Conversation
### What changes were proposed in this PR? Bind channel markers to the input port resolved by their control handler. Lifecycle callbacks use that port instead of the last data tuple's port. Reject invalid markers before entering the executor. ### Any related issues, documentation, discussions? Related to apache#8319. ### How was this PR tested? From amber, on Python 3.12: ```text python -m pytest --tb=short -q src/test/python/core/architecture/handlers/control/test_channel_marker_handlers.py src/test/python/core/runnables/test_data_processor.py src/test/python/core/runnables/test_main_loop.py 66 passed python -m ruff check src/main/python src/test/python All checks passed python -m ruff format --check src/main/python src/test/python 215 files already formatted ``` The new handler-to-processor cases cover an empty port 1 finishing after a tuple on port 0, and port 1 starting before any tuple. Both fail when marker routing is reverted to the current data port. Invalid-marker tests failed before adding the type guard. The full run, `python -m pytest --tb=no -q`, finished with 1320 passed, 1 xfailed, and 1 failed. The failure is the unchanged `test_rest_catalog_round_trip` integration test, which targets a Lakekeeper service at localhost:8181. The full suite is not green. Excluding integration tests, `python -m pytest --tb=no -q -m "not integration"` passes: 1320 passed, 1 deselected, 1 xfailed. ### Was this PR authored or co-authored using generative AI tooling? Generated-by: OpenAI Codex (backported from commit 471e53c) Co-authored-by: Meng Wang <mengw15@uci.edu>
Contributor
Automated Reviewer SuggestionsBased on the
|
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## release/v1.3 #8621 +/- ##
===============================================
Coverage 94.06% 94.06%
Complexity 4808 4808
===============================================
Files 1194 1194
Lines 48601 48611 +10
Branches 5860 5862 +2
===============================================
+ Hits 45718 45728 +10
Misses 1429 1429
Partials 1454 1454
*This pull request uses carry forward flags. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
Contributor
|
| config | throughput | MB/s | latency | max Δ latest / 7d | |
|---|---|---|---|---|---|
| 🔴 | bs=10 sw=10 sl=64 | 363 | 0.222 | 25,903/35,881/35,881 us | 🔴 +9.9% / 🔴 +127.1% |
| 🟢 | bs=100 sw=10 sl=64 | 800 | 0.488 | 124,941/141,801/141,801 us | 🟢 -6.4% / 🔴 +30.4% |
| ⚪ | bs=1000 sw=10 sl=64 | 913 | 0.557 | 1,088,132/1,193,053/1,193,053 us | ⚪ within ±5% / 🔴 +15.0% |
Baseline details
Latest main 5036097 from same runner
| config | metric | PR | latest main | 7d avg | Δ latest | Δ 7d |
|---|---|---|---|---|---|---|
| bs=10 sw=10 sl=64 | throughput | 363 tuples/sec | 391 tuples/sec | 767.43 tuples/sec | -7.2% | -52.7% |
| bs=10 sw=10 sl=64 | MB/s | 0.222 MB/s | 0.238 MB/s | 0.468 MB/s | -6.7% | -52.6% |
| bs=10 sw=10 sl=64 | p50 | 25,903 us | 23,566 us | 12,880 us | +9.9% | +101.1% |
| bs=10 sw=10 sl=64 | p95 | 35,881 us | 36,553 us | 15,801 us | -1.8% | +127.1% |
| bs=10 sw=10 sl=64 | p99 | 35,881 us | 36,553 us | 19,767 us | -1.8% | +81.5% |
| bs=100 sw=10 sl=64 | throughput | 800 tuples/sec | 817 tuples/sec | 988.73 tuples/sec | -2.1% | -19.1% |
| bs=100 sw=10 sl=64 | MB/s | 0.488 MB/s | 0.499 MB/s | 0.603 MB/s | -2.2% | -19.1% |
| bs=100 sw=10 sl=64 | p50 | 124,941 us | 121,423 us | 102,684 us | +2.9% | +21.7% |
| bs=100 sw=10 sl=64 | p95 | 141,801 us | 151,454 us | 108,712 us | -6.4% | +30.4% |
| bs=100 sw=10 sl=64 | p99 | 141,801 us | 151,454 us | 118,731 us | -6.4% | +19.4% |
| bs=1000 sw=10 sl=64 | throughput | 913 tuples/sec | 922 tuples/sec | 1,022 tuples/sec | -1.0% | -10.7% |
| bs=1000 sw=10 sl=64 | MB/s | 0.557 MB/s | 0.563 MB/s | 0.624 MB/s | -1.1% | -10.7% |
| bs=1000 sw=10 sl=64 | p50 | 1,088,132 us | 1,087,929 us | 999,086 us | +0.0% | +8.9% |
| bs=1000 sw=10 sl=64 | p95 | 1,193,053 us | 1,137,583 us | 1,037,033 us | +4.9% | +15.0% |
| bs=1000 sw=10 sl=64 | p99 | 1,193,053 us | 1,137,583 us | 1,066,123 us | +4.9% | +11.9% |
Raw CSV
config_idx,batch_size,schema_width,string_len,num_batches,total_ms,total_tuples,total_bytes,tuples_per_sec,mb_per_sec,lat_p50_us,lat_p95_us,lat_p99_us
0,10,10,64,20,550.58,200,128000,363,0.222,25902.83,35880.75,35880.75
1,100,10,64,20,2499.20,2000,1280000,800,0.488,124941.47,141801.45,141801.45
2,1000,10,64,20,21907.96,20000,12800000,913,0.557,1088132.17,1193052.57,1193052.57
mengw15
deleted the
backport/8432-preserve-channel-marker-ports-v1.3
branch
September 21, 2026 20:44
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
4 participants
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.
What changes were proposed in this PR?
Backport of #8432 to
release/v1.3: a clean cherry-pick of its squash commit, no adaptations — the backport commit reuses the squash message and author, as the automated fast path would. See #8432 for the change itself (channel markers bind to the port resolved by their control handler instead of the last data tuple's port, and invalid markers are rejected before entering the executor).Opened manually by the v1.3 release manager: the automated fast path cherry-picked this cleanly and then pushed it straight to
release/v1.3, where the Merge Queue ruleset rejected the push (GH013, run 35187408300). The Actions-app bypass meant to unblock that path (#8379) was never created — asfyaml rejects anIntegrationbypass actor — and the failing job's notification 403s for want ofpull-requests: write, so the loss left neither a backport PR nor a comment on #8432. See #8377.Source: 471e53c
Any related issues, documentation, discussions?
Backport of #8432. Related to #8319.
How was this PR tested?
The change is identical to #8432, which carries 235 lines of new pytest coverage (
test_channel_marker_handlers.py,test_data_processor.py,test_main_loop.py); the backport tree is verified byte-identical to cherry-picking the squash commit ontorelease/v1.3, and release-branch CI runs the full matrix on this PR. The two branch-level blockers that used to make pyamber jobs red here are already merged: #8549 (LICENSE-binary-python drift) and #8550 (RustFS as the default object store).Was this PR authored or co-authored using generative AI tooling?
Yes. Generated-by: Claude Code (claude-opus-5)