[NETPATH-366] Enrich dynamic network path with NPM domain cache #31685
Conversation
Test changes on VMUse this command from test-infra-definitions to manually test this PR changes on a VM: inv create-vm --pipeline-id=50331531 --os-family=ubuntuNote: This applies to commit 1578aa4 |
Regression DetectorRegression Detector ResultsMetrics dashboard Baseline: be4b703 Optimization Goals: ✅ Improvement(s) detected
|
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | pycheck_lots_of_tags | % cpu utilization | +2.34 | [-1.10, +5.79] | 1 | Logs |
| ➖ | quality_gate_logs | % cpu utilization | +0.77 | [-2.17, +3.72] | 1 | Logs |
| ➖ | otel_to_otel_logs | ingress throughput | +0.51 | [-0.17, +1.19] | 1 | Logs |
| ➖ | quality_gate_idle_all_features | memory utilization | +0.14 | [+0.05, +0.23] | 1 | Logs bounds checks dashboard |
| ➖ | file_to_blackhole_100ms_latency | egress throughput | +0.02 | [-0.75, +0.80] | 1 | Logs |
| ➖ | file_to_blackhole_300ms_latency | egress throughput | +0.01 | [-0.62, +0.64] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api | ingress throughput | +0.00 | [-0.11, +0.11] | 1 | Logs |
| ➖ | tcp_dd_logs_filter_exclude | ingress throughput | -0.00 | [-0.01, +0.01] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency_linear_load | egress throughput | -0.00 | [-0.47, +0.46] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency | egress throughput | -0.01 | [-0.83, +0.81] | 1 | Logs |
| ➖ | quality_gate_idle | memory utilization | -0.11 | [-0.16, -0.07] | 1 | Logs bounds checks dashboard |
| ➖ | file_to_blackhole_500ms_latency | egress throughput | -0.14 | [-0.90, +0.62] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency | egress throughput | -0.17 | [-0.96, +0.62] | 1 | Logs |
| ➖ | tcp_syslog_to_blackhole | ingress throughput | -0.24 | [-0.30, -0.17] | 1 | Logs |
| ➖ | file_tree | memory utilization | -1.21 | [-1.35, -1.06] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api_cpu | % cpu utilization | -1.73 | [-2.46, -1.00] | 1 | Logs |
| ✅ | basic_py_check | % cpu utilization | -5.74 | [-9.45, -2.03] | 1 | Logs |
Bounds Checks: ✅ Passed
| perf | experiment | bounds_check_name | replicates_passed | links |
|---|---|---|---|---|
| ✅ | file_to_blackhole_0ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_1000ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_1000ms_latency_linear_load | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_100ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_100ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_300ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_300ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_500ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_500ms_latency | memory_usage | 10/10 | |
| ✅ | quality_gate_idle | memory_usage | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_idle_all_features | memory_usage | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_logs | lost_bytes | 10/10 | |
| ✅ | quality_gate_logs | memory_usage | 10/10 |
Explanation
Confidence level: 90.00%
Effect size tolerance: |Δ mean %| ≥ 5.00%
Performance changes are noted in the perf column of each table:
- ✅ = significantly better comparison variant performance
- ❌ = significantly worse comparison variant performance
- ➖ = no significant change in performance
A regression test is an A/B test of target performance in a repeatable rig, where "performance" is measured as "comparison variant minus baseline variant" for an optimization goal (e.g., ingress throughput). Due to intrinsic variability in measuring that goal, we can only estimate its mean value for each experiment; we report uncertainty in that value as a 90.00% confidence interval denoted "Δ mean % CI".
For each experiment, we decide whether a change in performance is a "regression" -- a change worth investigating further -- if all of the following criteria are true:
-
Its estimated |Δ mean %| ≥ 5.00%, indicating the change is big enough to merit a closer look.
-
Its 90.00% confidence interval "Δ mean % CI" does not contain zero, indicating that if our statistical model is accurate, there is at least a 90.00% chance there is a difference in performance between baseline and comparison variants.
-
Its configuration does not mark it "erratic".
CI Pass/Fail Decision
✅ Passed. All Quality Gates passed.
- quality_gate_idle_all_features, bounds check memory_usage: 10/10 replicas passed. Gate passed.
- quality_gate_idle, bounds check memory_usage: 10/10 replicas passed. Gate passed.
- quality_gate_logs, bounds check lost_bytes: 10/10 replicas passed. Gate passed.
- quality_gate_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
|
/merge |
Devflow running:
|
What does this PR do?
This PR uses the DNS snooper as the first source of truth for network path tests.
Since the 1 minute TTL of the DNS snooper is shorter than the 5 minute query interval of a path test, the reverse DNS query is added as metadata to the test to avoid expiration issues.
Motivation
Data from the DNS snooper is preferable to reverse DNS, (if it is available) because it represents the actual DNS query the client made to get that IP.
Describe how you validated your changes
Run netpath collector test suite:
go test ./comp/networkpath/npcollector/npcollectorimplI also set up dynamic paths locally by setting
network_path.connections_monitoring.enabled: truein datadog.yaml, but there seems to be an issue inmainwith the system-probe remote interface so the actual traceroute doesn't work. However, I checked thatMetadata.ReverseDNSHostnameis being set properly.Possible Drawbacks / Trade-offs
Worth noting that the DNS snooper uses packet capture as opposed to eBPF, so these data sources could theoretically race (albeit unlikely because it runs within the 30s connection check). Would like someone to review the usage of
ScheduleConns(conns.Conns, conns.Dns)Additional Notes