Update Nextcloud 20.0.4 -> 20.0.5 gets stuck

Dear Linuxserver.io community,

I’m having problems with an update of Nextcloud. The code update went fine, but I’m getting errors when I try to update the database. The error is described in more detail in this Github issue. I’m using Swag as the reverse proxy.

Did anyone experience similar problems? Also, I don’t know exactly how to proceed. I inspected network and got the container IP addresses of the MariaDB container. Then I was able to ping it from within the Nextcloud container, so at least the two containers can see each other. I’m happy for all ideas! :slight_smile:

If it’s showing connection refused, It sounds like their could be an issue with your database server itself. Is it working correctly?

Thanks for your answer! There is really an error, apparently:

211030 08:18:03 mysqld_safe Starting mariadbd daemon with databases from /config/databases
2021-10-30  8:18:04 0 [Note] /usr/bin/mariadbd (mysqld 10.5.12-MariaDB-log) starting as process 31067 ...
2021-10-30  8:18:04 0 [Note] InnoDB: Uses event mutexes
2021-10-30  8:18:04 0 [Note] InnoDB: Compressed tables use zlib 1.2.11
2021-10-30  8:18:04 0 [Note] InnoDB: Number of pools: 1
2021-10-30  8:18:04 0 [Note] InnoDB: Using generic crc32 instructions
2021-10-30  8:18:04 0 [Note] mariadbd: O_TMPFILE is not supported on /var/tmp (disabling future attempts)
2021-10-30  8:18:04 0 [Note] InnoDB: Using Linux native AIO
2021-10-30  8:18:04 0 [Note] InnoDB: Initializing buffer pool, total size = 268435456, chunk size = 134217728
2021-10-30  8:18:04 0 [Note] InnoDB: Completed initialization of buffer pool
2021-10-30  8:18:04 0 [ERROR] InnoDB: Upgrade after a crash is not supported. The redo log was created with MariaDB 10.4.21.
2021-10-30  8:18:04 0 [ERROR] InnoDB: Plugin initialization aborted with error Generic error
2021-10-30  8:18:04 0 [Note] InnoDB: Starting shutdown...

Luckily I have a backup from before the update, but that will probably need some time. Is there a way to fix the database without a full restore?

Are you using our mariadb container aswell? Also, what host OS are you using?

Yes, I’m using the LinuxServer.io coontainer. OS is Debian 10.11

Ok, so the reason this is likely happened is, when you upgraded mariadb, there were outstanding transactions or nextcloud was writing to it at the time which coursed the corruption. Now, this can normally fix itself providing your only upgrade to a minor version of the software but as there was a major upgrade of our container recently, it can’t do both repair and upgrade at the same time.

You need to roll back to the previous working tag, likely version-110.4.21mariabionic, let the container start up cleanly, then stop the container - make sure anything using it has been shutdown first. Once this has been done, go back to :latest and mariadb should now boot backup correctly.

Thank you so much! This solved the issue. I will refer the Github issue to this thread and close it. Thanks again, you saved me! :slight_smile: