Hi community,
I’m one of the maintainers of Portabase, and this is my first time sharing about it on Lemmy.
Portabase is an open-source platform for database backup and restore.
It’s designed to be simple, reliable, and lightweight, without exposing your databases to public networks. It works via a central server and edge agents (like Portainer), making it perfect for self-hosted or edge environments.
It currently supports 7 databases:
PostgreSQL, MariaDB, MySQL, SQLite, MongoDB, Redis and Valkey
Repository: https://github.com/Portabase/portabase
(we hit 500 stars recently!)
Key features:
- Logical backups for PostgreSQL, MySQL, MariaDB, MongoDB, SQLite, Redis, Valkey
- Multiple storage backends: local filesystem, S3, Cloudflare R2, Google Drive
- Notifications via Discord, Telegram, Slack, webhooks, etc.
- Cron-based scheduling with flexible retention strategies
- Agent-based architecture for secure, edge-friendly deployments
- Ready-to-use Docker Compose setup and Helm Chart
What’s coming next:
- Increasing test coverage
- Extending database support
I’d love to hear from you: which database would you like to see supported next in Portabase?
Personally, I think going local file config first was the right decision, depending on what the target audience is. For homelab users with one or two databases it may seem like extra work. For professionals or advanced users who like, or even need, to automate using tools like Ansible etc this makes it easier to deploy on larger scale. You can of course allow users to update settings after the fact through the UI while keeping the json file. I have dropped projects in the past due to the exact issue that its only configurable to the UI that makes bulk operations or automation hard or impossible.
I’m currently using shell scripts (pg_dump) and crontabs to back up my databases, so thanks, will check it out!
Thanks for sharing your use case. To be transparent, we didn’t consider automation when we created the local file config, but it’s indeed the perfect use case for this mode of operation. In any case, we’ll keep the current behaviour to avoid breaking changes. This will be introduced as an additional feature.