Switch from duplicity to restic for backups #257
Merged
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
restic is a deduplicating backup tool: each new snapshot directly
references unchanged blocks in previous snapshots. It eliminates the
traditional distinction between full and incremental backups, where each
incremental backup is essentially a diff - and restoring data requires
fetching the last full backup and then playing all the increments onto
it.
It's also encrypted and compressed, making it more or less a strict
improvement on my existing duplicity-based backup solution.
Unfortunately, restic doesn't support cold storage, as it needs to be
able to read arbitrarily old blocks when restoring a snapshot. This
means I can't use AWS Glacier any more; I've decided to move to
Backblaze B2. However, duplicity doesn't really support cold storage
either, I've just never had need to restore a backup sufficiently old
that it had been rotated into cold storage; and if I ever did I'd have
to manually transfer it into hot storage (which I don't know how to
do)...
So, the amount of "hot" data will go up. This could increase cloud
spend, however B2's hot storage is cheaper than S3's, and also the total
amount of data should be far lower due to the deduplication. So we'll
see.