fix(replay): stop a failed network body read from breaking the host fetch - #4711
Merged
dustinbyrne merged 5 commits intoSep 1, 2026
Merged
Conversation
…etch On Safari, canceling a reader whose underlying transfer already died rejects with `TypeError: Load failed`. `_tryReadBodyStreaming` discarded that promise with `void reader?.cancel()`, so the rejection escaped as an uncaught error and broke the function's never-reject contract. Attach a no-op rejection handler to the cancel() promise, and guard the response body read in the fetch patch so a body-read failure can never propagate into a request the page already got a good response for. Generated-By: PostHog Desktop Task-Id: 327ac857-0f25-45f1-867d-b3c6f1297a7e
Contributor
Author
🦔 PostHog Review reviewed this pull requestNothing worth raising this time, so here's a calming picture instead:
|
Contributor
Replay incident risk checkThis diff touches code involved in past incidents. This is a heads-up, not a verdict: read the matched sections of INCIDENTS.md and answer their review questions before merging. For a judgment on whether this diff has the same failure mode, run the |
Contributor
Contributor
posthog-js Compliance ReportDate: 2026-09-01 17:37:31 UTC ✅ All Tests Passed!26/26 tests passed Capture Tests✅ 26/26 tests passed View Details
|
Contributor
|
Size Change: +855 B (0%) Total Size: 20.8 MB 📦 View Changed
ℹ️ View Unchanged
|
Contributor
Author
Replay incident risk — review questions answeredThe bot matched Class 1 (version skew) and Class 3 (network primitive wrappers). The diff adds only defensive error handling; it does not change request construction, body forwarding, any persisted state, or any cross-boundary signature. Class 1 — version skew across the lazy-load boundary
Class 3 — network primitive wrappers
Standing-policy flag for a human reviewerClass 3 records a standing policy: no AI-generated PRs in the wrappers. This PR is AI-generated and edits |
dustinbyrne
approved these changes
Sep 1, 2026
…-body-read-a8b381
dustinbyrne
marked this pull request as ready for review
September 1, 2026 17:21
dustinbyrne
requested review from
TueHaulund,
arnohillen,
fasyy612 and
ksvat
and removed request for
a team
September 1, 2026 17:21
dustinbyrne
enabled auto-merge (squash)
September 1, 2026 17:22
dustinbyrne
deleted the
posthog-self-driving/fixreplay-stop-a-failed-body-read-a8b381
branch
September 1, 2026 17:36
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
1 participant
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.
Problem
TypeError: Load failedfrom the replay network capture plugin — a failed response body read escaping into the page's ownfetch()call._tryReadBodyStreamingpromises it is guaranteed to resolve, never reject. Every read path honors that except one:void reader?.cancel()discarded the promisecancel()returns, and the surroundingtry/catchcannot catch an async rejection. On Safari, canceling a reader whose transfer already died rejects withTypeError: Load failed— the exact type, message, and browser observed.catch, so a body-read failure could reject the patchedfetchand break a request the page had already received a good response for. Commitf07e241bexists specifically so network capture never breaks the host page; this path defeated that.streamNetworkBody— but the failure class is the reason to fix it.Changes
cancel()promise, so a reader that rejects on cancel can never escape the never-reject helper.fetch._tryReadBodyStreamingresolves normally and raises no unhandled rejection whencancel()rejects with the Safari error.Release info Sub-libraries affected
Libraries affected
Checklist
If releasing new changes
pnpm changesetto generate a changeset file🤖 Agent context
Autonomy: Fully autonomous
void reader?.cancel()line; I confirmeddone()callscancel()on every settle (including success), so the rejected promise floats regardless of read outcome. I verified the regression test fails without the fix (unhandledTypeError) and passes with it.try/catchthat swallows and logs, so it needed no change.Created with PostHog Desktop from this inbox report.