martialonline/workflow-status

Workflow Status Action

workflow-status's Issues

Workflow Fails After Repository Ownership changed

A job is failing after the ownership of the repository was moved from one organization to another.

jq: error (at <stdin>:4): Cannot iterate over null (null)
Error: Process completed with exit code 5.

The relevant job that is failing:

  report:
    name: Report to Slack
    needs: [
      build_android,
      build_ios,
      quality
    ]
    runs-on: ubuntu-latest

    steps:
      - uses: martialonline/workflow-status@v4
        id: check
      - name: Slack Notification
        uses: rtCamp/action-slack-notify@v2
        env:
          SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}

The Job was working totally fine before transferring ownership and is working fine in other repositories across different organizations. Is there potentially a necessary setting not correct in the new organisation or potentially some cached things trying to point to the old repository?

Any help would be appreciated

More output from the action

Run martialonline/workflow-status@v4
  env:
    APP_DIRECTORY: packages/app
Run url="${GITHUB_API_URL}/repos"
  url="${GITHUB_API_URL}/repos"
  repo="${GITHUB_REPOSITORY}"
  run_id="${GITHUB_RUN_ID}"
  token="***"
  
  failure=$(curl -s -H "Authorization: token ${token}" "${url}/${repo}/actions/runs/${run_id}/jobs" | \
  jq -r '.jobs[] | select(.status == "completed" and .conclusion == "failure").conclusion' | \
  wc -l)
  
  cancelled=$(curl -s -H "Authorization: token ${token}" "${url}/${repo}/actions/runs/${run_id}/jobs" | \
  jq -r '.jobs[] | select(.status == "completed" and .conclusion == "cancelled").conclusion' | \
  wc -l)
  
  if [ "${failure}" -gt 0 ]; then
    status="failure"
  elif [ "${cancelled}" -gt 0 ]; then
    status="cancelled"
  else 
    status="success"
  fi
  
  echo "status=${status}" >> $GITHUB_OUTPUT
  shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0}
  env:
    APP_DIRECTORY: packages/creator
jq: error (at <stdin>:[4](https://github.com/{user}/{repo}/actions/runs/{runId}/job/{jobId}#step:3:5)): Cannot iterate over null (null)
Error: Process completed with exit code [5](https://github.com/{user}/{repo}/actions/runs/{runId}/job/{jobId}#step:3:6).

jq errors on no jobs found

It doesn't look like there are jobs found with the curl command and then an error occurs when jq tries to parse the jobs array.

Run url="${GITHUB_API_URL}/repos"
  url="${GITHUB_API_URL}/repos"
  repo="${GITHUB_REPOSITORY}"
  run_id="${GITHUB_RUN_ID}"
  
  failure=$(curl -s "${url}/${repo}/actions/runs/${run_id}/jobs" | \
  jq -r '.jobs[] | select(.status == "completed" and .conclusion == "failure").conclusion' | \
  wc -l)
  
  cancelled=$(curl -s "${url}/${repo}/actions/runs/${run_id}/jobs" | \
  jq -r '.jobs[] | select(.status == "completed" and .conclusion == "cancelled").conclusion' | \
  wc -l)
  
  if [ "${failure}" -gt 0 ]; then
    status="failure"
  elif [ "${cancelled}" -gt 0 ]; then
    status="cancelled"
  else 
    status="success"
  fi
  
  echo "::set-output name=status::${status}"
  shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0}

jq: error (at <stdin>:4): Cannot iterate over null (null)
Error: Process completed with exit code 5.

dfgtgt

why list the last job as dependency?

GitHub action will run jobs in parallel without needs

why we just need to use the last job as dependency?