POOLS

参考文档：https://docs.ceph.com/docs/master/rados/operations/pools/

列出集群的池

[ceph@ceph02 ~]$ ceph osd lspools

创建pool

#语法：
ceph osd pool create {pool-name} [{pg-num} [{pgp-num}]] [replicated] \
     [crush-rule-name] [expected-num-objects]
ceph osd pool create {pool-name} [{pg-num} [{pgp-num}]]   erasure \
     [erasure-code-profile] [crush-rule-name] [expected_num_objects] [--autoscale-mode=<on,off,warn>]

#举例：
[ceph@ceph02 ~]$ ceph osd pool create test11 100
pool 'test11' created

将pool关联到应用程序

#语法：
ceph osd pool application enable {pool-name} {application-name}

#举例：
[ceph@ceph02 ~]$ ceph osd pool application enable test11 rbd
enabled application 'rbd' on pool 'test11'

设置pool的配额

#语法：
ceph osd pool set-quota {pool-name} [max_objects {obj-count}] [max_bytes {bytes}]

#举例：
[ceph@ceph02 ~]$ ceph osd pool set-quota test11 max_objects 100
set-quota max_objects = 100 for pool test11

#注：要删除配额，请将其值设置为0。

删除pool

#语法：
ceph osd pool delete {pool-name} [{pool-name} --yes-i-really-really-mean-it]

#举例：
[ceph@ceph02 ~]$ ceph osd pool delete mytest mytest --yes-i-really-really-mean-it
Error EPERM: pool deletion is disabled; you must first set the mon_allow_pool_delete config option to true before you can destroy a pool

#根据错误提示，在ceph-deploy修改ceph.conf,然后把配置文件同步到其它节点
[ceph@ceph01 ceph-cluster]$ vi ceph.conf
[mon]
mon allow pool delete = true
[ceph@ceph01 ceph-cluster]$ ceph-deploy --overwrite-conf config push ceph{01,02,03,04,05,06}
#在mon节点上重启一下mon服务
[ceph@ceph01 ceph-cluster]$ sudo systemctl restart ceph-mon.target
[ceph@ceph02 ~]$ sudo systemctl restart ceph-mon.target
[ceph@ceph03 ~]$ sudo systemctl restart ceph-mon.target
#再删除pool
[ceph@ceph02 ~]$ ceph osd pool delete mytest mytest --yes-i-really-really-mean-it
pool 'mytest' removed

重命名pool

#语法
ceph osd pool rename {current-pool-name} {new-pool-name}

举例：
[ceph@ceph02 ~]$ ceph osd pool rename test11 test22
pool 'test11' renamed to 'test22'

统计pool信息

[ceph@ceph02 ~]$ rados df

#要获取特定池或全部池的I/O信息
ceph osd pool stats [{pool-name}]

pool快照管理

[ceph@ceph02 ~]$ rados -p test put testfile /etc/hosts   #把hosts文件写到对象testfile
[ceph@ceph02 ~]$ rados -p test ls                        #查看对象
testfile
[ceph@ceph02 ~]$ rados -p test mksnap snaphost           #创建名为snaphost的快照
created pool test snap snaphost
[ceph@ceph02 ~]$ rados -p test lssnap                    #查看快照
1       snaphost        2020.04.22 04:14:47
1 snaps
[ceph@ceph02 ~]$ rados -p test ls
testfile
[ceph@ceph02 ~]$ rados -p test rm testfile               #删除对象
[ceph@ceph02 ~]$ rados -p test ls                        #很奇怪，删除后ls还显示有这个对象
testfile
[ceph@ceph02 ~]$ rados -p test get testfile host         #验证一下这个对象是否存在，尝试把对象文件get下来，提示没有这个文件
error getting test/testfile: (2) No such file or directory
[ceph@ceph02 ~]$ rados rollback -p test testfile snaphost #回滚快照
rolled back pool test to snapshot snaphost
[ceph@ceph02 ~]$ rados -p test ls                         #查看这个pool里的对象，显示有testfile这个对象
testfile
[ceph@ceph02 ~]$ rados -p test get testfile host          #把这个对象get下来，验证回滚是否成功
[ceph@ceph02 ~]$ ls
host

获取对象副本数

[ceph@ceph01 ceph-cluster]$ ceph osd dump | grep 'replicated size'
pool 2 'test22' replicated size 2 min_size 2 crush_rule 0 object_hash rjenkins pg_num 128 pgp_num 128 autoscale_mode warn last_change 47 lfor 0/0/43 flags hashpspool,pool_snaps stripe_width 0
pool 3 'test' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 128 pgp_num 128 autoscale_mode warn last_change 48 flags hashpspool,pool_snaps stripe_width 0

GET POOL VALUES

#语法：
ceph osd pool get {pool-name} {key}
｛key｝有以下值： size|min_size|pg_num|pgp_num|crush_rule|hashpspool|nodelete|nopgchange|nosizechange|write_fadvise_dontneed|noscrub|nodeepscrub|hit_set_type|hit_set_period|hit_set_count|hit_set_fpp|use_gmt_hitset|target_max_objects|target_max_bytes|cache_target_dirty_ratio|cache_target_dirty_high_ratio|cache_target_full_ratio|cache_min_flush_age|cache_min_evict_age|erasure_code_profile|min_read_recency_for_promote|all|min_write_recency_for_promote|fast_read|hit_set_grade_decay_rate|hit_set_search_last_n|scrub_min_interval|scrub_max_interval|deep_scrub_interval|recovery_priority|recovery_op_priority|scrub_priority|compression_mode|compression_algorithm|compression_required_ratio|compression_max_blob_size|compression_min_blob_size|csum_type|csum_min_block|csum_max_block|allow_ec_overwrites|fingerprint_algorithm|pg_autoscale_mode|pg_autoscale_bias|pg_num_min|target_size_bytes|target_size_ratio

SET POOL VALUES

#语法：
ceph osd pool set {pool-name} {key} {value}

#举例：
[ceph@ceph03 ~]$ ceph osd pool set test size 2        #设置对象副本数
set pool 3 size to 2

ERASURE CODE

新建的pool默认都是 replicated（复制池） 的，而erasure类型的pool池可以用来代替以节省空间，最简单的erasure编码池等效于RAID5，但至少需要三个主机。
我们来看一下默认的erasure的配置文件：

[ceph@ceph06 ~]$ ceph osd erasure-code-profile ls              #查看集群内的存在的erasure配置文件
default
[ceph@ceph06 ~]$ ceph osd erasure-code-profile get default
k=2                       #K原始数据盘个数或恢复数据需要的磁盘个数
m=1                       #M校验盘个数或允许出故障的盘个数
#使用编码算法，通过K个原始数据生成K+M个新数据，通过任何K个新数据都能还原出原始的K个数据，即允许M个数据盘出故障，数据仍然不会丢失
plugin=jerasure           #jerasure plugin是最通用和灵活的插件,封装了Jerasure库。
technique=reed_sol_van    #使用的编码算法为reed_sol_van

新建一个ERASURE CODE PROFILES文件

[ceph@ceph06 ~]$ ceph osd erasure-code-profile set myprofile \
    k=2 \
    m=1 \
    crush-failure-domain=rack    #将创建一个CRUSH规则集，确保没有两个块存储在同一个机架中
[ceph@ceph06 ~]$ ceph osd erasure-code-profile ls
default
myprofile                        #看到自定义的erasure code文件已被创建

创建erasure pool

[ceph@ceph06 ~]$ ceph osd pool create ecpool 12 12 erasure myprofile
pool 'ecpool' created
[ceph@ceph06 ~]$ echo ABCDEFGHI | rados --pool ecpool put NYAN -
[ceph@ceph06 ~]$ rados --pool ecpool01 get NYAN -
ABCDEFGHI

删除erasure code profiles配置文件

[ceph@ceph05 ~]$ ceph osd erasure-code-profile ls
default
myprofile
[ceph@ceph05 ~]$ ceph osd erasure-code-profile rm myprofile        #因为ecpool引用了myprofile配置文件，删除失败
Error EBUSY: ecpool pool(s) are using the erasure code profile 'myprofile'