Discussion items

  • Pushed some commits to 2.0 progress
  • Reached the 10000 messages on Slack
  • Add extractors depends on which version of Clowder
  • Improving script to extractor notification to slack
    • Send notification only once a day if everything is fine
    • Send success after failure
Maxwell Burnette
  • Two PRs open
    • Rabbitmq monitoring tool
      • GET file/:id
        • returns 500 if id is wrong (fix)
        • return 403 if api key is wrong
  • Pyclowder
    • use internal IP instead of public
  • 2.0 commits
    • different code to serialize JSON
    • Is there a way to have multiple implicit conversion for different versions of the output
  • Cleaned up PR on JOB_ID in extractors
    • Change data in UI to "No Job ID" for those that don't have it and in popup explain what that means
  • Need to figure out how to do 
  • Needs reviews for the Transformation catalogs
  • Users having permission issue with volume
    • On Linux permissions of folders in container don't match users id in OS
    • Mounting volume from home directory will have different permissions
    • Mark Fredricksen will update docs for users who want to mount external directors Linux
  • Don't use latest tag in docker. Use explicit versions.
    • Docker will not re-pull a container if marked at latest
    • Docker swarm will use the sha as extension on it to make sure that they run the same version on each node
    • Kubernetes always pull down latest
  • We should start building extractors by importing pyclowder using pip and not starting with pyclowder 
    • Update documentation
    • Upgrade old extractors
      • When we move extractors to github?