[Bugs] [Bug 1180015] New: reboot node with some glusterd glusterfsd glusterfs services.

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1180015

            Bug ID: 1180015
           Summary: reboot node with some glusterd glusterfsd glusterfs
                    services.
           Product: GlusterFS
           Version: 3.6.1
         Component: glusterd
          Severity: high
          Assignee: bugs at gluster.org
          Reporter: helloparadise at 163.com
                CC: bugs at gluster.org, gluster-bugs at redhat.com



Description of problem:
three nodes make up one cluster, create a  redundancy volume named "test" in
this cluster, then ,start this volume 

. after i excute "init 6" or "reboot" in node2 ,i restart this volume "test"
and relevant services often  have some 

trouble and issue, result in volume test can not work.
  i would consult whether excute "reboot" or "init 6" operation in the
condition of volume not be stopped is right? 
  for example 
  all xattr of brick dir(xfs file system) lose ,and  client mount this volume
fail. 


Version-Release number of selected component (if applicable):


How reproducible:
node1
[root at node-1]#gluster volume create test redundancy 1 node1:/brick1
node2:/brick2 node3:/brick3
[root at node-1]#/sbin/mount.digioceanfs 127.0.0.1:/test /cluster2/test
then ,/cluster2/test/samba this dir is export dir of samba,and the windows
client connectting to this export dir of samba
node2
[root at node-2]/usr/sbin/digioceanfs --volfile-server=127.0.0.1
--volfile-id=/test /cluster2/test
node3
[root at node-3]/usr/sbin/digioceanfs --volfile-server=127.0.0.1
--volfile-id=/test /cluster2/test

Steps to Reproduce:
1. now i reboot node2
[root at node-2]reboot 
2. after node2 server start just have some issue as below
3.

Actual results:
ndoe3 brick glusterfsd log
[2015-01-08 11:50:16.053723] E [posix.c:5618:init] 0-test-posix: Extended
attribute trusted.digioceanfs.
volume-id is absent
[2015-01-08 11:50:16.053776] E [xlator.c:425:xlator_init] 0-test-posix:
Initialization of volume 'test-p
osix' failed, review your volfile again
[2015-01-08 11:50:16.053795] E [graph.c:322:digioceanfs_graph_init]
0-test-posix: initializing translato
r failed
[2015-01-08 11:50:16.053807] E [graph.c:525:digioceanfs_graph_activate]
0-graph: init failed

check brick xattr
Brick3: node-2.aaa.bbb.ccc:/digioceanfs/wwn-0x6000c29a2efdcab8ebe89f9298246f79
[root at node-3 ~]# getfattr -d -m ".*" -e hex
/digioceanfs/wwn-0x6000c299dcd6abf74489faac4a2c0afe

have no xattr.
glusterfsd progress not be started
node1
[root at node-1 ~]# df -h 
df command can be hang-up can not finish.
Expected results:


Additional info:
this condition often happen probability

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.