Automated backups have stopped

[Isambard]

Automated backups have stopped. But manually triggered ones still work. Is there a way to debug the problem? Frequency is set to daily. There was a change up upgrade to admin sidebar, but backups still worked for a few days after this change.

[merefield (Robert)]

Same happened to me last night.

There should be a link to the log in the notification if the automated backup failed.

My log tells me that upon trying to zip there was not enough space left. Probably not cleared the old database file? I’ll check.

[merefield (Robert)]

No, for me it was too many old images.

In any case try:

./launcher cleanup app whilst your site is still running and this is likely to help in any case.

[Isambard]

There was no failure message. It was as if it never attempted to do the backup. There’s no shortage of disk space. There are only 6 backups as I have the setting enabled that has it auto-delete older backups when there are too many.

Is there an easy way to trigger the backup externally. Maybe I’ll script it in cron on the host instead.

[Canapin (Coin-coin le Canapin)]

@Isambard

Is there an easy way to trigger the backup externally

Yes, you can run backups from the command line:

@supermathie

You have a scheduler outside of the container available, right?

You can schedule it there, e.g.:

# put in host crontab
# run backup once every 4 hours on the hour
0 */4 * * * docker exec app discourse backup --sql-only

[Isambard]

docker exec app discourse backup --sql-only

works with a discourse site with one forum. How do you specify the forum to backup with a multi-site setup?

EDIT: this command works

cd /var/www/discourse && RAILS_DB=chosenforum RAILS_ENV=production sudo -H -E -u discourse bundle exec script/discourse backup --sql-only

or with full docker command:

docker exec containerID bash -c 'cd /var/www/discourse; RAILS_DB=chosenforum RAILS_ENV=production sudo -H -E -u discourse bundle exec script/discourse backup --sql-only'