tuesday at work: I want to get the DNS automation for these NLBs hammered out before we roll to production, because having to make manual changes when the NLBs need to be re-created, is going to prolong an incident at some point
thursday at work: someone deletes one of the two staging kubernetes clusters, requiring all the NLBs to be re-created
friday at work: "hey so it looks like mail isn't transiting correctly in that staging env, can you look into it?" / looks into it / none of the NLBs exist because their names were hardcoded instead of being DNS-automated / we look directly into the camera