thank you very much, this clear my confused.
在2016年04月08 14时13分, "Christian Balzer"<chibi@xxxxxxx>写道:
Hello,
On Fri, 8 Apr 2016 13:55:03 +0800 (CST) powerhd wrote:
>
>
> hi all:
> I have a question about monitor node, what is the maximum numbers of
> monitor node and how many monitor node need in a production
> environment? The official docs only recommends that monitor need odd
> number, so how do I calculate the numbers of monitor in a cluster, for
> example, a cluster with 100 osd, how many numbers of monitor node is
> recommended? Is there a document or something? Can someone help me ?
> Thanks!!!
The official documentation also says:
---
For an initial deployment of a multi-node Ceph cluster, it is advisable to deploy three monitors, increasing the number two at a time if a valid need for more than three exists.
---
If you're fine with being able to just loose one MON at any given time, 3
is fine.
I try to deploy 5 monitors in production, so that I can do maintenance on
one machine and still have another crash during that time w/o impacting the
cluster.
There is no real correlation between OSDs (or even storage nodes) to
MONs. People with VERY large deployments may have more than 5 monitors,
there is a point of diminishing return, so more is not better.
When deploying MONs you should of course make sure that you distribute
them correctly across your failure domains, not all on the same switch, in
the same rack, etc.
Christian
--
Christian Balzer Network/Systems Engineer
chibi@xxxxxxx Global OnLine Japan/Rakuten Communications
http://www.gol.com/
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
