On Wed, Mar 1, 2017 at 10:15 AM, Jimmy Goffaux <jimmy@xxxxxxxxxx> wrote: > > > Hello, > > I have a cluster CEPH (10.2.5-1trusty) I use the various possibilities: > > -Block > > - Object > > - CephFS > > > > > > root@ih-par1-cld1-ceph-01:~# cat /etc/ceph/ceph.conf > [....] > mon_host = 10.4.0.1, 10.4.0.3, 10.4.0.5 > [....] > public_network = 10.4.0.0/24 > cluster_network = 192.168.33.0/24 > [....] > > I have dedicated servers for the storage Block, Object and other servers for > CephFS (Full SSD): > > > > > root@ih-par1-cld1-ceph-01:~# ceph osd tree > ID WEIGHT TYPE NAME UP/DOWN REWEIGHT > PRIMARY-AFFINITY > -6 2.79593 root ssdforcephfs > -7 0.46599 host ih-prd-cephfs-02 > 32 0.23299 osd.32 up 1.00000 > 1.00000 > 33 0.23299 osd.33 up 1.00000 > 1.00000 > -8 0.46599 host ih-prd-cephfs-03 > 34 0.23299 osd.34 up 1.00000 > 1.00000 > 35 0.23299 osd.35 up 1.00000 > 1.00000 > -9 0.46599 host ih-prd-cephfs-05 > 36 0.23299 osd.36 up 1.00000 > 1.00000 > 37 0.23299 osd.37 up 1.00000 > 1.00000 > -10 0.46599 host ih-prd-cephfs-01 > 38 0.23299 osd.38 up 1.00000 > 1.00000 > 39 0.23299 osd.39 up 1.00000 > 1.00000 > -11 0.46599 host ih-prd-cephfs-04 > 40 0.23299 osd.40 up 1.00000 > 1.00000 > 41 0.23299 osd.41 up 1.00000 > 1.00000 > -12 0.46599 host ih-prd-cephfs-07 > 42 0.23299 osd.42 up 1.00000 > 1.00000 > 43 0.23299 osd.43 up 1.00000 > 1.00000 > -1 116.47998 root default > -2 43.67999 host ih-par1-cld1-ceph-01 > 0 3.64000 osd.0 up 1.00000 > 1.00000 > 2 3.64000 osd.2 up 1.00000 > 1.00000 > 6 3.64000 osd.6 up 1.00000 > 1.00000 > 8 3.64000 osd.8 up 1.00000 > 1.00000 > 15 3.64000 osd.15 up 1.00000 > 1.00000 > 16 3.64000 osd.16 up 1.00000 > 1.00000 > 19 3.64000 osd.19 up 1.00000 > 1.00000 > 22 3.64000 osd.22 up 1.00000 > 1.00000 > 24 3.64000 osd.24 up 1.00000 > 1.00000 > 26 3.64000 osd.26 up 1.00000 > 1.00000 > 28 3.64000 osd.28 up 1.00000 > 1.00000 > 30 3.64000 osd.30 up 1.00000 > 1.00000 > -3 43.67999 host ih-par1-cld1-ceph-03 > 1 3.64000 osd.1 up 1.00000 > 1.00000 > 3 3.64000 osd.3 up 1.00000 > 1.00000 > 5 3.64000 osd.5 up 1.00000 > 1.00000 > 7 3.64000 osd.7 up 1.00000 > 1.00000 > 13 3.64000 osd.13 up 1.00000 > 1.00000 > 4 3.64000 osd.4 up 1.00000 > 1.00000 > 20 3.64000 osd.20 up 1.00000 > 1.00000 > 23 3.64000 osd.23 up 1.00000 > 1.00000 > 25 3.64000 osd.25 up 1.00000 > 1.00000 > 27 3.64000 osd.27 up 1.00000 > 1.00000 > 29 3.64000 osd.29 up 1.00000 > 1.00000 > 31 3.64000 osd.31 up 1.00000 > 1.00000 > -5 29.12000 host ih-par1-cld1-ceph-05 > 9 3.64000 osd.9 up 1.00000 > 1.00000 > 10 3.64000 osd.10 up 1.00000 > 1.00000 > 11 3.64000 osd.11 up 1.00000 > 1.00000 > 12 3.64000 osd.12 up 1.00000 > 1.00000 > 14 3.64000 osd.14 up 1.00000 > 1.00000 > 17 3.64000 osd.17 up 1.00000 > 1.00000 > 18 3.64000 osd.18 up 1.00000 > 1.00000 > 21 3.64000 osd.21 up 1.00000 > 1.00000 > > > > > > I use OpenNebula for the use of RDB on the public network: 10.4.0.0/16. > > I shall like separating the RDB, RGW network and CephFS... I have my > customers CephFS who can accèder to all the network RBD, the hypervisors > OpenNebula > > Example: > > Customer A (CephFS, path: /client1) = > Reaches at present all the network > 10.4.0.0/16 > Customer B (CephFS, path: /client2) = > Reaches at present all the network > 10.4.0.0/16 > > How is it possible to separate networks: RBD, RGW and have multiple access > networks for CephFS? CephFS clients talk directly to OSDs, like RBD clients -- if you want to avoid giving your CephFS clients access to your Ceph public network then the simplest way to accomplish that is to access the filesysten via an NFS server. OTOH if you just want your CephFS clients to be on the public network, but be unable to reach the non-CephFS clients (i.e. hypervisors), then that just calls for a firewall. John > > I hope to have been clear:/ > > Thank you > > > > _______________________________________________ > 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
 |