3 ms·

This article is too simplistic to take seriously. Basically it says: "staging is expensive (but $100k !?) and is not the same as prod because data (how many tes

This article is too simplistic to take seriously. Basically it says: "staging is expensive (but $100k !?) and is not the same as prod because data (how many tests fail because of this?), so I'll use feature flags and test in prod".

blntechie 6y ago
Data consistency between prod and stage is also simple to solve for use cases where it’s not an issue. For example, we have a job which restores each day’s prod backup to staging.
- devxpy 6y ago
  I do this, easy to implement, and it works pretty well - dropdb $PGDATABASE || true createdb -T template0 $PGDATABASE pg_dump $DUMP_DATABASE | psql -q $PGDATABASE
  - devxpy 6y ago
    whoa, am i doing something wrong? please share thoughts