Q1:Create a Job that run 10 time with 2 jobs running in parallel
主要参数：
# kubectl explain job.spec.parallelism
KIND: Job
VERSION: batch/v1
FIELD: parallelism
DESCRIPTION:
Specifies the maximum desired number of pods the job should run at any
given time. The actual number of pods running in steady state will be less
than this number when ((.spec.completions - .status.successful) <
.spec.parallelism), i.e. when the work left to do is less than max
parallelism. More info:
https://kubernetes.io/docs/concepts/workloads/controllers/jobs-run-to-completion/
# kubectl explain job.spec.completions
KIND: Job
VERSION: batch/v1
FIELD: completions
DESCRIPTION:
Specifies the desired number of successfully finished pods the job should
be run with. Setting to nil means that the success of any pod signals the
success of all pods, and allows parallelism to have any positive value.
Setting to 1 means that parallelism is limited to 1 and the success of that
pod signals the success of the job. More info:
https://kubernetes.io/docs/concepts/workloads/controllers/jobs-run-to-completion/
Job Yaml范例：
apiVersion: batch/v1
kind: Job
metadata:
name: date
spec:
parallelism: 2
completions: 10
template:
spec:
containers:
- name: date
image: busybox:v1
command: ["/bin/sh","-c","date&&/bin/sleep 15"]
restartPolicy: Never
Q2:Find which Pod is taking max CPU
# kubectl top pods -A
NAMESPACE NAME CPU(cores) MEMORY(bytes)
default busybox-httpd-1-7d9fcbc69-kh6jm 0m 0Mi
default busybox-httpd-1-7d9fcbc69-lqzk8 0m 0Mi
default busybox-httpd-2-79bc8f7fcd-gwz6l 0m 0Mi
default busybox-httpd-3-64594f984b-r7nhd 0m 0Mi
default task-pv-pod-2 0m 1Mi
default testpod-h9v4r 0m 0Mi
default testpod-hsfrp 0m 0Mi
default testpod-m9nbn 0m 0Mi
default testpod-nfb6h 0m 0Mi
default testpod-qzf7q 0m 0Mi
default testpod-ttq9k 0m 0Mi
kube-system calico-kube-controllers-8646dd497f-wqzwj 3m 11Mi
kube-system calico-node-5ckj5 13m 86Mi
kube-system calico-node-hklwn 18m 67Mi
Q3:List all PersistentVolumes sorted by their name
# kubectl get pv --sort-by=.metadata.name -A
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
ceph-pool1-lun1 10Gi RWO Retain Bound default/pvc1 manual 16h
ceph-pool1-lun2 15Gi RWO Retain Bound default/pvc2 manual 16h
ceph-pool1-lun3 10Gi RWO Retain Bound default/pvc5 manual 16h
ceph-pool1-lun4 15Gi RWO Retain Bound default/pvc3 manual 16h
ceph-pool1-lun5 10Gi RWO Retain Bound default/pvc4 manual 16h
Q4:Create a NetworkPolicy to allow connect to port 8080 by busybox pod only
1、需要安装支持网络策略的插件
2、使用标注将namespace default设置为缺省隔离
kubectl annotate ns default "net.beta.kubernetes.io/network-policy={\"ingress\": {\"isolation\": \"DefaultDeny\"}}"
3、配置NetworkPolicy
kind: NetworkPolicy
apiVersion: networking.k8s.io/v1
metadata:
name: busybox-access-8080
spec:
podSelector: #匹配目标
matchLabels:
version: test1
ingress: #匹配源
- from:
- podSelector:
matchLabels:
run: busybox
ports:
- protocol: TCP
port: 8080
Q5:fixing broken nodes, see https://kubernetes.io/docs/concepts/architecture/nodes/
Q6:etcd backup
参考资料：
Kubernetes的所有配置数据都存储在ETCD中，发生无法恢复的灾难时，可以使用ETCD备份来恢复所有数据。
ETCD会定期创建快照，但最好还是进行单独备份。建议也备份/etc/etcd/etcd.conf配置文件。
备份方法：
v2备份则可以备份V2和V3的数据。
v3备份只备份v3数据，不备份v2数据。
V2备份，在线执行备份，不中断ETCD群集操作
export ETCDCTL_API=2
alias etcdctl2=' etcdctl \
--cert-file=/etc/kubernetes/pki/etcd/peer.crt \
--key-file=/etc/kubernetes/pki/etcd/peer.key \
--ca-file=/etc/kubernetes/pki/etcd/ca.crt \
--endpoints=https://etcd01:2379,https://etcd02:2379,https://etcd03:2379'
etcdctl2 backup --data-dir /var/lib/etcd/default.etcd/ --backup-dir /backupdir
备份后是多个文件的目录结构
# tree /backupdir/
/backupdir/
└── member
├── snap
│ ├── 000000000000005c-0000000000356803.snap
│ └── db
└── wal
└── 0000000000000000-0000000000000000.wal
备份恢复：
1.停止所有主机上的ETCD
2.清除所有主机上的/var/lib/etcd/member
3.将备份复制到第一个ETCD主机上的/var/lib/etcd/member
4.在第一台ETCD主机上启动ETCD –force-new-cluster
5.将第一台ETCD主机上的正确对等URL设置为节点的IP，而不是127.0.0.1。
6.将下一个主机添加到群集
7.在下一台主机上启动etcd，将 –initial-cluster= "第一台+第二台"
8.重复5和6，直到所有ETCD节点连接
9.正常重启ETCD（使用现有设置）
V3备份，备份后是单个文件etcdbackup.20190612
export ETCDCTL_API=3
alias etcdctl3=' etcdctl \
--cert=/etc/kubernetes/pki/etcd/peer.crt \
--key=/etc/kubernetes/pki/etcd/peer.key \
--cacert=/etc/kubernetes/pki/etcd/ca.crt \
--endpoints=https://etcd01:2379,https://etcd02:2379,https://etcd03:2379'
etcdctl3 snapshot save /etcdbackup.20190612
1.停止所有主机上的ETCD
2.清除所有主机上的/var/lib/etcd/member
3.将备份复制到每一台etcd主机
4.在每一台etcd主机上'source /etc/default/etcd',并执行如下命令
etcdctl3 snapshot restore /etcdbackup.20190612 \
--name $ETCD_NAME --initial-cluster "$ETCD_INITIAL_CLUSTER" \
--initial-cluster-token “$ETCD_INITIAL_CLUSTER_TOKEN” \
--initial-advertise-peer-urls $ETCD_INITIAL_ADVERTISE_PEER_URLS \
--data-dir $ETCD_DATA_DIR
备份方法：
检查etcd是否存在v2的数据，可以看到当前集群并没有v2的数据
#export ETCDCTL_API=2
#alias etcdctl2=' etcdctl \
--cert-file=/etc/kubernetes/pki/etcd/peer.crt \
--key-file=/etc/kubernetes/pki/etcd/peer.key \
--ca-file=/etc/kubernetes/pki/etcd/ca.crt \
--endpoints=https://etcd01:2379,https://etcd02:2379,https://etcd03:2379'
#etcdctl2 ls /
检查etcd是否存在v3的数据，可以看到数据都在v3中
#export ETCDCTL_API=3
#alias etcdctl3=' etcdctl \
--cert=/etc/kubernetes/pki/etcd/peer.crt \
--key=/etc/kubernetes/pki/etcd/peer.key \
--cacert=/etc/kubernetes/pki/etcd/ca.crt \
--endpoints=https://etcd01:2379,https://etcd02:2379,https://etcd03:2379'
#etcdctl3 get / --prefix --keys-only
备份etcd v3的数据
#etcdctl3 snapshot save /etcdbackup.20190612
Q7:You have a Container with a volume mount. Add a init container that creates an empty file in the volume. (only trick is to mount the volume to init-container as well)
https://kubernetes.io/docs/concepts/workloads/pods/init-containers/
apiVersion: v1
kind: Pod
metadata:
name: myapp-pod
labels:
app: myapp
spec:
initContainers:
- name: init-myapp
image: busybox
command: ['sh', '-c', 'echo 1111 > /var/www/html/index.html ']
volumeMounts:
- mountPath: /var/www/html
name: volume1
containers:
- name: myapp
image: busybox
command: ['sh', '-c', '/bin/httpd -f -h /var/www/html']
volumeMounts:
- mountPath: /var/www/html
name: volume1
volumes:
- emptyDir:
name: volume1
Q8:When running a redis key-value store in your pre-production environments many deployments are incoming from CI and leaving behind a lot of stale cache data in redis which is causing test failures. The CI admin has requested that each time a redis key-value-store is deployed in staging that it not persist its data.
Create a pod named non-persistent-redis that specifies a named-volume with name app-cache, and mount path /data/redis. It should launch in the staging namespace and the volume MUST NOT be persistent.
Create a Pod with EmptyDir and in the YAML file add namespace: CI
kubectl create ns ci
apiVersion: v1
kind: Pod
metadata:
name: non-persistent-redis
labels:
app: myapp
namespace: ci
spec:
nodeName: node01
containers:
- name: redis
image: redis
volumeMounts:
- mountPath: /data/redis
name: app-cache
workingDir: /data/redis
env:
- name: data
value: /data/redis/
volumes:
- emptyDir:
name: app-cache
Q9: Find the error message with the string “Some-error message here”.
https://kubernetes.io/docs/concepts/cluster-administration/logging/
see kubectl logs and /var/log for system services
#kubectl logs pod-name
#cat /var/log/messages
Q10:Run a Jenkins Pod on a specified node only.
https://kubernetes.io/docs/tasks/administer-cluster/static-pod/
Create the Pod manifest at the specified location and then edit the systemd service file for kubelet(/etc/systemd/system/kubelet.service) to include --pod-manifest-path=/specified/path. Once done restart the service.
可以在pod.spec.nodeName 指定在哪个node上run
Q11:Create an Ingress resource, Ingress controller and a Service that resolves to cs.rocks.ch.
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: ingress-test
annotations:
nginx.ingress.kubernetes.io/rewrite-target: /
nginx.ingress.kubernetes.io/ssl-redirect: "false"
spec:
rules:
- host: cs.rocks.ch
http:
paths:
- path: /
backend:
serviceName: test1
servicePort: 80