Engineering
The systems and practices engineering uses to diagnose and fix customer-facing issues.
Questions
- Should support teams have access to production logs?Yes, with read-only access and sensible redaction.
- How do you connect support tickets to incidents?Connect your help desk to your incident tool so support sees active incidents before replying, then link every affected ticket to one parent.
- How do you find the root cause of a customer-reported bug?Start from the customer's exact failing request and work backward: find it in logs and traces, identify the failing service and code path, then check what changed there recently.
Terms
- Distributed TracingFollowing one request across every service it touches, with timing for each hop.
- Feature FlagA switch that turns code paths on or off per customer without a deploy.
- HotfixAn urgent fix shipped outside the normal release cycle.
- IncidentAn unplanned disruption or degradation of a service that needs an urgent response.
- Just-in-Time Access (JIT)Temporary, scoped access to production systems, granted only when an investigation needs it.
- LogsTimestamped records of events a system emits as it runs.
- ObservabilityHow well you can understand a system's internal state from the data it emits.
- On-CallThe rotation of engineers responsible for responding to urgent issues.
- PostmortemA blameless review of what went wrong and how to prevent it recurring.
- RegressionA bug where something that used to work stops working after a change.
- Root Cause Analysis (RCA)Finding the underlying reason a problem happened, not just its symptoms.
- Service Level Objective (SLO)An internal reliability target, like 99.9% of requests succeeding.
- UptimeThe share of time a service is available and working, usually shown as a percentage.
- X-Request-IDAn HTTP header carrying a unique ID for one request, so it can be found in logs later.
- YAML ConfigurationHuman-readable config files that define how software and infrastructure behave.