1. Develop a service checklist:
a. What services are provided by the server?
b. Who are the customers of each service?
c. What software package(s) provide which service?
2. Verify that each software package will work with the new OS, or plan a software upgrade path.
3. For each service, develop a test to verify that it is working.
4. Write a back-out plan, with specific triggers.
5. Select a maintenance window.
6. Announce the upgrade as appropriate.
7. Execute the tests developed earlier to make sure that they are still valid.
8. Lock out users.
9. Do the upgrade with someone watching/helping (mentoring).
10. Repeat all the tests developed earlier. Follow the usual debugging process.
11. If tests fail, or other events occur that trigger the back-out plan, execute the back-out plan.
12. Let users back in.
13. Communicate completion/back-out to the customers.
14. Analyze what went right and what didn’t; modify the checklist to reflect the experience.
And you thought it was all about using an install disc, didn’t you? Let’s look a little more closely at each step.