2020-01-25 05:52:48 +01:00
|
|
|
---
|
2020-01-25 06:02:56 +01:00
|
|
|
title: How to backup to a removable drive or an intermittent server
|
2020-01-25 05:52:48 +01:00
|
|
|
---
|
|
|
|
## Occasional backups
|
|
|
|
|
|
|
|
A common situation is backing up to a repository that's only sometimes online.
|
|
|
|
For instance, you might send most of your backups to the cloud, but
|
|
|
|
occasionally you want to plug in an external hard drive or backup to your
|
|
|
|
buddy's sometimes-online server for that extra level of redundancy.
|
|
|
|
|
|
|
|
But if you run borgmatic and your hard drive isn't plugged in, or your buddy's
|
|
|
|
server is offline, then you'll get an annoying error message and the overall
|
|
|
|
borgmatic run will fail (even if individual repositories complete just fine).
|
|
|
|
|
|
|
|
So what if you want borgmatic to swallow the error of a missing drive
|
|
|
|
or an offline server, and continue trucking along? That's where the concept of
|
|
|
|
"soft failure" come in.
|
|
|
|
|
|
|
|
## Soft failure command hooks
|
|
|
|
|
|
|
|
This feature leverages [borgmatic command
|
|
|
|
hooks](https://torsion.org/borgmatic/docs/how-to/add-preparation-and-cleanup-steps-to-backups/),
|
|
|
|
so first familiarize yourself with them. The idea is that you write a simple
|
|
|
|
test in the form of a borgmatic hook to see if backups should proceed or not.
|
|
|
|
|
|
|
|
The way the test works is that if any of your hook commands return a special
|
|
|
|
exit status of 75, that indicates to borgmatic that it's a temporary failure,
|
|
|
|
and borgmatic should skip all subsequent actions for that configuration file.
|
|
|
|
If you return any other status, then it's a standard success or error. (Zero is
|
|
|
|
success; anything else other than 75 is an error).
|
|
|
|
|
|
|
|
So for instance, if you have an external drive that's only sometimes mounted,
|
|
|
|
declare its repository in its own [separate configuration
|
|
|
|
file](https://torsion.org/borgmatic/docs/how-to/make-per-application-backups/),
|
|
|
|
say at `/etc/borgmatic.d/removable.yaml`:
|
|
|
|
|
|
|
|
```yaml
|
|
|
|
location:
|
|
|
|
source_directories:
|
|
|
|
- /home
|
|
|
|
|
|
|
|
repositories:
|
|
|
|
- /mnt/removable/backup.borg
|
|
|
|
```
|
|
|
|
|
|
|
|
Then, write a `before_backup` hook in that same configuration file that uses
|
|
|
|
the external `findmnt` utility to see whether the drive is mounted before
|
|
|
|
proceeding.
|
|
|
|
|
|
|
|
```yaml
|
|
|
|
hooks:
|
|
|
|
before_backup:
|
|
|
|
- findmnt /mnt/removable > /dev/null || exit 75
|
|
|
|
```
|
|
|
|
|
|
|
|
What this does is check if the `findmnt` command errors when probing for a
|
|
|
|
particular mount point. If it does error, then it returns exit code 75 to
|
|
|
|
borgmatic. borgmatic logs the soft failure, skips all further actions in that
|
|
|
|
configurable file, and proceeds onward to any other borgmatic configuration
|
|
|
|
files you may have.
|
|
|
|
|
|
|
|
You can imagine a similar check for the sometimes-online server case:
|
|
|
|
|
|
|
|
```yaml
|
|
|
|
location:
|
|
|
|
source_directories:
|
|
|
|
- /home
|
|
|
|
|
|
|
|
repositories:
|
|
|
|
- me@buddys-server.org:backup.borg
|
|
|
|
|
|
|
|
hooks:
|
|
|
|
before_backup:
|
|
|
|
- ping -q -c 1 buddys-server.org > /dev/null || exit 75
|
|
|
|
```
|
|
|
|
|
|
|
|
## Caveats and details
|
|
|
|
|
|
|
|
There are some caveats you should be aware of with this feature.
|
|
|
|
|
|
|
|
* You'll generally want to put a soft failure command in the `before_backup`
|
|
|
|
hook, so as to gate whether the backup action occurs. While a soft failure is
|
|
|
|
also supported in the `after_backup` hook, returning a soft failure there
|
|
|
|
won't prevent any actions from occuring, because they've already occurred!
|
|
|
|
Similiarly, you can return a soft failure from an `on_error` hook, but at
|
|
|
|
that point it's too late to prevent the error.
|
|
|
|
* Returning a soft failure does prevent further commands in the same hook from
|
|
|
|
executing. So, like a standard error, it is an "early out". Unlike a standard
|
|
|
|
error, borgmatic does not display it in angry red text or consider it a
|
|
|
|
failure.
|
|
|
|
* The soft failure only applies to the scope of a single borgmatic
|
|
|
|
configuration file. So put anything that you don't want soft-failed, like
|
|
|
|
always-online cloud backups, in separate configuration files from your
|
|
|
|
soft-failing repositories.
|
|
|
|
* The soft failure doesn't have to apply to a repository. You can even perform
|
|
|
|
a test to make sure that individual source directories are mounted and
|
|
|
|
available. Use your imagination!
|
2020-01-27 20:07:07 +01:00
|
|
|
* The soft failure feature also works for `before_prune`, `after_prune`,
|
|
|
|
`before_check`, and `after_check` hooks. However it is not implemented for
|
|
|
|
`before_everything` or `after_everything`.
|
2020-01-25 05:52:48 +01:00
|
|
|
|
|
|
|
## Related documentation
|
|
|
|
|
|
|
|
* [Set up backups with borgmatic](https://torsion.org/borgmatic/docs/how-to/set-up-backups/)
|
|
|
|
* [Make per-application backups](https://torsion.org/borgmatic/docs/how-to/make-per-application-backups/)
|
|
|
|
* [Add preparation and cleanup steps to backups](https://torsion.org/borgmatic/docs/how-to/add-preparation-and-cleanup-steps-to-backups/)
|
|
|
|
* [Monitor your backups](https://torsion.org/borgmatic/docs/how-to/monitor-your-backups/)
|