fix(perf): ASTFAssociation rules= + OFFLINE TRex profile validation#107
Merged
Conversation
… port= + add OFFLINE TRex-profile validation to end the one-error-per-EC2-run cycle. Run 28549496895 (post-#106) hit 'unexpected keyword argument port'. Rather than guess again, I stood up the TRex control-plane profile module locally (its ArgVerify is pure Python — no NIC/server) and introspected the real API: - ASTFAssociation(rules=...) — NOT port=; the port goes on ASTFAssociationRule. - ASTFTCPClientTemplate DOES accept port= (my #106 guess there was fine). Fixed the association and confirmed build_tcp_profile() builds a valid ASTFProfile for all payload sizes OFFLINE in <1s. scripts/perf-tests/test/validate_tcp_profile.sh reproduces this locally (sparse-clones the TRex control-plane, shims cgi/zmq for py3.13/arm64, builds the profile). This catches ASTF-API errors instantly instead of via ~30-min EC2 runs — the fix for the 'testing at deployment is expensive' problem. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Synthetic Performance Results (run)Commit: ✅ synthetic UDP socket bound to 10.0.0.1:9000 (MAC: 02:00:00:00:00:01) Synthetic UDP Performance ResultsMeasures framework overhead: sync IPv4 Baseline
IPv6
IPv6 vs IPv4 Comparison (sync path)
IPv4 avg sync/async ratio: 0.9x, worst: 1.0x | IPv6 vs IPv4 worst ratio: 1.19x (OK)
|
Synthetic Performance Results — Graviton (run)Commit: ✅ synthetic UDP socket bound to 10.0.0.1:9000 (MAC: 02:00:00:00:00:01) Synthetic UDP Performance ResultsMeasures framework overhead: sync IPv4 Baseline
IPv6
IPv6 vs IPv4 Comparison (sync path)
IPv4 avg sync/async ratio: 1.0x, worst: 1.1x | IPv6 vs IPv4 worst ratio: 1.31x (OK)
|
[CI] Stage: DeployInfrastructure ready.
|
[CI] Stage: DeployInfrastructure ready.
|
[CI] Stage: SummaryAll tests PASSED. ARP seeding: kernel /proc/net/arp (automatic)
|
1 similar comment
[CI] Stage: SummaryAll tests PASSED. ARP seeding: kernel /proc/net/arp (automatic)
|
✅ Integration Tests Passed — Graviton (run)Branch: Test Results
Application Logs (last 20 lines)receiver-echo-server.log sender-echo-server.log sender-test-client.log |
✅ Integration Tests Passed (Run 28551110519)Branch: Test Results
Application Logs (last 20 lines)receiver-echo-server.log sender-echo-server.log sender-test-client.log receiver-test-client-iperf.log sender-test-client-iperf.log Full Application Logs (last 200 lines each)receiver-echo-server.logsender-echo-server.logsender-test-client.logreceiver-test-client-iperf.logsender-test-client-iperf.log
|
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.
Follow-up to #106. Run 28549496895 hit
unexpected keyword argument 'port'. This is the 5th TRex-profile API error found one-at-a-time on EC2 — so instead of guessing again, I built the offline validator we discussed.The fix (ground-truthed, not guessed)
Stood up the TRex control-plane profile module locally (its
ArgVerifyis pure Python — no NIC/server) and introspected the real API:ASTFAssociation(rules=...)— does not takeport=; the port goes onASTFAssociationRule(port=...).ASTFTCPClientTemplatedoes acceptport=(my fix(perf): TRex ASTFIPGen.glob type + client ip_gen (profile build) #106 change there was fine).Fixed the association, and confirmed
build_tcp_profile()builds a validASTFProfilefor all payload sizes offline in <1s.The tool (ends the whack-a-mole)
scripts/perf-tests/test/validate_tcp_profile.sh— sparse-clones the TRex control-plane, shimscgi(py3.13) +zmq(arm64), and builds the profile locally. Any future ASTF-API mistake is caught in <1s instead of a ~30-min EC2 run. Directly addresses the "testing at deployment is expensive" concern. Follow-up: wire it into CI as a per-PR gate.Merge → the next EC2 run should finally get past profile-build to real CPS/throughput (or a genuine runtime issue, no longer a profile-build error).
🤖 Generated with Claude Code