I
Immich•2y ago
volcs0

Photos not being tagged

A lot of my photos are not getting faces tagged. I can tag them manually, and it will merge them, but this is not happening automatically. These are clear, good pictures. When I run "Recognize Faces" it will look like it's running - it will find 20,000+ faces to tag. But after I let it run, the faces still aren't tagged. I've deleted the Typesense image and rebuilt the stack, but no luck. I'm on unRAID using the Docker Compose stack. Thanks.
No description
25 Replies
volcs0
volcs0OP•2y ago
The job finished - and the photos from the last week are all untagged. It found the faces, but they are not tagged. I could tag and merge them, but I'm wondering why it stopped working?
Alex Tran
Alex Tran•2y ago
I suspect there is either problem with typesense or the models aren’t download correctly in the machine learning container Can you help to check the logs of those services?
volcs0
volcs0OP•2y ago
So, on my share, the "immich_machine_learning" folder is empty. I'm assuming it should not be?
volcs0
volcs0OP•2y ago
No description
volcs0
volcs0OP•2y ago
Here is the machine learning log:
No description
volcs0
volcs0OP•2y ago
And TypeSense:
No description
Alex Tran
Alex Tran•2y ago
They look normal to me
volcs0
volcs0OP•2y ago
Microservices has a bunch of these errors:
No description
Alex Tran
Alex Tran•2y ago
Ah interesting, somethings wrong with redis Restart the stack, make sure the redis container get remove and then bring them all backup Then I think it will work Try with one single photo with faces first to confirm that it works
volcs0
volcs0OP•2y ago
OK - will do. IN the meantime, here is the full log from the docker compose https://pastebin.com/F3LgkvHa
Pastebin
immich_microservices | immich_microservices | [Nest] 7 - 0...
Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.
volcs0
volcs0OP•2y ago
How can I run face recognition on a single photo? I see the refresh metadata and the refresh thumbnail. And now it's just found every face in the picture twice. With different thumbnail photos for each face (and names for none of them). Any advice is appreciated.
Alex Tran
Alex Tran•2y ago
By uploading a photo
volcs0
volcs0OP•2y ago
ah. ok
Alex Tran
Alex Tran•2y ago
You can reset everything by run detect face on all, I think the data might be messed up when your redis was down
volcs0
volcs0OP•2y ago
new pic imported, face found, but no name.
volcs0
volcs0OP•2y ago
Log here: https://pastebin.com/BRAVZVAW
Pastebin
immich_server | [Nest] 7 - 09/17/2023, 3:58:48 PM L...
Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.
volcs0
volcs0OP•2y ago
I will re-run full face detect. Should I do the "all" where it resets everyone's names again? (I have about 200 names entered ) When I run ALL and it asks me to clear all faces and re-run, it errors.
volcs0
volcs0OP•2y ago
No description
volcs0
volcs0OP•2y ago
That's the main log I tried just re-running "missing faces" and it found 120,000 - and ran, and now I just have a lot of faced doubled and no names. Maybe it's hopelessly messed up?
Alex Tran
Alex Tran•2y ago
Hhmm I think this related to one of the bug we discovered recently
volcs0
volcs0OP•2y ago
@Alex sorry to bother you again - resetting everything doesn't seem to work (it doesn't happen). What is the way to really reset it if clicking the "all faces" button doesn't work?
Alex Tran
Alex Tran•2y ago
I will have to look into this later to see where the bug is
volcs0
volcs0OP•2y ago
ok Thanks Alex.
terion
terion•2y ago
I have similar issue. Also microservices are having tons of logs like that 
[Nest] 7  - 09/19/2023, 8:19:17 PM   ERROR [JobService] Unable to run job handler: TypeError: fetch failed
[Nest] 7  - 09/19/2023, 8:19:17 PM   ERROR [JobService] TypeError: fetch failed
    at Object.fetch (node:internal/deps/undici/undici:11576:11)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
    at async MachineLearningRepository.post (/usr/src/app/dist/infra/repositories/machine-learning.repository.js:27:21)
    at async FacialRecognitionService.handleRecognizeFaces (/usr/src/app/dist/domain/facial-recognition/facial-recognition.services.js:105:23)
    at async /usr/src/app/dist/domain/job/job.service.js:105:37
    at async Worker.processJob (/usr/src/app/node_modules/bullmq/dist/cjs/classes/worker.js:346:28)
    at async Worker.retryIfFailed (/usr/src/app/node_modules/bullmq/dist/cjs/classes/worker.js:531:24)
[Nest] 7  - 09/19/2023, 8:19:17 PM   ERROR [JobService] Object:
{
  "id": "cb8f94a9-432a-4aee-beb8-2024fb0fe731"
}

[Nest] 7  - 09/19/2023, 8:19:17 PM   ERROR [JobService] Unable to run job handler: TypeError: fetch failed
[Nest] 7  - 09/19/2023, 8:19:17 PM   ERROR [JobService] TypeError: fetch failed
    at Object.fetch (node:internal/deps/undici/undici:11576:11)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
    at async MachineLearningRepository.post (/usr/src/app/dist/infra/repositories/machine-learning.repository.js:27:21)
    at async FacialRecognitionService.handleRecognizeFaces (/usr/src/app/dist/domain/facial-recognition/facial-recognition.services.js:105:23)
    at async /usr/src/app/dist/domain/job/job.service.js:105:37
    at async Worker.processJob (/usr/src/app/node_modules/bullmq/dist/cjs/classes/worker.js:346:28)
    at async Worker.retryIfFailed (/usr/src/app/node_modules/bullmq/dist/cjs/classes/worker.js:531:24)
[Nest] 7  - 09/19/2023, 8:19:17 PM   ERROR [JobService] Object:
[Nest] 7  - 09/19/2023, 8:19:17 PM   ERROR [JobService] Unable to run job handler: TypeError: fetch failed
[Nest] 7  - 09/19/2023, 8:19:17 PM   ERROR [JobService] TypeError: fetch failed
    at Object.fetch (node:internal/deps/undici/undici:11576:11)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
    at async MachineLearningRepository.post (/usr/src/app/dist/infra/repositories/machine-learning.repository.js:27:21)
    at async FacialRecognitionService.handleRecognizeFaces (/usr/src/app/dist/domain/facial-recognition/facial-recognition.services.js:105:23)
    at async /usr/src/app/dist/domain/job/job.service.js:105:37
    at async Worker.processJob (/usr/src/app/node_modules/bullmq/dist/cjs/classes/worker.js:346:28)
    at async Worker.retryIfFailed (/usr/src/app/node_modules/bullmq/dist/cjs/classes/worker.js:531:24)
[Nest] 7  - 09/19/2023, 8:19:17 PM   ERROR [JobService] Object:
{
  "id": "cb8f94a9-432a-4aee-beb8-2024fb0fe731"
}

[Nest] 7  - 09/19/2023, 8:19:17 PM   ERROR [JobService] Unable to run job handler: TypeError: fetch failed
[Nest] 7  - 09/19/2023, 8:19:17 PM   ERROR [JobService] TypeError: fetch failed
    at Object.fetch (node:internal/deps/undici/undici:11576:11)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
    at async MachineLearningRepository.post (/usr/src/app/dist/infra/repositories/machine-learning.repository.js:27:21)
    at async FacialRecognitionService.handleRecognizeFaces (/usr/src/app/dist/domain/facial-recognition/facial-recognition.services.js:105:23)
    at async /usr/src/app/dist/domain/job/job.service.js:105:37
    at async Worker.processJob (/usr/src/app/node_modules/bullmq/dist/cjs/classes/worker.js:346:28)
    at async Worker.retryIfFailed (/usr/src/app/node_modules/bullmq/dist/cjs/classes/worker.js:531:24)
[Nest] 7  - 09/19/2023, 8:19:17 PM   ERROR [JobService] Object:
[09/19/23 20:18:27] INFO     Starting gunicorn 21.2.0
[09/19/23 20:18:27] INFO     Listening at: http://0.0.0.0:3003 (9)
[09/19/23 20:18:27] INFO     Using worker: uvicorn.workers.UvicornWorker
[09/19/23 20:18:27] INFO     Booting worker with pid: 10
[09/19/23 20:18:35] INFO     Created in-memory cache with unloading disabled.
[09/19/23 20:18:35] INFO     Initialized request thread pool with 4 threads.
[09/19/23 20:18:27] INFO     Starting gunicorn 21.2.0
[09/19/23 20:18:27] INFO     Listening at: http://0.0.0.0:3003 (9)
[09/19/23 20:18:27] INFO     Using worker: uvicorn.workers.UvicornWorker
[09/19/23 20:18:27] INFO     Booting worker with pid: 10
[09/19/23 20:18:35] INFO     Created in-memory cache with unloading disabled.
[09/19/23 20:18:35] INFO     Initialized request thread pool with 4 threads.
it seems no models are being loaded in ML?
terion
terion•2y ago
ok, nevermind It was this: https://github.com/immich-app/immich/issues/3927
GitHub
[BUG] machineLearning.url must be a URL address (Immich Server Erro...
The bug With the implementation of v1.76 and the ability to adjust Facial Recognition settings, I went to just slightly lower the min detection score to see what different results I would get. Open...

Did you find this page helpful?