i have a 7 server
Server1:mon,mgr,mds
Server 2:mon,mgr,mds
Server3:mon,mgr,mds
Server4:osd.0,osd.1
Server5:osd.2,osd.3
Server6osd.4,osd.5
Server7:mon,mgr,mds
each osd size 4T
and :
ceph osd tree
ID CLASS WEIGHT TYPE NAME STATUS REWEIGHT PRI-AFF
-1 21.83098 root default
-3 7.27699 host cephnode01
0 hdd 3.63899 osd.0 up 1.00000 1.00000
1 hdd 3.63899 osd.1 up 1.00000 1.00000
-5 7.27699 host cephnode02
2 hdd 3.63899 osd.2 up 1.00000 1.00000
3 hdd 3.63899 osd.3 up 1.00000 1.00000
-7 7.27699 host cephnode03
4 hdd 3.63899 osd.4 up 1.00000 1.00000
5 hdd 3.63899 osd.5 up 1.00000 1.00000
Every 2.0s: ceph -s
Tue Aug 14 09:52:47 2018
cluster:
id: 6b28cabd-b94a-4a74-8d9e-00b8c9a61de4
health: HEALTH_WARN
Reduced data availability: 128 pgs inactive, 128 pgs incomplete
services:
mon: 5 daemons, quorum
cephnode01,cephmon01,cephnode02,cephnode03,cephmon03
mgr: cephmds01(active), standbys: cephmon03, cephmon02, cephmon01
mds: cephfs-1/1/1 up {0=cephmon03=up:active}, 3 up:standby
osd: 6 osds: 6 up, 6 in
data:
pools: 2 pools, 256 pgs
objects: 279 objects, 1.0 GiB
usage: 8.1 GiB used, 22 TiB / 22 TiB avail
pgs: 50.000% pgs not active
128 incomplete
128 active+clean
During power failure
health: HEALTH_WARN
Reduced data availability: 128 pgs inactive, 128 pgs incomplete
when i restart all servers Loss of all data and i must remove pool and recreate pool
below erorr displayed
health: HEALTH_WARN
Reduced data availability: 128 pgs inactive, 128 pgs incomplete
when i change crush map #rule from host to osd then health ok.
but i need host .
i used blow command for create pool .
i used erasure coding .
ceph osd pool create ECtemppool 128 128 erasure default
ceph osd erasure-code-profile set default crush-failure-domain=osd crush-
root=default k=4 m=2 --force
rados -p ECtemppool ls
ceph osd pool create cephfs_metadata 128
ceph osd pool set ECtemppool allow_ec_overwrites true
ceph fs new cephfs cephfs_metadata ECtemppool
best regards
i want a problem: below command is work!! :
ceph osd erasure-code-profile set EC-temp-pool
ceph osd erasure-code-profile set EC-temp-pool ruleset-failure-domain=host k=4 m=2 --force
ceph osd pool create ECtemppool 128 128 erasure EC-temp-pool
ceph osd pool create cephfs_metadata 128
ceph osd pool set ECtemppool allow_ec_overwrites true
ceph fs new cephfs cephfs_metadata ECtemppool