Pipeline Trigger fails with 500 error code

[Tom Maxted]

Hi there, I recently followed the examples in https://bitbucket.org/atlassian/trigger-pipeline/src/master/ to create a trigger from one pipeline, to call another.

The trigger successfully calls the second pipeline (and the first pipeline finishes), but then the second pipeline instantly fails with a 'Configuration Error':

Response Summary: HttpResponseSummary{httpStatusCode=500, httpStatusMessage=Internal Server Error, bodyAsString={"code":500,"message":"There was an error processing your request. It has been logged (ID f8fab38297b09cfc)."}} (command VARIABLE_SERVICE_CREATE_PIPELINE_VARIABLE, error key='unexpected.response.body') 

The first bitbucket-pipelines.yml:

options:

  docker: true

pipelines:

  default:

    - step:

        name: Build and Push Docker Image

        script:

          - if [[ -z "$DOCKER_USER" ]]; then echo "DOCKER_USER is not set"; exit 1; fi

          - if [[ -z "$DOCKER_PASSWORD" ]]; then echo "DOCKER_PASSWORD is not set"; exit 1; fi

          - export IMG_NAME=$DOCKER_USER/pipeline-img-test:$BITBUCKET_COMMIT

          - docker build -t $IMG_NAME .

          - docker login -u $DOCKER_USER -p $DOCKER_PASSWORD

          - docker push $IMG_NAME

    - step:

        name: Trigger deployment pipeline

        script:

          - if [[ -z "$BITBUCKET_USERNAME" ]]; then echo "BITBUCKET_USERNAME is not set"; exit 1; fi

          - if [[ -z "$BITBUCKET_APP_PASSWORD" ]]; then echo "BITBUCKET_APP_PASSWORD is not set"; exit 1; fi

          - pipe: atlassian/trigger-pipeline:4.3.1

            variables:

              BITBUCKET_USERNAME: "$BITBUCKET_USERNAME"

              BITBUCKET_APP_PASSWORD: "$BITBUCKET_APP_PASSWORD"

              REPOSITORY: "pipeline-deploy-test"

              CUSTOM_PIPELINE_NAME: "deploy"

              PIPELINE_VARIABLES: >

                [{

                  "key": "IMG_NAME",

                  "value": "$IMG_NAME",

                  "secured": true

                },

                {

                  "key": "COMMIT_SHA",

                  "value": "$BITBUCKET_COMMIT",

                  "secured": true

                }]

              WAIT: "false"

The second bitbucket-pipelines.yml:

pipelines:

  custom:

    deploy:

      # REQUIRED ENVIRONMENT VARIABLES:

      #   PIPELINES_AWS_ACCESS_KEY_ID

      #   PIPELINES_AWS_SECRET_ACCESS_KEY

      #   PRODUCTION_CLUSTER_1_NAME

      #   PRODUCTION_CLUSTER_1_REGION

      #   COMMIT_SHA

      #   IMG_NAME

      - step:

          name: Deploy to production cluster

          script:

            - if [[ -z "$PIPELINES_AWS_ACCESS_KEY_ID" ]]; then echo "PIPELINES_AWS_ACCESS_KEY_ID is not set"; exit 1; fi

            - if [[ -z "$PIPELINES_AWS_SECRET_ACCESS_KEY" ]]; then echo "PIPELINES_AWS_SECRET_ACCESS_KEY is not set"; exit 1; fi

            - if [[ -z "$PRODUCTION_CLUSTER_1_NAME" ]]; then echo "PRODUCTION_CLUSTER_1_NAME is not set"; exit 1; fi

            - if [[ -z "$PRODUCTION_CLUSTER_1_REGION" ]]; then echo "PRODUCTION_CLUSTER_1_REGION is not set"; exit 1; fi

            - if [[ -z "$COMMIT_SHA" ]]; then echo "COMMIT_SHA is not set"; exit 1; fi

            - if [[ -z "$IMG_NAME" ]]; then echo "IMG_NAME is not set"; exit 1; fi

            # setup AWS credentials

            - export AWS_ACCESS_KEY_ID=${PIPELINES_AWS_ACCESS_KEY_ID}

            - export AWS_SECRET_ACCESS_KEY=${PIPELINES_AWS_SECRET_ACCESS_KEY}

            # setup kubeconfig. FUTURE: apply to all production clusters

            - export PRODUCTION_CLUSTER_NAME=${PRODUCTION_CLUSTER_1_NAME}

            - export PRODUCTION_CLUSTER_REGION=${PRODUCTION_CLUSTER_1_REGION}

            - aws eks update-kubeconfig --name ${PRODUCTION_CLUSTER_NAME} --region ${PRODUCTION_CLUSTER_REGION} --kubeconfig ./kubeconfig

            # substitute the config's fields & apply the config

            - cat deployment-template.yaml | envsubst > deployment.yaml

            - kubectl apply -f deployment.yaml --kubeconfig ./kubeconfig

Can anyone see any issues with my config, or is this an internal Atlassian issue?

Thank you for any help,

Tom

[Tom Maxted · Accepted solution]

The problem was with my setup and the error message not helping. I didn't realise that environment variables are not persistent across steps, and I needed to add the line

- export IMG_NAME=$DOCKER_USER/pipeline-img-test:$BITBUCKET_COMMIT

into the step titled 'Trigger deployment pipeline'. This now works.