Can this be fine to link this with bug 1019954 which treats this case as a skip, and that would lead admin to query logs to figure out which all files are not synced.

Looks like there is no further code fix required as customer is okay with the existing behavior as described by Amar in comment#4. Since Kevin has checked with customer and given a go, moving it to QE to confirm that the behavior is as described in comment #4.

Msg from QE:

This is error message you get if the slave already has data,

"slave-node::slave is not empty. Please delete existing files in slave-node::slave and retry, or use force to continue without deleting the existing files.
geo-replication command failed
"

Marking this bug for documentation (also changing the component).

Based on Comment 7, I have added a note at: http://documentation-devel.engineering.redhat.com/docs/en-US/Red_Hat_Storage/2.1/html/Administration_Guide/chap-User_Guide-Geo_Rep-Starting.html#chap-User_Guide-Geo_Rep-Starting-Start

Hence, moving this bug to ON_QA.

Hi Mohit,

Could you review the doc (link in Comment 8) and let us know if it addresses your concern?

Regards,
Anjana

When a user tries to create a geo-rep session with the slave which already has data, geo-rep create throws an error like the one below.


[root@pythagoras ]# gluster v geo master euclid::slave create push-pem
euclid::slave is not empty. Please delete existing files in euclid::slave and retry, or use force to continue without deleting the existing files.
geo-replication command failed


Since this is documented in http://documentation-devel.engineering.redhat.com/docs/en-US/Red_Hat_Storage/2.1/html/Administration_Guide/chap-User_Guide-Geo_Rep-Starting.html#chap-User_Guide-Geo_Rep-Starting-Start I am moving this bug to verified.

If there is something else needed, please open a new bug.