[Gluster-users] Upgrade to 4.1.1 geo-replication does not work

Marcus Pedersén marcus.pedersen at slu.se
Mon Jul 16 19:59:31 UTC 2018 

Hi Kotresh,

I have been testing for a bit and as you can see from the logs I sent before permission is denied for geouser on slave node on file:

/var/log/glusterfs/cli.log

I have turned selinux off and just for testing I changed permissions on /var/log/glusterfs/cli.log so geouser can access it.

Starting geo-replication after that gives response successful but all nodes get status Faulty.


If I run: gluster-mountbroker status

I get:

+-----------------------------+-------------+---------------------------+--------------+--------------------------+
|             NODE            | NODE STATUS |         MOUNT ROOT        |    GROUP     |          USERS           |
+-----------------------------+-------------+---------------------------+--------------+--------------------------+
| urd-gds-geo-001.hgen.slu.se |          UP | /var/mountbroker-root(OK) | geogroup(OK) | geouser(urd-gds-volume)  |
|       urd-gds-geo-002       |          UP | /var/mountbroker-root(OK) | geogroup(OK) | geouser(urd-gds-volume)  |
|          localhost          |          UP | /var/mountbroker-root(OK) | geogroup(OK) | geouser(urd-gds-volume)  |
+-----------------------------+-------------+---------------------------+--------------+--------------------------+


and that is all nodes on slave cluster, so mountbroker seems ok.


gsyncd.log logs an error about /usr/local/sbin/gluster is missing.

That is correct cos gluster is in /sbin/gluster and /urs/sbin/gluster

Another error is that SSH between master and slave is broken,

but now when I have changed permission on /var/log/glusterfs/cli.log I can run:

ssh -oPasswordAuthentication=no -oStrictHostKeyChecking=no -i /var/lib/glusterd/geo-replication/secret.pem -p 22 geouser at urd-gds-geo-001 gluster --xml --remote-host=localhost volume info urd-gds-volume

as geouser and that works, which means that the ssh connection works.


Is the permissions on /var/log/glusterfs/cli.log changed when geo-replication is setup?

Is gluster supposed to be in /usr/local/sbin/gluster?


Do I have any options or should I remove current geo-replication and create a new?

How much do I need to clean up before creating a new geo-replication?

In that case can I pause geo-replication, mount slave cluster on master cluster and run rsync , just to speed up transfer of files?


Many thanks in advance!

Marcus Pedersén


Part from the gsyncd.log:

[2018-07-16 19:34:56.26287] E [syncdutils(worker /urd-gds/gluster):749:errlog] Popen: command returned error    cmd=ssh -oPasswordAuthentication=no -oStrictHostKeyChecking=no -i /var/lib/glusterd/geo-replicatio\
n/secret.pem -p 22 -oControlMaster=auto -S /tmp/gsyncd-aux-ssh-WrbZ22/bf60c68f1a195dad59573a8dbaa309f2.sock geouser at urd-gds-geo-001 /nonexistent/gsyncd slave urd-gds-volume geouser at urd-gds-geo-001::urd-gds-volu\
me --master-node urd-gds-001 --master-node-id 912bebfd-1a7f-44dc-b0b7-f001a20d58cd --master-brick /urd-gds/gluster --local-node urd-gds-geo-000 --local-node-id 03075698-2bbf-43e4-a99a-65fe82f61794 --slave-timeo\
ut 120 --slave-log-level INFO --slave-gluster-log-level INFO --slave-gluster-command-dir /usr/local/sbin/ error=1
[2018-07-16 19:34:56.26583] E [syncdutils(worker /urd-gds/gluster):753:logerr] Popen: ssh> failure: execution of "/usr/local/sbin/gluster" failed with ENOENT (No such file or directory)
[2018-07-16 19:34:56.33901] I [repce(agent /urd-gds/gluster):89:service_loop] RepceServer: terminating on reaching EOF.
[2018-07-16 19:34:56.34307] I [monitor(monitor):262:monitor] Monitor: worker died before establishing connection        brick=/urd-gds/gluster
[2018-07-16 19:35:06.59412] I [monitor(monitor):158:monitor] Monitor: starting gsyncd worker    brick=/urd-gds/gluster  slave_node=urd-gds-geo-000
[2018-07-16 19:35:06.99509] I [gsyncd(worker /urd-gds/gluster):297:main] <top>: Using session config file       path=/var/lib/glusterd/geo-replication/urd-gds-volume_urd-gds-geo-001_urd-gds-volume/gsyncd.conf
[2018-07-16 19:35:06.99561] I [gsyncd(agent /urd-gds/gluster):297:main] <top>: Using session config file        path=/var/lib/glusterd/geo-replication/urd-gds-volume_urd-gds-geo-001_urd-gds-volume/gsyncd.conf
[2018-07-16 19:35:06.100481] I [changelogagent(agent /urd-gds/gluster):72:__init__] ChangelogAgent: Agent listining...
[2018-07-16 19:35:06.108834] I [resource(worker /urd-gds/gluster):1348:connect_remote] SSH: Initializing SSH connection between master and slave...
[2018-07-16 19:35:06.762320] E [syncdutils(worker /urd-gds/gluster):303:log_raise_exception] <top>: connection to peer is broken
[2018-07-16 19:35:06.763103] E [syncdutils(worker /urd-gds/gluster):749:errlog] Popen: command returned error   cmd=ssh -oPasswordAuthentication=no -oStrictHostKeyChecking=no -i /var/lib/glusterd/geo-replicatio\
n/secret.pem -p 22 -oControlMaster=auto -S /tmp/gsyncd-aux-ssh-K9mB6Q/bf60c68f1a195dad59573a8dbaa309f2.sock geouser at urd-gds-geo-001 /nonexistent/gsyncd slave urd-gds-volume geouser at urd-gds-geo-001::urd-gds-volu\
me --master-node urd-gds-001 --master-node-id 912bebfd-1a7f-44dc-b0b7-f001a20d58cd --master-brick /urd-gds/gluster --local-node urd-gds-geo-000 --local-node-id 03075698-2bbf-43e4-a99a-65fe82f61794 --slave-timeo\
ut 120 --slave-log-level INFO --slave-gluster-log-level INFO --slave-gluster-command-dir /usr/local/sbin/ error=1
[2018-07-16 19:35:06.763398] E [syncdutils(worker /urd-gds/gluster):753:logerr] Popen: ssh> failure: execution of "/usr/local/sbin/gluster" failed with ENOENT (No such file or directory)
[2018-07-16 19:35:06.771905] I [repce(agent /urd-gds/gluster):89:service_loop] RepceServer: terminating on reaching EOF.
[2018-07-16 19:35:06.772272] I [monitor(monitor):262:monitor] Monitor: worker died before establishing connection       brick=/urd-gds/gluster
[2018-07-16 19:35:16.786387] I [monitor(monitor):158:monitor] Monitor: starting gsyncd worker   brick=/urd-gds/gluster  slave_node=urd-gds-geo-000
[2018-07-16 19:35:16.828056] I [gsyncd(worker /urd-gds/gluster):297:main] <top>: Using session config file      path=/var/lib/glusterd/geo-replication/urd-gds-volume_urd-gds-geo-001_urd-gds-volume/gsyncd.conf
[2018-07-16 19:35:16.828066] I [gsyncd(agent /urd-gds/gluster):297:main] <top>: Using session config file       path=/var/lib/glusterd/geo-replication/urd-gds-volume_urd-gds-geo-001_urd-gds-volume/gsyncd.conf
[2018-07-16 19:35:16.828912] I [changelogagent(agent /urd-gds/gluster):72:__init__] ChangelogAgent: Agent listining...
[2018-07-16 19:35:16.837100] I [resource(worker /urd-gds/gluster):1348:connect_remote] SSH: Initializing SSH connection between master and slave...
[2018-07-16 19:35:17.260257] E [syncdutils(worker /urd-gds/gluster):303:log_raise_exception] <top>: connection to peer is broken


________________________________
Från: gluster-users-bounces at gluster.org <gluster-users-bounces at gluster.org> för Marcus Pedersén <marcus.pedersen at slu.se>
Skickat: den 13 juli 2018 14:50
Till: Kotresh Hiremath Ravishankar
Kopia: gluster-users at gluster.org
Ämne: Re: [Gluster-users] Upgrade to 4.1.1 geo-replication does not work

Hi Kotresh,
Yes, all nodes have the same version 4.1.1 both master and slave.
All glusterd are crashing on the master side.
Will send logs tonight.

Thanks,
Marcus

################
Marcus Pedersén
Systemadministrator
Interbull Centre
################
Sent from my phone
################

Den 13 juli 2018 11:28 skrev Kotresh Hiremath Ravishankar <khiremat at redhat.com>:
Hi Marcus,

Is the gluster geo-rep version is same on both master and slave?

Thanks,
Kotresh HR

On Fri, Jul 13, 2018 at 1:26 AM, Marcus Pedersén <marcus.pedersen at slu.se<mailto:marcus.pedersen at slu.se>> wrote:

Hi Kotresh,

i have replaced both files (gsyncdconfig.py<https://review.gluster.org/#/c/20207/1/geo-replication/syncdaemon/gsyncdconfig.py> and repce.py<https://review.gluster.org/#/c/20207/1/geo-replication/syncdaemon/repce.py>) in all nodes both master and slave.

I rebooted all servers but geo-replication status is still Stopped.

I tried to start geo-replication with response Successful but status still show Stopped on all nodes.

Nothing has been written to geo-replication logs since I sent the tail of the log.

So I do not know what info to provide?


Please, help me to find a way to solve this.


Thanks!


Regards

Marcus


________________________________
Från: gluster-users-bounces at gluster.org<mailto:gluster-users-bounces at gluster.org> <gluster-users-bounces at gluster.org<mailto:gluster-users-bounces at gluster.org>> för Marcus Pedersén <marcus.pedersen at slu.se<mailto:marcus.pedersen at slu.se>>
Skickat: den 12 juli 2018 08:51
Till: Kotresh Hiremath Ravishankar
Kopia: gluster-users at gluster.org<mailto:gluster-users at gluster.org>
Ämne: Re: [Gluster-users] Upgrade to 4.1.1 geo-replication does not work

Thanks Kotresh,
I installed through the official centos channel, centos-release-gluster41.
Isn't this fix included in centos install?
I will have a look, test it tonight and come back to you!

Thanks a lot!

Regards
Marcus

################
Marcus Pedersén
Systemadministrator
Interbull Centre
################
Sent from my phone
################

Den 12 juli 2018 07:41 skrev Kotresh Hiremath Ravishankar <khiremat at redhat.com<mailto:khiremat at redhat.com>>:
Hi Marcus,

I think the fix [1] is needed in 4.1
Could you please this out and let us know if that works for you?

[1] https://review.gluster.org/#/c/20207/

Thanks,
Kotresh HR

On Thu, Jul 12, 2018 at 1:49 AM, Marcus Pedersén <marcus.pedersen at slu.se<mailto:marcus.pedersen at slu.se>> wrote:

Hi all,

I have upgraded from 3.12.9 to 4.1.1 and been following upgrade instructions for offline upgrade.

I upgraded geo-replication side first 1 x (2+1) and the master side after that 2 x (2+1).

Both clusters works the way they should on their own.

After upgrade on master side status for all geo-replication nodes is Stopped.

I tried to start the geo-replication from master node and response back was started successfully.

Status again .... Stopped

Tried to start again and get response started successfully, after that all glusterd crashed on all master nodes.

After a restart of all glusterd the master cluster was up again.

Status for geo-replication is still Stopped and every try to start it after this gives the response successful but still status Stopped.


Please help me get the geo-replication up and running again.


Best regards

Marcus Pedersén


Part of geo-replication log from master node:

[2018-07-11 18:42:48.941760] I [changelogagent(/urd-gds/gluster):73:__init__] ChangelogAgent: Agent listining...
[2018-07-11 18:42:48.947567] I [resource(/urd-gds/gluster):1780:connect_remote] SSH: Initializing SSH connection between master and slave...
[2018-07-11 18:42:49.363514] E [syncdutils(/urd-gds/gluster):304:log_raise_exception] <top>: connection to peer is broken
[2018-07-11 18:42:49.364279] E [resource(/urd-gds/gluster):210: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-hjRhBo/7e5534547f3675a710a107722317484f.sock geouser at urd-gds-geo-000 /nonexistent/gsyncd --session-owner 5e94eb7d-219f-4741-a179-d4ae6b50c7ee --local-id .%\
2Furd-gds%2Fgluster --local-node urd-gds-001 -N --listen --timeout 120 gluster://localhost:urd-gds-volume   error=2
[2018-07-11 18:42:49.364586] E [resource(/urd-gds/gluster):214:logerr] Popen: ssh> usage: gsyncd.py [-h]
[2018-07-11 18:42:49.364799] E [resource(/urd-gds/gluster):214:logerr] Popen: ssh>
[2018-07-11 18:42:49.364989] E [resource(/urd-gds/gluster):214:logerr] Popen: ssh>                  {monitor-status,monitor,worker,agent,slave,status,config-check,config-get,config-set,config-reset,voluuidget,d\
elete}
[2018-07-11 18:42:49.365210] E [resource(/urd-gds/gluster):214:logerr] Popen: ssh>                  ...
[2018-07-11 18:42:49.365408] E [resource(/urd-gds/gluster):214:logerr] Popen: ssh> gsyncd.py: error: argument subcmd: invalid choice: '5e94eb7d-219f-4741-a179-d4ae6b50c7ee' (choose from 'monitor-status', 'monit\
or', 'worker', 'agent', 'slave', 'status', 'config-check', 'config-get', 'config-set', 'config-reset', 'voluuidget', 'delete')
[2018-07-11 18:42:49.365919] I [syncdutils(/urd-gds/gluster):271:finalize] <top>: exiting.
[2018-07-11 18:42:49.369316] I [repce(/urd-gds/gluster):92:service_loop] RepceServer: terminating on reaching EOF.
[2018-07-11 18:42:49.369921] I [syncdutils(/urd-gds/gluster):271:finalize] <top>: exiting.
[2018-07-11 18:42:49.369694] I [monitor(monitor):353:monitor] Monitor: worker died before establishing connection       brick=/urd-gds/gluster
[2018-07-11 18:42:59.492762] I [monitor(monitor):280:monitor] Monitor: starting gsyncd worker   brick=/urd-gds/gluster  slave_node=ssh://geouser@urd-gds-geo-000:gluster://localhost:urd-gds-volume
[2018-07-11 18:42:59.558491] I [resource(/urd-gds/gluster):1780:connect_remote] SSH: Initializing SSH connection between master and slave...
[2018-07-11 18:42:59.559056] I [changelogagent(/urd-gds/gluster):73:__init__] ChangelogAgent: Agent listining...
[2018-07-11 18:42:59.945693] E [syncdutils(/urd-gds/gluster):304:log_raise_exception] <top>: connection to peer is broken
[2018-07-11 18:42:59.946439] E [resource(/urd-gds/gluster):210: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-992bk7/7e5534547f3675a710a107722317484f.sock geouser at urd-gds-geo-000 /nonexistent/gsyncd --session-owner 5e94eb7d-219f-4741-a179-d4ae6b50c7ee --local-id .%\
2Furd-gds%2Fgluster --local-node urd-gds-001 -N --listen --timeout 120 gluster://localhost:urd-gds-volume   error=2
[2018-07-11 18:42:59.946748] E [resource(/urd-gds/gluster):214:logerr] Popen: ssh> usage: gsyncd.py [-h]
[2018-07-11 18:42:59.946962] E [resource(/urd-gds/gluster):214:logerr] Popen: ssh>
[2018-07-11 18:42:59.947150] E [resource(/urd-gds/gluster):214:logerr] Popen: ssh>                  {monitor-status,monitor,worker,agent,slave,status,config-check,config-get,config-set,config-reset,voluuidget,d\
elete}
[2018-07-11 18:42:59.947369] E [resource(/urd-gds/gluster):214:logerr] Popen: ssh>                  ...
[2018-07-11 18:42:59.947552] E [resource(/urd-gds/gluster):214:logerr] Popen: ssh> gsyncd.py: error: argument subcmd: invalid choice: '5e94eb7d-219f-4741-a179-d4ae6b50c7ee' (choose from 'monitor-status', 'monit\
or', 'worker', 'agent', 'slave', 'status', 'config-check', 'config-get', 'config-set', 'config-reset', 'voluuidget', 'delete')
[2018-07-11 18:42:59.948046] I [syncdutils(/urd-gds/gluster):271:finalize] <top>: exiting.
[2018-07-11 18:42:59.951392] I [repce(/urd-gds/gluster):92:service_loop] RepceServer: terminating on reaching EOF.
[2018-07-11 18:42:59.951760] I [syncdutils(/urd-gds/gluster):271:finalize] <top>: exiting.
[2018-07-11 18:42:59.951817] I [monitor(monitor):353:monitor] Monitor: worker died before establishing connection       brick=/urd-gds/gluster
[2018-07-11 18:43:10.54580] I [monitor(monitor):280:monitor] Monitor: starting gsyncd worker    brick=/urd-gds/gluster  slave_node=ssh://geouser@urd-gds-geo-000:gluster://localhost:urd-gds-volume
[2018-07-11 18:43:10.88356] I [monitor(monitor):345:monitor] Monitor: Changelog Agent died, Aborting Worker     brick=/urd-gds/gluster
[2018-07-11 18:43:10.88613] I [monitor(monitor):353:monitor] Monitor: worker died before establishing connection        brick=/urd-gds/gluster
[2018-07-11 18:43:20.112435] I [gsyncdstatus(monitor):242:set_worker_status] GeorepStatus: Worker Status Change status=inconsistent
[2018-07-11 18:43:20.112885] E [syncdutils(monitor):331:log_raise_exception] <top>: FAIL:
Traceback (most recent call last):
  File "/usr/libexec/glusterfs/python/syncdaemon/syncdutils.py", line 361, in twrap
    except:
  File "/usr/libexec/glusterfs/python/syncdaemon/monitor.py", line 428, in wmon
    sys.exit()
TypeError: 'int' object is not iterable
[2018-07-11 18:43:20.114610] I [syncdutils(monitor):271:finalize] <top>: exiting.

---
När du skickar e-post till SLU så innebär detta att SLU behandlar dina personuppgifter. För att läsa mer om hur detta går till, klicka här <https://www.slu.se/om-slu/kontakta-slu/personuppgifter/>
E-mailing SLU will result in SLU processing your personal data. For more information on how this is done, click here <https://www.slu.se/en/about-slu/contact-slu/personal-data/>

_______________________________________________
Gluster-users mailing list
Gluster-users at gluster.org<mailto:Gluster-users at gluster.org>
https://lists.gluster.org/mailman/listinfo/gluster-users



--
Thanks and Regards,
Kotresh H R


---
När du skickar e-post till SLU så innebär detta att SLU behandlar dina personuppgifter. För att läsa mer om hur detta går till, klicka här <https://www.slu.se/om-slu/kontakta-slu/personuppgifter/>
E-mailing SLU will result in SLU processing your personal data. For more information on how this is done, click here <https://www.slu.se/en/about-slu/contact-slu/personal-data/>

---
När du skickar e-post till SLU så innebär detta att SLU behandlar dina personuppgifter. För att läsa mer om hur detta går till, klicka här <https://www.slu.se/om-slu/kontakta-slu/personuppgifter/>
E-mailing SLU will result in SLU processing your personal data. For more information on how this is done, click here <https://www.slu.se/en/about-slu/contact-slu/personal-data/>



--
Thanks and Regards,
Kotresh H R


---
När du skickar e-post till SLU så innebär detta att SLU behandlar dina personuppgifter. För att läsa mer om hur detta går till, klicka här <https://www.slu.se/om-slu/kontakta-slu/personuppgifter/>
E-mailing SLU will result in SLU processing your personal data. For more information on how this is done, click here <https://www.slu.se/en/about-slu/contact-slu/personal-data/>

---
När du skickar e-post till SLU så innebär detta att SLU behandlar dina personuppgifter. För att läsa mer om hur detta går till, klicka här <https://www.slu.se/om-slu/kontakta-slu/personuppgifter/>
E-mailing SLU will result in SLU processing your personal data. For more information on how this is done, click here <https://www.slu.se/en/about-slu/contact-slu/personal-data/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20180716/edf68eab/attachment.html>

More information about the Gluster-users mailing list