Repository navigation
Conversation
…faking them (#701) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Refs #701 (ask 4: test-mode knobs for timed behaviour).
What
New section 'Timed windows' in
skills/unhappy-path-testing.md(36 lines) plus a CHANGELOG line. Table of levers (timing constants viamxcli run --local --constant, fast tick plus due check for scheduled events, event-driven waits, window as K x knob, parallel by partition, one real-timings run), with what each cuts and its limit, and four rules that keep the fast mode honest (same code path, guard the knob, prove the window can go red, check the scheduled-events log line).Evidence
Field project with an offline PWA, an edge hub per site, a central app and an ERP stand-in: 31 s backoff, 120 s and 60 s absence windows, 15 min schedule, about 6 min per case.
Estimated gain
Retry and window waits roughly 10 to 15x shorter: one case 6 min to under 1 min, a suite of eight 20 to 40 min to 5 to 8 min. Boot, login, seed unchanged; the real-timings run adds its wait back once. Not run against a live runtime.
Found while reading
testing-shape.mdsaysrun --localstarts with scheduled events off;mxcli run --helpon nightly-20261105 says they run (ALL). Not changed here, needs a check on a runtime. No on-demand scheduled-event trigger was found in mxcli help, so the skill does not assume one.Overlap: #690 (unit-test boots), #698 (time-cost guard) and #700 (pipeline modes) do not touch e2e unhappy timing.
Time cost: none, skill text only.
🤖 Generated with Claude Code