Populate Network Path metadata for CNM connections

[AlexandreYang]

Summary

• document the CNM -> Network Path enrichment design in Confluence
• return per-connection npmodel.NetworkPath values from NPCollector scheduling paths
• populate Connection.NetworkPath.HasTest on evaluated CNM connections using the collector result
• bump github.com/DataDog/agent-payload/v5 to the companion payload schema with Connection.NetworkPath.HasTest

Companion PR

• agent-payload schema PR: Add Network Path connection metadata agent-payload#495

Health Platform issue state note

The agent-payload bump also regenerates the Health Platform proto definitions. In that regenerated code, ISSUE_STATE_NEW and ISSUE_STATE_ONGOING are deprecated; ISSUE_STATE_ACTIVE and ISSUE_STATE_RESOLVED are the canonical values going forward.

This PR updates Agent Health Platform code and tests to emit and assert ISSUE_STATE_ACTIVE for active issues, while still loading legacy persisted new/ongoing states as active. This is not part of the CNM Network Path runtime behavior, but it is required so the Agent continues to pass staticcheck after consuming the updated agent-payload module.

Tests

• git diff --check
• GOPROXY=https://proxy.golang.org,direct dda inv go.tidy
• GOPROXY=https://proxy.golang.org,direct dda inv test --targets=./pkg/process/checks --extra-args='-run TestScheduleNetworkPathSetsConnectionMetadata'
• GOPROXY=https://proxy.golang.org,direct dda inv test --targets=./comp/networkpath/npcollector/impl --extra-args='-run Test_npCollectorImpl_ScheduleNetworkPathTests'

Note: parallel dda inv test attempts for npcollector can fail before running tests due to a local dda temporary uv-* cleanup race; the same command passed when rerun by itself.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 7bc40e5460

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Avoid collecting decisions for ignored scheduling calls

When Network Path scheduling runs from the current production call sites I checked (comp/netflow/flowaggregator/aggregator.go:204, pkg/network/sender/sender_linux.go:324, and pkg/process/checks/net.go:185), the returned decisions are discarded. This append still allocates a backing slice for every yielded connection; in the NetFlow path it happens once per flow, so enabling network_path.netflow_monitoring adds an allocation on every flow even though no enrichment is populated yet. Please keep the old no-result scheduling path or make decision collection opt-in until a caller consumes it.

Useful? React with 👍 / 👎.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 11ed778918

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Plumb Network Path decisions through direct sender

Setting NetworkPath here only enriches the process-agent connections check path. In environments with network_config.direct_send: true, ConnectionsCheck.IsEnabled returns false (pkg/process/checks/net.go:143) and the Linux direct sender is the producer; that path still calls ScheduleNetworkPathTests and discards the returned decisions (pkg/network/sender/sender_linux.go:324), while encodeConnection never writes a network-path field. Those direct-send CNM deployments will schedule tests but continue sending connections without network_path.has_test, so please carry the decisions into the direct sender encoder as well.

Useful? React with 👍 / 👎.

[datadog-prod-us1-4]

🎯 Code Coverage (details)
• Patch Coverage: 72.73%
• Overall Coverage: 51.10% (-4.59%)

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 5de868d | Docs | Datadog PR Page | Give us feedback!}

[dd-octo-sts]

Files inventory check summary

File checks results against ancestor 1d93f4e9:

Results for datadog-agent_7.82.0~devel.git.442.5de868d.pipeline.121530583-1_amd64.deb:

No change detected

[dd-octo-sts]

Static quality checks

✅ Please find below the results from static quality gates
Comparison made with ancestor 1d93f4e
📊 Static Quality Gates Dashboard
🔗 SQG Job

Successful checks

Info

	Quality gate	Change	Size (prev → curr → max)
✅	agent_deb_amd64	+64.82 KiB (0.01% increase, -0.64% of buffer)	748.299 → 748.362 → 758.200
✅	agent_deb_amd64_fips	+68.79 KiB (0.01% increase, -1.12% of buffer)	703.865 → 703.932 → 709.840
✅	agent_heroku_amd64	+20.25 KiB (0.01% increase, -0.30% of buffer)	308.717 → 308.736 → 315.230
✅	agent_msi	+64.5 KiB (0.01% increase, -0.49% of buffer)	617.659 → 617.722 → 630.600
✅	agent_rpm_amd64	+64.82 KiB (0.01% increase, -0.64% of buffer)	748.283 → 748.346 → 758.170
✅	agent_rpm_amd64_fips	+68.79 KiB (0.01% increase, -1.12% of buffer)	703.849 → 703.916 → 709.840
✅	agent_rpm_arm64	+57.26 KiB (0.01% increase, -0.96% of buffer)	723.822 → 723.878 → 729.660
✅	agent_rpm_arm64_fips	+61.23 KiB (0.01% increase, -1.02% of buffer)	683.023 → 683.083 → 688.860
✅	agent_suse_amd64	+64.82 KiB (0.01% increase, -0.64% of buffer)	748.283 → 748.346 → 758.170
✅	agent_suse_amd64_fips	+68.79 KiB (0.01% increase, -1.12% of buffer)	703.849 → 703.916 → 709.840
✅	agent_suse_arm64	+57.26 KiB (0.01% increase, -0.96% of buffer)	723.822 → 723.878 → 729.660
✅	agent_suse_arm64_fips	+61.23 KiB (0.01% increase, -1.02% of buffer)	683.023 → 683.083 → 688.860
✅	docker_agent_amd64	+64.79 KiB (0.01% increase, -0.95% of buffer)	807.121 → 807.184 → 813.790
✅	docker_agent_arm64	+57.24 KiB (0.01% increase, -0.76% of buffer)	807.677 → 807.733 → 815.030
✅	docker_agent_jmx_amd64	+64.79 KiB (0.01% increase, -0.97% of buffer)	998.018 → 998.082 → 1004.550
✅	docker_agent_jmx_arm64	+57.24 KiB (0.01% increase, -0.75% of buffer)	987.227 → 987.283 → 994.710
✅	docker_cluster_agent_amd64	+16.15 KiB (0.01% increase, -0.81% of buffer)	208.530 → 208.546 → 210.470
✅	iot_agent_deb_amd64	+24.22 KiB (0.05% increase, -3.45% of buffer)	45.695 → 45.719 → 46.380
✅	iot_agent_deb_arm64	+20.22 KiB (0.05% increase, -1.51% of buffer)	42.414 → 42.434 → 43.720
✅	iot_agent_deb_armhf	+24.16 KiB (0.05% increase, -3.10% of buffer)	43.199 → 43.223 → 43.960
✅	iot_agent_rpm_amd64	+24.22 KiB (0.05% increase, -3.46% of buffer)	45.696 → 45.720 → 46.380
✅	iot_agent_suse_amd64	+24.22 KiB (0.05% increase, -3.45% of buffer)	45.695 → 45.718 → 46.380

11 successful checks with minimal change (< 2 KiB)

	Quality gate	Current Size
✅	docker_cluster_agent_arm64	221.753 MiB
✅	docker_cws_instrumentation_amd64	7.447 MiB
✅	docker_cws_instrumentation_arm64	6.877 MiB
✅	docker_dogstatsd_amd64	39.032 MiB
✅	docker_dogstatsd_arm64	37.180 MiB
✅	docker_host_profiler_amd64	305.308 MiB
✅	docker_host_profiler_arm64	316.411 MiB
✅	dogstatsd_deb_amd64	29.763 MiB
✅	dogstatsd_deb_arm64	27.822 MiB
✅	dogstatsd_rpm_amd64	29.763 MiB
✅	dogstatsd_suse_amd64	29.763 MiB

[cit-pr-commenter-54b7da]

Regression Detector

Regression Detector Results

Metrics dashboard
Target profiles
Run ID: 29accbf1-d6cd-452f-9c59-2e77ad2143d2

Baseline: 1d93f4e
Comparison: 5de868d
Diff

❌ Experiments with retried target crashes

This is a critical error. One or more replicates failed with a non-zero exit code. These replicates may have been retried. See Replicate Execution Details for more information.

• quality_gate_idle

Optimization Goals: ✅ Improvement(s) detected

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
✅	quality_gate_metrics_logs	memory utilization	-6.71	[-6.94, -6.48]	1	Logs bounds checks dashboard
✅	quality_gate_logs	% cpu utilization	-47.13	[-47.95, -46.30]	1	Logs bounds checks dashboard

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
➖	quality_gate_idle_all_features	memory utilization	-2.76	[-2.79, -2.73]	1	Logs bounds checks dashboard
➖	quality_gate_idle	memory utilization	-3.13	[-3.18, -3.08]	1	Logs bounds checks dashboard
➖	quality_gate_security_mean_fs_load	memory utilization	-3.63	[-3.66, -3.59]	1	Logs bounds checks dashboard
➖	quality_gate_security_no_fs_load	memory utilization	-3.69	[-3.79, -3.60]	1	Logs bounds checks dashboard
➖	quality_gate_security_idle	memory utilization	-3.71	[-3.77, -3.66]	1	Logs bounds checks dashboard
✅	quality_gate_metrics_logs	memory utilization	-6.71	[-6.94, -6.48]	1	Logs bounds checks dashboard
✅	quality_gate_logs	% cpu utilization	-47.13	[-47.95, -46.30]	1	Logs bounds checks dashboard

Bounds Checks: ✅ Passed

perf	experiment	bounds_check_name	replicates_passed	observed_value	links
✅	quality_gate_idle	intake_connections	10/10	3 ≤ 4	bounds checks dashboard
✅	quality_gate_idle	memory_usage	10/10	145.67MiB ≤ 154MiB	bounds checks dashboard
✅	quality_gate_idle	total_bytes_received	10/10	576.98KiB ≤ 819.20KiB	bounds checks dashboard
✅	quality_gate_idle_all_features	intake_connections	10/10	3 ≤ 4	bounds checks dashboard
✅	quality_gate_idle_all_features	memory_usage	10/10	489.64MiB ≤ 495MiB	bounds checks dashboard
✅	quality_gate_idle_all_features	total_bytes_received	10/10	0.89MiB ≤ 1.25MiB	bounds checks dashboard
✅	quality_gate_logs	intake_connections	10/10	4 ≤ 6	bounds checks dashboard
✅	quality_gate_logs	memory_usage	10/10	180.93MiB ≤ 195MiB	bounds checks dashboard
✅	quality_gate_logs	missed_bytes	10/10	0B = 0B	bounds checks dashboard
✅	quality_gate_logs	total_bytes_received	10/10	264.19MiB ≤ 292MiB	bounds checks dashboard
✅	quality_gate_metrics_logs	cpu_usage	10/10	332.60 ≤ 2000	bounds checks dashboard
✅	quality_gate_metrics_logs	intake_connections	10/10	3 ≤ 6	bounds checks dashboard
✅	quality_gate_metrics_logs	memory_usage	10/10	391.67MiB ≤ 430MiB	bounds checks dashboard
✅	quality_gate_metrics_logs	missed_bytes	10/10	0B = 0B	bounds checks dashboard
✅	quality_gate_metrics_logs	total_bytes_received	10/10	0.86GiB ≤ 1.04GiB	bounds checks dashboard
✅	quality_gate_security_idle	cpu_usage	10/10	28.38 ≤ 40	bounds checks dashboard
✅	quality_gate_security_idle	memory_usage	10/10	293.69MiB ≤ 330MiB	bounds checks dashboard
✅	quality_gate_security_mean_fs_load	cpu_usage	10/10	60.62 ≤ 80	bounds checks dashboard
✅	quality_gate_security_mean_fs_load	memory_usage	10/10	275.47MiB ≤ 310MiB	bounds checks dashboard
✅	quality_gate_security_no_fs_load	cpu_usage	10/10	23.09 ≤ 40	bounds checks dashboard
✅	quality_gate_security_no_fs_load	memory_usage	10/10	283.52MiB ≤ 320MiB	bounds checks dashboard

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:

1. Its estimated |Δ mean %| ≥ 5.00%, indicating the change is big enough to merit a closer look.

2. 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.

3. Its configuration does not mark it "erratic".

Replicate Execution Details

We run multiple replicates for each experiment/variant. However, we allow replicates to be automatically retried if there are any failures, up to 8 times, at which point the replicate is marked dead and we are unable to run analysis for the entire experiment. We call each of these attempts at running replicates a replicate execution. This section lists all replicate executions that failed due to the target crashing or being oom killed.

Note: In the below tables we bucket failures by experiment, variant, and failure type. For each of these buckets we list out the replicate indexes that failed with an annotation signifying how many times said replicate failed with the given failure mode. In the below example the baseline variant of the experiment named experiment_with_failures had two replicates that failed by oom kills. Replicate 0, which failed 8 executions, and replicate 1 which failed 6 executions, all with the same failure mode.

Experiment	Variant	Replicates	Failure	Logs	Debug Dashboard
experiment_with_failures	baseline	0 (x8) 1 (x6)	Oom killed		Debug Dashboard

The debug dashboard links will take you to a debugging dashboard specifically designed to investigate replicate execution failures.

❌ Retried Normal Replicate Execution Failures (non-profiling)

Experiment	Variant	Replicates	Failure	Debug Dashboard
quality_gate_idle	baseline	7	Oom killed	Debug Dashboard

❌ Retried Profiling Replicate Execution Failures (ddprof)

Note: Profiling replicas may still be executing. See the debug dashboard for up to date status.

Experiment	Variant	Replicates	Failure	Debug Dashboard
quality_gate_idle	baseline	10	Oom killed	Debug Dashboard
quality_gate_idle_all_features	baseline	10	Oom killed	Debug Dashboard
quality_gate_idle_all_features	comparison	10	Oom killed	Debug Dashboard
quality_gate_logs	baseline	10	Oom killed	Debug Dashboard
quality_gate_logs	comparison	10	Oom killed	Debug Dashboard
quality_gate_metrics_logs	baseline	10	Oom killed	Debug Dashboard
quality_gate_metrics_logs	comparison	10	Oom killed	Debug Dashboard
quality_gate_security_idle	baseline	10	Oom killed	Debug Dashboard
quality_gate_security_no_fs_load	comparison	10	Oom killed	Debug Dashboard

CI Pass/Fail Decision

✅ Passed. All Quality Gates passed.

• quality_gate_idle_all_features, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_idle_all_features, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_idle_all_features, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check missed_bytes: 10/10 replicas passed. Gate passed.
• quality_gate_security_idle, bounds check cpu_usage: 10/10 replicas passed. Gate passed.
• quality_gate_security_idle, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_security_mean_fs_load, bounds check cpu_usage: 10/10 replicas passed. Gate passed.
• quality_gate_security_mean_fs_load, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_security_no_fs_load, bounds check cpu_usage: 10/10 replicas passed. Gate passed.
• quality_gate_security_no_fs_load, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check cpu_usage: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check missed_bytes: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_idle, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_idle, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_idle, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.

[qsl73]

I do not have all the context so feel free to ignore.

What if instead of returning a slice we return an iterator with the npmodel.NetworkPathConnection like:

ScheduleNetworkPathTests(conns iter.Seq[npmodel.NetworkPathConnection]) iter.Seq2[npmodel.NetworkPathConnection, npmodel.NetworkPath]

That way we can get rid of the two arrays synced together: networkPaths and scheduledConns.

Implementation — yield each pair lazily as connections are processed:

func (s *npCollectorImpl) scheduleNetworkPathTests(...) iter.Seq2[npmodel.NetworkPathConnection, npmodel.NetworkPath] {
    return func(yield func(npmodel.NetworkPathConnection, npmodel.NetworkPath) bool) {
        // ... getVPCSubnets, startTime setup ...
        for conn := range conns {
            var path npmodel.NetworkPath
            if s.shouldScheduleNetworkPathForConn(conn, origin, vpcSubnets) {
                pathtest := s.makePathtest(conn, origin)
                path = npmodel.NetworkPath{HasTest: true, TestIdentity: makeTestIdentity(...)}
                if err := s.scheduleOne(&pathtest); err != nil {
                    s.logger.Errorf(...)
                }
            }
            if !yield(conn, path) {
                return
            }
        }
        // emit timing/count metrics
    }
}

Consumer in net.go — drops to one parallel slice instead of two, gets npc directly from the range:

func (c *ConnectionsCheck) scheduleNetworkPath(conns *model.Connections) {
    var scheduledModelConns []*model.Connection
    connsIter := func(yield func(npmodel.NetworkPathConnection) bool) {
        for _, conn := range conns.Conns {
            // ... build npc, parse IPs ...
            if !yield(npc) { return }
            scheduledModelConns = append(scheduledModelConns, conn)
        }
    }
    i := 0
    for npc, path := range c.npCollector.ScheduleNetworkPathTests(connsIter) {
        if i >= len(scheduledModelConns) { break }
        scheduledModelConns[i].NetworkPath = &model.NetworkPath{HasTest: path.HasTest}
        if path.HasTest {
            scheduledModelConns[i].NetworkPath.TestIdentity = path.TestIdentity
            logNetworkPathConnection(npc, path)
        }
        i++
    }
}

[dd-octo-sts]

This pull request has been automatically marked as stale because it has not had activity in the past 15 days.

It will be closed in 30 days if no further activity occurs. If this pull request is still relevant, adding a comment or pushing new commits will keep it open. Also, you can always reopen the pull request if you missed the window.

Thank you for your contributions!