Re: False positives in git diff-index

From: Alexander Gladysh <agladysh@gmail.com>
Date: Jan 4, 2011, 12:01 UTC
Message-ID: <AANLkTinfbyve-k8xBzDb1sTcXhJGvL_B+auuA8BQSUy2@mail.gmail.com>
In-Reply-To: <AANLkTi=Po7zA1YG-VdN6cZEV+ZF3GYNM9W9CLVXFaE5Z@mail.gmail.com>

On Tue, Jan 4, 2011 at 14:47, Zenaan Harkness <zen@freedbms.net> wrote:

> On Tue, Jan 4, 2011 at 20:45, Alexander Gladysh <agladysh@gmail.com> wrote:
>> Nobody is interested?

> Your problem set appears that you have a rather gnarly corner case
> issue, arising from your custom build processes. Although git really
> is amazing, I believe you may well be pushing git to its technological
> limits.

Committing few megabytes of data several times per second is technological limits? I do not believe so.

> So your problem could be quite hard to debug, whilst being distinctly
> difficult to ascertain the root causes.

> It also appears that your custom complicated build process is likely
> protecting, or at least integral to, your high value corporate process
> assets.

Show 9 quoted lines

> So _in this case_ you would be remiss to not find a suitable
> consultant to provide professional and discreet assistance - perhaps
> GitHub.com, as GitHub’s Tender provides both public and _private_
> support issue posting, and customized and private training if you and/
> or your colleagues require; you might contact GitHub direct (
> https://github.com/contact ) as their Support page does not link
> directly to support contract information; oh, and GitHub supports a
> lot of community projects too: their support for our community ought
> be supported.

> <disclaimer> I am _not_ affiliated with GitHub, I do work full time
> with a human rights association in Australia.

Thank you for your opinion.

I view this particular situation as follows:

1. I found a reproducible case for a hard to catch bug in Git. (This
is a bug in Git, not in my build process.) This bug in its
intermittent form annoyed me for quite some time — several months at
least — and is likely to annoy other users. (I'm not *that* unique!)

2. I can live happily with sleep(0.2) in my deployment code (while
this is not very satisfying, it is acceptable — certainly cheaper than
a paid consultant).

3. I'm willing to help Git developers with catching this bug for
mutual benefit — I will get rid of annoying issue and make my
deployment code more robust. Git will, well, be a bit more robust as
well.

4. The sole reason I'm pinging back on this bug report is that I'm
afraid to accidentally lose the data snapshot (or something in
environment) that makes the issue reproducible.

5. If no one is interested, well, that's opensource :-) No hard feelings.

Alexander.