Liveblog - The accidental sysadmin.

Tags: liveblog, DevOps,

liveblog: the accidental sysadmin (starting at 28 minutes into the session)

No staging environment - he refers to the pot roast story in Steve Mc Connels’ code complete

You can’t do everything, but specialization is also painful, so what to do?

  • In a startup there are not enough people to specialize
  • Agree interfaces
  • Some specialization, some more depth of knowledge

They are going to hire their first operations person, because there are too many issues there at the moment.

(In another startup when my co-cto suggested hiring a tester, that was where I drew the line - the problems we had could be solved without increasing specialisation at that point in time. I would hesitate less on operations).

  • Instrument all the things
  • Put instrumentation in from day one

They use cucumber to specify their infrastructure code.

Subscribe to our RSS feed