[prev in list] [next in list] [prev in thread] [next in thread] 

List:       ceph-users
Subject:    [ceph-users] Re: What exactly does the number of monitors depends on
From:       Tyler Stachecki <stachecki.tyler () gmail ! com>
Date:       2022-01-27 13:19:52
Message-ID: CAC6wqPXRnYDCPu1m7qpVyUELiFjEzFwVGYhHKRn6Z4jW7aJ_+Q () mail ! gmail ! com
[Download RAW message or body]

They maintain and distribute the OSD map, i.e. the state of the OSDs.

Generally 3 or 5 recommended as there always needs to be a quorum of mons
available or the cluster can stall.

With 3 mons, you can bring down one for servicing, patching, etc. but if
another mon fails while you intentionally take one down, that leaves only
one mon up and it will not have quorum, and not be able to function.

The mons also have to do CPU intensive work with PGs when you are changing
charteristics of the pools, so for large clusters you will want enough of
them such that you have enough headroom for the PG operations while still
distributing maps. Look at the mon_cpu_threads tunable in the code.

So for large clusters, 5 mons is advised to provide a little more buffer...
you may feel that 3 is fine for smaller clusters, though.

On Thu, Jan 27, 2022, 7:56 AM Zhenshi Zhou <deaderzzs@gmail.com> wrote:

> Hi all,
> 
> What makes me confused recently is that how to decide the number of
> monitors in a cluster.
> 
> In my opinion, it depends on the size of the cluster while my colleague
> says it should be at least 5 in a cluster. He sends me the redhat document
> <
> https://access.redhat.com/documentation/en-us/red_hat_ceph_storage/5/html-single/installation_guide/index
> 
> > 
> and
> it is indeed recommended to have 5 monitors in a cluster(see section 3.12).
> 
> However it also recommends deploying at least 3 monitors, deploying 5
> monitors for the cluster exceeding 750 osds.
> 
> So what exactly does the number of monitors depends on?
> 
> Thanks
> _______________________________________________
> ceph-users mailing list -- ceph-users@ceph.io
> To unsubscribe send an email to ceph-users-leave@ceph.io
> 
_______________________________________________
ceph-users mailing list -- ceph-users@ceph.io
To unsubscribe send an email to ceph-users-leave@ceph.io


[prev in list] [next in list] [prev in thread] [next in thread]