1/16/2018 Follow Up Discussion

Owned by Scott Seabolt
Last updated: Jan 18, 2018
3 min read


Minutes:

  • Discussion is open to the ONAP community.
  • Aaron Hay followed up with Wind River (Stephen Gooch) on the availability of adding the VM necessary to conduct the test.
  • The team still has the capability of creating the environment.  APPC's existing env got blown away but the stack is still there.
  • For Beijing, APPC will use the Kubernetes env, Hector Anapan may have an env with Kubernetes.
    • AI:  Hector Anapan will setup a call for OOM and Kubernetes. 
    • APPC team has the ability to setup up to 10 instances.
      • Allocation can be done dynamically.
      • With the Resiliency/Kubernetes env APPC will rquire more than 10 instances.
      • AI:  Aaron Hay to ask Stephen to bump APPC allocation up to 20.
    • Automated failover in a single site for Resiliency (Germane to Stability in that the env will be shared)
      • APPC needs a robust HealthCheck to determine if the cluster is healthy.
  • What collaboration is needed to ensure lab isn't touched during the testing period.
    • Need to closely collaborate with WindRiver to make sure they know when APPC is running the 72 hour testing.
      • WindRiver is working on improvements to their notification system.
    • ONAP Discuss subscribe list is where Stephen typically posts any downtime for the lab
  • Follow up with APPC team on possible VNF options for the Configure and ConfigModify.
    • One possibility is to setup APPC to support vFirewall/PacketGen to work with LCM Provider instead of the legacy API Provider.
    • The vFirewall will be available.  Need a more thorough eval on how to adapt to LCM.
      • AI:  Patrick Brady to research if pgstreams can be supported through LCM.
  • Layout plan for establishing data in A&AI in regards to the test VNFs/VMs.
    • Due TBD.
  • Document the Test Approach including the design, environment and expected results.
    • Due TBD.
  • APPC Benchmarking and Jmeter
    • ONAP component development will do the initial testing prior to the Integration team conducting their tests.
    • Integration testing uses Jmeter to test since it is standard.
    • Jmeter has good REST capabilities.
    • The team will need to do a little work to get DMaaP integrated into the Jmeter test.
      • Randa Maher asked:
        • Does Jmeter profile memory to see if you have a memory leak?
        • APPC needs to define a set of metrics we will collect on this stability run, like CPU usage, etc..
        • AI Ryan Young - Investigate capabilities of Jmeter and provide a list of capabilities.
    • APPC Metrics logging
    • AI:  Randa Maher to create EPIC for testing 1/17/2018 (Complete)
      • AI:  Scott Seabolt to create stories and tasks for stability testing.
    • Where does the code/config items go for this test?
      • For now this code will go into the Deployment repo.


Action Items to be completed by our next call 1/23/2018:

  • AI:  Hector Anapan will setup a call for OOM and Kubernetes. 
  • AI:  Aaron Hay to ask Stephen to bump APPC allocation up to 20.
  • AI:  Patrick Brady to research if pgstreams can be supported through LCM.
  • AI:  Ryan Young - Investigate capabilities of Jmeter and provide a list of capabilities.
  • AI:  Scott Seabolt to follow up with Taka as he is familiar with Elk. 
  • AI:  Scott Seabolt - Determine how much logging space would be required?
  • AI:  Patrick Brady, what is SONAR Regression test.
    • What is it and how can it be used for the Stability Testing.
  • AI:  Randa Maher to create EPIC for testing 1/17/2018 (Complete)
    • AI:  Scott Seabolt to create stories and tasks for stability testing.
  • AI:  Scott Seabolt to setup a follow up call for next week.