[Gluster-users] Solving healing and split-brain problems

Pranith Kumar Karampuri pkarampu at redhat.com
Thu Nov 21 09:02:10 UTC 2013


hi Alan,
      Could you please send me
1) logs from all the machines.
2) heal info full output
3) gluster volume info <volname> output.
4) "getfattr -d -m. -e hex <backend-brick-dir>/file-path" and "stat <backend-brick-dir>/file-path" outputs for the files on which the self-heals are attempted. 

Pranith

----- Original Message -----
> From: "Alan Orth" <alan.orth at gmail.com>
> To: gluster-users at gluster.org
> Sent: Thursday, November 21, 2013 12:28:59 PM
> Subject: Re: [Gluster-users] Solving healing and split-brain problems
> 
> Hi, I've got a similar issue on CentOS 6.4 + GlusterFS 3.4.0.
> 
> Yesterday I shut down one node in a replicated volume for some hardware
> maintenance, and after bringing it up a few minutes later the healing
> started automatically.  The healing has been going for 18 hours and
> seems to be in a loop or something.
> 
> # gluster volume heal homes info
> ...
> <gfid:cab1f7ca-da02-4b39-8a38-f0628a6cfe2c>
> <gfid:05566262-3492-403c-983e-a0490009fccd>
> <gfid:60fd0cc9-8827-4ca9-bc17-a9b6766b3e9b>
> <gfid:66713aa4-18db-433a-a83f-6d9e807d7675>
> <gfid:94c2b221-3c53-4c47-8735-f62cc89902ac>
> <gfid:44d1e6de-8def-46a4-9015-87f0d26c26bc>
> /segoli/BECA/GMCompletedRuns/TINA_BECA/RG1_2013-11-20_1_Tina_BECA2
> /segoli/BECA/GMCompletedRuns/TINA_BECA/Robert1_Tina_2013-11-12_1_Tina_BECA2
> /segoli/BECA/GMCompletedRuns/TINA_BECA/Olutayo_AFLP3_2013-11-12_1_Tina_BECA2
> /segoli/BECA/GMCompletedRuns/TINA_BECA/Olutayo_AFLP4_2013-11-12_1_Tina_BECA2
> 
> My gut says that there is split brain, but:
> 
> # gluster volume heal homes info split-brain | grep entries
> Number of entries: 0
> Number of entries: 0
> 
> I'm not sure how to interpret this.
> 
> Thanks,
> 
> Alan
> 
> On 11/05/2013 04:55 AM, Vijay Bellur wrote:
> > On 11/04/2013 11:20 PM, Hoggins! wrote:
> >> Hello list,
> >>
> >> I have a replicated (two nodes) volume called "thedude1".
> >>
> >> When I issue the commands:
> >>
> >> ~$ gluster volume heal thedude1 full
> >> ~$ gluster volume heal thedude1 info
> >>
> >>
> >> Question is : how do I manage this issue ? I know how to handle it with
> >> "regular" files, but not with raw gfids.
> >> A quick howto would be the best.
> >>
> >
> > Could this be of help:
> >
> > https://github.com/gluster/glusterfs/blob/master/doc/split-brain.md
> >
> > Thanks,
> > Vijay
> >
> >
> >
> > _______________________________________________
> > Gluster-users mailing list
> > Gluster-users at gluster.org
> > http://supercolony.gluster.org/mailman/listinfo/gluster-users
> 
> --
> Alan Orth
> alan.orth at gmail.com
> http://alaninkenya.org
> http://mjanja.co.ke
> "I have always wished for my computer to be as easy to use as my telephone;
> my wish has come true because I can no longer figure out how to use my
> telephone." -Bjarne Stroustrup, inventor of C++
> GPG Public Key: 0xf92c4bd91084bb5de14e20be9470dd588dd1026c
> 
> 
> 
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://supercolony.gluster.org/mailman/listinfo/gluster-users



More information about the Gluster-users mailing list