Re: about split-brain

[Ravishankar N <ravishankar@xxxxxxxxxx>]

On 08/21/2014 08:23 AM, everybody_enjoy@xxxxxxx wrote:

hi all:

I have a problem about split-brain, details as below:

PC1 and PC2 is AFR, stored a file 1.txt

1. write "123" to 1.txt, then 1.txt in pc1 is same with 1.txt in PC2, content is "123"

2. PC1  network disconnect, write "45" to 1.txt, then  content of 1.txt in pc1 is "123",  

  

   content of 1.txt in pc2 is "12345"

3. PC2  network disconnect, but PC1 network is good, write "67" to 1.txt, then  content of 1.txt in pc1 is "12367",  

  

   content of 1.txt in pc1 is "12345"

4. so split brain occur

is this a bug or design this?

The steps you have described is exactly how data split-brain occurs.

how glusterfs avoid this? Have some ways to solve? this is my questions. thank you

As of today, glusterfs gives EIO (input/output error) when you try to access a split-brained file. You need to resolve the split-brain [1] before the file becomes accessible again. If you want to *avoid* split-brain, you need to enable quorum [2].

[1] https://github.com/gluster/glusterfs/blob/master/doc/split-brain.md
[2] https://access.redhat.com/documentation/en-US/Red_Hat_Storage/2.1/html/Administration_Guide/ch10s10.html
     https://access.redhat.com/documentation/en-US/Red_Hat_Storage/2.0/html/Administration_Guide/sect-User_Guide-Managing_Volumes-Quorum.html

Hope this helps.

---

everybody_enjoy@xxxxxxx

_______________________________________________
Gluster-devel mailing list
Gluster-devel@xxxxxxxxxxx
http://supercolony.gluster.org/mailman/listinfo/gluster-devel

_______________________________________________
Gluster-devel mailing list
Gluster-devel@xxxxxxxxxxx
http://supercolony.gluster.org/mailman/listinfo/gluster-devel