main photos timeline shows no photos after upgrading from 1.29.0 to 1.30.0
After upgrading from 1.29.0 to 130.0 I see no photos in my main photos timeline on the web interface.
Photos on all our phones sycn to an album and new photos show up there just not on the main timeline.
Ive tested with two different user accounts in firefox and chrome along with incognito mode.
168 Replies
:wave: Hey @icedragon2002002,
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. :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..env
downgrading back to 1.29.0 fixes the issue
upgrading back to 1.30 breaks it again
noted that redis and database images have changed so ive updated those lines in my compose file
anything from the browser console log?
F12 > console
nothing that I can see as an issue
oh wait
you said console log lol
preloaded with link preload was not used within a few seconds. Make sure all attributes of the preload tag are set correctly.
Can you share a screenshot of your developer console?
of what specifically?
the console log page when you reload the page
this is the exact error minus my domain that im seeing in those console logs
Hmm
Would it be possible to try with a direct local IP?
herm. thinking proxy eh
ill have to edit my compose since im not allowing the port out even on the host level
trying to clear my browser history real quick and then ill check your local ip idea
Not really, just try to troubleshooting from the ground up 😄
sad face
with authentik in the way I cannot do that
hmm
Can you run this on your browser?
https://yourdomain.com/api/timeline/buckets?isArchived=false&size=MONTH&withPartners=true&withStacked=truethat spit out a bunch of junk
looks like a lot of time buckets
loging in using localhost:2283 yeilds the same results
HMMMM
same resource error in the console output
Can you check the server logs please?
@shadow Sorry to bring you in, do you have any thoughts where we can start troubleshooting this issue?
doesnt seem to be much in the server logs either
can you do
docker compose down, then docker compose up. After it finished starting up, navigate to the web and see if any error spit out from the logstandby
same thing
no difference
though it did seem to fix one of my issues with dockge not allowing me console access to the immich_server and only allowing me console access to redis, database, and machine learning
interesting. dockge things the database is still starting
a1df8744dccd tensorchord/pgvecto-rs:pg14-v0.2.0 "docker-entrypoint.s…" 2 minutes ago Up 2 minutes (health: starting) 5432/tcp
I brought in the expert, let's see what he can suggest later today
does that last docker ps I pasted above mean anything to you?
Walking to car now. What if you scroll a little. Mouse wheel or resize window?
I have run now. Be back in an hour or so
nothing changes when I do that btw
Clear cache?
it does seem like the database isnt fully starting
You have external libs?
compose and .env posted above. No im not using external libraries if thats what youre asking
folder mapping is using the custome folder locations but all resides on the same raid array
just different folders
If that's the case, you wouldn't be able to access your instance at all
its weird. i just downgraded back to 1.29 and now a different part of immich is stuck in the "starting" stage
though it seems to be working just fine
Looks like the server isn’t displaying all the services running. Maybe db migration didn’t finish?
Restart the db?
Do you have a db backup from before 130?
should have tons
The server log you posted looks like db migrations didn’t finish.
And maybe it’s in a weird in between state now
Can you run the
docker compose down && docker compose up in a terminal so that you know exactly what is happening?
we are looking for the starting up success message from the serverdid that earlier but would be happy to do so again
did you put the
-d in?You can do docker compose logs immich_server (but lowercase)
make sure you see something similar to this
yes
Not sure if there’s a way to increase verbosity on the db migrations. But I think that’s the problem
Can you check the database logs?
how does one do that?
whelp im done. does seem like the DB is no bueno though
guess im starting all over
re deploying immich using the backup restore instructions results in a somewhat functional immich instance yet the database still seems to be struggling
ill play some more later and if anything will just stick to 1.29.0 since that seems to be functional
#17113
another one?
Yes :Cryge:
Can you check if the mobile app shows the photos?
Can you help grabbing the mobile app log as well?
My db seems to be dead but working
I'll try the new image shortly and if that doesn't work I guess I'll start from scratch
Mobile app shows photos just fine, uploads just fine, and delete just fine
Hmm
Upgrading didn't resolve
Can you try changing the size of the browser, especially the height
I've zoomed in and out with no change
well I nuked my instance of immich. changed my .env to pull 1.29.0. re deployed immich and restored my database without error but still the db isnt healthy
but I can see my pictures in the timeline
I do have one error in the server logs though
Where’s the error ..?
edited my response above
sorry didnt notice it didnt copy and pasted the last thing I copied earlier
ok now its edited
Those are fine, log out and log back in
I think that’s just from an old login session, unrelated
so immich works at 1.29.0 at least in terms of my timeline but it seems the database is not happy anymore and ive gone so far as to restore to a backup created on the 20th
ok so final try and test. Here is what ive done to try and clear up this mess.
after doing this im left with a mosty functional immich but a very unhappy database
the only thing I didnt try is revert my database image to the older version
hi - I'm around now - do you have time to debug?
What do you mean when you say the DB isn’t healthy can we explore that
Also attach the output of
docker compose logs database
Is your DB_DATA_LOCATION start with a . or a ./ or just a plain alphabet character? meaning, are you using a volume or a bind-mount for the db?
Is the DB on a local drive? or is it on a NFS/SMB share?db is local on zfs
db location doesnt start with any of those characters you referenced.
db location along with all file / folder mount paths are absolute so no relative pathing is being used
I just did the most extreme thing and re deployed immich back to 1.28.0 and restored a db backup from 3/4 just before the release of 1.29.0
Ok, so that’s a bind mount. That should be ok. Zfs is a little picky with dbs, have you tuned the dataset for the db. (Probably unrelated)
I dont believe ive done anything specific to db tuning
about the most universal thing ive done is disable atime
oh I kinda fibbed
my db and the main portion of immich is on a nvme drive ext4
the upload, db backcups and profile is on the zfs array
Ah ok. Any logs from the db? Do you ever see the success message in the server logs? )compare the backups/old versions with new ones)
even after restoring to 1.28 it seems my database is stuck in a starting state
yet immich is still functional
I don’t see how it can be functional and in a starting state. Do you have any tranparent cache servers, load balancers, or other servers in between the docker server and your browser?
the only thing I have set up between immich and my browser is a reverse proxy server and authentik
Maybe try using a different drive for the db path to isolate hardware issues ?
well if it were a drive issue I would have a lot worse things happening
drive in question is my main boot drive
Reverse proxy does no caching?
and im not getting any smart alerts from it
cache assets is checked
just disabled and will check
If you have local access, can you use the internal ip? Connect directly to the server?
already tried and tested that
proxy / network doesnt seem to be an issue here
I feel like the web ui is cached, but the api server depends on the db for anything, so are any of the /api network requests returning 200?
not sure how to check that
forgive me but my knowedge doesnt extend to the api side
Go to dev console in your browser and the network tab. You can filter by /api
if it helps ive literally taken new photos on my phone. watched them upload to the server. and appear on the server so i dont think cache is the issue here
Hm, yeah sounds like it’s working. Why do you say the db is unhappy ? What do you mean? Do you see success in the server logs?
thats the only error I see in the console for the browser on immich
These wouldn’t be errors in the console. Go to network then you’ll see all the requests the browser makes
this is why
health is stuck at starting
it never leaves that state
Oh. Lemme check something. I think that’s ok
Let’s see
docker inspect immich_postgres please
It shouldn’t be starting for 14 minutes
Also docker logs immich_postgresWhat docker version and compose version you have
Docker version 28.0.1, build 068a01e
Do you use external library?
Docker Compose version v2.33.1
Can you double check the exclusion pattern quickly?
no
Okay
what do you mean
Nvm, it only applies to external library
ok then yes 0 external libraries on this deployment
docker —version
Docker version 28.0.1, build 068a01e
here is my system info if its of any help
im going to move forward with restoring back to 1.29 real quick
The db health check happens every 5 minutes by defuslt. So it probably just says starting since it hadn’t been up that long… had it?
tell you what. ill restore this version and go buy beer and let you know what the start up state of my db is
Double check your docker compose matches the latest from the git repo
K
ok 1.29 is up and running
Tho, if you’re able to upload from mobile, this is probably fine. It is a mystery
db still in a starting state
Web works?
going to go buy beer and let you know what it looks like once I get back
Ignore the health check for the db for now
yes web works 100%
Ok. Also
Do you have a touch screen laptop?
Or whatever your using
what an odd question
no I do not
To view the desktop web
There was a pr to hide the scrubber on touch devices
standard home built desktop / server. no touch screens here
Wondering if it affected touch laptops
before I go let me have you chew on my compose and .env files
your heathlcheck statement is truncated seems like
as a kind note. anything under /home/user/data is my zfs array
is what it should be
your TZ var has a leading space, which is probably bad too
edited the TZ and the test variables and spun the server back up
hay now
the DB is showing as healthy
ok so should we try upgrading to 1.30?
edit 1.30.1
if the web is working, 100%, then yes, i think try the upgrade
what browser, again?
web looks good
firefox
k, go for it
standby
no dice on a full fix here guys
everything looks healthy but we are back to this bs
ok, try 2 things, and when you have beer, I can screenshare with you if you want
resize the width and height of the window
then, scroll with mouse wheel, or two finger gesture
ok let me go get that beer and we can do the screen share
resizing didnt work btw
afk for a bit
k
ok im back
@shadow you still there?
ya
I found a few bugs, don't think its related, but worth a shot to try out, if you are up for it
modify your .env, the IMMICH_VERSION to be
IMMICH_VERSION=pr-17076
this is a bit of a longshotI'll try it now
machine learning and server threw errors when trying to upgrade
nvm I missed a 7
no dice. still nothing in the photos timeline
I’m basically out for tonight. Have you tried other browsers? Your phone/tablet?
phone app works fine
tried both chrome and firefox
Phone browser that is
cleared browser history / cache
3 different systems as well
I’m out of ideas. I think I need to see it. But can’t tonight
duck duck go on mobile displays the same thing
no problem bud. just happy you looked into it. besides im not the only one having issues with this latest update
will just revert to 1.29 for now and wait for a future fix
Yes. That worries me too
Can you visit the demo instance and see if it works for you
https://demo.immich.app
checking
confirmed working
So that’s running latest code. That makes me think it’s environmental
What if you create a brand new totally separate from scratch instance. And see if that works.
give me 5 min and ill have it done
cant seem to get it going
its stepping on my main instance of immich it seems
last ditch effort, can you try capture like 20 new photos today and upload it to your instance, see if it triggers the timeline loading?
ill give it a try first thing in the AM. im calling it for the night. Appreciate yalls efforts though
thanks, have a great night
Hey one thing I also can think of. Do you have a different computer that you can try access your instance on?
Found the reason!
There are invalid dates in the database. There were buckets with invalid date previously. I think with the new timeline rewritten, it causes some issue with group's date time formatting and hence not rendering anything
You can use this thread to check and resolve the invalid date https://github.com/immich-app/immich/issues/16046#issuecomment-2657784399
is it me or does my output not fit the criteria in that posting you just supplied
or is it the 0001 dates that are screwing things up
yes sir
You can copy those files to a new place, delete those record from the database
then your timeline should be back, then you can reupload them
it will be good to go again
now to figure out how to delete these
delete from assets a where a."fileModifiedAt" < '1000-01-01 00:00:00';I get a syntax error
Can you capture a screenshot?
will do so in a moment
let me copy these files to a different location
quick question. that command listed above. thats only removing the assets from the database correct?
should I remove them from the database and then move the files or move the files out of those directories and then remove them from the database
Correct
the error im getting
the exact command im using
ah. Do this first
then
fixed
Nice, timeline is showing up now?
correct
:peepoClap:
so the files that it just removed and that I just moved out of those directories
do I need to edit them or just re upload them?
you can drag and drop to reupload them
you can if you want, but there is no harm reuploading them now
fantastic
thank you kindly
No problem, glad we figure it out
Hey @Alex -- just wanted to say thanks for the clear instructions above. I had the exact same issue @icedragon2002002 had. I had one single asset with incorrect dates that was causing my entire timeline to go blank after updating to 1.32. Appreciate the effort here! 💯
My story: I run Immich in a container on Truenas Scale. I do updates when they roll out, and at some point I lost my timeline. It was empty on the web and phone app. New photos did not appear, but I could find them in folder view. Truenas allowed me to roll back to an earlier version, and my timeline magically reappeared. Looking down to the bottom, I saw four pics that I had uploaded from scanned photographs. They had mangled dates. So I gave them correct dates by "Change date" in Immich. Then I updated Immich to the latest version and the timeline remained, all present and correct. Conclusion, a pic with bad date/time will bork your timeline.
I've lost dates for at least 17 images during my time with Google Photos. 9 of them had dates as 12/31/0, the rest are spread around 1970. Changing the ones set to year 0 to 1970 fixed this issue for me too.
Though some notes:
I tried to run the db command suggested in the other thread to find old dates out of curiosity and I got this error: :Shrug_3: Any ideas? Everything seems to be working with the db
psql: error: connection to server on socket "/var/run/postgresql/.s.PGSQL.5432" failed: FATAL: role "postgres" does not exist
The web timeline should handle these dates like the mobile app does or at least warn people if it sees them or throw a console error or anything! Silent failure and having to dig through threads doesn't feel like a good answer to me, though I appreicate everybody's help figuring it out.
How I got here: I was on an old version of Truenas that used helm charts and they stopped upgrading the immich version, so I was stuck on 1.119 until today when I upgraded to 24.10 Electric Eel which transitions to docker apps. So I have been stuck on the old version for ~7 months now and the only reason I upgraded was because my Archived folder stopped showing any results on Android and iOS, but the desktop version still showed it. Unfortunately, my archived files still don't appear on mobile after the server upgrade, though newly archived files do show up. Was there some migration that didn't run maybe?