chore(release): bump 1.11.14 -> 1.11.15 - #671
Merged
Merged
Conversation
Cuts a patch release from current main so the wedge-recovery fix from PR #669 (`fix(daemon): pipe-pool depletion + compile-path wedge recovery`) reaches PyPI. 1.11.14 was published 2026-06-04 23:24 UTC; #669 merged 2026-06-05 07:02 UTC, which means every 1.11.14 user is still seeing the steady-state daemon wedge under heavy parallel-build load. No code changes in this commit — release-auto.yml triggers off the Cargo.toml `workspace.package.version` bump. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This was referenced Jun 5, 2026
zackees
added a commit
to FastLED/FastLED
that referenced
this pull request
Jun 5, 2026
zccache 1.11.15 (zackees/zccache#671) is the first release to ship the daemon-wedge recovery from zackees/zccache#669: - Layer A — wrap path detects daemon wedge via 90s configurable recv timeout, force-kills the wedged daemon, falls back to ephemeral compile so the build completes - Layer B — same wedge detection on ephemeral paths (fast-fail) - Layer C — `IpcListener::accept()` on Windows no longer silently shrinks the named-pipe pool on transient OS errors 1.11.14 users (this repo's current pin) are still seeing the steady- state daemon wedge under heavy parallel-build load (~670 TUs reproduces it reliably on Windows), because the recovery layers from #669 only ship in 1.11.15. Forcing the minimum to 1.11.15 prevents new contributors from picking up the broken-on-Windows 1.11.14 baseline. NOTE: This PR updates `pyproject.toml` only — `uv.lock` regeneration should happen after 1.11.15 lands on PyPI (run `uv lock` then squash into this commit, or land as a follow-up). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
zackees
added a commit
to FastLED/fbuild
that referenced
this pull request
Jun 5, 2026
zccache 1.11.15 (zackees/zccache#671) is the first release to ship the daemon-wedge recovery from zackees/zccache#669: - Layer A — wrap path detects daemon wedge via 90s configurable recv timeout, force-kills the wedged daemon, falls back to ephemeral compile so the build completes - Layer B — same wedge detection on ephemeral paths (fast-fail) - Layer C — `IpcListener::accept()` on Windows no longer silently shrinks the named-pipe pool on transient OS errors 1.11.14 users (this repo's current pin) are still seeing the steady- state daemon wedge under heavy parallel-build load (~670 TUs reproduces it reliably on Windows), because the recovery layers from #669 only ship in 1.11.15. `>=1.11.15` still allows 1.11.x and 1.12.x; it just forbids the known- broken-on-Windows 1.11.14. Sister PR: FastLED/FastLED#2835 — bumps the same pin in FastLED. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
pull Bot
pushed a commit
to manmuqingshan/FastLED
that referenced
this pull request
Jun 6, 2026
zccache 1.11.15 (zackees/zccache#671) is the first release to ship the daemon-wedge recovery from zackees/zccache#669: - Layer A — wrap path detects daemon wedge via 90s configurable recv timeout, force-kills the wedged daemon, falls back to ephemeral compile so the build completes - Layer B — same wedge detection on ephemeral paths (fast-fail) - Layer C — `IpcListener::accept()` on Windows no longer silently shrinks the named-pipe pool on transient OS errors 1.11.14 users (this repo's current pin) are still seeing the steady- state daemon wedge under heavy parallel-build load (~670 TUs reproduces it reliably on Windows), because the recovery layers from FastLED#669 only ship in 1.11.15. Forcing the minimum to 1.11.15 prevents new contributors from picking up the broken-on-Windows 1.11.14 baseline. NOTE: This PR updates `pyproject.toml` only — `uv.lock` regeneration should happen after 1.11.15 lands on PyPI (run `uv lock` then squash into this commit, or land as a follow-up). Co-Authored-By: Claude Opus 4.7 (1M context) <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
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.
Patch release from current main so the wedge-recovery fix from #669 reaches PyPI.
Context
fix(daemon): pipe-pool depletion + compile-path wedge recovery) merged 2026-06-05 07:02 UTC — about 8 hours after 1.11.14 shippedrecv timed out after 300s(1.11.14, Windows) #666 under heavy parallel-build load (~670 TUs), because the recovery layers from fix(daemon): pipe-pool depletion + compile-path wedge recovery (refs #666) #669 ship in this release for the first timeVerification this is a clean release
Cargo.tomlworkspace version (1.11.14 -> 1.11.15) and the 4 matchingCargo.lockentriesworkspace.package.versionbump on push to main, so merging this PR auto-cuts the PyPI releaseWhat 1.11.15 includes vs 1.11.14
Just PR #669:
wedge_recv_timeout(default 90 s), force-kills the wedged daemon, falls back to ephemeral compile so the user's build completesIpcListener::accept()on Windows no longer silently shrinks the pre-created named-pipe pool on transientServerOptions::create()failures. Bounded-backoff retry + emergency-create fallback when the pool ever empties. Replaces a panic with graceful recovery.The back-pressure refactor proposed in #670 is intentionally not in 1.11.15 - that work is incomplete (Phase 1+2 partial; Phases 3-4 TODO) and warrants its own release cycle once Phases 3-4 land.
Co-Authored-By: Claude Opus 4.7 (1M context) noreply@anthropic.com