[merge] errors in deleting obsolete packs should not be fatal (was Re: Problems with NFS mounted filestore)

[Martin Pool]

On Mon, May 26, 2008 at 11:02 PM, John Arbash Meinel
<john at arbash-meinel.com> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Martin Pool wrote:
> | On Sun, May 25, 2008 at 6:57 PM, Russel Winder
> | <russel.winder at concertant.com> wrote:

> |> However, the branch consistently fails to work when run on a machine
> |> which mounts the filestore over NFS.  I have just undertaken the branch
> |> on the fileserver itself and that is fine.  I don't know if this is a
> |> bzr or bzr-svn thing, nor whether it is a known issue, i.e. should I
> |> report a bug?
> |
> | Thanks for the report.  This is a new bug.

By this I meant: it is a newly-reported bug; the potential for it to
occur was there since packs were first merged before 1.0.  This .nfs*
thing is a known NFS behaviour but doesn't occur all that often.

> The patch seems okay, though it may mean lots of 'delete()' calls for a
> Remote
> transport which could do it in 1 call. (Then again, this sort of thing
> should be
> a higher-level logical 'set-pack-names' sort of call anyway.)

I thought about that, but as far as I can see there are no transports
that actually implement delete_multi.  As you say we probably want a
higher-level operation.  If we did implement it (maybe pipelined on
sftp or ftp) we'd need to prescribe more specifically what should
happen if some deletions fail.

-- 
Martin <http://launchpad.net/~mbp/>