Re: Repository corruption

From: Junio C Hamano <gitster@pobox.com>
Date: Mar 16, 2008, 23:14 UTC
Message-ID: <7v7ig2feil.fsf@gitster.siamese.dyndns.org>
In-Reply-To: <20080316230151.GB6803@machine.or.cz>

Petr Baudis <pasky@suse.cz> writes:

Show 7 quoted lines

> My idea I got few days ago when looking at the borken repositories is
> that it was caused by interrupted git-repack, since there were stale
> tmp_pack* files left in the repository - it seems like it removed old
> "redundant" packs even though the new one weren't properly completed.
> But that is just a speculation, and I did not get around to look at the
> script yet whether this could really happen, or if something like this
> got fixed recently.

Hmm, unless you are talking about ef07618 (git-repack: Properly abort in corrupt repository, 2005-11-21) as "recent", I do not see anything suspicious.

We may want to refine the first comment in the remove-redundant loop to clarify, though. We do check if elements of $existing are actually redundant or not in the loop against $fullbases these days, since ce85907 (Only repack active packs by skipping over kept packs., 2006-10-29).