[Gluster-users] "Structure needs cleaning" error

[Patrick Haley]

Hi,

I went on the particular server on which the file in question
resides.  I check the RAID and all the disks show up as OK.
I looked in dmesg and did not see anything on xfs.  To confirm
this absence, I did "grep -in xfs dmesg" and that also came
up empty. "grep -in fs dmesg" (no "x") returns

161:checking if image is initramfs... it is
196:usbcore: registered new driver usbfs
256:VFS: Disk quotas dquot_6.5.1
453:EXT3-fs: mounted filesystem with ordered data mode.
494:EXT3 FS on sda1, internal journal
496:EXT3 FS on sda6, internal journal
497:EXT3-fs: mounted filesystem with ordered data mode.
499:EXT3 FS on sda3, internal journal
500:EXT3-fs: mounted filesystem with ordered data mode.
502:EXT3 FS on sda2, internal journal
503:EXT3-fs: mounted filesystem with ordered data mode.


-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Pat Haley                          Email:  phaley at mit.edu
Center for Ocean Engineering       Phone:  (617) 253-6824
Dept. of Mechanical Engineering    Fax:    (617) 253-8125
MIT, Room 5-222B                   http://web.mit.edu/phaley/www/
77 Massachusetts Avenue
Cambridge, MA  02139-4301



Quoting Joe Landman <landman at scalableinformatics.com>:

> On 02/26/2012 12:54 PM, Patrick Haley wrote:
>>
>> Hi,
>>
>> We have recently upgraded our gluster to 3.2.5 and have
>> encountered the following error.  Gluster seems somehow
>> confused about one of the files it should be serving up,
>> specifically
>> /projects/philex/PE/2010/Oct18/arch07/BalbacFull_250_200_03Mar_3.png
>>
>> If I go to that directory and simply do an ls *.png I get
>>
>> ls: BalbacFull_250_200_03Mar_3.png: Structure needs cleaning
>
> This is usually what happens when you have an underlying xfs file 
> system as your backing store, and have had a failure which has shut 
> the xfs file system down.
>
> Have a look in dmesg for the servers, and see if xfs has shut down.
>
> Reasons for xfs shutdown include a) bugs in xfs/kernel, b) storage 
> device/RAID failure.
>
> Our experience is that "b" happens far more often than "a", though 
> "a" does happen (especially on some kernels).
>
> -- 
> Joseph Landman, Ph.D
> Founder and CEO
> Scalable Informatics Inc.
> email: landman at scalableinformatics.com
> web  : http://scalableinformatics.com
>        http://scalableinformatics.com/sicluster
> phone: +1 734 786 8423 x121
> fax  : +1 866 888 3112
> cell : +1 734 612 4615
>
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
>