Immich not recovering storage space (devs help please!)
So this is something Ive posted on a few times now on reddit. Previously its when I renamed an external folder, causing Immich to reprocess everything. With this new event that happened today it is even more blatant of an issue and I'm really hoping to get some help or see if Immich dev team can look into this.
Background - Been using Immich 8mo now, using an external library with about 80k assets, on my NAS. I never upload anything directly to immich or use the storage templates. Im on 1.135.3, and update approximately every version, checking for any release note changes needed. Outside of this, Ive had a problem free experience. Using immich has been rock solid. Just trying to recover my unused space
Today I, through a windows duplication app, deleted out 4000 videos/pics from the external library folder. I dont know exactly how much space this was but about 100gigs deleted from the external library. Prior to doing any deletions I looked in the lower left of the webui and it was 244gig of space used. After rescanning/processing the jobs its at 263gb (and still rising I believe).
Additionally none of those items have shown up in my trash like i had seen in the past.
Can someone help me understand how removing 100gb from my external library has INCREASED my Immich storage size by 15gigs or more? How would it increase at all? And how do I ensure the items I deleted, get removed in every possible way from Immich? (thumbnails, references, etc).
When I first posted on this I had 80gig left on my SSD but now I am getting down to 35gig because things dont seem to clear themselves out. I suppose I could start all over and reprocess everything but I know that will take a week or two to run.
142 Replies
:wave: Hey @RoachForLife,
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
- Container Logs:
docker compose logs docs
- Container Status: docker ps -a docs
- Reverse Proxy: https://immich.app/docs/administration/reverse-proxy
- Code Formatting https://support.discord.com/hc/en-us/articles/210298617-Markdown-Text-101-Chat-Formatting-Bold-Italic-Underline#h_01GY0DAKGXDEHE263BCAYEGFJA
Checklist
I have...
1. :ballot_box_with_check: verified I'm on the latest release(note that mobile app releases may take some time).
2. :ballot_box_with_check: read applicable release notes.
3. :ballot_box_with_check: reviewed the FAQs for known issues.
4. :ballot_box_with_check: reviewed Github for known issues.
5. :ballot_box_with_check: tried accessing Immich via local ip (without a custom reverse proxy).
6. :ballot_box_with_check: uploaded the relevant information (see below).
7. :ballot_box_with_check: 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.NOTE - Believing this may happen, I did grab some IDs of items that were deleted (a pic and 3gig video). When I go to the URL for these it says Not found or no asset.read access (400). Is there a way I can check if a thumbnail or similar file taking of space may exist still? Or through the DB via cli? Is there a job or DB command to clear out the junk?
Additional
Unfortunately I didnt check the number of assets in the Immich Admin area prior but I do use Homepage (self hosted app) which is connected to Immich via API. Prior to starting it showed 64,337 photos and 6,304 videos. After deleting the files, in the Immich WebUI it shows the same numbers.
Steps performed
1 Checked the file size in lower left of Immich WebUI, showed 244gig of 300gig used
2 Used AllDup windows app connected to my NAS folder and over an hour, removed 4000dupes, about 100gigs.
3 Rescanned and reprocessed via the Admin > Jobs menu
4 Refreshing WebUI the same value in lower left continue to rise. At 265gbg currently.
5 In the past, the deleted files would show in the Trash on the left side of the menu. They have not shown up. I have my Trash Settings to enable Trash and number of days is set to 30 currently
Docker compose file
name: immich
services:
immich-server:
container_name: immich_server
image: ghcr.io/immich-app/immich-server:${IMMICH_VERSION:-release}
#combined into single compose
#devices:
# - /dev/dri:/dev/dri
extends:
file: hwaccel.transcoding.yml
service: quicksync # set to one of [nvenc, quicksync, rkmpp, vaapi, vaapi-wsl] for accelerated transcoding
volumes:
# Do not edit the next line. If you want to change the media storage location on your system, edit the value of UPLOAD_LOCATION in the .env file
- ${UPLOAD_LOCATION}:/usr/src/app/upload
- /etc/localtime:/etc/localtime:ro
- /mnt/immich:/photos
env_file:
- .env
ports:
- 2283:2283
depends_on:
- redis
- database
restart: always
healthcheck:
disable: false
immich-machine-learning:
container_name: immich_machine_learning
# For hardware acceleration, add one of -[armnn, cuda, openvino] to the image tag.
# Example tag: ${IMMICH_VERSION:-release}-cuda
image: ghcr.io/immich-app/immich-machine-learning:${IMMICH_VERSION:-release}
#extends:
# uncomment this section for hardware acceleration - see https://immich.app/docs/features/ml-hardware-acceleration
#file: hwaccel.ml.yml
#service: openvino # set to one of [armnn, cuda, openvino, openvino-wsl] for accelerated inference - use the
-wsl version for WSL2 where applicable
volumes:
- model-cache:/cache
env_file:
- .env
restart: always
healthcheck:
disable: false
redis:
container_name: immich_redis
image: docker.io/redis:6.2-alpine@sha256:eaba718fecd1196d88533de7ba49bf903ad33664a92debb24660a922ecd9cac8
healthcheck:
test: redis-cli ping || exit 1
restart: always
database:
container_name: immich_postgres
image: ghcr.io/immich-app/postgres:14-vectorchord0.3.0-pgvectors0.2.0
environment:
POSTGRES_PASSWORD: ${DB_PASSWORD}
POSTGRES_USER: ${DB_USERNAME}
POSTGRES_DB: ${DB_DATABASE_NAME}
POSTGRES_INITDB_ARGS: --data-checksums
# Uncomment the DB_STORAGE_TYPE: 'HDD' var if your database isn't stored on SSDs
# DB_STORAGE_TYPE: 'HDD'
volumes:
# Do not edit the next line. If you want to change the database storage location on your system, edit the value of DB_DATA_LOCATION in the .env file
- ${DB_DATA_LOCATION}:/var/lib/postgresql/data
restart: always
volumes:
model-cache: null
networks: {}
df -h. Running this on Proxmox LXC using Dockge. Storing Immich on SSD (top one below) and my files on at a HDD called Atlas, mapped thru conf files ini proxmox.
root@Immich:~# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/mapper/pve-vm--110--disk--0 301G 267G 20G 94% /
atlas 48T 21T 27T 45% /mnt/immich
none 492K 4.0K 488K 1% /dev
udev 16G 0 16G 0% /dev/dri/card0
tmpfs 16G 0 16G 0% /dev/shm
tmpfs 6.3G 580K 6.3G 1% /run
tmpfs 5.0M 0 5.0M 0% /run/lock
overlay 301G 267G 20G 94% /var/lib/docker/overlay2/bcf4733102c42e080a7e6b7474b3674a09cac9840ea10f5cd8ac618b07dcdf04/merged
overlay 301G 267G 20G 94% /var/lib/docker/overlay2/0df6ca00c187e4feae7860f3f8fdbb2f3e60e474811fa03031341fd0a088fb8e/merged
overlay 301G 267G 20G 94% /var/lib/docker/overlay2/67cae80d009d8195d8bb34c4643b37e167fde26231e608e875ac4b26245494ff/merged
overlay 301G 267G 20G 94% /var/lib/docker/overlay2/fbecc231804cd5f4204b1b342b9a3ee7549cd6e6f7f1fb6bca70d5d2764e54f5/merged
overlay 301G 267G 20G 94% /var/lib/docker/overlay2/b65306d13897bc20549e7177d32274e54ad1d57c3f58c1125fc28a7d9b33cd19/merged
overlay 301G 267G 20G 94% /var/lib/docker/overlay2/b60b56a14ebea51138fb5e50058e78bde214217a285847eb46235405b8d965c8/merged
Having trouble getting the logs
Logs (from Dockge)
I should note that all of this happened on 6/24/25 (later down in the logs)
The storage size display is the total used on disk. Not just immich usage
So to really figure out what’s going on you will need to dig into the folder level
We also don’t recommend LXC so any weird issues we would recommend switch to a supported VM deployment
Can you post your env file contents?
This issue seems to only happen when deleting things. Previously I had renamed a folder which had 300gigs of photos in it, so immich reprocessed (adding space) but never cleared the old space. Not sure how it being a LXC would impact its ability to reach from the storage space but thats beyond my technical capabilities
Pretty massively. The storage interface layer for a LXC is tied into the host FS especially if using a pass through mount
Also the fact I deleted 4000 assets yet immichs external library metrics remain unchanged, is concerning
Grsabbing the env stuff now
Regardless, weird issues = recommend to use a VM
'# You can find documentation for all the supported env variables at https://immich.app/docs/install/environment-variables
The location where your uploaded files are stored
UPLOAD_LOCATION=./library
The location where your database files are stored
DB_DATA_LOCATION=./postgres
To set a timezone, uncomment the next line and change Etc/UTC to a TZ identifier from this list: https://en.wikipedia.org/wiki/List_of_tz_database_time_zones#List
TZ=Etc/UTC
The Immich version to use. You can pin this to a specific version like "v1.71.0"
IMMICH_VERSION=release
Connection secret for postgres. You should change it to a random password
Please use only the characters
A-Za-z0-9, without special characters or spaces
DB_PASSWORD=PASS
The values below this line do not need to be changed
###################################################################################
DB_USERNAME=postgres
DB_DATABASE_NAME=immich
TZ=US/Eastern'
Oops, that didnt post right (sorry)
Host filesystem is ZFS by the way
Immich server is on SSD (local-lvm). The external library is on ZFS (HDD)What is the absolute path of the library/ folder?
Reading your post there’s a lot of half points of data. We haven’t really figured out what’s going on yet
I cant browse that easily so would need a few to get into the console to check
I think you need to increase your logging / monitoring and figure out what exactly is growing / changing
For all I know it could just be old docker images from upgrades
Only thing that is on this SSD is Immich, and nothing has been done to it for the last 1 week. Also the server space was checked moments before I began deleting the assets
Ok, well we will still need to track the folder sizes to know the source
It could be a lack of the FS being trimmed within the guest as well
Is that a function of the filesystem itself or Immich? (sorry I am not familiar with that)
The FS
Theres a lot of things that can go wrong to cause this that have nothing to do with immich
I see, I'm with ya
So not that i would look forward to doing this but prior to moving this to the SSD I used to have both Immich and the External Library on the same FS (albeit still 2 locations as I didnt want my LXC backup, backing up 2TB everynight), but it would take the SSD out of the equation. I could spin the older immich LXC up, update it to latest Immich and bring over current DB. Would there be some benefit to doing any of that?
No, i doubt the ssd is the issue
I would start by monitoring the size nightly and see how it changes
Either a cron job for ex
Was just thinking maybe the SSD specifically isnt trimming whereas the zfs may?
I have been monitoring it, it hasnt increased this whole week until this deletion I did a few hours ago
You can run trim in the lxc i think
That you would have to google to research
I suppose there are 2 issues here. Why is it increasing is 1, but also why would removing 4000 photos not decrease it? The latter is what Im initially more concerned with
Also would try to prune docker images
4000 photos is not much thumbs, and we dont even know if thats why it grew
There are way too many possibilities right now to speculate. You need to bisect and identify the problem
This is sysadmin stuff still, not immich issue yet
I just wish I knew more about linux lol. I will post on the proxmox forums and see if that gets me anywhere
Im realizing that I think the increase in space is because of the Extract Metadata job. When its running the server space increases
Also whats odd is when I click 'Rescan' on External Libraries job it shows Active 2, then 1, then finishes in like 20 seconds.
Normally it would run through for a bit
Theres been improvements lately to make it much faster
How many external libs do you have? Can you share a screenshot?
Just one, sure
And no other users?
Oh ok, maybe its just that (the improvements). I havent really looked at the rescan timing in a good month or so
Nope, just me
Can you go into the LXC and run
find /mnt/immich -type f | wc -l ?I can see more logs, let me send this too. The bottom is when is hit RESCANE
You have many more files than you have external assets, why is that?
I dont know? Where are you seeing that?
Is that a reference to where is checked 80332 assets? And my #s show around 70k?
The find command has 102k assets
or, files. could be anything
I think I may have mentioned but 2 mo ago, I renamed an ext library folder at the root level, so all of those files got reprocessed in immich (and the file size never went down). Im wondering if those files are now doubled on the immich server, which is in line with what im experienced.
I think my old reddit post had some photo/vid qty #s let me check
No, that wouldn't explain why the external lib itself has more assets
immich doesn't write its thumbnails or anything to that volume
Oh sorry I misunderstand
On the external library I do have folders that are ignored via the exclusions
Im guessing that is why, I can run a check to see how many files are in those folders
ok, that probably explains it
I don't think there is anything else to do for now. you should just start tracking the growth of various folders in the LXC, and maybe start preparing to switch to a VM
because we really can't reproduce or accept bug reports on LXC, and I've never seen this before. I am fairly sure it is a server misconfiruation issue
If I do a VM, can I port my DB over and have it just work, or will I need to reprocess everything? I dread that because it will take a week or so
Which VM OS do you suggest and version?
BTW looks like the 2 temp upload folders I use are about 25k files. These are excluded from processing in Immich
Most of us use Debian 12
yes you can transfer
Ok thanks
I was finally able to get to the SSD via Samba. Heres what I found
-Immich is showing server space is 272gig being used
-Immich library is 235g (200g encoded video and 28gig thumbs, 7gig backups, 3gig postgress folder
Im checking the other folders outside of immich, I suppose it should be the difference of about 34gig. Not sure if that sounds high when this is a basic linux box with just dockge and immich on it
I suspect you have old docker images yeah
That seems a bit high
I dont mean to get too off the topic of immich (and will post elsewhere too) but I did see that the \bin (14gb) and \usr folder (17gig) show this weird nesting of folders a billion times
Would depend on what exactly is in there
Sorry each seem to have these contents:
but its repeated so many times
40 levels its saying
Thats weird. Not immich though
Never seen that
Thanks, yeh maybe its the container like you said. Anyhow I appreciate you helping me today
Sorry one last thing, I see the \var folder looks what I believe to be references to the search model I use in Immich. Its showing as a whopping 100gig. Perhaps this is where the space is going
And Im so sorry for all the questions but considering a restore. There is no way I can just grab my library folders where immich resides and copy that from one server to another, with all the encoded videos and thumbs in tack so they dont need to reprocess? Or I have to trigger the dump and then ro then run the command? Just trying to avoid reprocessing all the encoded videos and thumbs
If you use a DB dump then you should be able to just move the whole UPLOAD_LOCATION, yes
@RoachForLife what other containers do you have running? Because that X11 nesting is just you putting
env=env/X11 or something like that somewhereOnly other container I have is Album Auto Creator which is auto creates albums based on the folder of my nas. I see that X11 thing is on a few of my machines that use dockge so maybe its related to that. Ill have to check
I did find something new out though. So I had my older Immich box that lives on my HD (instead of this one that is on my SSD), which was current as of Feb so wouldnt require as much processing. But it was BEFORE I did this massive deletion of 4000 assets and before 2months ago when I renamed a root folder in my ext library causing immich to expand seeing it was 'new' because the path changed, but never recovered from. Anyhow I got it to the current version, used identical compose, imported my settings from the first immich, and it points to the same ext library. So its effectively identical. Let it run and now it has everything loaded.
I ran TreeSize on both machines and the Immich folder has a 35gig delta. In this photo the top one is the original machine I posted on (SSD) and bottom is one I just reprocessed (HDD). 240gig vs 205gig
Also odd though, my External library numbers are actually quite a bit HIGHER on the new box than the old. I feel like these numbers arent getting reported correctly
Old (SSD) is 64k photos, 6k videos
New is 74k photos, 7k videos
Double checked both immich, album creator, and dockge compose and env files, nothing like this in any of them
Did you set
DISPLAY in your ssh conf or something??? I didnt but also not sure how or what that is? I can check if if you wouldnt mind assisting on where to go
BTW further drilling into the differences of the 2 systems, the main diff is in the encoded-video folder (sorry I didnt post in screen above). But the old machine (ssd) is 202gig for encoded-video, and the new (hdd) is 165gig
I just checked /etc/ssh/ssh_config on one of the boxes and dont see anything about 'Display'. Only 5 are uncommented, PasswordAuth, Port, SendEnv Lang, HashKnownHosts, and GSSAPIAuth
This is because your model-cache volume is set to null
I have no idea why you would do that but I assume you’re forcing a new ML model to get downloaded to a new anon volume every restart
If Im doing it, its not on purpose 🙂
I did change my model like a month ago when you guys dropped some new ones
but that was just thru the webui
It’s in your compose. Compare the last lines to our compose on GitHub
ok thanks will check that now
Feel free to share the compose/env of your other containers and I'll take a quick peek to see if I can't spy the X11 madness
Yep I see that. So in the compose its just blank. Maybe dockge turned that into 'null'?
Just remove the
null partOk just did that on both instances, should that clear the data on its own or will I need to delete something manually?
You should do docker volume prune -a
I wonder if that model cache has any relation to the 40 instances of X11 folder?
No
That’s all on you haha
dumb question, but could that docker volume prune -a be the missing link from my original question as to why files werent being removed? Not sure if you saw my last image but I reprocessed on a new serer with identical settings and ext library and the encoded vid folder is 40gig diff
No, that shouldn’t need to be run in a normal install
ok im going to do more treesize comparisons between each of these steps to document any changes it does
Ran the command but saying reclaimed 0B
Maybe a different command i should be running?
running docker system prune now instead, total server space is decreasing. ill ping back when done
@Mraedis thanks. Here is Immich yamls and thd one for Dockge. Thats only stuff I have on my system. Thanks for looking over
@Zeus After that prune system my var folder went from 100gig to 24gig. Sweet. No change in the Immich folder but I guess that was expected. Odd that this folder went down 75gig but it only reclaimed 25gig from the msg
nothing special in the composes, what about
grep -rni ~/.config/ -e X11 does that yield anything ?
And without the .config part?
on the x11 thing, I came across this, apparently its from a package? https://raspberrypi.stackexchange.com/questions/12670/how-to-cleanly-remove-x11
Side note, I got WinMerge and Im comparing the immich library folder for both machines to see where the differences are. Then maybe that will help identify whats going on
No that's not what it's from
Your problem is you have something copying your X11 to X11
Oh gotcha, shrug...
Going back to the original issue, still feels like something odd is happening here. When I compare the Immich folders, where the settings are identicial and source is identical, it produces 2 difference folder sizes, partricularly on the encoded video folder. Earlier, Zeus was pointing to the fact I am using a LXC in proxmox which is likely the issue. I just wonder if there is more to it.
Equally the fact that me deleting items from my ext library is not updating the library metrics in the immich webui. All of it feels off, like something isnt happening in the background to refresh things properly
Equally the fact that me deleting items from my ext library is not updating the library metrics in the immich webui. All of it feels off, like something isnt happening in the background to refresh things properly
We still don't have any more info than before to say this i s due to immich
but I think it's time to swithc to a VM if it's still frustrating you
I will be doing that at some point. Just unsure how (if I were on a VM), how you would even troubleshoot this
well, first of all your LXC has other issues unrelated to immich confounding the issue
but you still haven't done what we asked which is to identify exactly what folder is growing and when
Throwing another idea out there. So you saw the other day I have 2 of my own upload/temp style folders, which is about 1tb in raw data. This is excluded from Immich via external library pattern exclusion. The workflow I go thru is I dump new photos into these photos (plus a lot of old ones I havent sorted yet) and then move into the other ext library folders where immich picks them up. Is it possible that im working in an excluded folder and then into immichs folders, its doing something weird with them?
no, that should be fine
I mean its the qty of folders in the encodedvideo folder
its just there are more of them
If i have 2 systems, referinging the same data source, same everything, and one has 165gig of encodedvideo folder and one has 205gig of encodedvideo folder, doesnt that show something isnt right?
depends on your settings for video encoding
i imported the settings from system 1 to system 2 so they should be identical, no?
You're making so many assumptions, mate
the exclusion patterns could differ.. the settings could differ.. certain files could be visible or not in each system
How could any of the settings differ if i exported from system 1 and imported to system 2?
Difference in the encodedvideo folder. Left is SSD and right is the new one on HDD
37k files vs 21k files. Same ext library data source,
I am not sure. We don't have any known bugs regarding this. deleted files are only purged after 30 days typically
SSD
HDD
Ok so maybe I truly just need to wait
Maybe some consideration could be given to having a job or a way to force the deletion ad hoc?
you can always delete the full encoded video folder and re encode and see what happens
I do recall when I first had this issue 2 mo ago, a dev on reddit said the 30 days, but 30 days later, there was no change. even 60 days later, no change
Yeh but that would take forever but I guess youre right
TBH the reason im making all these posts is less to fix my issue (although of course I care about it) but I want to understand it and if its a bug its something that the dev group could fix long term. This has come up a few times as I have either deleted things in bulk (using dupe finder in windows) or the one time i had to rename a root folder in my ext library (changed the name slightly) which also caused this. I realize Im not on a VM and that may be the whole thing, just feels like more to it
In the example Im comparing i have effectively started over like you suggested and the results show a 10k file difference between the two. Something tells me if i deleted the encodedvideo folder and start over Id end up at 21k videos, simiilar to what i have on system #2 (HDD) i referenced
Is this at all a possibility? Unsure why youd make folks wait a week if they wanted to run it immediately? or give someone a script or a job to run to do this?
I'm not sure, I don't work on that part of immich. You can open a feature request if you want, if it doesn't already exist
ok thats a good idea
Not to go on a tangent but you just made me think of something, and something i posted on reddit the other day, maybe you might know. Motion photos. Immich takes a motion photo and splits that between the photo and the video, is that correct? If say, I didnt want any motion photos to be stored (and had 50k of them lol so doing them on the phone didnt make sense), could I just delete the corresponding encodedvideo, or would immich fail?
you can always delete encoded videos you just might have to re encode them, yep
Hmm, maybe the FR will be to ignore motion photo encodedvideos when transcoding. Not sure if that would be possible but that might work
Do you know if I were to just delete a motion photos encodedvideo, would it still function in the webui if not re-transcoded? Like would it give an error or like an icon that i couldnt see the photo until the transcoding happened?
I'm not sure tbh. it definitely can't play it but I do not know how it would show up
ok fair enough
thanks again for the help and info!
sure, let us know if you find out more
@Zeus So someone on another thread suggested (i think you more or less did the same) to turn transcoding OFF, process everything, and see if any files are left over in the encoded videos folder. So I did that and 113gig still remain. I would have thought this would have dropped to 0 if encodedvideo was off (and using external library). Here is the before and after disabling transcoding
ALso at this point, can I delete the contents of encoded-video and turn transcoding back on and reprocess everything?
You can always delete everything inside encoded-video, except for the .immich file, if you then trigger encode on all @RoachForLife 🙂
(by always I mean it won't break your immich install... a lot)
Won't break it a lot lol. I'm definitely planning to do this later today and let it run overnight. I have lime 10k videos and know it will take some time to process.
I guess I'm wondering if there is also anything I can be doing to help the immich team with troubleshooting this issue and why it's not recovering space. Leaving behind 113gig of encoded videos (which is about half of my previous folder size) seems like something is really not working with Immich and how it operates with people using external libraries. It's not cleaning up after itself if a file is moved (or renamed in my case)
My case may be a bit more edge as this mainly is due to me renaming the root folder for my daughter which has about half of all the videos of my entire system (she about 4 so as you can imagine we take way too many pics and vids) but still, I'd love to see this issue be corrected so reprocesinf the entire thing is not needed. If, based on where I'm at in the process, can help immich in some way, please let me know
Going back to your suggestion, I see there is one sub folder in encoded video which is some sort of guild, and then a million sibfolders under that. I imagine that guid looking folder needs to stay intact too?
that's all normal yes
I'm not toooo sure the guid folder should stay intact 🤔
but everything under it should be throwaway
yeh that was my plan, thanks. Im making copies of the current state of files in case I can be of help later
When you say this is normal, so immich should not be deleting the encoded videos (and photos) of assets that move in the ext library?
IE video was in daughter\folderXYZ and then folderXYZ renamed to \daughter\folderABC, forcing immich to reprocess the video in ABC. Shouldnt it delete the missing reference in XYZ?
Before when I had Immich on a spinning HD and had 40tb of space to use, I never gave it a thought. Now I have it on a 500gb NVME drive the space used by immich is more critical. I would think/hope that immich would be smart enough to release the file no longer is there (it does put it in the trash and throws a huge red banner saying its missing) to then delete the corresponding photo and video in its library after, no?
It should keep them for 30 days in the trash
If you empty the trash it should clear out the old encodes
It's been 3mo since I did this change. Still there
I did say should
I had emptied the trash right after doing it as well.
One thing to note, hitting empty trash does not remove these missing references from the trash. If for example I deleted things via immich (so not moving in the ext library but straight up deleting) and also move some in windows. I then have a mix of 2 types of trash files. One that deletes something and one that is a missing reference...
Clicking empty trash will only process the ones for deletion. The ones with the bad reference remain
To delete those I have to select all and then click the delete button (in trash page)
Not a big deal but I'm wondering if the fact it's separated like that has any linkage to this
And to my main point, the "should" part you just said is what I'm hoping the immich dev team can research and fix so it's a "always" instead of "should". And again, I'm happy to assist if needed
I suppose an idea might be some sort of job that can run (scheduled or even manual) to scan the ext library and compare to the dB and library files and delete any orphaned files.
I wonder if the issue here is that it deleted all references from the DB after you changed the path and it also cleared the 'pending deletes' in the trash 🤔
@Mraedis Yeh perhaps. I wonder if because I went to the trash, selected all and hit the delete icon in the trash, if that messed things up? Maybe I just truly need to wait? Would seem odd but just a guess as to what caused this from deleting itself. In my case I wanted to reclaim my space right away.
I'm guessing that potentially a lot of people could be affected by this (not deleting its files in the library I mean) since on reddit, i feel maybe half of the users use via an external library (based on a few polls ive seen and comments). Anyhow, I've brought this up several times in the past 3 or 4 months but now have more 'proof' if you will. Is this something that can be escalated to the dev team to look into a bit more? The hopes are simply to make Immich a better product as I plan to use this for a long long time
I'll see if I can reproduce it 🙂
Thanks bud, I appreciate
@Mraedis No pressure but just for my own reference. Would this ticket be the best place for any future updates or findings you come across?
Possibly related but I have another ticket about 'reprocess specific transcoded video' and another admin said that there are mp4 files (which are set to not transcode) in the db. Possibly related?
I totally forgot after realizing I don't have an easy external library in my test setup 👀