Attendees

NameRolePresent

Mensah, Jacob

NIH/NCI [C]x
Hartman, Phil 
NIH/NCI [C]x
Wei, Lu
NIH/NCI [C]x
Pan, Sue
NIH/NCI [C] 
Stockwell, MichaelNIH/NCI [C] 
Bauer, Scott  Mayox
Stancl, Craig
Mayox

Endle,  Cory

Mayox

Links

Agenda

 

Discuss Mayo’s current use of Docker and determine if/how this can be supported at NCI.

Discussion Points:
  • Overview of what LeEVS team currently has running locally.
    • LexEVS has multiple components (separate GIT repositories)
    • Current Docker environment configured to build and run tests.
      • Containers communicate with each other for testing purposes
    • Jenkins configures the containers.
    • Test results are collated for review.
  • Discussion of Goals
    • CI on NIH server (same thing that we do locally)
      • NIH has Jenkins already, so not a problem.
  • Hardware Requirements for Docker
    • Minimally, a larger VM
      • Memory, CPU Cores
    • Currently running on one physical server.
  • NIH is looking into Docker Swarm/Cluster.

Decision Points:

  • Configure one physical server
    • Blade Server
  • Will need to work with team to determine configuration and set up of environment.

Discuss the possibility of creating reusable Docker containers based on NCI’s technology stack and placing them on NCI’s Nexus server where we can pull into our project.
Discussion Points:
  • NIH reusable containers to be on Nexus.
  • NIH base image built
    • This is an ongoing project to test the base line.
  • Repository is ready for pilot (NOT production)
    • LexEVS usecase is very good for pilot.
    • May want to look at getting LexEVS into a docker cluster.
  • Users will be able to manage own projects
    • This will still allow the LexEVS team to have access.

Decision Points:

  • NCI team is ready to take us on as a pilot.
  • As part of this, use evaluate and determine needs for docker cluster.
  • Provide Jenkins configuration details to Phil. May be able to push docker container to NCI docker hub. Optionally, we can provide screen share and share configuration.
  • NCI will need to potentially get the publish results plugin.
Discuss the possibility of deploying our application to the different NCI tiers as Docker containers.
Discussion Points:
  • This is NCI goal.
  • Currently working with other projects with goal of getting to production.

Decision Points:

The use of Jenkin’s for running LexEVS builds and tests
Discussion Points:
  • Toky and LexEVS team worked together to get Jenkins configuration set up on NCI server.
    • Partially configured, but not completed before she left.
  • Phil said they can preserve what Toky completed.
  • Phil can import our configuration and replace what Toky did.
  • This is a separate Jenkins.

Decision Points:

  • LexEVS team to export the Jenkins configuration for Phil.
Next Steps

Next Steps:

  • Jacob main touchpoint
  • Configuration details via email with Wei and Phil.
  • Continue to provide updates during the Wednesday systems call.
  • Additional meetings as needed.
  • Respond to TODos listed above.
  • Create a ticket for Jacob: Setup Docker and Jenkins for LexEVS.