"C. Michael Pilato" <cmpilato@collab.net> writes:
> Ben Collins-Sussman <sussman@collab.net> writes:
> > > Am I missing something obvious?
> >
> > I don't think a libsvn_fs_fs repository can't be backed up while "hot".
> > That's one of the advantages of a BerkeleyDB database.
>
> Er. You don't think a libsvn_fs_fs repository *can* be backed up
> while "hot". Trust me -- I know what you think.
But, I think fs_fs can be backed up using normal filesystem tools. I
should just go read a doc or ask Josh or Greg to make sure, but I
would think that this is one of the advantages of fs_fs.
Tristan Seligmann wrote:
> I don't see how you can reliable grab a consistent snapshot of a set of
> changing files with normal filesystem tools; you'd either need to access
> the files through some special interface, or temporarily disable access
> to the repository.
You do it the same way journaling filesystems do it, by ordering
writes in such a way that any time you make a copy, what you get is
consistent. (The copy may also contain some "extra" data, which is
the unfinished portion of a revision in progress, but that doesn't
matter -- the copy is still consistent, because every revision that it
claims is complete *is* complete.)
-Karl
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Wed May 19 18:37:46 2004