Skip to content

Latest commit

 

History

History
35 lines (25 loc) · 2.87 KB

File metadata and controls

35 lines (25 loc) · 2.87 KB

debugging — Domain Index

Route here for: diagnosing a failure — finding what is wrong and why. Fixing the diagnosed fault belongs to the owning domain.

Match your situation to a "load when" line; load only matching pages.

methodology

Page Load when
reproduce-first A bug is reported or behavior is wrong and you are about to investigate or fix; deciding what to capture when full reproduction is impossible (prod-only, timing-dependent, one-off crash)
isolate-by-bisection A bug reproduces but its location is unknown; it worked before / works in env A but not env B / fails with one input but not another — binary-searching versions (git bisect), code paths, data, or environment diffs
hypothesis-testing You have a suspect cause and are about to "try a fix"; several suspects compete and you must pick what to test next; verifying that a fix that "worked" actually addressed the mechanism
verify-the-fix You believe a bug is fixed and are about to close or ship it; the bug "cannot be reproduced anymore" after changes; a previously fixed bug came back; deciding what must pass (repro re-run, both directions, regression test) and what to clean up before closing

signals

Page Load when
stack-traces Interpreting an exception/stack trace from a log, test failure, or error tracker; wrapped/rethrown causes, NPE on a chained call, async traces missing the origin, minified JS frames
reading-error-messages Any error output you are about to act on — compiler/build failure, runtime error without a stack trace, config/startup error, a wall of errors from one run; deciding which of many errors to fix first (cascades); about to search for a fix or evaluate a fix found online
logs-and-correlation Diagnosing a failure that spans services/components using logs; reconstructing a timeline, following one request via (or without) a correlation id, finding the first error in a cascade

performance

Page Load when
profile-before-optimizing Something is slow (endpoint, job, test suite, page) and you are about to optimize; choosing CPU profiling vs wall-clock tracing; cold vs warm measurement; verifying a speedup (a single slow SQL statement → databases/query-optimization/reading-execution-plans)

concurrency

Page Load when
intermittent-failures A failure happens only sometimes: passes on retry, fails under load, fails only in CI, flaky test, occasional prod-only error — amplifying it into an on-demand reproduction