jlesage
Crash when opening a file
I've installed your container, but when i try to load a video it just crashes. I've tried some mp4s and webms with no luck. Any idea why?
https://user-images.githubusercontent.com/61975168/223448772-b971cafc-1985-4abc-a906-de44bd00988c.mp4
Log file: _handbrake-handbrake-1_logs.txt
Also the permissions on the volumes folders should be ok (container running with pid and gid 1000 which is the same for the folders)
I have the same exact problem. The same log output. The automated encoding from /watch folder works for me, but not if I press and use the web GUI. Have you tried this?
I am also experiencing this restarting problem when opening a file through the UI. It always seems to crash when generating the 9th of 10 preview images. The last working version for me is v1.25.1. Version v23.02.1 and newer seem to be affected.
Can you check the log file located at ghb/Activity.log* under the directory you mapped to /config?
Do you see the crash with all video files ? Or some are working ?
Hello,
Same problem here : proxmox/ubuntu/handbrake -> handbrake goes wild and crash on any opened source. The only "logs" i could find relative to the crash is this at the container level (same as the others) :
[supervisor ] all services started. [xvnc ] Mon May 15 20:20:43 2023 [xvnc ] Connections: accepted: /tmp/vnc.sock [xvnc ] SConnection: Client needs protocol version 3.8 [xvnc ] SConnection: Client requests security type None(1) [xvnc ] VNCSConnST: Server default pixel format depth 24 (32bpp) little-endian rgb888 [xvnc ] VNCSConnST: Client pixel format depth 24 (32bpp) little-endian bgr888 [xvnc ] ComparingUpdateTracker: 0 pixels in / 0 pixels out [xvnc ] ComparingUpdateTracker: (1:-nan ratio) [xvnc ] ComparingUpdateTracker: 0 pixels in / 0 pixels out [xvnc ] ComparingUpdateTracker: (1:-nan ratio) [app ] libdvdread: Can't open file VIDEO_TS.IFO. [supervisor ] service 'app' exited (got signal SIGSEGV). [supervisor ] service 'app' exited, shutting down... [supervisor ] stopping service 'autovideoconverter'... [supervisor ] service 'autovideoconverter' exited (got signal SIGTERM). [supervisor ] stopping service 'openbox'... [supervisor ] service 'openbox' exited (with status 0). [supervisor ] stopping service 'nginx'... [supervisor ] service 'nginx' exited (with status 0). [xvnc ] Mon May 15 20:20:52 2023 [xvnc ] VNCSConnST: closing /tmp/vnc.sock: Clean disconnection [xvnc ] EncodeManager: Framebuffer updates: 160 [xvnc ] EncodeManager: Tight: [xvnc ] EncodeManager: Solid: 208 rects, 14.8916 Mpixels [xvnc ] EncodeManager: 3.25 KiB (1:17899.4 ratio) [xvnc ] EncodeManager: Bitmap RLE: 19 rects, 15.506 kpixels [xvnc ] EncodeManager: 621 B (1:100.245 ratio) [xvnc ] EncodeManager: Indexed RLE: 337 rects, 285.878 kpixels [xvnc ] EncodeManager: 59.1465 KiB (1:18.9472 ratio) [xvnc ] EncodeManager: Tight (JPEG): [xvnc ] EncodeManager: Full Colour: 360 rects, 2.64658 Mpixels [xvnc ] EncodeManager: 1.36172 MiB (1:7.41709 ratio) [xvnc ] EncodeManager: Total: 924 rects, 17.8396 Mpixels [xvnc ] EncodeManager: 1.42325 MiB (1:47.8225 ratio) [xvnc ] Connections: closed: /tmp/vnc.sock [xvnc ] ComparingUpdateTracker: 18.8226 Mpixels in / 9.60027 Mpixels out [xvnc ] ComparingUpdateTracker: (1:1.96063 ratio) [supervisor ] stopping service 'xvnc'... [xvnc ] ComparingUpdateTracker: 0 pixels in / 0 pixels out [xvnc ] ComparingUpdateTracker: (1:-nan ratio) [supervisor ] service 'xvnc' exited (with status 0). [finish ] executing container finish scripts... [finish ] all container finish scripts executed.
No nginx error log, neither for hb.
Regards
After some more testing :
There is a problem on QEMU/KVM (proxmox).
If the docker host processors type is set to any QEMU type, like :
-> it will crash.
Resolved (in my case) by setting the "real" processor type (intel/amd/...).
Hope it might help others.
I was experiencing the same issue and @Ysval 's fix worked for me.
My container server is running in proxmox VM and was set to the default processor type of kvm64. Switching it to type "host" fixed the issue.
I also am running in a Proxmox VE VM. After changing the processor type from kvm64 to host, the problem is gone.
Thanks everyone!
Would ofc. still be preferable to not have a crash be expected behaviour ;)
Same problem/symptoms...
I'm trying to manually process a single MKV file by "Opening Source" (not the watch folder).
Handbrake scans the file, then disconnects/exits/crashes.
When looking at the log, it appears Handbrake exits when trying to process a non-existing VIDEO_TS.IFO file. There's nothing in my watch folder.
[autovideoconverter] Processing watch folder '/watch'...
[autovideoconverter] Watch folder '/watch' processing terminated.
**[app ] libdvdread: Can't open file VIDEO_TS.IFO.**
[supervisor ] service 'app' exited (got signal SIGSEGV).
[supervisor ] service 'app' exited, shutting down...
[supervisor ] stopping service 'autovideoconverter'...
[supervisor ] service 'autovideoconverter' exited (got signal SIGTERM).
[supervisor ] stopping service 'openbox'...
[supervisor ] service 'openbox' exited (with status 0).
[supervisor ] stopping service 'nginx'...
[supervisor ] service 'nginx' exited (with status 0).
[xvnc ] Thu Jul 6 14:07:26 2023
[xvnc ] VNCSConnST: closing /tmp/vnc.sock: Clean disconnection
[xvnc ] EncodeManager: Framebuffer updates: 296
[xvnc ] EncodeManager: Tight:
[xvnc ] EncodeManager: Solid: 224 rects, 7.35295 Mpixels
[xvnc ] EncodeManager: 3.5 KiB (1:8207.17 ratio)
[xvnc ] EncodeManager: Bitmap RLE: 25 rects, 13.132 kpixels
[xvnc ] EncodeManager: 780 B (1:67.7282 ratio)
[xvnc ] EncodeManager: Indexed RLE: 304 rects, 274.428 kpixels
[xvnc ] EncodeManager: 61.9863 KiB (1:17.3514 ratio)
[xvnc ] EncodeManager: Tight (JPEG):
[xvnc ] EncodeManager: Full Colour: 285 rects, 1.76825 Mpixels
[xvnc ] EncodeManager: 1.07826 MiB (1:6.2588 ratio)
[xvnc ] EncodeManager: Total: 838 rects, 9.40876 Mpixels
[xvnc ] EncodeManager: 1.14296 MiB (1:31.4108 ratio)
[xvnc ] Connections: closed: /tmp/vnc.sock
[xvnc ] ComparingUpdateTracker: 17.457 Mpixels in / 6.74748 Mpixels out
[xvnc ] ComparingUpdateTracker: (1:2.58719 ratio)
[supervisor ] stopping service 'xvnc'...
[xvnc ] ComparingUpdateTracker: 0 pixels in / 0 pixels out
[xvnc ] ComparingUpdateTracker: (1:-nan ratio)
[supervisor ] service 'xvnc' exited (with status 0).
[finish ] executing container finish scripts...
[finish ] all container finish scripts executed.
Are you running Docker in a VM ?
Via Unraid/Portainer.
AMD Athlon II X4 635 linux x86_64 Slackware 15.0 x86_64 (post 15.0 -current)
Is unRAID running in a VM ?
No, unRAID is not running inside a VM. I'm running bare metal.
Are you running Docker in a VM ?
Any ideas on this issue? I am running Docker in a VM with "Default (kvm64)" set due to a complicated QSV issue.
I'm having the exact same symptoms as the people above.
Tdarr and Jellyfin are not having this issue when encoding using ffmpeg, suggesting that there's another issue with this particular container.
EDIT: Linking this likely-related issue: https://github.com/jlesage/docker-handbrake/issues/250
Any ideas on this issue?
Yes. A workaround for your concrete issue exists. A fix does not.
a VM with "Default (kvm64)" set due to a complicated QSV issue
I'm sorry, but you just explained why this is happening - it has already been pointed out in this thread. Expected behaviour with this configuration is currently a hard crash on file open. The only known workaround at present is to not use the emulated processor types in your VM configuration.