From: Tomas Pospisek Date: Mon, 24 May 2010 14:46:44 GMT Subject: Re: Bug? file at the same time being deleted and not registered Message-ID: In-Reply-To: On Mon, 24 May 2010, Tomas Pospisek wrote: > On Mon, 24 May 2010, Jeff King wrote: > >> On Mon, May 24, 2010 at 10:29:02AM +0200, Tomas Pospisek wrote: >> >>>> If it's empty or 32 bytes, then that explains what status is reporting >>>> (but the question still remains how we got into that state). >>> >>> There's no .git/index file there: >>> >>> $ ls -l .git/ >>> total 36 >>> drwxr-xr-x 2 tpo tpo 4096 2010-05-23 21:36 branches >>> -rw-r--r-- 1 tpo tpo 4 2010-05-23 21:36 COMMIT_EDITMSG >>> -rw-r--r-- 1 tpo tpo 73 2010-05-23 21:36 description >>> -rw-r--r-- 1 tpo tpo 23 2010-05-23 21:36 HEAD >>> drwxr-xr-x 2 tpo tpo 4096 2010-05-23 21:36 hooks >>> drwxr-xr-x 2 tpo tpo 4096 2010-05-23 21:36 info >>> drwxr-xr-x 3 tpo tpo 4096 2010-05-23 21:36 logs >>> drwxr-xr-x 7 tpo tpo 4096 2010-05-24 09:26 objects >>> drwxr-xr-x 4 tpo tpo 4096 2010-05-23 21:36 refs >> >> Thanks. That means "git status" is at least reporting the right thing. >> Now we just need to figure out why, when the strace of commit shows it >> being written and renamed into place, the index file is missing. >> >> I tried setting up a simple mhddfs mount to reproduce your problem, but >> everything works fine for me. What version of mhddfs are you using? I'm >> using version 0.1.28. I wonder if git is somehow triggering an mhddfs >> bug. Looking through the svn logs for mhddfs, between 0.1.27 and 0.1.28, >> there is a commit with message "fixed rename bug". > > Oh wow (trying to reproduce the bug just from the strace). It's mhddfs. Look: > > $ touch bla > $ touch bla.lock > $ echo asdf > bla.lock > $ mv bla.lock bla > $ ls > $ # nothing here > > That's a quite catastrophic bug in mhddfs :-((( ! Argh. > > Moving this over to mhddfs. OK, I'm waering sackcloth and ashes. The problem was a config fault of mine - one of the backing fs/partitions was included *twice* in the mhdffs fs setup in /etc/fstab. And as of now mhdffs doesn't detect a user composing the mhddfs filesystem in a braindammaged way. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=582888 *t