Hi,

I'm new with ceph and I'm trying to understand how replicating data
between two regions/zones work.

I've read this http://ceph.com/docs/master/radosgw/federated-config/
and this 
http://www.sebastien-han.fr/blog/2013/01/28/ceph-geo-replication-sort-of/
and tried that http://blog.kri5.fr/?p=21

Here are my thoughts and please let me know if I'm wrong.

1. Data can be synchronized between two zones - a master and a slave.
Can it be reversed? Let's say the master fails and I want to use the
slave as the replacement.
2. Data can't be synchronized between two regions. Two regions can
only share metadata.
Why is that? I know Python and I've seen that radosgw-agent throws an
exception in this case ('data sync can only occur between zones in the
same region'). But I'm curious what is the reason to do so. Are there
any plans to change this behaviour?
3. One cluster can live in many datacenters, but only when they are
close and the latency is low.

Unfortunately I haven't found any other resources about multiregion
config. Perhaps someone could recommend me some other resouces to
read...


Thanks,

Patrycja Szabłowska
_______________________________________________
ceph-users mailing list
[email protected]
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

Reply via email to