plakar check performs a health check on snapshots without restoring them to the filesystem.
By default, it performs a full check making sure that it can retrieve everything it needs for a full restore, validating that the checksums for every chunk and object match. This check is similar to restoring a snapshot, computing a checksum of the files as they are being restored and removing them once it’s done, but does it in a more efficient way.
When the -fast option is provided,
it performs a fast check by asking the repository if it holds the resources needed,
but without retrieving them and validating checksums.
This check is much faster,
doesn’t imply much data transfer,
but is only suitable if there’s confidence that the repository is trusted and not corrupted.
plakar check snapshotID
plakar check snapshotID:/path/to/file
Performing a check then a fast check on snapshot b3:
$ plakar check b3 && echo ok
ok
$ plakar check -fast b3 && echo ok
ok
Performing a check on a file within snapshot b3:
$ plakar check b3:/private/etc/passwd && echo ok
ok
$ plakar check -fast b3:/private/etc/passwd && echo ok
ok