环境配置,三个节点,Ceph三副本
查看Ceph最小副本数
[root@node-1 ~]# ceph osd dump | grep -E " size | min_size "
pool 1 '.rgw.root' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 3 flags hashpspool stripe_width 0 application rgw
pool 2 'volumes' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 64 pgp_num 64 last_change 228 lfor 0/59 flags hashpspool stripe_width 0 application cinder-volume
pool 3 'compute' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 16 pgp_num 16 last_change 372 lfor 0/57 flags hashpspool stripe_width 0
pool 4 'rbd' replicated size 3 min_size 1 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 236 flags hashpspool stripe_width 0
pool 5 'ssdpool' replicated size 3 min_size 1 crush_rule 1 object_hash rjenkins pg_num 128 pgp_num 128 last_change 63 lfor 0/62 flags hashpspool stripe_width 0
pool 6 'metrics' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 35 flags hashpspool stripe_width 0
pool 7 'backups' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 229 flags hashpspool stripe_width 0 application cinder-backup
pool 8 'default.rgw.control' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 42 flags hashpspool stripe_width 0 application rgw
pool 9 'default.rgw.meta' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 44 flags hashpspool stripe_width 0 application rgw
pool 10 'default.rgw.log' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 46 flags hashpspool stripe_width 0 application rgw
pool 11 'default.rgw.buckets.index' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 53 flags hashpspool stripe_width 0 application rgw
pool 12 'default.rgw.buckets.data' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 8 pgp_num 8 last_change 55 flags hashpspool stripe_width 0 application rgw
1.查看rbd pool
rbd ls
2.在rbd pool里创建一个1G的文件
rbd create -s 1G rbd/test
3.Disable node-6的一个osd
kubectl edit nodes node-6
4.查看rbd bench的使用手册
rbd help bench
5.查看rbd/test大小
rbd du rbd/test
6.写满rbd/test
rbd bench --io-size 4096 --io-type write rbd/test
7.Enable node-6的osd
kubectl edit nodes node-6
环境配置,四个节点,Ceph三副本
1.查看ceph osd tree
[root@node-1 ~]# ceph osd tree
ID CLASS WEIGHT TYPE NAME STATUS REWEIGHT PRI-AFF
-2 0.49213 root ssdpool
-3 0.12303 host node-1_ssdpool
1 hdd 0.12303 osd.1 down 0 1.00000
-5 0.12303 host node-2_ssdpool
3 hdd 0.12303 osd.3 up 1.00000 1.00000
-17 0.12303 host node-3_ssdpool
5 hdd 0.12303 osd.5 up 1.00000 1.00000
-4 0.12303 host node-4_ssdpool
7 hdd 0.12303 osd.7 up 1.00000 1.00000
-1 0.85931 root default
-14 0.21483 host node-1
0 ssd 0.21483 osd.0 down 0 1.00000
-28 0.21483 host node-2
2 ssd 0.21483 osd.2 up 1.00000 1.00000
-6 0.21483 host node-3
4 ssd 0.21483 osd.4 up 1.00000 1.00000
-13 0.21483 host node-4
6 ssd 0.21483 osd.6 up 1.00000 1.00000
2.Out 一个 osd
[root@node-1 ~]# ceph osd out 3
marked out osd.3.
3.ceph处于recovery状态
[root@node-1 ~]# ceph -s
cluster:
id: 3927e644-f606-4cf6-b166-1beccc4c3047
health: HEALTH_WARN
1465/33586 objects misplaced (4.362%)
Reduced data availability: 8 pgs inactive, 29 pgs peering
services:
mon: 3 daemons, quorum node-1,node-3,node-4
mgr: node-4(active), standbys: node-3, node-1
osd: 8 osds: 6 up, 5 in; 84 remapped pgs
flags nodeep-scrub
rbd-mirror: 1 daemon active
rgw: 3 daemons active
data:
pools: 12 pools, 360 pgs
objects: 16.79k objects, 46.8GiB
usage: 99.8GiB used, 818GiB / 918GiB avail
pgs: 12.778% pgs not active
1465/33586 objects misplaced (4.362%)
276 active+clean
46 remapped+peering
37 active+remapped+backfill_wait
1 active+remapped+backfilling
io:
client: 550KiB/s wr, 0op/s rd, 46op/s wr
recovery: 17.0MiB/s, 5objects/s
4.加入osd
[root@node-1 ~]# ceph osd in 3
marked in osd.3.