Hi, If you run this below command on master gluster vol geo-rep <mastervol> <slave-vol> config slave-gluster-command-dir <gluster-binary-location> on slave run "which gluster" to know gluster-binary-location on slave It will make the same entry in gsyncd.conf file please recheck and confirm both entries are same and also can you confirm that both master and slave have same gluster version. - Sunny On Mon, Jul 23, 2018 at 5:50 PM Maarten van Baarsel <mrten_glusterusers at ii.nl> wrote:> > On 23/07/18 13:48, Sunny Kumar wrote: > > Hi Sunny, > > thanks again for replying! > > > >> Can I test something else? Is the command normally run in a jail? > > > Please share gsyncd.log form master. > > [2018-07-23 12:18:19.773240] I [monitor(monitor):158:monitor] Monitor: starting gsyncd worker brick=/var/lib/gluster slave_node=gluster-4.glstr > [2018-07-23 12:18:19.832611] I [gsyncd(agent /var/lib/gluster):297:main] <top>: Using session config file path=/var/lib/glusterd/geo-replication/gl0_gluster-4.glstr_glbackup/gsyncd.conf > [2018-07-23 12:18:19.832674] I [gsyncd(worker /var/lib/gluster):297:main] <top>: Using session config file path=/var/lib/glusterd/geo-replication/gl0_gluster-4.glstr_glbackup/gsyncd.conf > [2018-07-23 12:18:19.834259] I [changelogagent(agent /var/lib/gluster):72:__init__] ChangelogAgent: Agent listining... > [2018-07-23 12:18:19.848596] I [resource(worker /var/lib/gluster):1345:connect_remote] SSH: Initializing SSH connection between master and slave... > [2018-07-23 12:18:20.387191] E [syncdutils(worker /var/lib/gluster):301:log_raise_exception] <top>: connection to peer is broken > [2018-07-23 12:18:20.387592] E [syncdutils(worker /var/lib/gluster):747:errlog] Popen: command returned error cmd=ssh -oPasswordAuthentication=no -oStrictHostKeyChecking=no -i /var/lib/glusterd/geo-replication/secret.pem -p 22 -oControlMaster=auto -S /tmp/gsyncd-aux-ssh-nN8_GE/2648484453eaadd9d3042ceba9bafa6a.sock georep at gluster-4.glstr /nonexistent/gsyncd slave gl0 georep at gluster-4.glstr::glbackup --master-node gluster-3.glstr --master-node-id 9650e965-bf4f-4544-a42b-f4d540d23a1f --master-brick /var/lib/gluster --local-node gluster-4.glstr --local-node-id 736f6431-2f9c-4115-9790-68f9a88d99a7 --slave-timeout 120 --slave-log-level INFO --slave-gluster-log-level INFO --slave-gluster-command-dir /usr/sbin/ error=1 > [2018-07-23 12:18:20.388887] I [repce(agent /var/lib/gluster):80:service_loop] RepceServer: terminating on reaching EOF. > [2018-07-23 12:18:21.389723] I [monitor(monitor):266:monitor] Monitor: worker died in startup phase brick=/var/lib/gluster > > repeated again and again. > > Maarten.
On 23/07/18 15:28, Sunny Kumar wrote:> Hi, > If you run this below command on master > > gluster vol geo-rep <mastervol> <slave-vol> config > slave-gluster-command-dir <gluster-binary-location> > > on slave run "which gluster" to know gluster-binary-location on slaveDone that, repeatedly, no change :(> It will make the same entry in gsyncd.conf file please recheck and(what gsyncd.conf? the one in /etc or someplace else?)> confirm both entries are same and also can you confirm that both > master and slave have same gluster version.slave: root at gluster-4:~$ /usr/sbin/gluster --version glusterfs 4.0.2 master: root at gluster-3:/home/mrten# /usr/sbin/gluster --version glusterfs 4.0.2 Looking at the slaves' cli.log: [2018-07-23 15:53:26.187547] I [cli.c:767:main] 0-cli: Started running /usr/sbin/gluster with version 4.0.2 [2018-07-23 15:53:26.187611] I [cli.c:646:cli_rpc_init] 0-cli: Connecting to remote glusterd at localhost [2018-07-23 15:53:26.229756] I [MSGID: 101190] [event-epoll.c:609:event_dispatch_epoll_worker] 0-epoll: Started thread with index 1 [2018-07-23 15:53:26.229871] W [rpc-clnt.c:1739:rpc_clnt_submit] 0-glusterfs: error returned while attempting to connect to host:(null), port:0 [2018-07-23 15:53:26.229963] I [socket.c:2625:socket_event_handler] 0-transport: EPOLLERR - disconnecting now [2018-07-23 15:53:26.230640] I [cli-rpc-ops.c:8785:gf_cli_mount_cbk] 0-cli: Received resp to mount [2018-07-23 15:53:26.230825] I [input.c:31:cli_batch] 0-: Exiting with: -1 there's a weird warning there with host:(null), port:0 M.