Git Stability

[Curt Sampson <cjs%NetBSD.org@localhost>]

On 2010-01-13 10:01 +0100 (Wed), S.P.Zeidler wrote:

> If it hasn't happened to one repo in over 4 years and to another twice in
> 3 months, that does constitute a -slight- difference in stability.

Certainly. What went wrong with git, if I may ask?

> You don't have an index without which your entire repo becomes
> unreadable in cvs....

If you're talking about what's usually referred to as the "index" in git
(the .git/index file), you don't have such a thing in git bare repo,
which is what the master repo should be. The index file is used only in
repos with a working copy to track information about the working copy.
There's nothing especially important in it; if it becomes corrupted,
you can just remove it and do a "git reset" to create a new one. The
only thing you might lose is a few uncommitted changes that are not file
contents.

cjs
-- 
Curt Sampson         <cjs%cynic.net@localhost>         +81 90 7737 2974
             http://www.starling-software.com
The power of accurate observation is commonly called cynicism
by those who have not got it.    --George Bernard Shaw