集群管理
获取集群信息
包括集群名称、地址、卷数量、节点数量及使用率等
./udbto-cli cluster info
[Cluster]
Cluster name : udbto
Master leader : 192.168.2.162:15010
Master-3 : 192.168.2.185:15010
Master-1 : 192.168.2.162:15010
Master-2 : 192.168.2.181:15010
Auto allocate : Enabled
MetaNode count (active/total) : 3/3
MetaNode used : 0 GB
MetaNode available : 9 GB
MetaNode total : 10 GB
DataNode count (active/total) : 3/3
DataNode used : 139 GB
DataNode available : 330 GB
DataNode total : 469 GB
Volume count : 1
Allow Mp Decomm : Enabled
EbsAddr : http://192.168.2.162:8500
LoadFactor : 0
DpRepairTimeout : 2h0m0s
DataPartitionTimeout : 20m0s
volDeletionDelayTime : 48 h
MetaNodeGOGC : 100
DataNodeGOGC : 100
EnableAutoDecommission : false
AutoDecommissionDiskInterval : 10s
EnableAutoDpMetaRepair : false
AutoDpMetaRepairParallelCnt : 100
MarkDiskBrokenThreshold : 0%
DecommissionDpLimit : 10
DecommissionDiskLimit : 1
DpBackupTimeout : 168h0m0s
ForbidWriteOpOfProtoVersion0 : false
LegacyDataMediaType : 1
FlashNodeHandleReadTimeout : 1000 ms
FlashNodeReadDataNodeTimeout : 3000 ms
BatchCount : 0
MarkDeleteRate : 0
DeleteWorkerSleepMs: 0
AutoRepairRate : 0
MaxDpCntLimit : 3000
MaxMpCntLimit : 300
获取集群状态
按区域获取元数据和数据节点的使用量、状态等
./udbto-cli cluster stat
[Cluster Status]
DataNode Status:
TOTAL/GB USED/GB INCREASED/GB USED RATIO
469 139 0 0.296
MetaNode Status:
TOTAL/GB USED/GB INCREASED/GB USED RATIO
10 0 0 0.039
Zone List:
ZONE NAME ROLE TOTAL/GB USED/GB AVAILABLE/GB USED RATIO TOTAL NODES WRITEBLE NODES
default DATANODE 469.86 139.23 330.63 0.3 3 3
METANODE 10.28 0.4 9.88 0.04 3 3
冻结/解冻集群
是否冻结集群，设置为 true 冻结后，当partition写满，集群不会自动分配新的partition
./udbto-cli cluster freeze [true/false]
Freeze/Unfreeze cluster successful!
设置内存阈值
设置集群中每个 MetaNode 的内存阈值。当内存使用率超过该阈值时，上面的 meta partition 将会被设为只读。[float] 应当是一个介于0和1之间的小数.
./udbto-cli cluster threshold [float]
MetaNode threshold is set to 0.8!
设置卷延迟删除时间
volDeletionDelayTime代表启用延迟卷删除后，卷将在多少小时后被永久删除。在此之前，卷将被标记为删除状态，可以恢复, 以小时为单位, 默认48小时。
./udbto-cli cluster volDeletionDelayTime [VOLDELETIONDELAYTIME]
master volDeletionDelayTime is set to 48 h!
设置集群的参数
./udbto-cli cluster cluster set [flags]
Flags:
--autoRepairRate string DataNode auto repair rate
--batchCount string MetaNode delete batch count
--deleteWorkerSleepMs string MetaNode delete worker sleep time with millisecond. if 0 for no sleep
-h, --help help for set
--loadFactor string Load Factor
--markDeleteRate string DataNode batch mark delete limit rate. if 0 for no infinity limit
--maxDpCntLimit string Maximum number of dp on each datanode, default 3000, 0 represents setting to default