ceph常用操作
//修复一批数据不一致的pg
ceph health detail|grep acting|awk -F' ' '{print $2}'|xargs -n1 ceph pg repair
//修复osd上too many repair read
ceph tell osd.* clear_shards_repaired
//查看ceph集群的健康状态
ceph health
//查看ceph集群的健康详情信息
ceph health detail
ceph -w可以查看日志
ceph -s
//打印ceph 集群上的osd
ceph osd dump|grep kubernetes
//打印ceph 集群上的pg
ceph pg dump|head -10
打印pool列表
ceph osd lspools
创建pool
ceph osd pool create test-pool 128
设置pool最大允许的对象
ceph osd pool set-quota test-pool max_objects 100
设置pool最大容量
ceph osd pool set-quota test-pool max_bytes $((10 * 1024 * 1024 * 1024))
重命名pool
ceph osd pool rename test-pool test-pool-new
删除pool
ceph osd pool delete test-pool test-pool --yes-i-really-really-mean-it
查看pool状态信息
rados df
通过以下语法可以设置pool的元数据
ceph osd pool set {pool-name} {key} {value}
比如设置pool的冗余副本数量为:3
ceph osd pool set test-pool size 3
通过get操作能够获取pool的配置值,比如获取当前pg_num
ceph osd pool get test-pool pg_num
获取当前副本数
ceph osd pool get test-pool size
卷操作:
1. 创建新卷:rbd create kubernetes/rbdname -s 100G
2. 扩容新卷:rbd resize kubernetes/rbdname -s 200G
3. 删除新卷:rbd remove kubernetes/rbdname
rbd info kubernetes/rbdname
rbd ls kubernetes
rbd diff kubernetes/rbdname
rbd showmapped