The fanbase is still large, but the Lemmy community hasn’t quite caught up yet, and now there is a transitional period where the audience is smaller.
The fanbase is still large, but the Lemmy community hasn’t quite caught up yet, and now there is a transitional period where the audience is smaller.
Please do it more often if you have users other than yourself. One backup on the same server is barely a backup at all.
Fair enough. I’ll look into automating it using some sort of storage from another provider.
Even just a cronjob or scheduled task to download the backups to a machine at another location would be a big improvement. Then you can do it far more often because it’s automated.
But personally I like to have both a copy on a PC and a cloud backup, in addition to the server.
I’m using the easy Lemmy script to run the docker instance. How do I take a backup of a running docker instance.
The backups I’ve done so far are full shard backups. But I don’t have a way to automate that.
The page here explains getting a database dump on a running instance (and how to restore): https://join-lemmy.org/docs/administration/backup_and_restore.html
Then just back up the other files in the volumes directory where Lemmy is installed (everything except postgres, which is what the database dump does).
The pictrs volume includes both the uploaded images and the image cache. I have no idea how to separate out the uploaded images so you don’t have to back up the cache, I just back it all up.
this is the bash script I use to create backups
#! /bin/bash # https://join-lemmy.org/docs/administration/backup_and_restore.html#a-sample-backup-script now=$(date +"%Y-%m-%d_%H.%M.%S") cd ~/lemmy && (docker-compose exec -T postgres pg_dumpall -c -U lemmy 1> dump.sql 2> dump.errors) cd ~/lemmy && zip -r9 ~/bak-lemmy-$now.zip ./ --exclude "volumes/postgres/*" rm -f ~/lemmy/dump.sqlit creates very small zip files as a result so it’s very efficient
I made a cron for it to run every 3 hours, like
I figured out how to do this with docker container, but that’s not ideal for a script.
Using docker compose it just fails with: Service “postgres” is not running container #1
I can see lemmy-easy-deploy if I do: docker compose ls
The service name is postgres in the docker-compose.yml file. Any idea what the issue might be?
Where is this lemmy-easy-deploy? I haven’t seen that before, maybe if I read how it works I can figure out what’s wrong
https://github.com/ubergeek77/Lemmy-Easy-Deploy
I think you might just need to change the
cds to go into the correct directory where the activedocker-compose.ymlfile is, which should be in the folder calledliveBackblaze is fairly cheap but can be slow to get data from.