flink 1.12.0 kubernetes-session部署问题

本人第一次尝试在k8s上部署flink，版本用的是1.12.0，jdk是1.8.0_275，scala是2.12.12，在我的mac机器上安装有minikube单机环境，以下是实验步骤： 

git clone https://github.com/apache/flink-dockercdflink-docker/1.12/scala_2.12-java8-debian  docker build --tag flink:1.12.0-scala_2.12-java8 . 

cd flink-1.12.0  ./bin/kubernetes-session.sh \ -Dkubernetes.container.image=flink:1.12.0-scala_2.12-java8 \ -Dkubernetes.rest-service.exposed.type=NodePort \ -Dtaskmanager.numberOfTaskSlots=2 \ -Dkubernetes.cluster-id=flink-session-cluster 

显示JM启起来了，但无法通过web访问 

2020-12-27 22:08:12,387 INFO org.apache.flink.kubernetes.KubernetesClusterDescriptor [] - Create flink session cluster session001 successfully, JobManager Web Interface: http://192.168.99.100:8081 

通过 kubectl get pods 命令查看到pod一直处理ContainerCreating状态 

NAME READY STATUS RESTARTS AGE 

flink-session-cluster-858bd55dff-bzjk2 0/1 ContainerCreating 0 5m59s 

kubernetes-dashboard-1608509744-6bc8455756-mp47w 1/1 Running 0 6d14h 

于是通过 kubectl describe pod flink-session-cluster-858bd55dff-bzjk2命令查看详细，结果如下： 

Name: flink-session-cluster-858bd55dff-bzjk2 

Namespace: default 

Priority: 0 

Node: minikube/192.168.99.100 

Start Time: Sun, 27 Dec 2020 22:21:56 +0800 

Labels: app=flink-session-cluster 

component=jobmanager 

pod-template-hash=858bd55dff 

type=flink-native-kubernetes 

Annotations:  

Status: Pending 

IP: 172.17.0.4 

IPs: 

IP: 172.17.0.4 

Controlled By: ReplicaSet/flink-session-cluster-858bd55dff 

Containers: 

flink-job-manager: 

Container ID:  

Image: flink:1.12.0-scala_2.12-java8 

Image ID:  

Ports: 8081/TCP, 6123/TCP, 6124/TCP 

Host Ports: 0/TCP, 0/TCP, 0/TCP 

Command: 

/docker-entrypoint.sh 

Args: 

native-k8s 

$JAVA_HOME/bin/java -classpath $FLINK_CLASSPATH -Xmx1073741824 -Xms1073741824 -XX:MaxMetaspaceSize=268435456 -Dlog.file=/opt/flink/log/jobmanager.log -Dlogback.configurationFile=file:/opt/flink/conf/logback-console.xml -Dlog4j.configuration=file:/opt/flink/conf/log4j-console.properties -Dlog4j.configurationFile=file:/opt/flink/conf/log4j-console.properties org.apache.flink.kubernetes.entrypoint.KubernetesSessionClusterEntrypoint -D jobmanager.memory.off-heap.size=134217728b -D jobmanager.memory.jvm-overhead.min=201326592b -D jobmanager.memory.jvm-metaspace.size=268435456b -D jobmanager.memory.heap.size=1073741824b -D jobmanager.memory.jvm-overhead.max=201326592b 

State: Waiting 

Reason: ImagePullBackOff 

Ready: False 

Restart Count: 0 

Limits: 

cpu: 1 

memory: 1600Mi 

Requests: 

cpu: 1 

memory: 1600Mi 

Environment: 

_POD_IP_ADDRESS: (v1:status.podIP) 

HADOOP_CONF_DIR: /opt/hadoop/conf 

Mounts: 

/opt/flink/conf from flink-config-volume (rw) 

/opt/hadoop/conf from hadoop-config-volume (rw) 

/var/run/secrets/kubernetes.io/serviceaccount from default-token-s47ht (ro) 

Conditions: 

Type Status 

Initialized True  

Ready False  

ContainersReady False  

PodScheduled True  

Volumes: 

hadoop-config-volume: 

Type: ConfigMap (a volume populated by a ConfigMap) 

Name: hadoop-config-flink-session-cluster 

Optional: false 

flink-config-volume: 

Type: ConfigMap (a volume populated by a ConfigMap) 

Name: flink-config-flink-session-cluster 

Optional: false 

default-token-s47ht: 

Type: Secret (a volume populated by a Secret) 

SecretName: default-token-s47ht 

Optional: false 

QoS Class: Guaranteed 

Node-Selectors:  

Tolerations: node.kubernetes.io/not-ready:NoExecute op=Exists for 300s 

node.kubernetes.io/unreachable:NoExecute op=Exists for 300s 

Events: 

Type Reason Age From Message 

---- ------ ---- ---- ------- 

Normal Scheduled 21m default-scheduler Successfully assigned default/flink-session-cluster-858bd55dff-bzjk2 to minikube 

Warning FailedMount 21m (x2 over 21m) kubelet MountVolume.SetUp failed for volume "flink-config-volume" : configmap "flink-config-flink-session-cluster" not found 

Warning FailedMount 21m (x2 over 21m) kubelet MountVolume.SetUp failed for volume "hadoop-config-volume" : configmap "hadoop-config-flink-session-cluster" not found 

Normal Pulling 13m (x4 over 21m) kubelet Pulling image "flink:1.12.0-scala_2.12-java8" 

Warning Failed 13m (x4 over 15m) kubelet Failed to pull image "flink:1.12.0-scala_2.12-java8": rpc error: code = Unknown desc = Error response from daemon: manifest for flink:1.12.0-scala_2.12-java8 not found: manifest unknown: manifest unknown 

Normal BackOff 13m (x5 over 15m) kubelet Back-off pulling image "flink:1.12.0-scala_2.12-java8" 

Warning Failed 11m (x5 over 15m) kubelet Error: ErrImagePull 

Warning Failed 100s (x53 over 15m) kubelet Error: ImagePullBackOff 

一开始怀疑本地镜像没有生成，于是通过 docker images 命令查看 

REPOSITORY TAG IMAGE ID CREATED SIZE 

flink 1.12.0-scala_2.12-java8 f7dd9b9e020b 12 hours ago 642MB 

显示镜像的确是存在的，这就奇怪了，为什么从本地pull镜像会失败呢？是哪里有问题了吗？minikube下，如何从本地web访问到k8s上运行的flink集群dashboard呢？ 

第一次用k8s，还请各位指点，谢谢！*来自志愿者整理的flink邮件归档