linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [GIT BISECT] first bad commit: 1f36f774 Switch !O_CREAT case to use of do_last()
@ 2010-03-24 15:49 Boaz Harrosh
  2010-03-24 16:00 ` Al Viro
  0 siblings, 1 reply; 50+ messages in thread
From: Boaz Harrosh @ 2010-03-24 15:49 UTC (permalink / raw)
  To: Al Viro, linux-fsdevel, pNFS Mailing List, Benny Halevy,
	"J. Bruce Fields" <bfiel

1f36f774b22a0ceb7dd33eca626746c81a97b6a5 is the first bad commit
commit 1f36f774b22a0ceb7dd33eca626746c81a97b6a5
Author: Al Viro <viro@zeniv.linux.org.uk>
Date:   Sat Dec 26 10:56:19 2009 -0500

    Switch !O_CREAT case to use of do_last()
    
    ... and now we have all intents crap well localized
    
    Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

:040000 040000 53c1effc5b22746bb83cdbc6a419bf898067882d 9606f7275fde188a056fd174aa1d622ac0630893 M      fs

I have a file corruption regression  bisected to above. (The one step before def4af30cf945a3735ffca865788ea84b30b25d9
was fine)

The test:

- I have an exofs filesystem mounted on /mnt/exofs
- []$ cd /mnt/exofs/some_linux_git; git status;
  All is fine
- []$ mount -t nfs4 -o minorversion=0 localhost:/ /mnt/nfs
  (Where etc/exports will export /mnt/exofs via nfs4.1)
- []$ cd /mnt/nfs/some_linux_git; git status;
  This will fail and will corrupt the .git/index file. Sometimes the file would be
  too short, and sometimes the file will become a directory (Yes really)

Simple revert of this patch over 2.6.34-rc2 will not work so I've not checked it.
A single threaded read like: "diff -r /mnt/nfs/some_linux_git /mnt/nfs/some_linux_git.back"
or even "diff -r /mnt/exofs/some_linux_git /mnt/nfs/some_linux_git" works fine.
A simple dd write test works fine as well.

So it looks like a race that only something like git can exercise, with multi threaded
reads and writes, through nfs/nfsd.

I know Benny also had failures with files export ontop of 2.6.34-rc2
(I looked at the patch, it is totally out of my league, please help)

I will try to reproduce this with a more conventional FS, or any thing else
I should test?

Thanks
Boaz

^ permalink raw reply	[flat|nested] 50+ messages in thread

end of thread, other threads:[~2010-03-25 18:33 UTC | newest]

Thread overview: 50+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-03-24 15:49 [GIT BISECT] first bad commit: 1f36f774 Switch !O_CREAT case to use of do_last() Boaz Harrosh
2010-03-24 16:00 ` Al Viro
2010-03-24 16:04   ` Boaz Harrosh
2010-03-24 16:07     ` Al Viro
2010-03-24 16:10       ` Boaz Harrosh
2010-03-24 16:39         ` Al Viro
2010-03-24 17:15           ` Boaz Harrosh
2010-03-24 17:32             ` [pnfs] " Boaz Harrosh
2010-03-24 17:47               ` Boaz Harrosh
2010-03-24 17:58                 ` Boaz Harrosh
2010-03-24 18:06                   ` Al Viro
2010-03-24 18:26                     ` Doug Nazar
2010-03-24 18:56                       ` Al Viro
2010-03-25  9:39                         ` Boaz Harrosh
2010-03-25 10:12                           ` Al Viro
2010-03-25 10:22                             ` Benny Halevy
2010-03-25 10:31                               ` Benny Halevy
2010-03-25 10:49                               ` Al Viro
2010-03-25 10:56                                 ` Benny Halevy
2010-03-25 11:00                                   ` Al Viro
2010-03-25 11:12                                     ` Benny Halevy
2010-03-25 11:13                                       ` Benny Halevy
2010-03-25 11:55                                 ` Al Viro
2010-03-25 13:00                                   ` Boaz Harrosh
2010-03-25 13:11                                     ` Boaz Harrosh
2010-03-25 10:54                             ` Al Viro
2010-03-25 11:19                               ` Benny Halevy
2010-03-25 12:07                               ` Benny Halevy
2010-03-25 12:18                                 ` Benny Halevy
2010-03-25 13:06                                   ` Al Viro
2010-03-25 13:30                                     ` Boaz Harrosh
2010-03-25 13:37                                       ` Al Viro
2010-03-25 13:45                                         ` Boaz Harrosh
2010-03-25 14:04                                           ` Al Viro
2010-03-25 14:27                                             ` Boaz Harrosh
2010-03-25 15:25                                               ` Al Viro
2010-03-25 17:28                                                 ` Boaz Harrosh
2010-03-25 17:59                                                   ` Trond Myklebust
2010-03-25 18:06                                                     ` Boaz Harrosh
2010-03-25 18:18                                                       ` Trond Myklebust
2010-03-25 18:33                                                         ` Boaz Harrosh
2010-03-25 13:52                                         ` Benny Halevy
2010-03-25 14:06                                           ` Al Viro
2010-03-25 14:07                                             ` Benny Halevy
2010-03-25 14:36                                               ` Benny Halevy
2010-03-24 18:02                 ` Trond Myklebust
2010-03-24 18:10                   ` Trond Myklebust
2010-03-25  9:13                     ` Boaz Harrosh
2010-03-25 15:44                       ` Trond Myklebust
2010-03-25 10:11                     ` Benny Halevy

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).