BUG: Immich ML python process stops responding

[fayeinmay]

Upstream issue:
immich-app/immich#27228
I assume this can be closed when the upstream issue is closed.

I just created this bug report here in case anyone else is running imagegenius docker-immich container and wonders about this issue.
Basically, if you see the errors below within your immich logs, the chance is pretty high that the issue is upstream. A fix can be found in the upstream issue, but TL;DR: add this ENV to your container: GUNICORN_CMD_ARGS=--no-control-socket

`[Nest] 267 - 05/10/2026, 5:41:58 PM WARN [Microservices:MachineLearningRepository] Machine learning request to "http://127.0.0.1:3003" failed: fetch failed

[Nest] 267 - 05/10/2026, 5:41:58 PM ERROR [Microservices:{"id":"30bf78f2-a9e3-43f6-a99d-2d0c5e705e8b"}] Unable to run job handler (SmartSearch): Error: Machine learning request '{"clip":{"visual":{"modelName":"ViT-B-16-SigLIP__webli"}}}' failed for all URLs
Error: Machine learning request '{"clip":{"visual":{"modelName":"ViT-B-16-SigLIP__webli"}}}' failed for all URLs
at MachineLearningRepository.predict (/app/immich/server/dist/repositories/machine-learning.repository.js:117:15)
at process.processTicksAndRejections (node:internal/process/task_queues:104:5)
at async MachineLearningRepository.encodeImage (/app/immich/server/dist/repositories/machine-learning.repository.js:135:26)
at async SmartInfoService.handleEncodeClip (/app/immich/server/dist/services/smart-info.service.js:91:27)
at async JobService.onJobRun (/app/immich/server/dist/services/job.service.js:51:30)
at async EventRepository.onEvent (/app/immich/server/dist/repositories/event.repository.js:91:13)
at async /app/immich/server/node_modules/.pnpm/bullmq@5.71.0/node_modules/bullmq/dist/cjs/classes/worker.js:570:32`