ffe2bdd1590 error 7 in materialsystem_client.so[7f15e0cf8000+17a000]
[350490.451370] csgo_linux64[6569]: segfault at fffffffffffffff8 ip 00007f494ae35895 sp 00007ffe65d60020 error 7 in materialsystem_client.so[7f494ade8000+17a000]

[jlanzobr]

I have this issue as well. Here's the output from Steam:

Adding process 9156 for game ID 730
assert_20181208235111_1.dmp[9157]: Finished uploading minidump (out-of-process): success = yes
assert_20181208235111_1.dmp[9157]: response: Discarded=1
assert_20181208235111_1.dmp[9157]: file ''/tmp/dumps/assert_20181208235111_1.dmp'', upload yes: ''Discarded=1''
pid 9157 != 9156, skipping destruction (fork without exec?)
Game removed: AppID 730 "", ProcID 8209
No cached sticky mapping in ActivateActionSet.JS method call WebChat.GetOverlayChatBrowserInfo with 1 arguments
JS method call Storage.SetString with 2 arguments

The dump is here: http://lanz.privatedns .org/assert_20181208235111_1.dmp [remove the space]

I run Arch Linux, fully updated as of 2018-12-09, and my system is a Ryzen 2700X with a Vega 64. Full specs: https://gist.github.com/jlanzobr/8c4f906ab00ff15b37b37da8d0ab9364

[devoutharpist]

Hi all, getting the same issue. Here is my gist: https://gist.github.com/devoutharpist/0d8951385a2e35ecf0866ce964570d7f

Seems to work if i choose Picnic... but i have had one crash before i could choose a starting location.

[kiroma]

Well this is odd, in my 6 or so games I have not encountered a single crash.
sysinfo gist

[M-Reimer]

I also have this issue. I already tried the workarounds suggested here.

• Letting the autopick do its job doesn't help
• Changing setting.mem_level to 0 doesn't help

I'm on Arch Linux with an Nvidia GTX1070 and up-to-date drivers.

[mellowagain]

Encounters on my machine as well.
sysinfo

[okaestne]

Same here, sysinfo: https://gist.github.com/2green2freak/8e77f1fa50a134df36a6a04d28a25783

[stephenangelico]

I've had a grand total of one crash since setting Medium shader detail. It was in the Crane area, possibly right on top of the crane. I've updated the initial post to reflect the crashes that have been experienced since the workaround was discovered.

[DarkArc]

Just tested with the update just deployed a few minutes ago, still broken.

[odysei]

Broken for me as well.
Kubuntu 18.10
Nvidia drivers 415.25 (GTX 1060)

[Managor]

I found a second spot where the crash happens every time

[blogdron]

Still not fixed the problem? LOL. Valve come on ))))))))))))))))))

[DarkArc]

Just tested with the update just deployed a few minutes ago, still broken.

[DarkArc]

Just tested with the update just deployed a few minutes ago, still broken.

[sylware]

Just did test the "radio" drop location, could not reproduce the crash.

[sylware]

crash on "radio" drop location happening again:
[38918.858039] csgo_linux64[4259]: segfault at fffffffffffffff8 ip 00007fbd08c4d895 sp 00007ffce250e9e0 error 7 in materialsystem_client.so[7fbd08c00000+17a000]

[DarkArc]

Switched GPUs, went from a RX 580 to an RTX 2080; deployment crash still occurs on the nvidia GPU with the 415 driver.

[nightsky30]

Same type of crash that has happened in DZ Blacksite has occurred twice in a matter of minutes while playing on Arms Race Lake both times.

[ 3588.761432] csgo_linux64[7026]: segfault at fffffffffffffff8 ip 00007fc59ea11b6a sp 00007fff8da65040 error 7 in materialsystem_client.so[7fc59e963000+17a000]
[ 3588.761436] Code: 48 8b 85 b8 fe ff ff 48 89 83 78 01 00 00 48 8d 43 f8 49 8b 94 24 e0 02 00 00 48 85 d2 74 51 48 89 02 49 89 84 24 e0 02 00 00 <48> c7 43 f8 00 00 00 00 48 81 c4 68 01 00 00 5b 41 5c 41 5d 41 5e

[ 3841.786349] csgo_linux64[7976]: segfault at fffffffffffffff8 ip 00007fcf4a79ceba sp 00007fffa3e98820 error 7 in materialsystem_client.so[7fcf4a6e9000+17a000]
[ 3841.786354] Code: 00 49 c7 44 24 28 00 00 00 00 49 89 44 24 30 49 8d 44 24 f8 48 8b 93 e0 02 00 00 48 85 d2 74 50 48 89 02 48 89 83 e0 02 00 00 <49> c7 44 24 f8 00 00 00 00 48 83 c4 18 5b 41 5c 41 5d 41 5e 41 5f

[kisak-valve]

Hello, per "– Increased size of rendering queues for OSX and Linux game clients to address crashes when playing Danger Zone." in the 2019-02-12 CS:GO update, please retest this issue.

[DarkArc]

I've tested several spots, I'm cautiously optimistic this is fixed :)

[nightsky30]

#NotCrashedYet

[Vash63]

Seems fixed for me as well, finally! Great update.

…

On Wed, 13 Feb 2019, 03:01 nightsky30 ***@***.*** wrote: #NotCrashedYet — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1908 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AC8WwcCajnMJtxxKEH0PJQ6qRAR11tZuks5vM3IQgaJpZM4ZJf5n> .

[Aroma1997]

This seems to have fixed it for me as well.

[stephenangelico]

I haven't had a crash yet under the same circumstances as when I originally opened the issue, but I want to do more thorough testing before declaring it fixed. In the short testing I did with @Rosuav, he had a crash (on his own, local server) just as we engaged in a 1v1. This was well after deployment, but we have no log so we don't know if it's related or even reproducible.

[pipebomb]

I've been able to play successfully earlier with the "lower down the graphics settings" workaround before. Just tested with High settings again in a couple of spots that always crashed for me (Silos, Boat Houses, Gas Station) - the crashes seem to no longer occur.
So it's safe to say (at least for me) that the issue has been fixed. It took some time but thanks!

[jaapz]

Thanks Valve for fixing this, and also for being responsive (and patient) on this public bug tracker.

[LiamDawe]

Can we go ahead and close this @kisak-valve ?

[blogdron]

Can we go ahead and close this ?

You can go retest it ))))))))))))))

[Rosuav]

I would love to retest it, if only Danger Zone matches would fire in Australia. Hopefully it'll be possible to find a match on the weekend, but during the week, I have had no luck whatsoever.

If only it were possible to have bots on Blacksite, it would be easier to test...

[kisak-valve]

Thanks for retesting, closing.