Hi German, For Data to be split over the racks you should set the crush rule set to 'step chooseleaf firstn 0 type rack' instead of 'step chooseleaf firstn 0 type host'
Thanks On Wed, Mar 23, 2016 at 3:50 PM, German Anders <[email protected]> wrote: > Hi all, > > I had a question, I'm in the middle of a new ceph deploy cluster and I've > 6 OSD servers between two racks, so rack1 would have osdserver1,3 and 5, > and rack2 osdserver2,4 and 6. I've edited the following crush map and I > want to know if it's ok and also if the objects would be stored one on each > rack-host. So, if I lost one rack, I had one copy on the other rack/server: > > *http://pastebin.com/raw/QJf1VeeJ <http://pastebin.com/raw/QJf1VeeJ>* > > Also If I need to run any command in order to 'apply' the new crush map to > the existing pools (actually only two): > > - 0 rbd (pg_num: 4096 | pgp_num: 4096 | size: 2 | min_size: 1) > - 1 cinder-volumes (pg_num: 4096 | pgp_num: 4096 | size: 2 | min_size: 1) > > # ceph --cluster cephIB osd tree > ID WEIGHT TYPE NAME UP/DOWN REWEIGHT PRIMARY-AFFINITY > -1 51.29668 root default > -8 26.00958 rack cage5-rack1 > -2 8.66986 host cibn01 > 0 0.72249 osd.0 up 1.00000 1.00000 > 1 0.72249 osd.1 up 1.00000 1.00000 > 2 0.72249 osd.2 up 1.00000 1.00000 > 3 0.72249 osd.3 up 1.00000 1.00000 > 4 0.72249 osd.4 up 1.00000 1.00000 > 5 0.72249 osd.5 up 1.00000 1.00000 > 6 0.72249 osd.6 up 1.00000 1.00000 > 7 0.72249 osd.7 up 1.00000 1.00000 > 8 0.72249 osd.8 up 1.00000 1.00000 > 9 0.72249 osd.9 up 1.00000 1.00000 > 10 0.72249 osd.10 up 1.00000 1.00000 > 11 0.72249 osd.11 up 1.00000 1.00000 > -4 8.66986 host cibn03 > 24 0.72249 osd.24 up 1.00000 1.00000 > 25 0.72249 osd.25 up 1.00000 1.00000 > 26 0.72249 osd.26 up 1.00000 1.00000 > 27 0.72249 osd.27 up 1.00000 1.00000 > 28 0.72249 osd.28 up 1.00000 1.00000 > 29 0.72249 osd.29 up 1.00000 1.00000 > 30 0.72249 osd.30 up 1.00000 1.00000 > 31 0.72249 osd.31 up 1.00000 1.00000 > 32 0.72249 osd.32 up 1.00000 1.00000 > 33 0.72249 osd.33 up 1.00000 1.00000 > 34 0.72249 osd.34 up 1.00000 1.00000 > 35 0.72249 osd.35 up 1.00000 1.00000 > -6 8.66986 host cibn05 > 48 0.72249 osd.48 up 1.00000 1.00000 > 49 0.72249 osd.49 up 1.00000 1.00000 > 50 0.72249 osd.50 up 1.00000 1.00000 > 51 0.72249 osd.51 up 1.00000 1.00000 > 52 0.72249 osd.52 up 1.00000 1.00000 > 53 0.72249 osd.53 up 1.00000 1.00000 > 54 0.72249 osd.54 up 1.00000 1.00000 > 55 0.72249 osd.55 up 1.00000 1.00000 > 56 0.72249 osd.56 up 1.00000 1.00000 > 57 0.72249 osd.57 up 1.00000 1.00000 > 58 0.72249 osd.58 up 1.00000 1.00000 > 59 0.72249 osd.59 up 1.00000 1.00000 > -9 25.28709 rack cage5-rack2 > -3 8.66986 host cibn02 > 12 0.72249 osd.12 up 1.00000 1.00000 > 13 0.72249 osd.13 up 1.00000 1.00000 > 14 0.72249 osd.14 up 1.00000 1.00000 > 15 0.72249 osd.15 up 1.00000 1.00000 > 16 0.72249 osd.16 up 1.00000 1.00000 > 17 0.72249 osd.17 up 1.00000 1.00000 > 18 0.72249 osd.18 up 1.00000 1.00000 > 19 0.72249 osd.19 up 1.00000 1.00000 > 20 0.72249 osd.20 up 1.00000 1.00000 > 21 0.72249 osd.21 up 1.00000 1.00000 > 22 0.72249 osd.22 up 1.00000 1.00000 > 23 0.72249 osd.23 up 1.00000 1.00000 > -5 8.66986 host cibn04 > 36 0.72249 osd.36 up 1.00000 1.00000 > 37 0.72249 osd.37 up 1.00000 1.00000 > 38 0.72249 osd.38 up 1.00000 1.00000 > 39 0.72249 osd.39 up 1.00000 1.00000 > 40 0.72249 osd.40 up 1.00000 1.00000 > 41 0.72249 osd.41 up 1.00000 1.00000 > 42 0.72249 osd.42 up 1.00000 1.00000 > 43 0.72249 osd.43 up 1.00000 1.00000 > 44 0.72249 osd.44 up 1.00000 1.00000 > 45 0.72249 osd.45 up 1.00000 1.00000 > 46 0.72249 osd.46 up 1.00000 1.00000 > 47 0.72249 osd.47 up 1.00000 1.00000 > -7 7.94737 host cibn06 > 60 0.72249 osd.60 up 1.00000 1.00000 > 61 0.72249 osd.61 up 1.00000 1.00000 > 62 0.72249 osd.62 up 1.00000 1.00000 > 63 0.72249 osd.63 up 1.00000 1.00000 > 64 0.72249 osd.64 up 1.00000 1.00000 > 65 0.72249 osd.65 up 1.00000 1.00000 > 66 0.72249 osd.66 up 1.00000 1.00000 > 67 0.72249 osd.67 up 1.00000 1.00000 > 68 0.72249 osd.68 up 1.00000 1.00000 > 69 0.72249 osd.69 up 1.00000 1.00000 > 70 0.72249 osd.70 up 1.00000 1.00000 > > > Ceph version 9.2.1 (752b6a3020c3de74e07d2a8b4c5e48dab5a6b6fd) > > > Thanks in advance, > > Best, > > *German* > > _______________________________________________ > ceph-users mailing list > [email protected] > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > >
_______________________________________________ ceph-users mailing list [email protected] http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
