使用生命周期挂钩运行脚本在kubernetes中无法正常运行

时间：2019-09-25 04:40:26

标签： postgresql kubernetes minikube kubernetes-pod lifecycle-hook

我是kubernetes的新手。我正在使用minikube学习kubernetes。我有一种情况，我必须在PostgreSQL中配置fdw（外部表概念）。

我正在尝试在kubernetes中实现此fdw概念。为此，我有两个Pod Meta Pod，用于创建外部表和服务器，Data Pod用于创建实际表。要创建一个meta pod，您应该有一个包含以下脚本的Docker映像。

Script.sh 

#!/bin/bash
psql -d metap -U papu -c "CREATE EXTENSION if not exists postgres_fdw;"
psql -d metap -U papu -c "CREATE SERVER if not exists dataserver FOREIGN DATA WRAPPER postgres_fdw OPTIONS (host 'datap.default.svc.cluster.local', dbname 'datap', port '5432');"
psql -d metap -U papu -c "CREATE USER MAPPING if not exists FOR ais SERVER dataserver OPTIONS (user 'papu', password 'papu');"
psql -d metap -U papu -c "CREATE FOREIGN TABLE if not exists dream (id integer, val text) SERVER dataserver OPTIONS (schema_name 'public', table_name 'dream');"

两个Pod的yaml文件如下所示。

p-config.yaml 

apiVersion: v1
kind: ConfigMap
metadata:
  name: p-config
  labels:
    app: post
data:
  POSTGRES_DB: datap
  POSTGRES_USER: papu
  POSTGRES_PASSWORD: papu

datap.yaml 

apiVersion: v1
kind: Service
metadata:
  name: datap
  labels:
    app: datap
spec:
  ports:
  - port: 5432
    name: datap
  clusterIP: None
  selector:
    app: datap
---
apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: datap
spec:
  serviceName: "datap"
  replicas: 1
  selector:
    matchLabels:
      app: datap
  template:
    metadata:
      labels:
        app: datap
    spec:
      containers:
      - name: datap
        image: postgres:latest
        envFrom:
          - configMapRef:
              name: p-config
        ports:
        - containerPort: 5432
          name: datap
        volumeMounts:
        - name: datap 
          mountPath: /var/lib/postgresql/data
          subPath: datap
  volumeClaimTemplates:
    - metadata:
        name: datap 
      spec:
        accessModes: [ "ReadWriteOnce" ]
        resources:
          requests:
            storage: 2Gi

我的用例是我已经启动并运行了数据容器，现在我必须通过在meta pod中运行脚本来动态创建外部表。

为此，我正在使用生命周期挂钩。当我运行此配置时， 将创建外部表并建立fdw连接 。 但是日志显示生命周期挂钩未执行。这是错误吗？还是我的配置有问题？

$ kubectl describe pod metap-0

Name:           metap-0
Namespace:      default
Priority:       0
Node:           minikube/10.0.2.15
Start Time:     Fri, 20 Sep 2019 15:50:41 +0530
Labels:         app=metap
                controller-revision-hash=metap-648ddb5465
                statefulset.kubernetes.io/pod-name=metap-0
Annotations:    <none>
Status:         Running
IP:             172.17.0.10
Controlled By:  StatefulSet/metap
Containers:
  metap:
    Container ID:  
    Image:          <script containing image >:latest
    Port:           5432/TCP
    Host Port:      0/TCP
    State:          Running
      Started:      Fri, 20 Sep 2019 15:51:29 +0530
    Last State:     Terminated
      Reason:       Completed
      Exit Code:    0
      Started:      Fri, 20 Sep 2019 15:51:14 +0530
      Finished:     Fri, 20 Sep 2019 15:51:15 +0530
    Ready:          True
    Restart Count:  2
    Environment:    <none>
    Mounts:
      /var/lib/postgresql/data from metap (rw,path="metap")
      /var/run/secrets/kubernetes.io/serviceaccount from default-token
Conditions:
  Type              Status
  Initialized       True 
  Ready             True 
  ContainersReady   True 
  PodScheduled      True 
Volumes:
  mpostgredb:
    Type:       PersistentVolumeClaim (a reference to a PersistentVolumeClaim in the same namespace)
    ClaimName:  metap-metap-0
    ReadOnly:   false
  default-token-r2ncm:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  default-token-r2ncm
    Optional:    false
QoS Class:       BestEffort
Node-Selectors:  <none>
Tolerations:     node.kubernetes.io/not-ready:NoExecute for 300s
                 node.kubernetes.io/unreachable:NoExecute for 300s
Events:
  Type     Reason               Age                From               Message
  ----     ------               ----               ----               -------
  Warning  FailedScheduling     71s (x2 over 71s)  default-scheduler  pod has unbound immediate PersistentVolumeClaims
  Normal   Scheduled            69s                default-scheduler  Successfully assigned default/metap-0 to minikube
  Warning  FailedPostStartHook  67s                kubelet, minikube  Exec lifecycle hook ([/bin/sh -c script.sh]) for Container "metap" in Pod "metap-0_default(6a367766-cd7e-4bab-826a-908e33622bcf)" failed - error: command '/bin/sh -c script.sh' exited with 2: psql: could not connect to server: No such file or directory
           Is the server running locally and accepting
           connections on Unix domain socket "/var/run/postgresql/.s.PGSQL.5432"?
psql: could not connect to server: No such file or directory
  Is the server running locally and accepting
  connections on Unix domain socket "/var/run/postgresql/.s.PGSQL.5432"?
psql: could not connect to server: No such file or directory
  Is the server running locally and accepting
  connections on Unix domain socket "/var/run/postgresql/.s.PGSQL.5432"?
psql: could not connect to server: No such file or directory
  Is the server running locally and accepting
  connections on Unix domain socket "/var/run/postgresql/.s.PGSQL.5432"?
, message: "psql: could not connect to server: No such file or directory\n\tIs the server running locally and accepting\n\tconnections on Unix domain socket \"/var/run/postgresql/.s.PGSQL.5432\"?\npsql: could not connect to server: No such file or directory\n\tIs the server running locally and accepting\n\tconnections on Unix domain socket \"/var/run/postgresql/.s.PGSQL.5432\"?\npsql: could not connect to server: No such file or directory\n\tIs the server running locally and accepting\n\tconnections on Unix domain socket \"/var/run/postgresql/.s.PGSQL.5432\"?\npsql: could not connect to server: No such file or directory\n\tIs the server running locally and accepting\n\tconnections on Unix domain socket \"/var/run/postgresql/.s.PGSQL.5432\"?\n"
  Warning  FailedPostStartHook  35s  kubelet, minikube  Exec lifecycle hook ([/bin/sh -c script.sh]) for Container "metap" in Pod "metap-0_default(6a367766-cd7e-4bab-826a-908e33622bcf)" failed - error: command '/bin/sh -c script.sh' exited with 1: psql: FATAL:  the database system is starting up
psql: FATAL:  the database system is starting up
ERROR:  server "dataserver" does not exist
ERROR:  server "dataserver" does not exist
, message: "psql: FATAL:  the database system is starting up\npsql: FATAL:  the database system is starting up\nERROR:  server \"dataserver\" does not exist\nERROR:  server \"dataserver\" does not exist\n"
  Normal   Killing  35s (x2 over 67s)  kubelet, minikube  FailedPostStartHook
  Warning  BackOff  33s (x2 over 34s)  kubelet, minikube  Back-off restarting failed container
  Normal   Created  21s (x3 over 68s)  kubelet, minikube  Created container metap
  Normal   Started  21s (x3 over 68s)  kubelet, minikube  Started container metap
  Normal   Pulling  21s (x3 over 68s)  kubelet, minikube  Pulling image " <script containing image >:latest"
  Normal   Pulled   21s (x3 over 68s)  kubelet, minikube  Successfully pulled image " <script containing image >:latest"



$ kubectl logs metap-0 
2019-09-20 10:21:29.500 UTC [1] LOG:  listening on IPv4 address "0.0.0.0", port 5432
2019-09-20 10:21:29.500 UTC [1] LOG:  listening on IPv6 address "::", port 5432
2019-09-20 10:21:29.502 UTC [1] LOG:  listening on Unix socket "/var/run/postgresql/.s.PGSQL.5432"
2019-09-20 10:21:29.514 UTC [22] LOG:  database system was shut down at 2019-09-20 10:21:15 UTC
2019-09-20 10:21:29.518 UTC [1] LOG:  database system is ready to accept connections

kubectl版本 

Client Version: version.Info{Major:"1", Minor:"15", GitVersion:"v1.15.3", GitCommit:"2d3c76f9091b6bec110a5e63777c332469e0cba2", GitTreeState:"clean", BuildDate:"2019-08-19T11:13:54Z", GoVersion:"go1.12.9", Compiler:"gc", Platform:"linux/amd64"}
Server Version: version.Info{Major:"1", Minor:"15", GitVersion:"v1.15.2", GitCommit:"f6278300bebbb750328ac16ee6dd3aa7d3549568", GitTreeState:"clean", BuildDate:"2019-08-05T09:15:22Z", GoVersion:"go1.12.5", Compiler:"gc", Platform:"linux/amd64"}

* Minikube版本* 

minikube version: v1.3.1
commit: ca60a424ce69a4d79f502650199ca2b52f29e631

2 个答案:

答案 0 :(得分：1)

使用初始化容器，您可以检查第二个容器是否已经启动并正在运行，然后运行脚本

    spec:
      initContainers:
      - name: check-second-ready
        image: postgres
        command: ['sh', '-c', 
          'until pg_isready -h connection/url/to/second/conatiner -p 5432; 
          do echo waiting for database; sleep 2; done;']
      containers:
          first container config...
------

答案 1 :(得分：0)

我转载了您的问题，发现问题出在生命周期挂钩配置中。

该脚本是在数据库准备就绪之前执行的，这就是它引发错误的原因： command '/bin/sh -c script.sh' exited with 2: psql: could not connect to server: No such file or directory,

failed - error: command '/bin/sh -c script.sh' exited with 1: psql: FATAL: the database system is starting up 您应该在“脚本开始处添加sleep，以使数据库有一定的准备时间。

示例：

#!/bin/bash
sleep 5
psql -d metap -U papu -c "CREATE EXTENSION if not exists postgres_fdw;"
psql -d metap -U papu -c "CREATE SERVER if not exists dataserver FOREIGN DATA WRAPPER postgres_fdw OPTIONS (host 'datap.default.svc.cluster.local', dbname 'datap', port '5432');"
psql -d metap -U papu -c "CREATE USER MAPPING if not exists FOR ais SERVER dataserver OPTIONS (user 'papu', password 'papu');"
psql -d metap -U papu -c "CREATE FOREIGN TABLE if not exists dream (id integer, val text) SERVER dataserver OPTIONS (schema_name 'public', table_name 'dream');
相关问题
最新问题