看看吧
glusterfs数据卷基本操作
一、如何查看Pod使用哪个pvc
查看pod的yaml，找到volume的定义部分，看pvc的名字
kubectl get pod/elasticsearch-0 -o yaml -n efk
输出如下
...
volumeMounts:
- mountPath: /usr/share/elasticsearch/data
name: elasticsearch
- mountPath: /usr/share/elasticsearch/config/elasticsearch.yml
name: efkelasticsearchv300
readOnly: true
subPath: elasticsearch.yml
- mountPath: /var/run/secrets/kubernetes.io/serviceaccount
name: default-token-8pqd4
readOnly: true
...
volumes:
- name: elasticsearch
persistentVolumeClaim:
claimName: elasticsearch-elasticsearch-0
...
可以看到使用的pvc名字就是elasticsearch-elasticsearch-0
二、如何查看pvc对应的pv
查看pvc：elasticsearch-elasticsearch-0
[root@kube-master-1 ~]$ kubectl get pvc/elasticsearch-elasticsearch-0 -n efk
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
elasticsearch-elasticsearch-0 Bound pvc-32ec397e-722b-11ea-983d-525400c29763 30Gi RWO glusterfs-20200316200453-a4cf6f 4h8m
所以pv的名字就是pvc-32ec397e-722b-11ea-983d-525400c29763
三、查看pv对应的glusterfs volume名字
查看pv的yaml，找到path部分
[root@kube-master-1 ~]$ kubectl get pv/pvc-32ec397e-722b-11ea-983d-525400c29763 -o yaml | grep path:
path: vol_00e9a96e51d2b6a911313e5b471360ce
所以该pv对应的glusterfs volume名字为vol_00e9a96e51d2b6a911313e5b471360ce
四、查看某个数据卷的状态
使用gluster volume status vol_name命令
[root@kube-master-1 ~]$ kubectl get pv/pvc-32ec397e-722b-11ea-983d-525400c29763 -o yaml | grep path:
path: vol_00e9a96e51d2b6a911313e5b471360ce
[root@kube-master-1 ~]$ gluster volume status vol_00e9a96e51d2b6a911313e5b471360ce
Status of volume: vol_00e9a96e51d2b6a911313e5b471360ce
Gluster process TCP Port RDMA Port Online Pid
------------------------------------------------------------------------------
Brick 192.168.21.232:/gluster/volumes/glust
erfs/brick_25b598b829cff6930692731ec5c77c3a
/brick 49155 0 Y 1307556
Brick 192.168.21.233:/gluster/volumes/glust
erfs/brick_bae202c07de80c5ac66ebc42cc5651e7
/brick 49155 0 Y 2036501
Brick 192.168.21.231:/gluster/volumes/glust
erfs/brick_7fdfd7c4ba26d1b0cc7544b884b55ca6
/brick 49155 0 Y 3580151
Self-heal Daemon on localhost N/A N/A Y 3580243
Quota Daemon on localhost N/A N/A Y 3580496
Self-heal Daemon on 192.168.21.233 N/A N/A Y 2036582
Quota Daemon on 192.168.21.233 N/A N/A Y 2036672
Self-heal Daemon on 192.168.21.232 N/A N/A Y 1307576
Quota Daemon on 192.168.21.232 N/A N/A Y 1307705
Task Status of Volume vol_00e9a96e51d2b6a911313e5b471360ce
------------------------------------------------------------------------------
There are no active volume tasks
正常情况下，每个 brick 的 Online 状态应该是 Y，如果需要存在 N 的，则是有异常
五、查看gluster节点状态
执行gluster peer status可以看到除了自己之外的其他节点状态，比如
[root@kube-master-1 ~]$ gluster peer
detach probe status
[root@kube-master-1 ~]$ gluster peer
detach probe status
[root@kube-master-1 ~]$ gluster peer status
Number of Peers: 2
Hostname: 192.168.21.232
Uuid: d689f3aa-ac14-45b6-b111-b5668abd4c75
State: Peer in Cluster (Connected)
Hostname: 192.168.21.233
Uuid: fa88a450-da1d-431c-a303-91223f469dae
State: Peer in Cluster (Connected)
peer 正常的状态应该是处于 Peer in Cluster
六、查看所有volume列表
执行gluster volume list查看当前集群存在的所有volume
[root@kube-master-1 ~]$ gluster volume list
compass-stack-heketi-back
compass-stack-heketi-main
vol_00e9a96e51d2b6a911313e5b471360ce
vol_364ca8493826e5ca072b449185be28de
vol_8c978e63897ef085f0b3135dbed32857
vol_965fbd0bce77dc5a8b8b9afb9a8c19ca
vol_a79edbce22e8d8842f3a92056da04c2c
vol_c8b4042091369542838b2ee819d530ba
vol_e9d5fea6f8d7b52ab196c239ca8225b1
vol_f2b4b3c08a99d06bead3a8368482c320
七、查看所有volume的状态
gluster volume status可以查看所有volume状态
[root@kube-master-1 ~]$ gluster volume status
Status of volume: compass-stack-heketi-back
Gluster process TCP Port RDMA Port Online Pid
------------------------------------------------------------------------------
Brick 192.168.21.231:/gluster/heketi/compas
s-stack-heketi-back 49152 0 Y 2353
Brick 192.168.21.232:/gluster/heketi/compas
s-stack-heketi-back 49152 0 Y 1352
Brick 192.168.21.233:/gluster/heketi/compas
s-stack-heketi-back 49152 0 Y 1349
Self-heal Daemon on localhost N/A N/A Y 3580243
Self-heal Daemon on 192.168.21.232 N/A N/A Y 1307576
Self-heal Daemon on 192.168.21.233 N/A N/A Y 2036582
Task Status of Volume compass-stack-heketi-back
------------------------------------------------------------------------------
There are no active volume tasks
Status of volume: compass-stack-heketi-main
Gluster process TCP Port RDMA Port Online Pid
......
八、启动某个数据卷
使用 gluster volume start vol_name 可以启动某个停止的数据卷；如果数据卷处于异常，可以加上 force 通过 gluster volume start vol_name force 强制启动，比如
# gluster volume start vol_6d9e24328c19610bb652537411089be4 force
volume start: vol_6d9e24328c19610bb652537411089be4: success
九、停止某个数据卷
使用 gluster volume stop vol_name 可以停止某个启动的数据卷；如果数据卷处于异常，可以加上 force 通过 gluster volume stop vol_name force 强制停止，比如
# gluster volume stop vol_6d9e24328c19610bb652537411089be4 force
Stopping volume will make its data inaccessible. Do you want to continue? (y/n) y
volume stop: vol_6d9e24328c19610bb652537411089be4: success