Bug 57631 - Kernel bug at extent-tree.c:5446
Summary: Kernel bug at extent-tree.c:5446
Status: RESOLVED CODE_FIX
Alias: None
Product: File System
Classification: Unclassified
Component: btrfs (show other bugs)
Hardware: All Linux
: P1 high
Assignee: Josef Bacik
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-05-06 05:34 UTC by Ákos Szőts
Modified: 2013-05-14 20:47 UTC (History)
0 users

See Also:
Kernel Version: 3.9
Subsystem:
Regression: Yes
Bisected commit-id:

Attachments
Picture about the stack trace (469.69 KB, image/jpeg)
2013-05-06 05:34 UTC, Ákos Szőts 		Details
Add an attachment (proposed patch, testcase, etc.)

Description Ákos Szőts 2013-05-06 05:34:17 UTC 
Created attachment 100771 [details]
Picture about the stack trace

When I tried to recover my partition from the space cache bug (see http://www.mail-archive.com/linux-btrfs@vger.kernel.org/msg24360.html ), I issued the command `mount -o clear_cache -t btrfs /dev/sda7 /aaa/aaa`. After 10 minutes the mount command segfaulted. I cloned the latest btrfs-progs from Josef's Github account and ran the `btrfsck --repair` twice on the partition.

Then I tried to mount the partition under my normal openSUSE system (with kernel 3.9) and the following BUG appeared during the boot-up process: Kernel bug at extent-tree.c:5446
I attached the picture I've taken.

I tried to boot in an openSUSE Rescue CD, an Ubuntu Live CD, but just a SystemRescueCD was actually working, but I can't reach my partition. However, I was able to make an image by `btrfs-image -c9 -t8`.
Comment 1 Josef Bacik 2013-05-06 13:00:18 UTC 
I'm pulling your image down now, I'll let you know what I find.
Comment 2 Josef Bacik 2013-05-06 13:59:56 UTC 
Alright good news is I know what this is, you got bitten by my tree logger bug.  The not great news is that it is going to require a fix to fsck to fix since the file system is technically corrupted (don't worry the data is there, the extents are just wrong).  So this is my new priority #1, and I already have your image so I have something to test my patch on.  I will hopefully have a patch for you in the next couple of days.
Comment 3 Josef Bacik 2013-05-14 18:50:43 UTC 
Ok I've fixed fsck to fix this particular corruption, please clone my tree and build

git://github.com/josefbacik/btrfs-progs.git

then run

./btrfsck --repair /dev/whatever

and you should be good to go.
Comment 4 Ákos Szőts 2013-05-14 20:09:35 UTC 
Thank you very much for your effort, it works flawlessly!
Comment 5 Josef Bacik 2013-05-14 20:47:53 UTC 
No problem.
Note You need to log in before you can comment on or make changes to this bug.