mirror of
https://github.com/NginxProxyManager/nginx-proxy-manager.git
synced 2026-04-25 09:25:55 +03:00
[GH-ISSUE #1769] Cannot Log into Admin After Upgrade to 2.9.15 #1309
Labels
No labels
awaiting feedback
bug
cannot reproduce
dns provider request
duplicate
enhancement
enhancement
enhancement
good first issue
help wanted
invalid
need more info
no certbot plugin available
product-support
pull-request
question
stale
troll
upstream issue
v2
v2
v2
v3
wontfix
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
starred/nginx-proxy-manager-NginxProxyManager#1309
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @jiriteach on GitHub (Jan 17, 2022).
Original GitHub issue: https://github.com/NginxProxyManager/nginx-proxy-manager/issues/1769
Updated via docker-compose to 2.9.15 from 2.9.14 and can no longer access the admin console.
Login page loads but cannot log in.
Docker logs -
I don't see anything further than this. Normally can take ~ 30s-1min before other bits are logged but nothing further showing up with 2.9.15.
Decided to roll back to 2.9.14 - pulled the 2.9.14 image and loaded - works fine!.
I've tried twice with 2.9.15 and no luck. Changes didn't show any big ones so any pointers as to what could cause this?
Thanks
@kjw1985 commented on GitHub (Jan 17, 2022):
Can confirm, update from 2.9.9 to 2.9.15 can't login. Downgraded to 2.9.14 and could login again.
@jc21 commented on GitHub (Jan 17, 2022):
I can't confirm. I went from 2.9.14 to 2.9.15 in my production environments just fine.
The current database version should not be none. What database are you using, mysql/maria or sqlite?
@jiriteach commented on GitHub (Jan 17, 2022):
Using SQLLite
@jc21 commented on GitHub (Jan 17, 2022):
Just tested on a sqlite setup, also don't have a problem logging in.
@jc21 commented on GitHub (Jan 17, 2022):
and yes I realise the database version is also none there...
@kjw1985 commented on GitHub (Jan 17, 2022):
I'm using jc21/mariadb-aria:10.4.15 as my database.
Hereby part of my logging:
@jiriteach commented on GitHub (Jan 17, 2022):
Will try again now and see what happens ...
@jiriteach commented on GitHub (Jan 17, 2022):
No go again -
Standard way i've been upgrading for a long while now.
Docker log -
Trying to log into admin - i'm seeing 502's in Chromes console.
Existing rules are process OK through as all my checks are running.
My docker-compose.yml
Reverting back to 2.9.14 and all working fine again.
@jiriteach commented on GitHub (Jan 17, 2022):
Found updated logs with errors -
@Manuel5cc commented on GitHub (Jan 17, 2022):
Same issue here. I'm using Mariadb as my database with the latest version from jc21's docker container. The upgrade seems to have created a loop that is generating high load on my server.
@jc21 commented on GitHub (Jan 17, 2022):
So the issue is
setuptools, version 60 is causing problems whereas version 58 is fine with the certbot plugins you're using. I'll get a new version out soon@Manuel5cc commented on GitHub (Jan 17, 2022):
In my case, downgrading (tried 2 or 3 releases behind) didn't fix the issue, as now it's stuck at this step:
[1/17/2022] [11:44:48 AM] [Setup ] › ℹ info Wrote JWT key pair to config file: /app/config/production.jsonand still cannot login.@jc21 commented on GitHub (Jan 17, 2022):
Please try with the
github-developdocker tag and let me know how you go@Manuel5cc commented on GitHub (Jan 17, 2022):
Using the develop branch gives me a migration error in the database.
[1/17/2022] [12:29:57 PM] [Global ] › ✖ error The migration directory is corrupt, the following files are missing: 20211010141200_ssl_passthrough_host.js [1/17/2022] [12:29:57 PM] [Migrate ] › ℹ info Current database version: 20211108145214I think going back and forward messed up something.
@ACodingGenie commented on GitHub (Jan 17, 2022):
I was facing this logging in issue last night too but rebuilding both MariaDB and Nginx Proxy Manager containers solved it. Can't speak for anyone else using other types of DBs though. I will update here if things so south again.
NPM Version:
v2.9.15MariaDB Version:
Version: '10.4.15-MariaDB'@jc21 commented on GitHub (Jan 17, 2022):
@Manuel5cc it looks like you've used a docker image tag from a PR that has not been merged, and that included a database migration. You will have to revert that migration.
this is MySQL specific:
but make sure you backup your MySQL data first, and next time you use a PR image backup everything beforehand and revert immediately after testing it.
@jiriteach commented on GitHub (Jan 17, 2022):
@jc21 - Wasn't sure if this was to everyone, but I tried this and no difference. Same as before - cannot log in and same errors. Reverted back to 2.9.14.
@kjw1985 commented on GitHub (Jan 18, 2022):
Tested the
github-developversion. I'm able to login. But I'm not really sure if the error in the log below should be expected.'''
[1/18/2022] [6:29:50 AM] [Global ] › ℹ info Manual db configuration already exists, skipping config creation from environment variables,
[1/18/2022] [6:30:04 AM] [IP Ranges] › ℹ info Fetching https://www.cloudflare.com/ips-v4,
[1/18/2022] [6:30:04 AM] [Setup ] › ℹ info Added Certbot plugins certbot-dns-transip~=0.3.3 ,
[1/18/2022] [6:29:51 AM] [Migrate ] › ℹ info Current database version: 20211108145214,
[1/18/2022] [6:30:04 AM] [Setup ] › ℹ info Logrotate Timer initialized,
[1/18/2022] [6:30:04 AM] [IP Ranges] › ℹ info Fetching IP Ranges from online services...,
[1/18/2022] [6:30:04 AM] [Setup ] › ℹ info Logrotate completed.,
[1/18/2022] [6:30:04 AM] [IP Ranges] › ℹ info Fetching https://ip-ranges.amazonaws.com/ip-ranges.json,
[1/18/2022] [6:30:04 AM] [IP Ranges] › ℹ info Fetching https://www.cloudflare.com/ips-v6,
[1/18/2022] [6:30:05 AM] [SSL ] › ℹ info Let's Encrypt Renewal Timer initialized,
[1/18/2022] [6:30:05 AM] [SSL ] › ℹ info Renewing SSL certs close to expiry...,
[1/18/2022] [6:30:05 AM] [IP Ranges] › ℹ info IP Ranges Renewal Timer initialized,
[1/18/2022] [6:30:05 AM] [SSL ] › ✖ error Error: Command failed: certbot renew --non-interactive --quiet --config "/etc/letsencrypt.ini" --preferred-challenges "dns,http" --disable-hook-validation ,
[1/18/2022] [6:30:05 AM] [Global ] › ℹ info Backend PID 244 listening on port 3000 ...,
An unexpected error occurred:,
ModuleNotFoundError: No module named 'version',
at ChildProcess.exithandler (node:child_process:397:12),
at ChildProcess.emit (node:events:390:28),
at maybeClose (node:internal/child_process:1064:16),
at Process.ChildProcess._handle.onexit (node:internal/child_process:301:5),
QueryBuilder#allowEagermethod is deprecated. You should useallowGraphinstead.allowEagermethod will be removed in 3.0,QueryBuilder#eagermethod is deprecated. You should use thewithGraphFetchedmethod instead.eagermethod will be removed in 3.0,QueryBuilder#omit is deprecated. This method will be removed in version 3.0,
Model#$omit is deprected and will be removed in 3.0.,
'''
@Manuel5cc commented on GitHub (Jan 18, 2022):
I finally decided to rebuild the container from zero using SQLITE instead of Mariadb and it seems to be working, however I haven't been able to generate wildcard certificates due to timeouts (no errors are shown in logs).
@jgunsing commented on GitHub (Jan 18, 2022):
Can confirm that the tag github-develop solved the issue
@jc21 commented on GitHub (Jan 18, 2022):
It appears that the
certbot-dns-transipplugin is installing v0.3.3 and theNo module named 'version'error seems to indicate that it's only written for python 2, not 3 as we've been using. The latest version is 0.4.3, hopefully that works.I've created #1773 to bump the version, but I am entirely unable to test it. Keep an eye on that PR for instructions to use the built docker image once it passes CI, and let me know how you go.
@jiriteach commented on GitHub (Jan 19, 2022):
@jc21 - Doesn't look to have fixed the issue with 2.9.15. Tried this PR and same as before.
Not much in the logs -
@kjw1985 commented on GitHub (Jan 20, 2022):
@jc21 Tested
jc21/nginx-proxy-manager:github-pr-1773, no errors anymore and able to login. Haven't tested if the certbot-dns-transip is working correctly within this image, as I'm not using it (yet) in the container.I know that certbot-dns-transip 0.4.3 should be able to get wildcard certificate as I've used this in another container recently.
See output logs:
@jiriteach commented on GitHub (Jan 20, 2022):
@jc21 - Any further ideas on what I can do to upgrade? Rolled back so using 2.9.14 ok. Thanks
@X4V1 commented on GitHub (Jan 23, 2022):
I'm also having the login issue with the version 2.9.15. Working fine on 2.9.14.
In my case I have a database version in my logs but later there are multiple errors. I don't know what it the root cause so I added a link to my logs in case it could help.
There is also a certbot issue (
Command failed: pip install certbot-dns-ovh==$(certbot --version | grep -Eo '[0-9](\.[0-9]+)+')) and a python error linked to cryptography-36.0.1.tar.gz.And also, thank you very much for the great work !
@D1no commented on GitHub (Jan 28, 2022):
For all the noobs like me who basically catched the wrong timing on trying
portainer,docker-composeandnginx-proxy-manager— here a docker compose file that works.Because the
bad gatewayissue can come upDB_MYSQL_HOST: 'db'"doublequotes instead of american'quotes😵💫
So this uses the previous version that works
2.9.14and routes the port correctly within docker compose by connecting to thedb-container.In Portainer: Go to
Stack->Add Stack-> Copy paste above and hitdeploy.@CharlPels commented on GitHub (Jan 29, 2022):
The transip plugin need to be upgraded to use rest api instead off soap
@kt-solutions commented on GitHub (Feb 10, 2022):
For me it worked to have this db-image instead of the latest:
jc21/mariadb-aria:10.4.15-innodb@jiriteach commented on GitHub (Feb 10, 2022):
I'm using a SQLLite db so still no solution.
@kevin-wynn commented on GitHub (Feb 16, 2022):
I too am getting this error now. From upgrading from 2.9.5 to latest, running on a pi3 it seems to no longer allow me to log into admin. I get the same looking logs as @Jiriteach throwing errors looking for Python.h. These are my logs:
I've tried downgrading to 2.9.14, 2.9.11, 2.9.5, 2.9.4. No luck moving forward with any of these versions now. I'm also using mariadb with mine, I've tried using 10.4.15-innodb version there, latest, etc. Nothing helping so far. NPM is still running smoothly behind the scenes, all my existing proxies and redirects are working as expected. Just unable to log in.
@jc21 commented on GitHub (Feb 16, 2022):
@kevin-wynn What about the
github-developtag?@kevin-wynn commented on GitHub (Feb 16, 2022):
@jc21 tried that one as well, didn't work. I can try it again and see what logs I get there though.
@kevin-wynn commented on GitHub (Feb 16, 2022):
@jc21 These are my logs using the
github-developtag...with a docker-compose file like this:
I get a 502 error when making the login post request to
/api/tokensAlso, if any consolation, I've tried letting the app recreate the admin and using the
changemepassword to no avil as well, this is my mysql table for the users:@jc21 commented on GitHub (Feb 16, 2022):
The 502 should show some kind of error in the docker logs related to that.
@kevin-wynn commented on GitHub (Feb 17, 2022):
@jc21 whenever I follow the docker logs and try to login all I see populate is
However I'm not certain if that was actually triggered from the 502 error as multiple more attempts to login didn't trigger anything.
Since this was JWT token related I tried clearing my
localStorageto see if that helped, it didn't@kevin-wynn commented on GitHub (Feb 17, 2022):
@jc21 Acutally, left the logs up and it looks like more errors populated in:
@jc21 commented on GitHub (Feb 17, 2022):
This pip stuff is becoming a bit ridiculous. It's dependency hell dealing with certbot. Nevertheless, that should not prevent anyone from logging in.
All I can think of doing is wiping out data in the
userandauthtables, restarting NPM and it will create that defaultadmin@example.com:changemeuser/auth for you. But as you said that doesn't work.FWIW I've not experienced this on any of my 3 production clusters running a combination of sqlite and MySQL backends.
@kevin-wynn commented on GitHub (Feb 17, 2022):
I can give wiping those tables clean, I had only tried setting the is_deleted flag on the one user. So maybe wiping those tables out entirely will help. I will give it a go.
I agree with you though, these pip errors don't seem like flags that would prevent logging in. But maybe they are hiding more errors... I will try wiping tables and maybe poke around some, maybe the node server isn't routing correctly?
@jiriteach commented on GitHub (Feb 17, 2022):
@jc21 and @kevin-wynn - Thanks guys and agree these dependencies are a nightmare. @kevin-wynn - Let me know if that works!
@kevin-wynn commented on GitHub (Feb 17, 2022):
@Jiriteach wiping the user and auth tables didn't work. I think what I'm going to try and do since I only have like 8 entries for proxies is grab those. Wipe everything out clean and start fresh and manually insert those db tables once I've logged in
@jiriteach commented on GitHub (Feb 17, 2022):
Good to know. Thanks.
I'd be keen to start again but just need a way to backup the db side with rules, custom certs etc. If there was an easy way to export/restore this into the db then keen to try!
@kevin-wynn commented on GitHub (Feb 17, 2022):
So fwiw, clearing the database didn't help. My guess is that this has nothing to do with the database at all at this point it might be something in the nginx configurations, as my nginx rules stayed in tack even clearing the db (I assume NPM only uses the database to show users data in the UI). I will start debugging the nginx rules to see if something is preventing an /api/tokens request from the UI
@kevin-wynn commented on GitHub (Feb 17, 2022):
Something also to note, trying to do a curl post to /api/tokens from within the docker container results in a 502 as well
@jc21 commented on GitHub (Feb 17, 2022):
The base of the api:
works on the
github-developdocker tag, but on the2.9.15tag.@kevin-wynn commented on GitHub (Feb 17, 2022):
Interesting, I just switched back to the
github-developbranch and gave it a test. 502 here@kevin-wynn commented on GitHub (Feb 17, 2022):
@Jiriteach While I haven't been able to find a real fix for this. I have gotten myself up and running with all my original data running on 2.9.15. Here are the steps I took to get where I'm at now:
Did a mysqldump of all my data
docker exec -it <mysql container> shmysqldump -u root -p ... npm setting access_list access_list_auth access_list_client certificate proxy_host redirection_host stream > dump.sqldocker-compose.ymlfile over to that new directorydocker-compose downthe existing stuff,cdto the new directory anddocker-compose up -dto spin it upcmd+ron the :81/login route to make sure you're not hitting a cached route and log in with theadmin@example.com:changemecredentialsletsencryptanddatadirectories from the existing set up to this new directoryAnd that's where I'm at now, things seem to be running. I'm able to log in and out fine, and all my existing content seems to be running fine now. But I haven't tested everything, just hit a few urls and saw them working is all really.
@jiriteach commented on GitHub (Feb 17, 2022):
Setup a replica environment so I don't mess with my core environment of NPM.
Db and data intact on a new RPI with new install of Docker and with the github-develop version.
Now that I have an enviroment to run diagnostics and tests - will try and investigate further as no big deal if i trash this one :)
Start-up and errors in the logs -
@jiriteach commented on GitHub (Feb 17, 2022):
@kevin-wynn - Thats awesome! Looks a goer. Will try that now.
@jiriteach commented on GitHub (Feb 17, 2022):
@kevin-wynn @jc21 - So its definately an issue with the existing version of the db so seems that an export/dumb and re-import is the way to go!
@jiriteach commented on GitHub (Feb 17, 2022):
@kevin-wynn @jc21 - Have narrowed it down. The issue is to do with a cert in my db. Removed all certs from the table and my existing db loads. Its easy for me to try and my sql is a sqlite db. I think I might know which cert it is - I'm using a Cloudflare origin so will check if this is the issue or if its any of the certs!
@jiriteach commented on GitHub (Feb 17, 2022):
@jc21 -
Here's the issue - existing wildcard certificates in the certificates table! That new version must have some checks etc.
Cert was valid but its a wildcard.
Remove that record from the certificates table and all works fine!
@jiriteach commented on GitHub (Feb 17, 2022):
Sigh - While this worked to get me access to the admin console and everything looked like it was working - I can no longer request SSL certificates via DNS challenge using Cloudflare with the latest build or github-develop. I get no error message back - simple a red banner. Looks like a timeout. I need to do more investigation as whats caused this now!
@jiriteach commented on GitHub (Feb 18, 2022):
@jc21 - OK here is my final diagnosis on this issue.
DNS Challenge with Cloudflare is actually not working with 2.9.15 and 2.9.16 and that the reason why I have been experiencing this issue. This is working with 2.9.14 but is now broken with anything above.
Decided to do a fresh (blank) install - and first task was to add an SSL certificate using my domain name with a DNS Challenge against Cloudflare. Enter the API Key and error!
I am using github-develop on a fresh install of NPM. This would be the cause of the issues and no fix for me unfortunately as even if start fresh - I cannot add validate any new certificates against Cloudflare since this DNS Challenge is broken. Looks like the library might be a Python3 and this is failing to install? Anything you can do to update the dependancy?
Thanks
@jc21 commented on GitHub (Feb 18, 2022):
and what happens if you run that command inside the container yourself?
that seems to be working for me?
@jiriteach commented on GitHub (Feb 18, 2022):
@jc21 - This fails with the same error as above. Still using github-develop.
Hmm what's different with your setup?
Here is the full log -
@X4V1 commented on GitHub (Feb 20, 2022):
I tried to update to version 2.9.16 and I'm also getting an python error during installation (from raspberry pi 3b+ wuth debian 11 32bits)
@jiriteach commented on GitHub (Feb 20, 2022):
Yup - downgrade to 2.9.14 or remove any certificates that do DNS challenges agains Cloudflare. This doesn't fix the underlying issue with Certbot but does help in trying source a fix.
@X4V1 commented on GitHub (Feb 20, 2022):
@Jiriteach Indeed I downgraded to 2.9.14 and everything's back to normal.
@HaoSs07 commented on GitHub (Feb 21, 2022):
note for those who use a raspberry pi, and end up here fallowing log errors, do not force the 64bit kernel ( using arm_64bit=1 in boot/config ) it will F tings up even if you revert back to 2.9.14
@Eeems commented on GitHub (Mar 3, 2022):
I've currently had to downgrade to 2.9.13 as I have the same issue on 2.9.14. Might be due to how I'm deploying this in proxmox though.
@twiesing commented on GitHub (Mar 10, 2022):
Same problem here with 2.9.15 and 2.9.16 😨 Using Armbian (armv7l).
@moracabanas commented on GitHub (Mar 18, 2022):
I run docker example on Raspberry pi 4 and I get 502 error from
latestdown to2.9.13I refreshed the data bind mount evey time and I can't login with default demo (sqlite)
@redpixy commented on GitHub (Mar 21, 2022):
same here. Experiencing the same issue when using those 2 versions.
@bencollinz commented on GitHub (Mar 29, 2022):
same. I've downgraded, trying 15, 14 and now 13, still can't login but everything runs in the background.
@bencollinz commented on GitHub (Mar 30, 2022):
Same. Nothing I've done has made any difference so far.
@FlyingPersian commented on GitHub (Mar 30, 2022):
So using DNS challenge with Cloudflare isn't working atm, even with 2.9.14?
@niawag commented on GitHub (Apr 1, 2022):
Hi all, same problem for me on Pi3, it works with 2.9.14 but not with 15, 16 or 18. I'm using sqlite default db. I'm having a pip issue :
Is there anything I could try to help debug this?
@mala-cimbra commented on GitHub (Apr 3, 2022):
I've found that the problem lies in the python cryptography library. There is not a compiled version for armv7l as stated here.
In the end, I've found a solution here by compiling the library inside the container.
Another solution, but I haven't tested it yet, is installing the certbot dns plugin from piwheels.
In my case:
@Europa2010AD commented on GitHub (Apr 4, 2022):
Just want to confirm that this problem persists in NPM 2.9.18. Running on latest official MariaDB.
@cmj2002 commented on GitHub (Apr 29, 2022):
This problem persists in NPM 2.9.18 with SQLite. I connected to the NPM via port 81 to disable and then re-enable the proxy for NPM, then the problem was solved.
@pinpspinps commented on GitHub (May 6, 2022):
Hi,
I confirm that @iomataani solution works at first but does not work anymore after restarting the container.
I also confirm that this problem persists in version 2.9.18.
@cmj2002 commented on GitHub (May 6, 2022):
@pinpspinps This is not surprising, because after the docker is recreated, all changes will disappear unless it's in a mounted volume.
I have not tried the solution yet, but if it works for you, maybe you can rewrite the Dockerfile before any release that fix the problem.
@pinpspinps commented on GitHub (May 12, 2022):
Is it not possible to integrate this change in the official branch?
@NSerbin commented on GitHub (Jun 12, 2022):
any news ?
@X4V1 commented on GitHub (Aug 8, 2022):
Still no news ?
@jc21 commented on GitHub (Nov 14, 2022):
For anyone having this problem with the cloudflare plugin, please try the
github-developdocker tag. For anyone else, please try running this command, with your problem plugin and let me know if it works, then I can add some exceptions.@HaoSs07 commented on GitHub (Nov 18, 2022):
So far all good, login works, nu Cloudflare errors.
@NSerbin commented on GitHub (Nov 24, 2022):
It worked for me... for cloudflare-dns the cmd is:
pip install certbot-dns-cloudflare --index-url https://www.piwheels.org/simple --prefer-binary@tmullin commented on GitHub (Feb 9, 2023):
Worked flawlessly for me after running this for Namecheap:
pip install certbot-dns-namecheap --index-url https://www.piwheels.org/simple --prefer-binaryI just set up NPM fresh today with minimal configuration (i.e. SQLite) and am on version 2.9.19.
@BartAgterbosch commented on GitHub (Jun 13, 2023):
Still not working to this day.... Getting a bad gateway message on the latest version still, the only thing that works is downgrading to 2.10.0 or lower, with all the possible vulnerabilities that are included....
@HaoSs07 commented on GitHub (Jun 13, 2023):
check logs. latest versions work fine for me
@BartAgterbosch commented on GitHub (Jun 13, 2023):
error create table
migrations(idint unsigned not null auto_increment primary key,namevarchar(255),batchint,migration_timetimestamp) - ER_CANT_CREATE_TABLE: Can't create tablenpm.migrations(errno: 13 "Permission denied")That is what seems to coincide with "Bad gateway" when I try to log in
@HaoSs07 commented on GitHub (Jun 13, 2023):
try restarting the DB
@BartAgterbosch commented on GitHub (Jun 13, 2023):
Tried that many many times, but on the latest it doesn't do anything, not sure from which version exactly that started and I used to have it set to always pull the latest, I can restart the db as many times as I want but the only thing that works is downgrading to 2.10.0..
From what I can see the db container is throwing a connection error: [Warning] Aborted connection 3 to db: 'name' user: 'user' host: '172.18.0.2' (Got an error reading communication packets)
I changed the db name and username in in the error to something else btw, they are in fact correct, the error also only ever happens when on the latest version, not on 2.10.0, also, the db is also set to retrieve the latest image, so the latest db and the 2.10.0 nginx works fine together, but latest db and latest nginx apparently does not, at least not in my case, restarting the db image doesn't work
@HaoSs07 commented on GitHub (Jun 13, 2023):
yah, this is new, my proxys work fine, can login in the manager and all the normal things but the log is throwing related error ( mine is on migration_lock):
[6/10/2023] [2:59:31 PM] [Global ] › ✖ error create table
migrations_lock(indexint unsigned not null auto_increment primary key,is_lockedint) - ER_CANT_CREATE_TABLE: Can't create tableDBNAME.migrations_lock(errno: 13 "Permission denied")@BartAgterbosch commented on GitHub (Jun 13, 2023):
Hm, then it sounds like I'm looking at the wrong error, but it seems to be the only error message in the entire log..
@bencollinz commented on GitHub (Jun 13, 2023):
This is the same problem with mine, continually. And same as you, have to downgrade/revert to get it working again.
2.9.18 is the last version that works for me.
@HaoSs07 commented on GitHub (Jun 13, 2023):
Fixed it,
just added
PUID: 1000
PGID: 1000
the id of my user to my environment: variables in the compose file. now it won't trow DB errors, ( function was added in 2.10 )
@bencollinz commented on GitHub (Jun 13, 2023):
on 2.10.3 this did not work for me. this is PUID: npm❌1000:1000::/tmp/npmuserhome:/bin/false and still getting
@tr7zw commented on GitHub (Jan 20, 2024):
Still an issue with v2.11.0 and SQLite file.
Downgrading to 2.10.0 did the trick for now.
@JimJafar commented on GitHub (Jan 20, 2024):
I've just come here with the same issue, also on
v2.11.0and using SQLite. I'm considering switching to MariaDB but I love the easy backup of a static file database. I'm using the PUID and PGID environment variables.I can confirm that downgrading to
v2.10.0allowed me to login again. Obviously that's not ideal, but at least it gets you back up and running.@github-actions[bot] commented on GitHub (Aug 14, 2024):
Issue is now considered stale. If you want to keep it open, please comment 👍
@github-actions[bot] commented on GitHub (Aug 18, 2025):
Issue was closed due to inactivity.