• Set expectations for customers that network unavailability will happen and when it will happen.
  • Give regular status updates at certain times during the window of downtime.
    • “Home Assistant is back up” for example.
  • Set a time window for the client and stick to it.
    • Regardless of the situation, if you don’t complete it within a certain time, pull the ripcord and rollback the changes.
      • What is your rollback plan.
  • Know ahead of time what you are going to test and have multiple test cases.
    • Laptop can connect, but desktop is not able to connect to the network for example.
  • Be prepared for any aftermath.
    • Changes cause unanticipated problems down the line.
  • Set aside time for the “oops” moments.
  • Communication path only with stakeholders, to avoid other parties blaming the changes made.
  • Check firewall rules after the network changes are made.
    • Firewalls very quickly see the new changes and can tell if changing from a /24 to a /21 IP scheme.

Updated: