How "Safe" Defaults Blow Up Production

Defaults Matter

Most incidents aren’t exotic edge cases — they’re human choices around configuration — and using defaults. We treat defaults something “obviously working”, therefore safe: then an empty partition key melts one broker and Airflow’s catchup=True quietly queues two years of backfills.

Three moves that hold under pressure:

• If a value is sensible for 80% or more users, make it the default. If no single value fits the majority, do not set a default — force an explicit choice.

• Reject ambiguous sentinels ("", *, null); choose safe strategies (sticky/round‑robin; explicit include lists).

• Make scope visible. Previews (“this will enqueue N runs”), dry-runs by default,

  and hard caps.

In Short: Bad defaults are worse than no defaults — a wrong paved road silently guides everyone over a cliff. The full post has a number of real stories and a copy-ready checklist: link below.

Quick question: Which default has bitten you lately? Share your stories in comments!