Marc Eisenbarth
2016-Feb-16 05:11 UTC
[Gluster-users] Failure Mode for Distributed Hot-Tier and Distributed-Replicated Cold-Tier
What happens if I have a distributed hot-tier and distributed-replicated cold-tier when one of the hot-tier volumes goes away and one of the distributed-replicated volumes also goes away? Is my entire system still in a read-write state? -Marc -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20160215/b37bf363/attachment.html>
knarra
2016-Feb-16 05:39 UTC
[Gluster-users] Failure Mode for Distributed Hot-Tier and Distributed-Replicated Cold-Tier
On 02/16/2016 10:41 AM, Marc Eisenbarth wrote:> What happens if I have a distributed hot-tier and > distributed-replicated cold-tier when one of the hot-tier volumes goes > away and one of the distributed-replicated volumes also goes away? Is > my entire system still in a read-write state? > > -Marc > > > > _______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://www.gluster.org/mailman/listinfo/gluster-usersHi Marc, When you say "one of the hot-tier volumes and one of the distributed-replicated volumes goes away" do you mean to say, one of the bricks in those volumes?If yes, then cold-tier volume is still in read-write state. But you will not able to read/write anything to the brick in the hot-tier which goes away since it is a distribute volume. Thanks kasturi. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20160216/40a04b6e/attachment.html>