I'm working on giving my engineers a "pre-flight" checklist for our production web sites. This is something they can do either after a build, or first thing Monday morning, or even every morning. It's a list of items not often checked like:
The idea is to be proactive and "own" the site rather than wait for something to happen, then check the logs and realize the error has been happening for several weeks.
- production error logs
- database tables looking for frequent or long running queries
- job logs
The idea is to be proactive and "own" the site rather than wait for something to happen, then check the logs and realize the error has been happening for several weeks.