Bug 606468 - gfs2_fsck ends on journal recovery
Summary: gfs2_fsck ends on journal recovery
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Enterprise Linux 5
Classification: Red Hat
Component: gfs2-utils
Version: 5.5
Hardware: All
OS: Linux
low
medium
Target Milestone: rc
: ---
Assignee: Robert Peterson
QA Contact: Cluster QE
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2010-06-21 17:46 UTC by Jaroslav Kortus
Modified: 2011-01-13 23:20 UTC (History)
3 users (show)

Fixed In Version: gfs2-utils-0.1.62-25.el5
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2011-01-13 23:20:58 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)
gfs2 metadata (1.12 MB, application/x-bzip2)
2010-06-21 17:47 UTC, Jaroslav Kortus
no flags Details
Prototype patch (2.80 KB, patch)
2010-06-21 22:18 UTC, Robert Peterson
no flags Details | Diff


Links
System ID Private Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2011:0135 0 normal SHIPPED_LIVE gfs2-utils bug fix update 2011-01-12 19:26:41 UTC

Description Jaroslav Kortus 2010-06-21 17:46:47 UTC
Description of problem:
gfs2_fsck checks corrupted filesystem, exists with EC=8 and does not provide any further output describing what has failed.

[root@a1 ~]# gfs2_edit restoremeta quota-crash.meta /dev/vedder/vedder0 
There are 53311488 blocks of 4096 bytes in the destination file system.

42004 metadata blocks (164MB) restored.    
File quota-crash.meta restore successful.
[root@a1 ~]# gfs2_fsck -y /dev/vedder/vedder0 
Initializing fsck
Recovering journals (this may take a while)
[root@a1 ~]#
[root@a1 ~]# gfs2_fsck -y /dev/vedder/vedder0 ; echo $?
Initializing fsck
Recovering journals (this may take a while)
8


Version-Release number of selected component (if applicable):
gfs2-utils-0.1.62-20.el5

How reproducible:
restore the metadata attached

Steps to Reproduce:
1. gfs2_edit restoremeta metadata /dev/mydevice
2. gfs2_fsck -y /dev/mydevice
3.
  
Actual results:
see above

Expected results:
gfs2_fsck goes through all phases and prints diagnostic reports if something fails

Additional info:
gfs2_fsck -n results may differ, probably due to journal recovery.

Comment 1 Jaroslav Kortus 2010-06-21 17:47:19 UTC
Created attachment 425705 [details]
gfs2 metadata

Comment 2 Robert Peterson 2010-06-21 22:18:50 UTC
Created attachment 425764 [details]
Prototype patch

I backtracked this problem.  It is a regression caused by
this commit:

http://git.fedoraproject.org/git/?p=cluster.git;a=commitdiff;h=7ccf5c0c60ff29fc6e8a1ef0fea01d510f2df79b

This is a replacement patch that fixes the problem.
However, even with this patch in place, there is a new
problem that appears on ia64 only:

fsck.gfs2(10238): unaligned access to 0x600000000006930b, ip=0x400000000004c730

I backtracked this new problem to function gfs2_bitfit.  I need
to look into how to best fix this tomorrow.

Comment 3 Robert Peterson 2010-06-21 22:19:53 UTC
Requesting ack flags to get this into 5.6.

Comment 5 Robert Peterson 2010-06-25 18:51:28 UTC
I pushed the patch to the RHEL56 branch of the cluster git tree
for inclusion into 5.6.  The patch was tested on roth-01.
No need the push any of the upstream branches; it's already there.
Changing status to POST.

Comment 6 Robert Peterson 2010-09-17 15:55:13 UTC
Build 2768496 successful.  Changing status to Modified.
This fix is in gfs2-utils-0.1.62-25.el5.

Comment 9 errata-xmlrpc 2011-01-13 23:20:58 UTC
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on therefore solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHBA-2011-0135.html


Note You need to log in before you can comment on or make changes to this bug.