Skip to content

Time Sensitive Triage

Isaac Zarsky edited this page Jun 29, 2018 · 1 revision
  1. Characterize the risk. This should be informed by, but not dictated by, opinions of others. Use your own sense.
  2. Perform the minimum viable triage. Rather than trying to fix the underlying issue, focus on stemming any damage. The goal is for you, the person dealing with the time pressure, not to have to embark on a long process of fixing the fundamental issue alone. For example, disable the next day's prod deploy, or roll back to a known good state.
  3. Let the team help you. You don't have to decide unilaterally what to do past these initial steps. If it can stay broken overnight, let it stay broken overnight.
Clone this wiki locally