On 02/05/2016 08:45 AM, songxin wrote:
Hi,
I use glusterfs (version
3.7.6) in replicate mode for sync between two boards
in a node.
When one of the board is
locked and replaced with new board and restarted we see that
sync is lost between the two boards.The mounted glusterfs
volume is not present on the replaced board.
Output of some of gluster
commands on the replaced board are as below.
002500> gluster volume
status c_glusterfs
Status of volume:
c_glusterfs
Gluster process TCP Port
RDMA Port Online Pid
------------------------------------------------------------------------------
Brick
192.32.0.48:/opt/lvmdir/c2/brick 49240 0 Y 1293
Task Status of Volume
c_glusterfs
------------------------------------------------------------------------------
There are no active volume
tasks
002500> gluster volume
info
Volume Name: c_glusterfs
Type: Distribute
Volume ID:
3625f7ff-2b92-4ac4-9967-7abf966eceef
Status: Started
Number of Bricks: 1
Transport-type: tcp
Bricks:
Brick1:
192.32.0.48:/opt/lvmdir/c2/brick
Options Reconfigured:
performance.readdir-ahead:
on
network.ping-timeout: 4
nfs.disable: on
In status info , we don't
see the gluster process of the replaced board. The gluster
process Brick 192.32.0.48:/opt/lvmdir/c2/brick is of the
other board which is not replaced ,
Output of the commands on
the other board are :
# gluster volume info
Volume Name: c_glusterfs
Type: Distribute
Volume ID:
3625f7ff-2b92-4ac4-9967-7abf966eceef
Status: Started
Number of Bricks: 1
Transport-type: tcp
Bricks:
Brick1:
192.32.0.48:/opt/lvmdir/c2/brick
Options Reconfigured:
performance.readdir-ahead:
on
network.ping-timeout: 4
nfs.disable: on
# gluster peer status
Number of Peers: 2
Hostname: 192.32.1.144
Uuid:
bbe2a458-ad3d-406d-b233-b6027c12174e
State: Peer in Cluster
(Connected)
Hostname: 192.32.1.144
Uuid:
bbe2a458-ad3d-406d-b233-b6027c12174e
State: Peer in Cluster
(Connected)
gluster peer status shows
the same host twice , the gluster process of same is missing
in the volume info, Also , the command gluster volume status
c_glusterfs hangs
From the logs of gluster at
/var/log/glusterfs , we observed some errors
cmd_history.log :
volume add-brick c_glusterfs
replica 2 192.32.1.144:/opt/lvmdir/c2/brick force : FAILED :
Locking failed on 192.32.1.144. Please check log file for
details.
cli.log:
[2016-01-30 04:32:40.179381]
I [cli.c:721:main] 0-cli: Started running gluster with
version 3.7.6
[2016-01-30 04:32:40.191715]
I [MSGID: 101190]
[event-epoll.c:632:event_dispatch_epoll_worker] 0-epoll:
Started thread with index 1
[2016-01-30 04:32:40.193246]
I [socket.c:2355:socket_event_handler] 0-transport:
disconnecting now
[2016-01-30 04:32:40.196551]
I [cli-rpc-ops.c:2465:gf_cli_add_brick_cbk] 0-cli: Received
resp to add brick
[2016-01-30 04:32:40.196684]
I [input.c:36:cli_batch] 0-: Exiting with: -1
Can any one help me to
analyze the reason？
I just replied on the bug you raised but this mail has more info. It
seems like the volume is Distribute volume which means it doesn't
sync. I also don't understand how you ended up in a situation where
two of the peers have same uuid and hostname. What are the steps you
took to get into this situation? What are the two bricks you want to
be in sync? May be we can help once you give this information.
Pranith
_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://www.gluster.org/mailman/listinfo/gluster-users
|
_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://www.gluster.org/mailman/listinfo/gluster-users
