They don’t go though the MONs for IOs but they need access to the MONs over the public network for authentication and to receive the cluster map.
JC
While moving. Excuse unintended typos.
Hi
You deploy 3 MONs on a production cluster for HA.
CephFS clients talk to MONs MDSs and OSDs over the public network.
CephFS is not NFS and you’ll need ganesha to enable NFS access into your Ceph File system. See http://docs.ceph.com/docs/master/cephfs/nfs/ Ganesha will access your ceph cluster over the public network like any regular ceph client. JC
While moving. Excuse unintended typos. thanks for the replies, I don't know that cephFS clients go through the MONs, they reach the OSDs directly. When I mentioned NFS, I meant NFS clients (ie. not cephFS clients) This should have been pretty straight forward. Anyone doing HA on the MONs? How do you mount the cephFS shares, surely you'd have a vip?
> On Sep 17, 2018, at 16:13, solarflow99 <solarflow99@xxxxxxxxx> wrote:
>
> Hi, I read through the various documentation and had a few questions:
>
> - From what I understand cephFS clients reach the OSDs directly, does the cluster network need to be opened up as a public network?
Client traffic only goes over the public network. Only OSD to OSD traffic (replication, rebalancing, recovery go over the cluster network)
>
> - Is it still necessary to have a public and cluster network when the using cephFS since the clients all reach the OSD's directly?
Separating the network is a plus for troubleshooting and sizing for bandwidth
>
> - Simplest way to do HA on the mons for providing NFS, etc?
Don’t really understand the question (NFS vs CephFS).
> _______________________________________________
> ceph-users mailing list
> ceph-users@xxxxxxxxxxxxxx
> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
|
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
