Repeated segmentation faults on animation playback #88690

New Issue

Michael Macha · 2021-05-30T20:19:24+02:00

Michael Macha commented

2021-05-30 20:19:24 +02:00

System Information
Operating system: Linux Mint 19
Graphics card: GTX 1660

Blender Version
Broken: 2.92
Worked: ??

Short description of error
I'm dealing with a periodic problem in a file which is, unfortunately, proprietary enough to be unsharable at this time; but I do have an error log and plenty of system statistics. Now and then, Blender will crash with a segmentation fault on playback of the animation. Included in the animation are two cloths, three metaballs attached to rigid bodies, a metarig and an additional custom armature (with IK on seven joints), seven different identifiable curves, and probably some other things I've forgotten about because I've clearly been working at it for a while. 502,527 verts, 751,012 faces, 996,278 tris. 402 MB file. For memory, I'm using 1.34/16 GB 2kHz system RAM, and 0.8/6 GB video card RAM (though I have seen it climb to around 1.2); so I know I'm not running out of RAM. While I am of course willing to check additional factors, I have doubts that it's a hardware issue—this is a spacious and well-powered desktop and I even clean this thing physically every three months.

Blender will continue to have issues with playback—though manually skipping frames by clicking on the timeline works fine—every time I reopen it, until I reboot. (It's occurred to me that this implies a kernel module, as resetting X alone does not help the situation. This screams Nvidia, but I'm already on the most recent stable release I know of at v.465.)

Exact steps for others to reproduce the error

This is fairly inconsistent until it happens once, but my error log is as follows.

# Blender 2.92.0, Commit date: 2021-02-24 16:25, Hash 02948a2cab44

# backtrace
./blender(BLI_system_backtrace+0x20) [0x865aed0]
./blender() [0xeb124a]
/lib/x86_64-linux-gnu/libc.so.6(+0x3f040) [0x7fb45646f040]
./blender() [0xf7cd3a]
./blender() [0xf7d406]
./blender() [0xf7d4cb]
./blender() [0xf7ef30]
./blender(BKE_rigidbody_rebuild_world+0x115) [0xf7f755]
./blender() [0x72ab732]
./blender() [0x72ab902]
./blender() [0x1226035]
./blender() [0x865eb13]
./blender() [0x1233735]
./blender() [0x12339eb]
./blender(BLI_task_pool_work_and_wait+0xaa) [0x865e95a]
./blender(_ZN7blender3deg23deg_evaluate_on_refreshEPNS0_9DepsgraphE+0x1e0) [0x72abcb0]
./blender(BKE_scene_graph_update_for_newframe+0xc8) [0xf84048]
./blender() [0x17b1e07]
./blender() [0x123d434]
./blender() [0x1240334]
./blender() [0x12415da]
./blender() [0x12418ec]
./blender(wm_event_do_handlers+0x86a) [0x124252a]
./blender(WM_main+0x20) [0x1238ce0]
./blender(main+0x31e) [0xde8b2e]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xe7) [0x7fb456451bf7]
./blender() [0xeadbe3]

# Python backtrace

Perhaps this means something to someone more familiar with the project, perhaps not? I am happy to provide any other data that is needed, and after the project is ready, the file itself; but we'll have to wait on that and it will likely be a few weeks. I'm aware that 2.93 is right around the corner, and this may well be related to something already resolved; but I'm not sure what to do with those memory references and this feels like the most applicable place to report this.

For what it's worth, I'm using Linux Mint 19 (fully updated) with the XFCE desktop environment. I do not generally have anything special running in the background, aside from maybe Steam at startup and perhaps Chrome (27 tabs, none of them doing anything more special than an occasional YouTube video). If anyone has so much as a suggestion as to how I could get more data about the crash, I would be very grateful!

My current workaround is to build additions to the scene in other files and append them into the main project. I might end up compositing multiple renders together later, too.

**System Information** Operating system: Linux Mint 19 Graphics card: GTX 1660 **Blender Version** Broken: 2.92 Worked: ?? **Short description of error** I'm dealing with a periodic problem in a file which is, unfortunately, proprietary enough to be unsharable at this time; but I do have an error log and plenty of system statistics. Now and then, Blender will crash with a segmentation fault on playback of the animation. Included in the animation are two cloths, three metaballs attached to rigid bodies, a metarig and an additional custom armature (with IK on seven joints), seven different identifiable curves, and probably some other things I've forgotten about because I've clearly been working at it for a while. 502,527 verts, 751,012 faces, 996,278 tris. 402 MB file. For memory, I'm using 1.34/16 GB 2kHz system RAM, and 0.8/6 GB video card RAM (though I have seen it climb to around 1.2); so I know I'm not running out of RAM. While I am of course willing to check additional factors, I have doubts that it's a hardware issue—this is a spacious and well-powered desktop and I even clean this thing physically every three months. Blender will continue to have issues with playback—though manually skipping frames by clicking on the timeline works fine—every time I reopen it, until I reboot. (It's occurred to me that this implies a kernel module, as resetting X alone does not help the situation. This *screams* Nvidia, but I'm already on the most recent stable release I know of at v.465.) **Exact steps for others to reproduce the error** This is fairly inconsistent until it happens once, but my error log is as follows. ``` # Blender 2.92.0, Commit date: 2021-02-24 16:25, Hash 02948a2cab44 # backtrace ./blender(BLI_system_backtrace+0x20) [0x865aed0] ./blender() [0xeb124a] /lib/x86_64-linux-gnu/libc.so.6(+0x3f040) [0x7fb45646f040] ./blender() [0xf7cd3a] ./blender() [0xf7d406] ./blender() [0xf7d4cb] ./blender() [0xf7ef30] ./blender(BKE_rigidbody_rebuild_world+0x115) [0xf7f755] ./blender() [0x72ab732] ./blender() [0x72ab902] ./blender() [0x1226035] ./blender() [0x865eb13] ./blender() [0x1233735] ./blender() [0x12339eb] ./blender(BLI_task_pool_work_and_wait+0xaa) [0x865e95a] ./blender(_ZN7blender3deg23deg_evaluate_on_refreshEPNS0_9DepsgraphE+0x1e0) [0x72abcb0] ./blender(BKE_scene_graph_update_for_newframe+0xc8) [0xf84048] ./blender() [0x17b1e07] ./blender() [0x123d434] ./blender() [0x1240334] ./blender() [0x12415da] ./blender() [0x12418ec] ./blender(wm_event_do_handlers+0x86a) [0x124252a] ./blender(WM_main+0x20) [0x1238ce0] ./blender(main+0x31e) [0xde8b2e] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xe7) [0x7fb456451bf7] ./blender() [0xeadbe3] # Python backtrace ``` Perhaps this means something to someone more familiar with the project, perhaps not? I am happy to provide any other data that is needed, and after the project is ready, the file itself; but we'll have to wait on that and it will likely be a few weeks. I'm aware that 2.93 is right around the corner, and this may well be related to something already resolved; but I'm not sure what to do with those memory references and this feels like the most applicable place to report this. For what it's worth, I'm using Linux Mint 19 (fully updated) with the XFCE desktop environment. I do not generally have anything special running in the background, aside from maybe Steam at startup and perhaps Chrome (27 tabs, none of them doing anything more special than an occasional YouTube video). If anyone has so much as a suggestion as to how I could get more data about the crash, I would be very grateful! My current workaround is to build additions to the scene in other files and append them into the main project. I might end up compositing multiple renders together later, too.

Michael Macha commented

2021-05-30 20:19:24 +02:00

Added subscriber: @Michael_Macha

Philipp Oeser commented

2021-06-01 12:17:31 +02:00

Added subscriber: @lichtwerk

Philipp Oeser commented

2021-06-01 12:17:31 +02:00

Changed status from 'Needs Triage' to: 'Needs User Info'

Philipp Oeser commented

2021-06-01 12:17:31 +02:00

In its current state, this is not something we can investigate further because we lack the time of creating reproduction scenarios ourselves.
So I am afraid we have to put this on hold until we either have reproduction steps that lead to this scenario or a file that shows this issue.

Will set to "Needs Information from User" for the time being but please be sure to share any furhter information to track down the bug/crash.

In its current state, this is not something we can investigate further because we lack the time of creating reproduction scenarios ourselves. So I am afraid we have to put this on hold until we either have reproduction steps that lead to this scenario or a file that shows this issue. Will set to "Needs Information from User" for the time being but please be sure to share any furhter information to track down the bug/crash.

Michael Macha commented

2021-06-01 16:49:47 +02:00

Will do, and I appreciate your attention. As I mentioned, it is likely an Nvidia driver issue and may be fixed with their next update; in the meanwhile I've just been slimming down my file but developing elements of the scene in new projects, and cross-compositing when I need extra details. It's workable, it's just a little strange.

(I'm just naturally assuming that you don't want to dig through a half-gig project file.)

Will do, and I appreciate your attention. As I mentioned, it is likely an Nvidia driver issue and may be fixed with their next update; in the meanwhile I've just been slimming down my file but developing elements of the scene in new projects, and cross-compositing when I need extra details. It's workable, it's just a little strange. (I'm just naturally assuming that you don't want to dig through a half-gig project file.)

Pratik Borhade commented

2021-06-09 13:09:58 +02:00

Added subscribers: @RogerBarr, @PratikPB2123

Pratik Borhade commented

2021-06-09 13:09:58 +02:00

@RogerBarr , any news?

Michael Macha commented

2021-06-12 02:03:45 +02:00

@PratikPB2123 I think you might mean vikinghelmet99. No news yet.It seems like it might be related to the numerous geometry node crashes. It hasn't happened in a while, and I have migrated to 2.93; so there's a chance that it was fixed with something else.

When I see it again, and hopefully have a better notion of what's causing the crash, I'll be sure to either update this or open a new ticket.

@PratikPB2123 I think you might mean vikinghelmet99. No news yet.It seems like it might be related to the numerous geometry node crashes. It hasn't happened in a while, and I have migrated to 2.93; so there's a chance that it was fixed with something else. When I see it again, and hopefully have a better notion of what's causing the crash, I'll be sure to either update this or open a new ticket.

Pratik Borhade commented

2021-06-12 07:08:29 +02:00

Removed subscriber: @RogerBarr

Pratik Borhade commented

2021-06-12 07:08:29 +02:00

When I see it again, and hopefully have a better notion of what's causing the crash, I'll be sure to either update this or open a new ticket.

Thanks for checking, if problem reappears then report here otherwise I will close the report after a week :)

(yes I mean @Michael_Macha, mistakenly added other subscriber :p)

>When I see it again, and hopefully have a better notion of what's causing the crash, I'll be sure to either update this or open a new ticket. Thanks for checking, if problem reappears then report here otherwise I will close the report after a week :) (yes I mean @Michael_Macha, mistakenly added other subscriber :p)

Pratik Borhade commented

2021-06-21 12:35:42 +02:00

Changed status from 'Needs User Info' to: 'Archived'

Pratik Borhade closed this issue

2021-06-21 12:35:42 +02:00

Pratik Borhade commented

2021-06-21 12:35:42 +02:00

Thanks for checking, if problem reappears then report here otherwise I will close the report after a week :)

No activity for more than a week. As per the tracker policy we assume the issue is gone and can be closed.

Thanks again for the report. If the problem persists please open a new report with the required information.

>Thanks for checking, if problem reappears then report here otherwise I will close the report after a week :) No activity for more than a week. As per the tracker policy we assume the issue is gone and can be closed. Thanks again for the report. If the problem persists please open a new report with the required information.

Sign in to join this conversation.

No Label

Download

What's New

Blender Studio

Manual

Developers Blog

Documentation

Benchmark

Blender Conference

Development Fund

One-time Donations

Repeated segmentation faults on animation playback #88690