15 Feb
2021
15 Feb
'21
9:49 p.m.
Hi Matthew,
I have always seen (and this is what we ourselves practice) that 3 is the
optimal number of monitors for all but the most extreme clusters.
We have never had an issue with that setup. We run monitors as VMs on
either Proxmox or VMWare, and back them up, and replicate as well, so there
is HA and the ability to bring a mon back from a short while ago, in which
case it will synchronize with the rest of the cluster quickly - we had
tested this successfully in the lab.
It does not seem like going to 5 mons is necessary for you.
--
Alex Gorbachev
I <https://alextelescope.blogspot.com>ntelligent Systems Services Inc.
STORCIUM
On Mon, Feb 15, 2021 at 11:18 AM Matthew Vernon <mv3(a)sanger.ac.uk> wrote:
On 12/02/2021 15:47, Freddy Andersen wrote:
I would say everyone recommends at least 3
monitors and since they
need to be 1,3,5 or 7 I always read that as 5 is the best number (if
you have 5 servers in your cluster).
We have 3 on all our clusters, and at the
risk of tempting fate, haven't
had any issues as a result...
[it's slightly fiddly to add more, since we give them a bunch of extra
storage than our other nodes since the Mon store can get pretty big in a
large cluster if you have to do a big rebalance]
Regards,
Matthew
--
The Wellcome Sanger Institute is operated by Genome Research
Limited, a charity registered in England with number 1021457 and a
company registered in England with number 2742969, whose registered
office is 215 Euston Road, London, NW1 2BE.
_______________________________________________
ceph-users mailing list -- ceph-users(a)ceph.io
To unsubscribe send an email to ceph-users-leave(a)ceph.io