nextcloud/tests/QA/010-restore-instance.md

# Restore instance

For the below to work, you need a backup archive of an AIO instance and the location on the test machine and the password for the backup archive. You can get one here: [backup-archive](./assets/backup-archive/)

- [ ] The section that allows to restore the whole AIO instance from backup should show three input fields: one that allows to enter a location where the backup archive is located and one that allows to enter a remote ssh path and one that allows to enter password of the archive. It should also show a short explanation regarding the path requirements
- [ ] First, check restoring from a local backup location:
    - [ ] Entering an incorrect path and/or password should let you continue and test your settings in the next step
        - [ ] Clicking on the test button should after a reload bring you back to the initial screen where it should say that the test was unsuccessful. Also you should be able to have a look at the backup container logs for investigation what exactly failed.
        - [ ] You should also now see the input boxes again where you can change the path and password, confirm it and bring you again to the screen where you can test your settings.
    - [ ] Entering the correct path to the backup archive and the correct password here should:
        - [ ] Should reload and should hide all options except the option to test the path and password
        - [ ] After the test you should see the options to check the integrity of the backup and a list of backup archives that you can choose from to restore your instance
        - [ ] Clicking on either option should show a window prompt that lets you cancel the operation
        - [ ] Clicking on the integrity check option should check the integrity and report that the backup integrity is good after a while which should then only show the option to choose the backup archive that should be restored
        - [ ] Choosing the restore option should finally restore your files.
        - [ ] After waiting a while it should reload the page and should show the usual container interface again with the state of your containers (stopped) and the option to start and update the containers again.
- [ ] Next, check restoring from a remote backup location via SSH. The remote borg repo URL must contain both `@` and `:`. The restore process works as follows:
    1. You enter a remote borg repo URL (e.g. `ssh://user@host:port/path/to/repo` or `user@host:/path/to/repo`) and the backup password.
    2. On the first connection attempt, a SSH key pair is generated automatically and the public key is displayed.
    3. You add the public key to the `~/.ssh/authorized_keys` file on the remote server so that AIO can connect to it.
    4. Once authorized, AIO can list and restore backups from the remote server.
    - [ ] Enter an invalid remote repo URL (e.g. `user` without `@` and `:`) which should send an error
    - [ ] Enter a valid remote borg repo URL and the correct backup password:
        - [ ] Should reload and should hide all options except the option to test the path and password
        - [ ] After the first failed connection attempt, the SSH public key for borg should be shown so it can be authorized on the remote server
        - [ ] After authorizing the key on the remote server, scroll down and click on the test button again. This time it should succeed and show the options to check the integrity and list backup archives
        - [ ] After the test you should see the options to check the integrity of the backup and a list of backup archives that you can choose from to restore your instance
        - [ ] Clicking on either option should show a window prompt that lets you cancel the operation
        - [ ] Clicking on the integrity check option should check the integrity and report that the backup integrity is good after a while which should then only show the option to choose the backup archive that should be restored
        - [ ] Choosing the restore option should finally restore your files.
        - [ ] After waiting a while it should reload the page and should show the usual container interface again with the state of your containers (stopped) and the option to start and update the containers again.

- [ ] Clicking on `Start and update containers` should show a window prompt that you should create a backup. Canceling should cancel the operation, confirming should reveal the big spinner again.
- [ ] After waiting a bit, all containers should be green and your instance should be fully functional again

You can now continue with [020-backup-and-restore.md](./020-backup-and-restore.md)