#metadata all fk'ed up

1 messages · Page 1 of 1 (latest)

jagged grotto
#

Timeline of events:

  1. Initially setup immich on 1TB internal ssd, realised it was not going to be enough and copied (rsycnc -avh /source /desination) the data to an external ssd
  2. started using the external ssd and realised it gets disconnected quite frequently, turned out usb protocol is not as stable for running 24 hours or maybe its just that extreme ssd that I had bought was the issue, (rsycnc -avh /source /desination) all the files from external ssd to an internal ssd of 2TB.
  3. Internal SSD (2tb) working fine for a while but later realised alot (mostly all images/videos) were duplicated (atleast 3 copies), spinned up a remote machine learning instance to clean up duplication,
  4. After remote ML setup (running on default immich ML configuration) still lot if not all of duplicates exist, timelines are all messed up (90% of the photos/videos are in 2025-2024 timeline)

How do I fix this issue? Is starting from scratch my only option?

potent salmonBOT
#

:wave: Hey @jagged grotto,

Thanks for reaching out to us. Please carefully read this message and follow the recommended actions. This will help us be more effective in our support effort and leave more time for building Immich immich.

References

#

Checklist

I have...

  1. :blue_square: verified I'm on the latest release(note that mobile app releases may take some time).
  2. :blue_square: read applicable release notes.
  3. :blue_square: reviewed the FAQs for known issues.
  4. :blue_square: reviewed Github for known issues.
  5. :blue_square: tried accessing Immich via local ip (without a custom reverse proxy).
  6. :blue_square: uploaded the relevant information (see below).
  7. :blue_square: tried an incognito window, disabled extensions, cleared mobile app cache, logged out and back in, different browsers, etc. as applicable

(an item can be marked as "complete" by reacting with the appropriate number)

Information

In order to be able to effectively help you, we need you to provide clear information to show what the problem is. The exact details needed vary per case, but here is a list of things to consider:

  • Your docker-compose.yml and .env files.
  • Logs from all the containers and their status (see above).
  • All the troubleshooting steps you've tried so far.
  • Any recent changes you've made to Immich or your system.
  • Details about your system (both software/OS and hardware).
  • Details about your storage (filesystems, type of disks, output of commands like fdisk -l and df -h).
  • The version of the Immich server, mobile app, and other relevant pieces.
  • Any other information that you think might be relevant.

Please paste files and logs with proper code formatting, and especially avoid blurry screenshots.
Without the right information we can't work out what the problem is. Help us help you ;)

If this ticket can be closed you can use the /close command, and re-open it later if needed.

jagged grotto
#
  1. I am on releaste 2.5.5
  2. this issue does not seem to be related new releases
  3. reran Extract metadata and sidecar metadata job, still did not help
  4. accessing the instance works locally, no issue on this
tiny cedar
#

Did you perhaps rsync and import/export the entire UPLOAD_LOCATION, which would include thumbnails (multiple copies of each asset)?

#

If you still have the external copy you can "start over" by simply copying only the actual library

jagged grotto
#

Thanks for the reply.

Yes, I had copied the entire UPLOAD_LOCATION, unfortunately the external copy is getting rsycned as a "backup" so I guess its all been overwritten already. This setup was working initially without any issues (regards to metadata, I was not using any remote ML server for jobs at the time. I had to introduce remote ML, only because there were duplicates)

tiny cedar
#

Well the internal deduplication tool relies on ML to look for similarity and defaults to the larger file so you could use that

#

Or you can really dig in the db to list assets that have a UUID as original filename and delete those

jagged grotto
#

Getting a lot of errors similar to the above, might be because of duplicates, not sure but checking if theres anything that can scan the files from start and push it accordingly:

tiny cedar
#

Those errors are thumbnails that it can't find

jagged grotto
#

I just uploaded some files not sure why they are erroring out, trying to rerun ML jobs and that is also not running

tiny cedar
jagged grotto
#

I start the thumbnail generation job on the latest assets that has been uploaded from my phone, get the below logs:

[Nest] 23 - 02/18/2026, 10:16:03 AM DEBUG [Api:LoggingInterceptor~vtp7lq5b] GET /api/assets/1ed59369-998c-44af-b1fb-096a79ccc9c8/thumbnail?size=preview&edited=true&c= 200 13.79ms ::ffff:192.168.18.73
[Nest] 23 - 02/18/2026, 10:16:03 AM DEBUG [Api:GlobalExceptionFilter~vtp7lq5b] HttpException(404): {"message":"Asset media not found","error":"Not Found","statusCode":404}

#

should I delete from the server and reupload the assets to regenerate the thumbnail? Thanks

tiny cedar
#

Do you mean delete everything or do you have a way of specifically uploading the broken ones?

jagged grotto
#

just delete the recently broken ones, reupload those

tiny cedar
#

That'd work fine I guess

#

Just make sure your system is stable going forward

jagged grotto
#

weird thing is, the thumbnail/asset will be visible if I hover on it.

tiny cedar
#

it's playing the encoded video

#

the actual thumbnail is broken

jagged grotto
#

yea thats the confusion because the encoded video/assets are there but the thumbnail generation itself is failing

#

just tried deleting one of the assets and now I cannot reupload it, not sure if this is the correct way to generate the thumbnails

tiny cedar
#

There's actually an option on the asset screen itself to force it, but that doesn't help if it can't find the asset

#

Did you clear out the trash?

jagged grotto
#

havent removed from the trash, do i have to?

tiny cedar
#

If you want to re-upload, yes