Xited with code 137

[CaitlinVa]

Which OS are you using?

• OS: Apple Sequoia 15.0.1

The program starts with no problem but when the model.bin is finalized I get te following error.
I use Docker Desktop.

[jhj0517]

Hi. I just tried to reproduce the error but failed.

Exited with code 137

According to the stackoverflow discussion here, code 137 means your device doesn't have enough RAM.

Docker uses a lot of RAM and your device may not be able to run the container.

Alternatively you can download Whisper-WebUI-Portable-Mac.zip and install it locally.

[CaitlinVa]

don't think it's ram related. I have 18GB RAM and I just monitored it it doesn't go over 13GB. This is what docker says about RAM and CPU usage:

[jhj0517]

OK, since you're using a Mac device (Apple Sequoia 15.0.1), check if you might have misconfigured the memory limit for your container. (Default setting is supposed to use Nvidia GPU, so Mac needs different setup) :

If these lines exist in your docker-compose.yaml, remove them all:

Whisper-WebUI/docker-compose.yaml

Lines 21 to 29 in cecc846

	# If you're not using nvidia GPU, Update device to match yours.
	# See more info at : https://docs.docker.com/compose/compose-file/deploy/#driver
	deploy:
	resources:
	reservations:
	devices:
	- driver: nvidia
	count: all
	capabilities: [ gpu ]

Also, remove these lines in requirements.txt if it exists.

Whisper-WebUI/requirements.txt

Lines 1 to 5 in cecc846

	# Remove the --extra-index-url line below if you're not using Nvidia GPU.
	# If you're using it, update url to your CUDA version (CUDA 12.1 is minimum requirement):
	# For CUDA 12.1, use : https://download.pytorch.org/whl/cu121
	# For CUDA 12.4, use : https://download.pytorch.org/whl/cu124
	--extra-index-url https://download.pytorch.org/whl/cu121

If you rebuild the image again it should work fine. Please let me know if you have the same problem.

[CaitlinVa]

The lines were already removed. If I don't remove them the docker file doesn't want to compose.

[jhj0517]

when the model.bin is finalized I get te following error.

I still think this is probably related to OOM, memory error.
Do you still get the same error when using smaller models, like tiny or tiny.en?

[CaitlinVa]

i've tested it with the medium model, this works. But I also have an laptop with only 8gb RAM and there the large model works.

[jhj0517]

I don't know why, but it seems that your Docker + Mac is somehow limiting the memory usage.

Try running image with --oom-kill-disable, like adding this to your docker-compose.yml:

oom_kill_disable: true

[CaitlinVa]

I get this error when running it:

[jhj0517]

I'm not sure if this will help, but according to the documentation,

You can set the memory limit manually in docker-composer.yml, probably something like 12g.

services:
  app:
    mem_limit: 12g