Cloudron Forum

A few days ago I had a somewhat "working" version of Azuracast actually running.

It was one of the apps I temporarily uninstalled to make space for solving the problem with the newly Rocket.Chat log-filling-the-disk problem.

I have now reinstalled Azuracast, but it refuses to start, giving this strange message in the logs

Jan 28 12:45:22 2025-01-28T11:45:22Z
Jan 28 12:45:22 Exception trace:
Jan 28 12:45:22 at /app/code/www/vendor/doctrine/dbal/src/Driver/PDO/Exception.php:28
Jan 28 12:45:22 Doctrine\DBAL\Driver\PDO\Exception::new() at /app/code/www/vendor/doctrine/dbal/src/Driver/PDO/Connection.php:62
Jan 28 12:45:22 Doctrine\DBAL\Driver\PDO\Connection->query() at /app/code/www/vendor/doctrine/dbal/src/Connection.php:772
Jan 28 12:45:22 Doctrine\DBAL\Connection->executeQuery() at /app/code/www/vendor/doctrine/migrations/lib/Doctrine/Migrations/Version/DbalExecutor.php:273
Jan 28 12:45:22 Doctrine\Migrations\Version\DbalExecutor->executeResult() at /app/code/www/vendor/doctrine/migrations/lib/Doctrine/Migrations/Version/DbalExecutor.php:150
Jan 28 12:45:22 Doctrine\Migrations\Version\DbalExecutor->executeMigration() at /app/code/www/vendor/doctrine/migrations/lib/Doctrine/Migrations/Version/DbalExecutor.php:72
Jan 28 12:45:22 Doctrine\Migrations\Version\DbalExecutor->execute() at /app/code/www/vendor/doctrine/migrations/lib/Doctrine/Migrations/DbalMigrator.php:88
Jan 28 12:45:22 Doctrine\Migrations\DbalMigrator->executePlan() at /app/code/www/vendor/doctrine/migrations/lib/Doctrine/Migrations/DbalMigrator.php:54
Jan 28 12:45:22 Doctrine\Migrations\DbalMigrator->executeMigrations() at /app/code/www/vendor/doctrine/migrations/lib/Doctrine/Migrations/DbalMigrator.php:136
Jan 28 12:45:22 Doctrine\Migrations\DbalMigrator->migrate() at /app/code/www/vendor/doctrine/migrations/lib/Doctrine/Migrations/Tools/Console/Command/MigrateCommand.php:219
Jan 28 12:45:22 Doctrine\Migrations\Tools\Console\Command\MigrateCommand->execute() at /app/code/www/vendor/symfony/console/Command/Command.php:279
Jan 28 12:45:22 Symfony\Component\Console\Command\Command->run() at /app/code/www/vendor/symfony/console/Application.php:1049
Jan 28 12:45:22 Symfony\Component\Console\Application->doRunCommand() at /app/code/www/vendor/symfony/console/Application.php:318
Jan 28 12:45:22 Symfony\Component\Console\Application->doRun() at /app/code/www/vendor/symfony/console/Application.php:169
Jan 28 12:45:22 Symfony\Component\Console\Application->run() at /app/code/www/bin/console:12
Jan 28 12:45:22 2025-01-28T11:45:22Z
Jan 28 12:45:22 In Connection.php line 57:
Jan 28 12:45:22
Jan 28 12:45:22 [PDOException (42S21)]
Jan 28 12:45:22 SQLSTATE[42S21]: Column already exists: 1060 Duplicate column name 'lyrics' 

SQLSTATE[42S21]: Column already exists: 1060 Duplicate column name 'lyrics'?

What could possibly occupy that column name from....other apps....or something else?

Sorry, I forgot to create the directory and make cloudron the owner. Done

@girish thanks a lot for helping.

I have had some frustrating hours, had to reinstall the whole cloudron in a 16GB ram droplet in order to be able to reinstall latest backup of Rocket.Chat, that backup was over 7GB, while only 2GB before the logtrouble startet.

Anyway, have it installed and working, and in Rocket.Chat made the exact setting you advice above. But in the Rocket.Chat terminal /app/data I can see no "app_logs" file?

Also when I go into mongodb and issue the commands you suggest, I get the following error message:

root@1116b395-909e-4f51-bd9e-9ad9bed3846e:/app/data# mongosh -u "${CLOUDRON_MONGODB_USERNAME}" -p "${CLOUDRON_MONGODB_PASSWORD}" ${CLOUDRON_MONGODB_HOST}:${CLOUDRON_MONGODB_PORT}/${CLOUDRON_MONGODB_DATABASE} db.getCollection('rocketchat_apps_logs').countDocuments()
bash: syntax error near unexpected token `('

I would like to have a look in mongodb, and delete logentries there if there are any. So what to do with these two issues?

@girish I have cloudron version 8.2.3, but can't find any "apps" konfiguration under "Settings". Where do I find the settings to change the log destination?

I have the same "filling up space" problem on two cloudrons I administer. They are not working, as space is totally filled up. Is there a fix? What can be done?

@joseph Thanks for your answer.

Hi team

It looks to me that cloudron packaged AzuraCast is the rolling release instead of stable. (I see it in the /admin/updates/ section.

Why not use the stable instead? There seems to be many bugs in the rolling release version that make it almost impossible to run a working radio station.

Any thoughts?

Same issue here too.

Problem solved. I deleted ALL config files and restarted the server.

Hi @girish

I have a similar problem (missing certificate so nginx does not start)

root@my:~# journalctl -u nginx -fa
-- Logs begin at Wed 2022-08-03 20:41:55 UTC. --
Aug 03 23:45:02 my nginx[17769]: nginx: [emerg] cannot load certificate key "/home/yellowtent/platformdata/nginx/cert/_.domainnamexxxx.info.key": PEM_read_bio_PrivateKey() failed (SSL: error:0906D06C:PEM routines:PEM_read_bio:no start line:Expecting: ANY PRIVATE KEY)

I find app config files in /home/yellowtent/platformdata/nginx/applications

But the nginx-log messages does not tell what app exactly is creating the problem.

May I delete all config files, or do I have to go "trial and error" with each file to figure it out?

Where do I find the nginx config files for the app relating to the missing certificate?

@nebulon Thank you for having a look at it.

@nebulon said in Increase character limit for toots from 500 to xxxx:

In your case I guess you could fork the mastodon app package and create your own with that patchset.

Holy moly, unknown territory

Hi @nebulon

I am not sure this is directly from upstream docs, but here is one suggestion:

https://draklyckan.se/2021/11/how-to-increase-the-character-limit-for-toots-in-mastodon/

The above article mention these sources:

https://github.com/mastodon/mastodon/issues/12265
https://github.com/kamysheblid/mastodon_max_characters
https://old.reddit.com/r/Mastodon/comments/hozl70/
https://docs.joinmastodon.org/admin/

Hi all

The default character limit for new toots is 500. But I would like to increase that limit for my special instance.

How can I do this? And hopefully make it a permanent change, not overwritten by updates?

Any help?

Re: Graphite stuck starting

Hi all. I have the same problem with graphite stuck starting. I have the same type of log entries to that of @eyecreate , and I have upped the allocated memory, and have tried the @nebulon suggested method of purging graphite database with no luck so far.

What to do?

I had to install 18.04, then install Cloudron 5.5.0, then restore the backed up cloudron, then upgrade it to 6.1.2, then make a backup.
Then destroy the 18.04 and install 20.04 instead, then restore the 6.1.2 backup....puhh. Done, working!

(Could not restore a 26GB Nextcload app, but no problem, had the files backed up locally)

Guess this thread can be marked as solved?

Hold on a minute....seems like restore worked well after all....strange thing....I check it out and report back....

@nebulon

Thank you. I tried using 18.04, installation went well, but when I picked the restore option, filled in the details for downloading my my DO Spaces backup, I tried once and forgot archive encryption key, then filled in the key, but now got error message:

"Already activated. Restore with a fresh Cloudron installation."

Maybe its easier to install a fresh cloudron, and have you guys connect it to the running subscription?

Please advice how to proceed.

Hmm.....

I actually reinstalled the server OS (Ubuntu 20.04) and started all over from scratch.

./cloudron-setup --provider generic --version 5.5.0

Then installation stopped and gave error msg. From /var/log/cloudron-setup.log:

Package mysql-server-5.7 is not available, but is referred to by another package.
This may mean that the package is missing, has been obsoleted, or
is only available from another source
However the following packages replace it:
  mariadb-server-10.3

E: Package 'mysql-server-5.7' has no installation candidate

So, as the cloudron I try to reinstall is a subscription cloudron (paid), but with no critical information stored on it, would it be easier to just install a fresh latest version cloudron, and have you guys (support) change the running subscription so it is connected to this new cloudron ID?

Hi

I try to restore a backed up cloudron (upgraded server OS), but setup the wrong cloudron version on my new server. Need to revert to older cloudron version that corresponds with the backup version.

Is there a way to force cloudron to revert to an earlier version, or do I simply need to uninstall the cloudron and install the correct version?

What is the best way to do it? Is there a working command for "uninstall cloudron", or do I have to reinstall the whole server OS and start over again?

Thanks

@girish said in New installation, not able to log in:

echo "changeme" | npm run reset-password -- -u root

Thank you @girish. I actually had a lot running on this cloudron, but could not get PeerTube to work even if I removed several other apps to free up memory.

But the above command for some strange reason fixed it and solved the problem. PeerTube is now working!

Thank you