Homarr•2w ago

Latest update broke my Homarr

Error message: Application error: a server-side exception has occurred (see the server logs for more information). Digest: 1516925094 Been spending the whole night trying to fix this, I am running the PVEHS Proxmox LXC version, since it does not run with docker I dont know how to fault seek since there does not seem to be any documentation about it. Running behind Nginx Proxy Manager, been trying different ports to no avail. I have done my research but since I am very tired after no sleep my explaining is limited, sorry!

73 Replies

Cakey Bot•2w ago

Thank you for submitting a support request. Depending on the volume of requests, our team should get in contact with you shortly.

⚠️ Please include the following details in your post or we may reject your request without further comment: - Log (See https://homarr.dev/docs/community/faq#how-do-i-open-the-console--log) - Operating system (Unraid, TrueNAS, Ubuntu, ...) - Exact Homarr version (eg. 0.15.0, not latest) - Configuration (eg. docker-compose, screenshot or similar. Use ``your-text`` to format) - Other relevant information (eg. your devices, your browser, ...)

Frequently Asked Questions | Homarr documentation

Can I install Homarr on a Raspberry Pi?

monkeOP•2w ago

2855-d7df72897ab6a82...

Manicraft1001•2w ago

!qa required_data

Cakey Bot•2w ago

Thank you for your question. Our support team has noted that your request does not meet the minimum required data so we are unable to answer your question. Please provide: - your Homarr logs - your setup configuration (eg. docker compose, docker run, helm chart, screenshot, ...) - A detailed description of your problem or question with additional images where necessary - Your operating system - Version of Homarr (do not say latest! say 0.15.2, ...) Failure to comply with this will result in a permanent closure of your support request.

Manicraft1001•2w ago

Please provide the server logs, the .js file is useless for debugging

Dr0PeR•2w ago

Same here after update from tteck script in proxmox:

Dr0PeR•2w ago

1.6.0 to 1.7.0

Dr0PeR•2w ago

2855-d7df72897ab6a82...

Dr0PeR•2w ago

debian lxc container, proxmox If you need more info, pls explain how to provide it

Dr0PeR•2w ago

I can access to /manage

Dr0PeR•2w ago

Meierschlumpf•2w ago

Server logs menas not the ones in the browser and rather those that are probably shown from your proxmox instance

Dr0PeR•2w ago

where are located the logs? what route?

Manicraft1001•2w ago

https://homarr.dev/docs/community/faq/#how-do-i-open-the-console--log

Frequently Asked Questions | Homarr documentation

Can I install Homarr on a Raspberry Pi?

Manicraft1001•2w ago

You posted the browser console. We need your logs. Since you use Proxmox, which is not officially supported, we can't help you with getting them

Dok•2w ago

I use ProxMox LXC as well and I am getting the same error. Here is the server log that I get when trying to login with the only user I have setup for Homarr which is the admin account. I can access manage and look at the server logs from Tools > Logs.

Dok•2w ago

Here is the error attached.

server_error.txt

Manicraft1001•2w ago

@CrazyWolf13 can you assist?

Dok•2w ago

if it helps, I migrated from 0.15 to 1.5 before this and upgraded to 1.6 last week which was working okay.

CrazyWolf13•2w ago

honestly I'm a bit unsure if this is really our script's fault, we really exactly run the same entrypoint as you do in the dockerfile, though I experienced the same issue. For me a manual db migration solved the issue, so probably running systemctl restart homarr should solve the issue.

Manicraft1001•2w ago

Do you disable migrations in the script or do you not run the db:migration:sqlite:run command?

CrazyWolf13•2w ago

yeah they run under the homarr.service, running journalctl -u homarr shows them 🙂

monkeOP•2w ago

I tried systemctl restart homarr, did not solve my issue

Dok•2w ago

how do you do the manual db migration? I have just tried systemctl restart homarr and also a full reboot of the LXC, did not help.

CrazyWolf13•2w ago

hmm not sure, we run this: https://github.com/community-scripts/ProxmoxVE/blob/main/install/homarr-install.sh#L91 copied over from https://github.com/homarr-labs/homarr/blob/dev/scripts/run.sh#L11

GitHub

homarr/scripts/run.sh at dev · homarr-labs/homarr

A modern and easy to use dashboard. 14+ integrations. 10K+ icons built in. Authentication out of the box. No YAML, drag and drop configuration. - homarr-labs/homarr

GitHub

ProxmoxVE/install/homarr-install.sh at main · community-scripts/Pro...

Proxmox VE Helper-Scripts (Community Edition) . Contribute to community-scripts/ProxmoxVE development by creating an account on GitHub.

monkeOP•2w ago

If I clean the cookies I can access the login page for Homarr But as soon as I login it gives me the digest error

Manicraft1001•2w ago

In my opinion this is wrong and you should not copy the migrations directory. Run the script instead. @Meierschlumpf can you jump in?

Meierschlumpf•2w ago

What do you mean with copy the migrations? They are not copied and rather the migrate.cjs script is run with the path of the migrations directory

CrazyWolf13•2w ago

I was also kinda confused, because I got the same error on my instance, though running the command

export DB_DIALECT='sqlite'
export AUTH_SECRET=$(openssl rand -base64 32)
node /opt/homarr_db/migrations/$DB_DIALECT/migrate.cjs /opt/homarr_db/migrations/$DB_DIALECT

export DB_DIALECT='sqlite'
export AUTH_SECRET=$(openssl rand -base64 32)
node /opt/homarr_db/migrations/$DB_DIALECT/migrate.cjs /opt/homarr_db/migrations/$DB_DIALECT

it seemed to fix the issue. Though I have no idea why this is not applied automatically, as we basically run the same startup script as you do.

Meierschlumpf•2w ago

I mean in the dockerfile the following envs are defined:

ENV DB_URL='/appdata/db/db.sqlite'
ENV DB_DIALECT='sqlite'
ENV DB_DRIVER='better-sqlite3'
ENV AUTH_PROVIDERS='credentials'
ENV NODE_ENV='production'

ENV DB_URL='/appdata/db/db.sqlite'
ENV DB_DIALECT='sqlite'
ENV DB_DRIVER='better-sqlite3'
ENV AUTH_PROVIDERS='credentials'
ENV NODE_ENV='production'

So I guess that should work?

CrazyWolf13•2w ago

yes our env looks like this:

DB_DRIVER='better-sqlite3'
DB_DIALECT='sqlite'
SECRET_ENCRYPTION_KEY='${SECRET_ENCRYPTION_KEY}'
DB_URL='/opt/homarr_db/db.sqlite'
TURBO_TELEMETRY_DISABLED=1
AUTH_PROVIDERS='credentials'
NODE_ENV='production'

DB_DRIVER='better-sqlite3'
DB_DIALECT='sqlite'
SECRET_ENCRYPTION_KEY='${SECRET_ENCRYPTION_KEY}'
DB_URL='/opt/homarr_db/db.sqlite'
TURBO_TELEMETRY_DISABLED=1
AUTH_PROVIDERS='credentials'
NODE_ENV='production'

CrazyWolf13•2w ago

Maybe I got the issue, the migration also needs the secret_key, which was not visible to us, as it's not used here: https://github.com/homarr-labs/homarr/blob/dev/scripts/run.sh I'll update our scripts to properly load the secret too.

GitHub

homarr/scripts/run.sh at dev · homarr-labs/homarr

A modern and easy to use dashboard. 14+ integrations. 10K+ icons built in. Authentication out of the box. No YAML, drag and drop configuration. - homarr-labs/homarr

monkeOP•2w ago

Would running the update script again fix the issue? Once you changed the script then I mean

CrazyWolf13•2w ago

Yes as soon as the PR gets merged, you can re-run the update command and it should fix it 🙂

monkeOP•2w ago

Good! Thank you for the help 🙂

CrazyWolf13•2w ago

No problem, thanks @Manicraft1001 for the ping 🙂

Dok•2w ago

Awesome. You are all awesome! 🙂 Thanks for figuring this out. @CrazyWolf13 quick question. if the update command detects that homarr is already at 1.7.0 version and won't do anything, how will it apply the fix? Is there way to force it manually? I have tried updating the run_homarr.sh by adding the following at the beginning, that seems to not help. Is that the only thing needed for the script to fix the migration? set -a source /opt/homarr/.env set +a Running the following does not fix the issue for me for some reason. root@homarr:/opt/homarr# export DB_DIALECT='sqlite' root@homarr:/opt/homarr# export AUTH_SECRET=$(openssl rand -base64 32) root@homarr:/opt/homarr# node /opt/homarr_db/migrations/$DB_DIALECT/migrate.cjs /opt/homarr_db/migrations/$DB_DIALECT Skipping seeding of group 'everyone' as it already exists Skipping seeding of onboarding as it already exists Skipping seeding for serverSetting as it already exists key=analytics Skipping seeding for serverSetting as it already exists key=crawlingAndIndexing Skipping seeding for serverSetting as it already exists key=board Skipping seeding for serverSetting as it already exists key=appearance Skipping seeding for serverSetting as it already exists key=culture Skipping seeding for serverSetting as it already exists key=search Migration complete

CrazyWolf13•2w ago

That means the migration was successful so my fix worked, now after restarting the homarr service it should work again.

Dok•7d ago

I gave up and just recreated the Homarr LXC 😄

ficcoficco•6d ago

@CrazyWolf13 i update right now, and problem isnt fixed, so, miss i something?

CrazyWolf13•6d ago

@ficcoficco You are right, weirdly. No idea why it worked for me on my prod instance. I've just been able to replicate your problem when upgrading from v1.5.0 to v1.7.9. as one can see in the below log, the migrations seems to be run successfully, however it does not fix, so I kindly request @Meierschlumpf / @Manicraft1001 To look again into this. We literally use the exact same migration command as you in the dockerfile and I really cannot get any valid point, why this issue should be on our scripts' side. https://pastebin.com/1QVAdgqm

Pastebin

Feb 25 23:04:51 homarr systemd[1]: Stopped homarr.service - Homarr ...

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.

monkeOP•6d ago

I did exactly the same haha

Manicraft1001•6d ago

@CrazyWolf13 do you... - run the migration on every Homarr startup? If no, please do so. - run the migration on the same sqlite as you use for the application?

CrazyWolf13•6d ago

@Manicraft1001 Yes, again, we use nearly exactly the same startup as your docker entrypoint does. https://github.com/community-scripts/ProxmoxVE/blob/main/install/homarr-install.sh#L87-L102 I think so? Our DB is located in /opt/homarr_db/db.sqlite

GitHub

ProxmoxVE/install/homarr-install.sh at main · community-scripts/Pro...

Proxmox VE Helper-Scripts (Community Edition) . Contribute to community-scripts/ProxmoxVE development by creating an account on GitHub.

Manicraft1001•6d ago

And from what version to what version does this problem happen? Can you reproduce it in Docker as well?

CrazyWolf13•6d ago

I've not tested that. But can ofc. I think 1.5 to 1.7 it was

Dok•6d ago

yeah 1.5 to 1.6 was fine. upgrading to 1.7 seem to not work.

Manicraft1001•6d ago

Would be great if you can try to reproduce. We are very busy at the moment If you can reproduce it, send me the DB from 1.6 and I'll try to see what is going on Oh, and please remove any personal data or urls from the DB. Use public URLs if possible, so you don't "leak" any of your private setup

ficcoficco•6d ago

im on 1.6 updated without problem from 1.5.. in my setup, problem is from 1.6 to 1.7

CrazyWolf13•6d ago

https://pastebin.com/ev65egYr

Pastebin

2025-02-26 19:37:27 homarr | Starting with UID='0', GID='0'2025-02...

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.

Manicraft1001•6d ago

This log is from...? Except line 2, it looks normal to me

CrazyWolf13•6d ago

Docker-log @Manicraft1001 Can't really reproduce on docker :/ I'll reproduce on LXC and send the db 🙂

Manicraft1001•6d ago

Thanks for debugging. Then it's likely an issue in the script. Good to know Do you properly persist /appdata?

CrazyWolf13•6d ago

While I totally agree, it's kind of a mystery to me, everything in that matter is so similar to your startup, though maybe something pops up in the db 🙂

Manicraft1001•6d ago

Not sure how Proxmox handles storage

CrazyWolf13•5d ago

@Manicraft1001 Most likely I found the issue. Why the heck is your migration now writing to a subfolder:

node /opt/homarr_db/migrations/$DB_DIALECT/migrate.cjs /opt/homarr_db/migrations/$DB_DIALECT

node /opt/homarr_db/migrations/$DB_DIALECT/migrate.cjs /opt/homarr_db/migrations/$DB_DIALECT

Since v1.5.0 or v1.6.0 this seems to create a new folder in /opt/homarr_db/migrations/migrations/sqlite instead of /opt/homarr_db/migrations/sqlite Did you change anything on homarr side in this matter? We temp. solved it with a recursive copy, though this seems a bit odd.

Meierschlumpf•5d ago

Actually this one is new since last week:

{
          DB_URL:
            // Fallback to the default sqlite file path in production
            commonEnv.NODE_ENV === "production" && isDriver("better-sqlite3")
              ? z.string().default("/appdata/db/db.sqlite")
              : z.string().nonempty(),
        }

{
          DB_URL:
            // Fallback to the default sqlite file path in production
            commonEnv.NODE_ENV === "production" && isDriver("better-sqlite3")
              ? z.string().default("/appdata/db/db.sqlite")
              : z.string().nonempty(),
        }

Previously we did not fallback to /appdata/db/db.sqlite, not sure if that is also relevant? Not sure why it would have two times migrations

CrazyWolf13•5d ago

hmm I see. Though the DB is picked up fine in our folder /opt/homarr_db/db.sqlite. Anyway, we fixed that by just copying the migration files a folder up. Just wanted to see if maybe something on your side changed.

Dok•3d ago

@CrazyWolf13 1.7 --> 1.8 broke again when I ran update command on the LXC. Is there anything specific I need to do to fix it? 😄

Meierschlumpf•3d ago

Oh no 🫣 Yet sgain a problem with migration?

CrazyWolf13•3d ago

restart the service a second time and it should work? e.g. reboot the lxc or the service homarr. tested and for me that worked 🙂 i'll take a look at doing this directly during update.

Dr0PeR•3d ago

Before update mine, I taked a backup, then update and same result as update from 1.6 to 1.7. Now I think I'll stay on 1.7 forever :'), It works fine so no problem for me, just, dont update and keep it outside internet until someone (homarr devs or lxc script maintainer) fix this.

CrazyWolf13•3d ago

it's fixed 🙂 didn't I post here? wait where did my message go?

Manicraft1001•3d ago

It was deleted because you mentioned too many people We block that due to recent spam attempts

CrazyWolf13•3d ago

Oh I see. can so shall I not ping them or can I split it up into multiple messages?

Manicraft1001•3d ago

Split up if possible

CrazyWolf13•3d ago

@Dok @Dr0PeR
Hi y'all A fix has been implemented and the next homarr release has been made, so now you can all update, using the update command or re-running the installation command on our website. If anyone still sees the application-error page after the update, just restart the homarr service or reboot the LXC. (PR merged.) @fcpwiz @ficcoficco @Magickaiser @monke @Snail02

Dr0PeR•3d ago

@CrazyWolf13 Ok, I'll try right now It's working ;D after the update you need to restart the container and that's it

Manicraft1001•3d ago

Nice to know. Can we mark this as resolved then?

Dr0PeR•3d ago

I think so. Thanks for all the support, both from the devs and the maintainer and the rest of the users who have helped.

Gaming

Programming

Latest update broke my Homarr

Did you find this page helpful?