On Jan 7, 2019, at 4:05, Radek Matěj ***@***.***> wrote: Fixes #386. I've spent enormous time investigating the conditions when backpressure occurs, yet I still have a lot of unanswered questions. Although I hate uncertainties, the formula I introduced is a verified hypothesis. I'm well prepared and open to any discussions about this topic. :) I would appreciate any feedback on the code, especially tests. You can view, comment on, or merge this pull request online at: #563 Commit Summary Test clone timeout Mock response instead test-specific URL Test resolved clone Create chai helper for testing promise timeout Move chai timeout helper to separate file Lower timeout value to speed up test Return mock response URL from mockResponse Add highWaterMark param to clone method Update information about clone parameter Add more extensive highWaterMark documentation Test described highWaterMark formula File Changes A CLONE-HIGHWATERMARK.md (108) M LIMITS.md (11) M src/body.js (10) M src/response.js (5) A test/chai-timeout.js (17) M test/server.js (14) M test/test.js (67) Patch Links: https://github.com/bitinn/node-fetch/pull/563.patch https://github.com/bitinn/node-fetch/pull/563.diff — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.