Description of problem:
When a dispersed volume is mounted by NFS, any write to it returns an error and nothing is written (file size is 0).
Version-Release number of selected component (if applicable): 3.6
Steps to Reproduce:
1. gluster volume create d3_1 disperse node1:/bricks/d3_1 node2:/bricks/d3_1 node3:/bricks/d3_1
2. gluster volume start d3_1
3. mount node1:/d3_1 /gluster/d3_1
4. dd if=/dev/zero of=/gluster/d3_1/test bs=1024k count=1k
dd: error writing ‘/gluster/d3_1/test’: Unknown error 527
470+0 records in
469+0 records out
491782144 bytes (492 MB) copied, 3.41857 s, 144 MB/s
The file should be successfully written
The problem is caused because NFS uses anonymous fd, even for writes. Anonymous fs are not correctly handled by ec xlator.
REVIEW: http://review.gluster.org/8367 (cluster/ec: Fix incorrect management of NFS requests) posted (#1) for review on master by Xavier Hernandez (email@example.com)
COMMIT: http://review.gluster.org/8367 committed in master by Vijay Bellur (firstname.lastname@example.org)
Author: Xavier Hernandez <email@example.com>
Date: Tue Jul 8 17:52:57 2014 +0200
cluster/ec: Fix incorrect management of NFS requests
Some operations, specially those comming from NFS, do not use a
regular fd and use an anonymous fd (i.e. a previous open call has
not been sent). Any context information created during open or
create will not be present on these fd's, so we simply return NULL
for contexts of those fd.
Also it seems that NFS can send write requests with a very big
buffer (higher that the default value of 128 KB). Some changes
have been made to correctly handle these large buffers.
Signed-off-by: Xavier Hernandez <firstname.lastname@example.org>
Tested-by: Gluster Build System <email@example.com>
Reviewed-by: Vijay Bellur <firstname.lastname@example.org>
This bug is getting closed because a release has been made available that should address the reported issue. In case the problem is still not fixed with glusterfs-3.7.0, please open a new bug report.
glusterfs-3.7.0 has been announced on the Gluster mailinglists , packages for several distributions should become available in the near future. Keep an eye on the Gluster Users mailinglist  and the update infrastructure for your distribution.