Gluster users - Mar 2019 - Geo-replication status always on 'Created'

Hi Maurya,

I guess you missed last trick to distribute keys in slave node. I see
this is non-root geo-rep setup so please try this:


Run the following command as root in any one of Slave node.

/usr/local/libexec/glusterfs/set_geo_rep_pem_keys.sh  <slave_user>
<master_volume> <slave_volume>

- Sunny

On Wed, Mar 20, 2019 at 10:47 PM Maurya M <mauryam at gmail.com>
wrote:
>
> Hi all,
>  Have setup a 3 master nodes - 3 slave nodes (gluster 4.1) for
geo-replication, but once have the geo-replication configure the status is
always on "Created',
> even after have force start the session.
>
> On close inspect of the logs on the master node seeing this error:
>
> "E [syncdutils(monitor):801:errlog] Popen: command returned error  
cmd=ssh -oPasswordAuthentication=no -oStrictHostKeyChecking=no -i
/var/lib/glusterd/geo-replication/secret.pem -p 22 azureuser at xxxxx.xxxx..xxx.
gluster --xml --remote-host=localhost volume info
vol_a5ae34341a873c043c99a938adcb5b5781      error=255"
>
> Any ideas what is issue?
>
> thanks,
> Maurya
>
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> https://lists.gluster.org/mailman/listinfo/gluster-users

Hi Sunil,
 I did run the on the slave node :
 /usr/libexec/glusterfs/set_geo_rep_pem_keys.sh azureuser
vol_041afbc53746053368a1840607636e97 vol_a5aee81a873c043c99a938adcb5b5781
getting this message "/home/azureuser/common_secret.pem.pub not present.
Please run geo-replication command on master with push-pem option to
generate the file"

So went back and created the session again, no change, so manually copied
the common_secret.pem.pub to /home/azureuser/ but still the
set_geo_rep_pem_keys.sh is looking the pem file in different name
:  COMMON_SECRET_PEM_PUB=${master_vol}_${slave_vol}_common_secret.pem.pub ,
change the name of pem , ran the command again :

 /usr/libexec/glusterfs/set_geo_rep_pem_keys.sh azureuser
vol_041afbc53746053368a1840607636e97 vol_a5aee81a873c043c99a938adcb5b5781
Successfully copied file.
Command executed successfully.


- went back and created the session , start the geo-replication , still
seeing the  same error in logs. Any ideas ?

thanks,
Maurya



On Wed, Mar 20, 2019 at 11:07 PM Sunny Kumar <sunkumar at redhat.com>
wrote:
> Hi Maurya,
>
> I guess you missed last trick to distribute keys in slave node. I see
> this is non-root geo-rep setup so please try this:
>
>
> Run the following command as root in any one of Slave node.
>
> /usr/local/libexec/glusterfs/set_geo_rep_pem_keys.sh  <slave_user>
> <master_volume> <slave_volume>
>
> - Sunny
>
> On Wed, Mar 20, 2019 at 10:47 PM Maurya M <mauryam at gmail.com>
wrote:
> >
> > Hi all,
> >  Have setup a 3 master nodes - 3 slave nodes (gluster 4.1) for
> geo-replication, but once have the geo-replication configure the status is
> always on "Created',
> > even after have force start the session.
> >
> > On close inspect of the logs on the master node seeing this error:
> >
> > "E [syncdutils(monitor):801:errlog] Popen: command returned error
>  cmd=ssh -oPasswordAuthentication=no -oStrictHostKeyChecking=no -i
> /var/lib/glusterd/geo-replication/secret.pem -p 22 azureuser at
xxxxx.xxxx..xxx.
> gluster --xml --remote-host=localhost volume info
> vol_a5ae34341a873c043c99a938adcb5b5781      error=255"
> >
> > Any ideas what is issue?
> >
> > thanks,
> > Maurya
> >
> > _______________________________________________
> > Gluster-users mailing list
> > Gluster-users at gluster.org
> > https://lists.gluster.org/mailman/listinfo/gluster-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.gluster.org/pipermail/gluster-users/attachments/20190321/899abe56/attachment.html>