Repository navigation
test(live): wait for the images facade however long the CPU takes - #48
Merged
Merged
Conversation
The engine candidate run of master-982-0518b34 failed its macOS live cells, pin and candidate alike: the facade call `POST /v1/images/generations` answers only once the picture exists, and on the macOS runner's CPU that took longer than global fetch waits for a response's headers (undici's headersTimeout, 300 s: `fetch failed`, UND_ERR_HEADERS_TIMEOUT). Linux passed. The test now sends that request through node:http, which sets no such limit; the facade's own 30-minute timeout still applies, and every assertion is the same. Reproduced on its own: against a server answering after 310 s, fetch fails at 301 s with UND_ERR_HEADERS_TIMEOUT and node:http gets the 200. 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Why
The engine candidate run of
master-982-0518b34(run 39623418838) failed on macOS only, pin and candidate alike (core_failure). The failing test wasa real stable-diffusion.cpp engine installs the tree, loads the model, generates, cancels and unloads, after 715 s, withTypeError: fetch failed, caused byHeadersTimeoutError (UND_ERR_HEADERS_TIMEOUT), at the call to the OpenAI facade (test/live/diffusion.test.ts:240). Linux x64 and arm64 passed.POST /v1/images/generationsanswers only once the picture exists. On the macOS runner's CPU that takes longer than 300 s, and globalfetch(undici) stops waiting for a response's headers after 300 s. The facade's own limit is 30 minutes (IMAGES_TIMEOUT_MS).What changes
test/live/diffusion.test.tssends that one request throughnode:http(postJson), which has no headers timeout. The request, the expected status, the body checks and the gallery count are unchanged. The facade's 30-minute timeout and the test's 45-minute timeout still apply.Checked
fetchfails at 301 s withUND_ERR_HEADERS_TIMEOUT, andnode:httpgets the 200 at 310 s.npm run engine:check -- --engine sd-cpp --tag master-982-0518b34on macOS arm64 → PASS. Pinmaster-883-137f740-a36f1b1a5/5, candidate 5/5. Surface 0 break, 6 warn.🤖 Generated with Claude Code