mirror of
https://github.com/BookStackApp/BookStack.git
synced 2026-02-06 09:09:38 +03:00
Suddenly get a 500 errors #4768
Reference in New Issue
Block a user
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 @ZuluPilgrim on GitHub (May 13, 2024).
Attempted Debugging
Searched GitHub Issues
Describe the Scenario
I was working in my bookstack when I suddenly could not save. I thought my vm had corrupted so did a restore but the site no longer loads when I look at the log files I see the below.
I have tried with both the URL and local IP.
Details
I haven't changed anything in my docker compose file or tried restarting the container so not sure hat has gone wrong, Also the restore would have taken it back to two days ago when it was working.
My compose file is
Exact BookStack Version
Not sure how to find this now.
Log Content
No response
Hosting Environment
Static hostname: webserver-01
Icon name: computer-vm
Chassis: vm
Machine ID: ed09422d73064b2f927c9abfdfc36784
Boot ID: e590bd70b8fc498c8c774ced9a2e2a4b
Virtualization: xen
Operating System: Ubuntu 22.04.4 LTS
Kernel: Linux 5.15.0-106-generic
Architecture: x86-64
Hardware Vendor: Xen
Hardware Model: HVM domU
@ssddanbrown commented on GitHub (May 13, 2024):
Hi @ZuluPilgrim, Looks like a common issue of the app not being able to access the database.
bookstack_app_data/www/.envfile, and where possible confirm the values of any options starting withDB_.@ZuluPilgrim commented on GitHub (May 13, 2024):
@ssddanbrown thanks for getting back to me. I just tried doing some troubleshooting while away from home and now I van access the site.
I tried editing the document I was working on and all seemed to go well until I hit save as the page eventually returned an unknown error. I then tried logging off and back on again. I got to the MFA page but then it all wen pear-shaped again.
Here is what you asked for
Which are correct as I can now browse to the site again.
It looks like I am going to need to wait until the MFA starts working again. Tried incognito mode and that made no difference.
Actually the whole site is now becoming unresponsive again.
I might wait until it is up again and then pull all the docs our and stand up a new instance.
Just got: 504 Gateway Time-out
VM has heaps of disk space and memory.
Also other docker containers are running fine on this VM.
Should I try upgrading to the latest Bookstack? If so, do I just remove existing container and then re-run docker compose as it stands?
@ssddanbrown commented on GitHub (May 14, 2024):
Okay, I'm a little confused on the state of things now and what's been changed so I'll put out some options:
Original Database Connection Error
If this was the issue things should not be intermittently working unless there's also a networking issue between containers or something like that.
The original error you had appeared to be a connection error to the database. Based on the details provided, it looks like there's been a database password change at some time since the .env file
DB_PASSWORDdoes not align with the composeDB_PASSoption. On both the database and app container, the password options can be a bit "sticky" (and don't act upon changes) so it's best to leave these as the original value (which I'd guess isdbpasswordunless the env has been manually changed since initial run)."An error has occurred" message
If you see a "An error has occurred" message, with a BookStack-style header bar still showing, this is an app-level error so should be logged to the app error log, found at
bookstack_app_data/www/laravel.log. Find the latest messages for that which have a timestamp that align with when you got the error to show, then share back the error.504 Gateway Time-out
Generally indicates a webserver/proxy level issue. Check the
bookstack_app_data/log/nginx/error.logfile for more details. If nothings there, check the logs for any proxy layers involved if you have any.IMO that risks just throwing more variables into the picture, think it'd be rare to be something based on the particular version of image you're using.
@ZuluPilgrim commented on GitHub (May 14, 2024):
I am not sure what was happening. As I work in IT I know things generally don't break without someone doing something but I really don't remember making any changes and I have three other instances on other VMs running with no issue.
I felt there was something in the container that was off so I pulled it down, pruned all aspects from docker and repulled it from the source.
Everything is now up and running again.
Thanks for walking me through this all.