查看某namespace的所有资源
    kubectl get all -n xxx

    pod删除之后，如果service或deployment等上层定义，即使删除了还是重新启动。
    如果批量一次删除没有顺序会删除不干净，还会存在。
    建议先删除deployment等其他，pod最后删除。
    实际操作建议：
    1）定义要操作的namespace变量
    [root@ceph-node2 ~]# NS=XXX
    2）输出删除指令，肉眼校验
    [root@ceph-node2 ~]# for s in kubectl get all -n$NS|grep -v ^NAME|awk '{print $1}';do echo .$s.;echo kubectl delete $s -n $NS;done
    3）执行删除，先删除deployment，rc，service等
    复制刚才的输出指令执行
    4）查看资源占用情况

    一个删除例子：
    [root@ceph-node2 ~]# kubectl get all -nmetallb-system
    NAME READY STATUS RESTARTS AGE
    pod/controller-cd8657667-l4f4m 1/1 Running 0 2d21h
    pod/speaker-9bkjh 1/1 Running 6 429d
    pod/speaker-cm2fr 1/1 Running 2 508d
    pod/speaker-d75q7 1/1 Running 2 508d
    pod/speaker-kd224 1/1 Running 5 508d
    pod/speaker-l288w 1/1 Running 4 508d
    pod/speaker-n8hgt 1/1 Running 3 508d
    pod/speaker-vzpbx 1/1 Running 1 243d
    NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE
    daemonset.apps/speaker 7 7 7 7 7 508d
    NAME READY UP-TO-DATE AVAILABLE AGE
    deployment.apps/controller 1/1 1 1 508d
    NAME DESIRED CURRENT READY AGE
    replicaset.apps/controller-cd8657667 1 1 1 508d
    [root@ceph-node2 ~]# kubectl delete deployment.apps/controller -nmetallb-system
    deployment.apps “controller” deleted
    [root@ceph-node2 ~]# kubectl get all -nmetallb-system
    NAME READY STATUS RESTARTS AGE
    pod/speaker-9bkjh 1/1 Running 6 429d
    pod/speaker-cm2fr 1/1 Running 2 508d
    pod/speaker-d75q7 1/1 Running 2 508d
    pod/speaker-kd224 1/1 Running 5 508d
    pod/speaker-l288w 1/1 Running 4 508d
    pod/speaker-n8hgt 1/1 Running 3 508d
    pod/speaker-vzpbx 1/1 Running 1 243d
    NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE
    daemonset.apps/speaker 7 7 7 7 7 508d
    [root@ceph-node2 ~]# kubectl delete daemonset.apps/speaker -nmetallb-system
    daemonset.apps “speaker” deleted
    [root@ceph-node2 ~]# kubectl get all -nmetallb-system
    No resources found.

    实际只执行删除了两个指令，就删除完了。