Re: Help with "fatal: unable to read ...." error during GC?

From: Jeff King <peff@peff.net>
Date: Aug 8, 2018, 18:24 UTC
Message-ID: <20180808182436.GA19096@sigill.intra.peff.net>
In-Reply-To: <b247434b62ccd30f32adbebb83fa6ea12b51b6ff.camel@mad-scientist.net>

On Wed, Aug 08, 2018 at 01:35:30PM -0400, Paul Smith wrote:

> Thanks for the note!  Unhappily for me none of these operations seem to
> find any actionable problems...
> [...]

Drat.

One other option is that it _could_ be related to the "old unreachable objects that are reachable from recent unreachable objects should be kept" code. That's supposed to quietly ignore broken links in unreachable objects, but there could be a bug.

Let's narrow it down first and make sure we're dying where I expect. Can you try:

  GIT_TRACE=1 git gc

and confirm the program running when the fatal error is produced?

From what you've shown it's going to be git-repack, but what I'm not clear on is whether it is repack itself that is complaining, or the pack-objects process it spawns. I'd guess the latter.

If so, can you try running it under gdb and getting a stack trace? Something like:

  gdb git
  [and then inside gdb...]
  set args pack-objects --all --reflog --indexed-objects foo </dev/null
  break die
  run
  bt

That might give us a clue where the broken object reference is coming from.

-Peff