Please enable JavaScript.

Coggle requires JavaScript to display documents.

LPQ - Live Stream LB (Dependencies (Prod Windows Patching (CAP ? (double…

- - - - How many environements will they want?
      - Zach , devops .. get with Diane
  - - - last bug fixes - 2-3
      - CAP Testing
        
        service running on data servers
        
        new VIPs - until they are targeting
        
        keep us informed
        
        ME
        
        Consider:
        
        Kafka - Datateam
        
        Connecting to existing - MD Headends
  - - - double check.. should be patched
  - - - Live node to .Net
        
        gives us req & matching response
      - ETA - next week
- - - - DAtes
      - Needs to get done.. not a releasable thing
    - - QA01 has node10 already
        
        do we need to change?
      - QA01 - replace build with true GB w/ Node 10
        
        Then - App/QA - TESTING IN QA01
        
        2 servers
        
        What does Justin need? in lists of new environments
        
        TIMING - E&O
        
        RedHat and certificate
        
        ticket open
        
        4 weeks trying to build out capsule
        
        when satellite server was done already
      - get QA01 up to up to good enought state for CIT2
        
        connect to bank PSP
        
        can be done simulatoneou w/ cap
        
        must be done prior to PROD Launch
      - CAP
        
        5 servers
        
        todo: Load testing
        
        sandbox - to test switch over
        
        Prod Switch Testing
        
        see prod release
        
        Cap - test Switch for prod
        
        TEST Roll Back
        
        Do we need an outage ?
        
        Can we gracefully reconnnect ?
        
        Does NOT Requre - project R
        
        no backend services are affected
        
        TEST - During a Ramp Up
        
        Seeing Both pools stay connected & new pool growing
    - - Execution:
        
        seperate set of POOL in F5 (of backend servers)
        
        Switch:
        
        priortize the second pool
        
        while old ones bleeed off
    - - 5-6 Node Boxes
      - Rough Timeline
        
        Month
  - - - current: same boxes different ports
      - Do we need seperate boxes
      - Dev team status:
        
        dev - mostly ready
        
        Need CAP
    - - Separate Boxes
        
        Need Created
        
        another 5
      - Testing
        
        LoadTesting
        
        one test - count of quotes coming through
        
        spot checks quotes have what we need
        
        pretty transparent testing
        
        only difference - split in front end (from pipeline of quotes)
        
        replay - known segment of the market ?
        
        compare old and new
        
        how many number of quotes
        
        have some of this - per what he developed for overnight run
        
        can fine tune
        
        TEST Plan
        
        Prod Switch & rollback
        
        full regression ?
      - timing
        
        CAP - 5 days?
        
        dependent upon Bank tie up
    - - another 5
        
        new VIP
        
        new DNS
      - release - Execution plan
      - Launched all the way to PROD
        
        for EO, vips etd
        
        Final release : CODE Relaese
        
        Rollback - Code rollback
- - - - could provide for E&O major changes, on weekdays not weekend
        
        i.e. take B environment offline from lcients
        
        run change..
        
        run proj R
        
        Switch back to B environemtn
- - - - One VIP and a set of i Rules in F5
  - - - Digest - Nginx config
        
        list of everything
        
        fact - finding discovery