Alexandr Porunov
2016-Nov-17 16:54 UTC
[Gluster-users] Can use geo-replication with distributed-replicated volumes?
Thank you I will wait for it Sincerely, Alexandr On Thu, Nov 17, 2016 at 6:43 PM, Bipin Kunal <bkunal at redhat.com> wrote:> I don't have URL handy right now. Will send you tomorrow. Texting from > mobile right now. > > Thanks, > Bipin > > On Nov 17, 2016 9:00 PM, "Alexandr Porunov" <alexandr.porunov at gmail.com> > wrote: > >> Thank you for your help! >> >> Could you please give me a link or some information about failover? How >> to change a master state to a slave state? >> >> Best regards, >> Alexandr >> >> On Thu, Nov 17, 2016 at 5:07 PM, Bipin Kunal <bkunal at redhat.com> wrote: >> >>> Please find my comments inline. >>> >>> On Nov 17, 2016 8:30 PM, "Alexandr Porunov" <alexandr.porunov at gmail.com> >>> wrote: >>> > >>> > Hello, >>> > >>> > I have several questions about Geo-replication. Please answer if you >>> can. >>> > >>> > 1. Can use geo-replication with distributed-replicated volumes? >>> Yes. You can. >>> > 2. Can we use less servers in a slave datacenter then in the master >>> datacenter? (I.e. if I replicate a distributed -replicated volume which >>> consists from 10 servers to the slave datacenter where only 5 servers). For >>> example use less replicas in the slave datacenter. >>> Yes. You are free to use. It is just recommended to have slave volume >>> size equal to master volume. >>> > 3. Are there a possibility to enable failover? I.e. when master >>> datacenter dies we change our slave to the master? >>> Yes. You can promote slave when master dies. And when Master comes back >>> you can failback to master. >>> > >>> > Sincerely, >>> > Alexandr >>> > >>> > >>> Thanks, >>> Bipin Kunal >>> _______________________________________________ >>> > Gluster-users mailing list >>> > Gluster-users at gluster.org >>> > http://www.gluster.org/mailman/listinfo/gluster-users >>> >> >>-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20161117/0cb43e7c/attachment.html>
Bipin Kunal
2016-Nov-18 09:56 UTC
[Gluster-users] Can use geo-replication with distributed-replicated volumes?
Unfortunately upstream doc is not up to date with failover and failback commands. But you can use downstream doc : https://access.redhat.com/documentation/en-US/Red_Hat_Storage/3.1/html-single/Administration_Guide/index.html#sect-Disaster_Recovery These steps should work fine for you. We will try to update upstream doc as early as possible. Thanks, Bipin Kunal On Thu, Nov 17, 2016 at 10:24 PM, Alexandr Porunov <alexandr.porunov at gmail.com> wrote:> Thank you I will wait for it > > Sincerely, > Alexandr > > On Thu, Nov 17, 2016 at 6:43 PM, Bipin Kunal <bkunal at redhat.com> wrote: >> >> I don't have URL handy right now. Will send you tomorrow. Texting from >> mobile right now. >> >> Thanks, >> Bipin >> >> >> On Nov 17, 2016 9:00 PM, "Alexandr Porunov" <alexandr.porunov at gmail.com> >> wrote: >>> >>> Thank you for your help! >>> >>> Could you please give me a link or some information about failover? How >>> to change a master state to a slave state? >>> >>> Best regards, >>> Alexandr >>> >>> On Thu, Nov 17, 2016 at 5:07 PM, Bipin Kunal <bkunal at redhat.com> wrote: >>>> >>>> Please find my comments inline. >>>> >>>> On Nov 17, 2016 8:30 PM, "Alexandr Porunov" <alexandr.porunov at gmail.com> >>>> wrote: >>>> > >>>> > Hello, >>>> > >>>> > I have several questions about Geo-replication. Please answer if you >>>> > can. >>>> > >>>> > 1. Can use geo-replication with distributed-replicated volumes? >>>> Yes. You can. >>>> > 2. Can we use less servers in a slave datacenter then in the master >>>> > datacenter? (I.e. if I replicate a distributed -replicated volume which >>>> > consists from 10 servers to the slave datacenter where only 5 servers). For >>>> > example use less replicas in the slave datacenter. >>>> Yes. You are free to use. It is just recommended to have slave volume >>>> size equal to master volume. >>>> > 3. Are there a possibility to enable failover? I.e. when master >>>> > datacenter dies we change our slave to the master? >>>> Yes. You can promote slave when master dies. And when Master comes back >>>> you can failback to master. >>>> > >>>> > Sincerely, >>>> > Alexandr >>>> > >>>> > >>>> Thanks, >>>> Bipin Kunal >>>> _______________________________________________ >>>> > Gluster-users mailing list >>>> > Gluster-users at gluster.org >>>> > http://www.gluster.org/mailman/listinfo/gluster-users >>> >>> >