docker 容器内包含 ssh 命令的 shell 脚本失败并显示 - ssh: not found

Question

我有一个在 docker 容器上运行的 spring boot java 应用程序，它尝试运行 shell 脚本。shell 脚本有一个 ssh 命令，运行时出现以下错误

    2020-08-12 09:22:29.425  INFO 1 --- [io-11013-exec-1] b.n.i.s.d.e.service.EmrManagerService    : Executing spark submit, calling shell script: /tmp/temp843155675494688636.sh 172.29.199.15
    2020-08-12 09:22:29.434 DEBUG 1 --- [io-11013-exec-1] b.n.i.s.d.e.service.EmrManagerService    : Starting Input Stream:
    2020-08-12 09:22:29.435  INFO 1 --- [io-11013-exec-1] b.n.i.s.d.e.service.EmrManagerService    : #1 arg: 172.29.199.15
    2020-08-12 09:22:29.436  INFO 1 --- [io-11013-exec-1] b.n.i.s.d.e.service.EmrManagerService    : Exist Value127
    2020-08-12 09:22:29.436 ERROR 1 --- [io-11013-exec-1] b.n.i.s.d.e.service.EmrManagerService    : Starting Error Stream:
    2020-08-12 09:22:29.436 ERROR 1 --- [io-11013-exec-1] b.n.i.s.d.e.service.EmrManagerService    : 
    /tmp/temp843155675494688636.sh: line         5: ssh: not found

当直接运行 jar 而不是作为 docker 容器运行时，相同的代码可以正常工作。这与 docker 容器中无法识别 ssh 有关吗？

    shell script - 
    #!/bin/bash
    
    
    echo "#1 arg:" $1
    ssh -i /home/dnaidaasd/aws-oneid-idaas-2020Q2.pem -oStrictHostKeyChecking=no hadoop@$1 '/etc/alternatives/jre/bin/java -Xmx1000m -server \ 
    -XX:OnOutOfMemoryError="kill -9 %p" -cp "/usr/share/aws/emr/instance \
    -controller/lib/*" -Dhadoop.log.dir=/mnt/var/log/hadoop/steps/s-100-120 \
    -Dhadoop.log.file=syslog -Dhadoop.home.dir=/usr/lib/hadoop \ 
    -Dhadoop.id.str=hadoop -Dhadoop.root.logger=INFO,DRFA -Djava.library.path=:/usr/lib/hadoop-lzo/lib/native:/usr/lib/hadoop/lib/native \
    -Dhadoop.policy.file=hadoop-policy.xml -Djava.net.preferIPv4Stack=true \
    -Djava.io.tmpdir=/mnt/var/lib/hadoop/steps/s-14611-353/tmp \
    -Dhadoop.security.logger=INFO,NullAppender \
    -Dsun.net.inetaddr.ttl=30 \
    org.apache.hadoop.util.RunJar /var/lib/aws/emr/step-runner/hadoop-jars/command-runner.jar spark-submit \
    --conf spark.hadoop.mapred.output.compress=true \
    --conf spark.hadoop.mapred.output.compression.codec=org.apache.hadoop.io.compress.GzipCodec \
    --class biz.neustar.idaas.services.dataprofile.ProfileMain \
    --name IdaasProfile --conf spark.dynamicAllocation.enabled=true \
    --conf spark.executor.instances=2 --conf spark.driver.memory=8G \
    --conf spark.executor.memory=4G --conf spark.executor.cores=1 \
    --conf spark.sql.catalogImplementation=hive \
    --jars s3://oneid-idaas-dev-us-east-1/dev/emr/TestIdaasProfile/spark-core_2.11-2.4.5.jar,s3://oneid-idaas-dev-us-east-1/dev/emr/TestIdaasProfile/spark-sql_2.11-2.4.5.jar,s3://oneid-idaas-dev-us-east-1/dev/emr/TestIdaasProfile/spark-mllib_2.11-2.4.5.jar,s3://oneid-idaas-dev-us-east-1/dev/emr/TestIdaasProfile/jackson-module-scala_2.11-2.6.7.1.jar,s3://oneid-idaas-dev-us-east-1/dev/emr/TestIdaasProfile/jackson-databind-2.6.7.jar  s3://oneid-idaas-dev-us-east-1/dev/emr/TestIdaasProfile/data-profile-14.0.jar' \
    $2 $3 $4

这个 shell 脚本被称为 -

public void executeSparkSubmit(String masterNodeIp, String pathToScript, String input_hive_table, String s3_output_path,  String output_hive_table ) throws IOException, InterruptedException, DataProfileServiceException {

        log.info("Executing spark submit, calling shell script: " + pathToScript + " " + masterNodeIp);
        ProcessBuilder pb = new ProcessBuilder("sh", pathToScript, masterNodeIp, input_hive_table, s3_output_path, output_hive_table);
        Process pr = pb.start();

内容Dockerfile是：

FROM openjdk:8-jdk-alpine
ADD ./data-profile-provider/build/libs/data-profile-provider-203.2.0-SNAPSHOT.jar data-profile.jar
EXPOSE 11013
ENTRYPOINT ["java", "-jar", "data-profile.jar", "application.properties"]

Answer 1

正如我所怀疑的 - 您的图像是基于 Alpine 的，并且 Alpine 默认情况下没有安装 SSH 客户端。

更正Dockerfile：

FROM openjdk:8-jdk-alpine
RUN apk add --no-cache openssh-client
ADD ./data-profile-provider/build/libs/data-profile-provider-203.2.0-SNAPSHOT.jar data-profile.jar
EXPOSE 11013
ENTRYPOINT ["java", "-jar", "data-profile.jar", "application.properties"]

编辑：我忘了补充一点，Alpine 也没有 Bash。幸运的是，您的应用程序调用您的脚本sh scriptname.sh- 否则您会收到bash: not found错误。