[Borgbackup] Recovery from missing segments

[David Gasaway]

On Wed, Jun 27, 2018 at 8:38 AM, David Gasaway <dave at gasaway.org> wrote:

> Yes, but I'm trying to find a check mode that will repair this
> condition without reading the segment files, as that would take a very
> long time.

Not that anyone cares, but I went ahead with a 'borg check --repair'.
It took over 3 days, thankfully with no network issues in that time.
The most stressful part was I had no good way to gauge how far it
progressed.  The command gave no output for 3 days, possibly because
it didn't find any issues until the end.  Perhaps something could be
added to write output every so many GiB or chunks.

> Hypothetical mode would simply
> check the filesystem for the expected segment files, and if not found,
> replace the associated chunks in the index with zero chunks so the
> chunks/segments get recreated at the next backup (assuming source data
> is still present).

On second thought, I may also need it to do something like check the
expected length of segment files in case a new chunk was added to an
existing segment, but the segment was "rolled back".  The more I think
about this though, what I'd probably want is a way to invalidate any
chunk created after a certain date/time.  Not sure if this is
possible.

Still, I've taken steps to avoid this in the future.  Most critically,
I upgraded s3ql which added new features to avoid this kind of loss.

-- 
-:-:- David K. Gasaway
-:-:- Email: dave at gasaway.org