Pod状态CrashLoopBackOff无法启动：initContainer正常主容器异常求助

问题背景 #

原本通过ConfigMap注入单个URL作为环境变量时，DaemonSet的Pod可正常启动。现在因存在多个URL，改为通过initContainer筛选有效URL并写入/tmp/ip.env，再在主容器中设置环境变量，但initContainer执行无报错，主容器却陷入CrashLoopBackOff状态。

原正常配置 #

apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: {{ .Chart.Name }}
  annotations:
    reloader.stakater.com/auto: "true"
spec:
  selector:
    matchLabels:
      some-service: {{ .Chart.Name }}
  template: #required fields
    spec:
      hostNetwork: yes
      containers:
       - name: {{ .Chart.Name }}
         image: "{{ .Values.image.repository }}:{{ .Values.image.tag | default .Chart.AppVersion }}"
         imagePullPolicy: {{ .Values.image.pullPolicy }}
         env:
         - name: HOST_URL
           valueFrom:
            configMapKeyRef:
              name: some-config
              key: some-url

修改后的配置 #

apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: {{ .Chart.Name }}
  annotations:
    reloader.stakater.com/auto: "true"
spec:
  selector:
    matchLabels:
      some-service: {{ .Chart.Name }}
  template: #required fields
    spec:
      hostNetwork: yes
      containers:
       - name: {{ .Chart.Name }}
         image: "{{ .Values.image.repository }}:{{ .Values.image.tag | default .Chart.AppVersion }}"
         imagePullPolicy: {{ .Values.image.pullPolicy }}
         command:
         - sh
         - -c
         - env HOST_URL=$(cat /tmp/ip.env)
         volumeMounts:
         - name: workdir
           mountPath: /tmp
      initContainers:
       - name: wait-for-url
         image: "{{ .Values.image.repository }}:{{ .Values.image.tag | default .Chart.AppVersion }}"
         volumeMounts:
          - mountPath: "/tmp"
            name: workdir
         env:
          - name: multipleUrls
            valueFrom:
             configMapKeyRef:
               name: some-config
               key: multipleUrls
         command: 
           - '/bin/bash'
           - '-c'
           - |
             
             echo "Actual Data: $multipleUrls"
             :
             :
             Logic to find the valid url and store it in the /tmp/ip.env file
             :
             :
             echo "Accessible Url is:$hosturl"
             echo $hosturl> /tmp/ip.env
             
             if [ -s /tmp/ip.env ]; then 
                echo "HOSTURL is: " `cat /tmp/ip.env` 
                exit 0;
             else 
                sleep 10 
                exit 1;
             fi
      volumes:
       - name: workdir
         emptyDir: {}

kubectl describe pod 输出 #

Name:         some-name-nckfs
Namespace:    default
Priority:     0
Node:         10.22.44.169/10.22.44.169
Start Time:   Sat, 29 Oct 2022 15:11:11 +0000
Labels:       controller-revision-hash=55ff545b89
              pod-template-generation=32
Annotations:  <none>
Status:       Running
IP:           10.22.44.169
IPs:
  IP:           10.22.44.169
Controlled By:  DaemonSet/some-something
Init Containers:
  wait-for-hosturl:
    Container ID:  docker://62e3b20b6eeeca4f37e78c02fa8d68b482bde1644fc30a0ed94f788c35d3cd86
    Image ID:      docker://sha256:805f01431e534713510826a63dab75873d45aa96c6a9614f2c9bfbeab800a774
    Port:          <none>
    Host Port:     <none>
    Command:
      /bin/bash
      -c

      ::
      ::

    State:          Terminated
      Reason:       Completed
      Exit Code:    0
      Started:      Sat, 29 Oct 2022 15:11:12 +0000
      Finished:     Sat, 29 Oct 2022 15:11:12 +0000
    Ready:          True
    Restart Count:  0
    Environment:
      multipleUrls:  <set to the key 'multipleUrls' of config map 'some-config'>  Optional: false
    Mounts:
      /tmp from workdir (rw)
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-vks68 (ro)
Containers:
  some-something:
    Container ID:  docker://1fd6d60a70a6e2c10339809e4bb3084738aee7500dc0087973801c1c3a5a497a
    Image ID:      docker://sha256:805f01431e534713510826a63dab75873d45aa96c6a9614f2c9bfbeab800a774
    Port:          9091/TCP
    Host Port:     9091/TCP
    Command:
      sh
      -c
      env HOST_URL=$(cat /tmp/ip.env)
    State:          Waiting
      Reason:       CrashLoopBackOff
    Last State:     Terminated
      Reason:       Completed
      Exit Code:    0
      Started:      Sat, 29 Oct 2022 15:57:42 +0000
    Finished:     Sat, 29 Oct 2022 15:57:42 +0000
    Ready:          False
    Restart Count:  14
    Environment:
    Mounts:
      /tmp from workdir (rw)
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-vks68 (ro)
Conditions:
  Type              Status
  Initialized       True
  Ready             False
  ContainersReady   False
  PodScheduled      True
Volumes:
  workdir:
    Type:       EmptyDir (a temporary directory that shares a pod's lifetime)
    Medium:
    SizeLimit:  <unset>
  default-token-vks68:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  default-token-vks68
    Optional:    false
QoS Class:       BestEffort
Node-Selectors:  <none>
Tolerations:     node.kubernetes.io/disk-pressure:NoSchedule op=Exists
                 node.kubernetes.io/memory-pressure:NoSchedule op=Exists
                 node.kubernetes.io/network-unavailable:NoSchedule op=Exists
                 node.kubernetes.io/not-ready:NoExecute op=Exists
                 node.kubernetes.io/pid-pressure:NoSchedule op=Exists
                 node.kubernetes.io/unreachable:NoExecute op=Exists
                 node.kubernetes.io/unschedulable:NoSchedule op=Exists
Events:
  Type     Reason     Age                    From               Message
  ----     ------     ----                   ----               -------
  Normal   Created    49m                    kubelet            Created container wait-for-hosturl
  Normal   Started    49m                    kubelet            Started container wait-for-hosturl
  Normal   Created    48m (x5 over 49m)      kubelet            Created container some-something
  Normal   Started    48m (x5 over 49m)      kubelet            Started container some-something
  Warning  BackOff    4m57s (x212 over 49m)  kubelet            Back-off restarting failed container

kubectl logs 输出 #

KUBERNETES_SERVICE_PORT=443
KUBERNETES_PORT=tcp://10.43.0.1:443
HOSTNAME=something
HOME=/root
KUBERNETES_PORT_443_TCP_ADDR=10.43.0.1
PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
KUBERNETES_PORT_443_TCP_PORT=443
KUBERNETES_PORT_443_TCP_PROTO=tcp
KUBERNETES_PORT_443_TCP=tcp://10.43.0.1:443
KUBERNETES_SERVICE_PORT_HTTPS=443
KUBERNETES_SERVICE_HOST=10.43.0.1
JAVA_HOME=/usr/lib/jvm/java-11-openjdk-amd64
PWD=/
HOST_URL=10.22.44.168

原因分析 #

主容器的command仅执行了env HOST_URL=$(cat /tmp/ip.env)，该命令执行完成后立即退出（返回码0），而Kubernetes容器需要长期运行的进程才能保持存活状态。原配置未指定command，会使用镜像默认的启动命令维持容器运行；修改后覆盖了默认命令，导致容器执行完env命令就终止，kubelet不断重启容器，最终出现CrashLoopBackOff。

解决方案 #

需要在设置环境变量后，执行镜像原本的启动命令，确保容器长期运行。有两种常见方式：

方式1：直接拼接命令 #

假设镜像默认启动命令为java -jar app.jar（需根据实际镜像调整），修改主容器的command：

command:
- sh
- -c
- "export HOST_URL=$(cat /tmp/ip.env) && exec java -jar app.jar"

使用exec让原启动进程成为容器的PID 1，确保信号能正确传递。

方式2：通过环境变量文件加载 #

先修改initContainer的写入逻辑，将环境变量以KEY=VALUE格式写入文件：

echo "HOST_URL=$hosturl" > /tmp/ip.env

然后主容器通过source加载环境变量并执行原命令：

command:
- sh
- -c
- "source /tmp/ip.env && exec <原镜像启动命令>"

内容的提问来源于stack exchange，提问作者Nagaraju Chitimilla