[AGENTRUN-1108] Skip portlist unit test when kernel return ESPIPE by pgimalac · Pull Request #46909 · DataDog/datadog-agent

pgimalac · 2026-02-25T13:54:56Z

What does this PR do?

Skip portlist unit test when kernel return ESPIPE.
Basically #46825 but for the unit test.

Motivation

Avoid flaky test.

Describe how you validated your changes

CI

Additional Notes

agent-platform-auto-pr · 2026-02-25T14:16:19Z

Files inventory check summary

File checks results against ancestor 77fc7adf:

Results for datadog-agent_7.78.0~devel.git.102.877c165.pipeline.98854485-1_amd64.deb:

No change detected

cit-pr-commenter-54b7da · 2026-02-25T14:49:15Z

Regression Detector

Regression Detector Results

Metrics dashboard
Target profiles
Run ID: 8ab7e820-0016-420d-b9f5-cc1f0d192637

Baseline: 77fc7ad
Comparison: 877c165
Diff

Optimization Goals: ✅ No significant changes detected

Experiments ignored for regressions

Regressions in experiments with settings containing erratic: true are ignored.

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
➖	docker_containers_cpu	% cpu utilization	-0.84	[-3.88, +2.20]	1	Logs

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
➖	quality_gate_logs	% cpu utilization	+1.95	[+0.48, +3.42]	1	Logs bounds checks dashboard
➖	ddot_logs	memory utilization	+0.29	[+0.23, +0.36]	1	Logs
➖	ddot_metrics_sum_cumulative	memory utilization	+0.21	[+0.05, +0.38]	1	Logs
➖	file_tree	memory utilization	+0.15	[+0.10, +0.19]	1	Logs
➖	ddot_metrics	memory utilization	+0.12	[-0.09, +0.34]	1	Logs
➖	file_to_blackhole_500ms_latency	egress throughput	+0.07	[-0.31, +0.46]	1	Logs
➖	file_to_blackhole_100ms_latency	egress throughput	+0.04	[-0.00, +0.08]	1	Logs
➖	uds_dogstatsd_to_api	ingress throughput	+0.01	[-0.12, +0.13]	1	Logs
➖	uds_dogstatsd_to_api_v3	ingress throughput	+0.00	[-0.13, +0.14]	1	Logs
➖	tcp_dd_logs_filter_exclude	ingress throughput	+0.00	[-0.08, +0.09]	1	Logs
➖	ddot_metrics_sum_delta	memory utilization	-0.01	[-0.21, +0.18]	1	Logs
➖	file_to_blackhole_1000ms_latency	egress throughput	-0.02	[-0.44, +0.39]	1	Logs
➖	quality_gate_idle	memory utilization	-0.02	[-0.07, +0.02]	1	Logs bounds checks dashboard
➖	otlp_ingest_metrics	memory utilization	-0.14	[-0.30, +0.01]	1	Logs
➖	file_to_blackhole_0ms_latency	egress throughput	-0.16	[-0.63, +0.32]	1	Logs
➖	ddot_metrics_sum_cumulativetodelta_exporter	memory utilization	-0.17	[-0.41, +0.06]	1	Logs
➖	quality_gate_metrics_logs	memory utilization	-0.26	[-0.49, -0.04]	1	Logs bounds checks dashboard
➖	docker_containers_memory	memory utilization	-0.28	[-0.36, -0.21]	1	Logs
➖	otlp_ingest_logs	memory utilization	-0.31	[-0.41, -0.22]	1	Logs
➖	quality_gate_idle_all_features	memory utilization	-0.47	[-0.50, -0.43]	1	Logs bounds checks dashboard
➖	tcp_syslog_to_blackhole	ingress throughput	-0.64	[-0.71, -0.57]	1	Logs
➖	uds_dogstatsd_20mb_12k_contexts_20_senders	memory utilization	-0.71	[-0.77, -0.66]	1	Logs
➖	docker_containers_cpu	% cpu utilization	-0.84	[-3.88, +2.20]	1	Logs

Bounds Checks: ✅ Passed

perf	experiment	bounds_check_name	replicates_passed	links
✅	docker_containers_cpu	simple_check_run	10/10
✅	docker_containers_memory	memory_usage	10/10
✅	docker_containers_memory	simple_check_run	10/10
✅	file_to_blackhole_0ms_latency	lost_bytes	10/10
✅	file_to_blackhole_0ms_latency	memory_usage	10/10
✅	file_to_blackhole_1000ms_latency	lost_bytes	10/10
✅	file_to_blackhole_1000ms_latency	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_500ms_latency	lost_bytes	10/10
✅	file_to_blackhole_500ms_latency	memory_usage	10/10
✅	quality_gate_idle	intake_connections	10/10	bounds checks dashboard
✅	quality_gate_idle	memory_usage	10/10	bounds checks dashboard
✅	quality_gate_idle_all_features	intake_connections	10/10	bounds checks dashboard
✅	quality_gate_idle_all_features	memory_usage	10/10	bounds checks dashboard
✅	quality_gate_logs	intake_connections	10/10	bounds checks dashboard
✅	quality_gate_logs	lost_bytes	10/10	bounds checks dashboard
✅	quality_gate_logs	memory_usage	10/10	bounds checks dashboard
✅	quality_gate_metrics_logs	cpu_usage	10/10	bounds checks dashboard
✅	quality_gate_metrics_logs	intake_connections	10/10	bounds checks dashboard
✅	quality_gate_metrics_logs	lost_bytes	10/10	bounds checks dashboard
✅	quality_gate_metrics_logs	memory_usage	10/10	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:

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_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.
quality_gate_logs, 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_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_all_features, bounds check memory_usage: 10/10 replicas passed. Gate passed.
quality_gate_idle_all_features, bounds check intake_connections: 10/10 replicas passed. Gate passed.
quality_gate_metrics_logs, bounds check lost_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 cpu_usage: 10/10 replicas passed. Gate passed.
quality_gate_metrics_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.

agent-platform-auto-pr · 2026-02-25T15:05:36Z

Static quality checks

✅ Please find below the results from static quality gates
Comparison made with ancestor 77fc7ad
📊 Static Quality Gates Dashboard
🔗 SQG Job

31 successful checks with minimal change (< 2 KiB)

	Quality gate	Current Size
✅	agent_deb_amd64	743.908 MiB
✅	agent_deb_amd64_fips	702.139 MiB
✅	agent_heroku_amd64	310.896 MiB
✅	agent_msi	607.752 MiB
✅	agent_rpm_amd64	743.892 MiB
✅	agent_rpm_amd64_fips	702.123 MiB
✅	agent_rpm_arm64	721.997 MiB
✅	agent_rpm_arm64_fips	683.276 MiB
✅	agent_suse_amd64	743.892 MiB
✅	agent_suse_amd64_fips	702.123 MiB
✅	agent_suse_arm64	721.997 MiB
✅	agent_suse_arm64_fips	683.276 MiB
✅	docker_agent_amd64	804.925 MiB
✅	docker_agent_arm64	807.769 MiB
✅	docker_agent_jmx_amd64	995.837 MiB
✅	docker_agent_jmx_arm64	987.463 MiB
✅	docker_cluster_agent_amd64	203.427 MiB
✅	docker_cluster_agent_arm64	217.835 MiB
✅	docker_cws_instrumentation_amd64	7.135 MiB
✅	docker_cws_instrumentation_arm64	6.689 MiB
✅	docker_dogstatsd_amd64	38.512 MiB
✅	docker_dogstatsd_arm64	36.812 MiB
✅	dogstatsd_deb_amd64	29.736 MiB
✅	dogstatsd_deb_arm64	27.893 MiB
✅	dogstatsd_rpm_amd64	29.736 MiB
✅	dogstatsd_suse_amd64	29.736 MiB
✅	iot_agent_deb_amd64	42.636 MiB
✅	iot_agent_deb_arm64	39.738 MiB
✅	iot_agent_deb_armhf	40.463 MiB
✅	iot_agent_rpm_amd64	42.637 MiB
✅	iot_agent_suse_amd64	42.637 MiB

On-wire sizes (compressed)

	Quality gate	Change	Size (prev → curr → max)
✅	agent_deb_amd64	+14.14 KiB (0.01% increase)	173.673 → 173.686 → 186.090
✅	agent_deb_amd64_fips	+21.19 KiB (0.01% increase)	164.599 → 164.620 → 180.330
✅	agent_heroku_amd64	neutral	74.983 MiB → 88.440
✅	agent_msi	+24.0 KiB (0.02% increase)	137.613 → 137.637 → 154.470
✅	agent_rpm_amd64	-31.06 KiB (0.02% reduction)	175.590 → 175.559 → 189.170
✅	agent_rpm_amd64_fips	-33.31 KiB (0.02% reduction)	167.408 → 167.375 → 181.060
✅	agent_rpm_arm64	-16.98 KiB (0.01% reduction)	157.827 → 157.810 → 170.020
✅	agent_rpm_arm64_fips	+16.27 KiB (0.01% increase)	150.647 → 150.663 → 164.130
✅	agent_suse_amd64	-31.06 KiB (0.02% reduction)	175.590 → 175.559 → 189.170
✅	agent_suse_amd64_fips	-33.31 KiB (0.02% reduction)	167.408 → 167.375 → 181.060
✅	agent_suse_arm64	-16.98 KiB (0.01% reduction)	157.827 → 157.810 → 170.020
✅	agent_suse_arm64_fips	+16.27 KiB (0.01% increase)	150.647 → 150.663 → 164.130
✅	docker_agent_amd64	neutral	266.263 MiB → 279.410
✅	docker_agent_arm64	neutral	253.537 MiB → 267.960
✅	docker_agent_jmx_amd64	neutral	334.908 MiB → 348.040
✅	docker_agent_jmx_arm64	+4.06 KiB (0.00% increase)	318.171 → 318.175 → 332.560
✅	docker_cluster_agent_amd64	neutral	71.194 MiB → 71.920
✅	docker_cluster_agent_arm64	neutral	66.835 MiB → 67.220
✅	docker_cws_instrumentation_amd64	neutral	2.995 MiB → 3.330
✅	docker_cws_instrumentation_arm64	neutral	2.726 MiB → 3.090
✅	docker_dogstatsd_amd64	neutral	14.903 MiB → 15.820
✅	docker_dogstatsd_arm64	neutral	14.238 MiB → 14.830
✅	dogstatsd_deb_amd64	neutral	7.855 MiB → 8.790
✅	dogstatsd_deb_arm64	neutral	6.744 MiB → 7.710
✅	dogstatsd_rpm_amd64	neutral	7.866 MiB → 8.800
✅	dogstatsd_suse_amd64	neutral	7.866 MiB → 8.800
✅	iot_agent_deb_amd64	neutral	11.241 MiB → 12.040
✅	iot_agent_deb_arm64	neutral	9.604 MiB → 10.450
✅	iot_agent_deb_armhf	neutral	9.805 MiB → 10.620
✅	iot_agent_rpm_amd64	neutral	11.255 MiB → 12.060
✅	iot_agent_suse_amd64	neutral	11.255 MiB → 12.060

rdesgroppes

Much appreciated mitigation!

…6909) ### What does this PR do? Skip portlist unit test when kernel return ESPIPE. Basically #46825 but for the unit test. ### Motivation Avoid flaky test. ### Describe how you validated your changes CI ### Additional Notes (cherry picked from commit 7ae91dc) ___ Co-authored-by: Pierre Gimalac <pierre.gimalac@datadoghq.com>

…PIPE (#46909)" This reverts commit 7ae91dc.

### What does this PR do? After a tentative fix, the present change implements the consensus established in a [review comment](#46927 (comment)). It replaces the fd-caching + seek pattern with plain open-per-poll: - drops `procNetFiles []*os.File` from `linuxImpl`, - opens each `/proc/net/{tcp,tcp4,udp,udp6}` anew in `AppendListeningPorts`, closed immediately after parsing, - eliminates `lseek` on `/proc/net/*` entirely, sidestepping the `ESPIPE` [regression introduced in Linux 6.6.102/6.12.42](https://lore.kernel.org/all/20250815195616.64497967@chagall.paradoxon.rec/). With no fds to release, `Close()` became a no-op on all platforms; it is therefore removed from `osImpl`, `Poller`, and all (empty) implementations, along with `TestClose` and the sole external call site in `port.go`. This allows to always enable portlist unit test by also reverting: - #46909. ### Motivation Triggered by: - #incident-50065 - #incident-50070 - #incident-50071 Co-authored-by: regis.desgroppes <regis.desgroppes@datadoghq.com>

… return ESPIPE (#46937) Backport 7ae91dc from #46909. ___ ### What does this PR do? Skip portlist unit test when kernel return ESPIPE. Basically #46825 but for the unit test. ### Motivation Avoid flaky test. ### Describe how you validated your changes CI ### Additional Notes Co-authored-by: dd-octo-sts[bot] <200755185+dd-octo-sts[bot]@users.noreply.github.com> Co-authored-by: Pierre Gimalac <pierre.gimalac@datadoghq.com>

…) ### What does this PR do? After a tentative fix, the present change implements the consensus established in a [review comment](#46927 (comment)). It replaces the fd-caching + seek pattern with plain open-per-poll: - drops `procNetFiles []*os.File` from `linuxImpl`, - opens each `/proc/net/{tcp,tcp4,udp,udp6}` anew in `AppendListeningPorts`, closed immediately after parsing, - eliminates `lseek` on `/proc/net/*` entirely, sidestepping the `ESPIPE` [regression introduced in Linux 6.6.102/6.12.42](https://lore.kernel.org/all/20250815195616.64497967@chagall.paradoxon.rec/). With no fds to release, `Close()` became a no-op on all platforms; it is therefore removed from `osImpl`, `Poller`, and all (empty) implementations, along with `TestClose` and the sole external call site in `port.go`. This allows to always enable portlist unit test by also reverting: - #46909. ### Motivation Triggered by: - #incident-50065 - #incident-50070 - #incident-50071 Co-authored-by: regis.desgroppes <regis.desgroppes@datadoghq.com> (cherry picked from commit f4d3c85) ___ Co-authored-by: Régis Desgroppes <rdesgroppes@gmail.com>

…on each poll (#46975) Backport f4d3c85 from #46927. ___ ### What does this PR do? After a tentative fix, the present change implements the consensus established in a [review comment](#46927 (comment)). It replaces the fd-caching + seek pattern with plain open-per-poll: - drops `procNetFiles []*os.File` from `linuxImpl`, - opens each `/proc/net/{tcp,tcp4,udp,udp6}` anew in `AppendListeningPorts`, closed immediately after parsing, - eliminates `lseek` on `/proc/net/*` entirely, sidestepping the `ESPIPE` [regression introduced in Linux 6.6.102/6.12.42](https://lore.kernel.org/all/20250815195616.64497967@chagall.paradoxon.rec/). With no fds to release, `Close()` became a no-op on all platforms; it is therefore removed from `osImpl`, `Poller`, and all (empty) implementations, along with `TestClose` and the sole external call site in `port.go`. This allows to always enable portlist unit test by also reverting: - #46909. ### Motivation Triggered by: - #incident-50065 - #incident-50070 - #incident-50071 ___ (for `7.77.x`, that's #incident-50132 and #incident-50133) Co-authored-by: florent.clarret <florent.clarret@datadoghq.com>

… return ESPIPE (#46939) Backport 7ae91dc from #46909. ___ ### What does this PR do? Skip portlist unit test when kernel return ESPIPE. Basically #46825 but for the unit test. ### Motivation Avoid flaky test. ### Describe how you validated your changes CI ### Additional Notes Co-authored-by: dd-octo-sts[bot] <200755185+dd-octo-sts[bot]@users.noreply.github.com> Co-authored-by: Pierre Gimalac <pierre.gimalac@datadoghq.com>

…) ### What does this PR do? After a tentative fix, the present change implements the consensus established in a [review comment](#46927 (comment)). It replaces the fd-caching + seek pattern with plain open-per-poll: - drops `procNetFiles []*os.File` from `linuxImpl`, - opens each `/proc/net/{tcp,tcp4,udp,udp6}` anew in `AppendListeningPorts`, closed immediately after parsing, - eliminates `lseek` on `/proc/net/*` entirely, sidestepping the `ESPIPE` [regression introduced in Linux 6.6.102/6.12.42](https://lore.kernel.org/all/20250815195616.64497967@chagall.paradoxon.rec/). With no fds to release, `Close()` became a no-op on all platforms; it is therefore removed from `osImpl`, `Poller`, and all (empty) implementations, along with `TestClose` and the sole external call site in `port.go`. This allows to always enable portlist unit test by also reverting: - #46909. ### Motivation Triggered by: - #incident-50065 - #incident-50070 - #incident-50071 Co-authored-by: regis.desgroppes <regis.desgroppes@datadoghq.com> (cherry picked from commit f4d3c85) ___ Co-authored-by: Régis Desgroppes <rdesgroppes@gmail.com>

…on each poll (#47028) Backport f4d3c85 from #46927. ___ ### What does this PR do? After a tentative fix, the present change implements the consensus established in a [review comment](#46927 (comment)). It replaces the fd-caching + seek pattern with plain open-per-poll: - drops `procNetFiles []*os.File` from `linuxImpl`, - opens each `/proc/net/{tcp,tcp4,udp,udp6}` anew in `AppendListeningPorts`, closed immediately after parsing, - eliminates `lseek` on `/proc/net/*` entirely, sidestepping the `ESPIPE` [regression introduced in Linux 6.6.102/6.12.42](https://lore.kernel.org/all/20250815195616.64497967@chagall.paradoxon.rec/). With no fds to release, `Close()` became a no-op on all platforms; it is therefore removed from `osImpl`, `Poller`, and all (empty) implementations, along with `TestClose` and the sole external call site in `port.go`. This allows to always enable portlist unit test by also reverting: - #46909. ### Motivation Triggered by: - #incident-50065 - #incident-50070 - #incident-50071 ___ (for `7.76.x`, that's #incident-50204 and #incident-50205) Co-authored-by: axel.vonengel <axel.vonengel@datadoghq.com>

chore: skip portlist unit test when kernel return ESPIPE

877c165

pgimalac requested a review from a team as a code owner February 25, 2026 13:54

pgimalac requested a review from Enzu83 February 25, 2026 13:54

pgimalac added changelog/no-changelog No changelog entry needed qa/no-code-change No code change in Agent code requiring validation team/agent-runtimes labels Feb 25, 2026

dd-octo-sts Bot added the internal Identify a non-fork PR label Feb 25, 2026

github-actions Bot added the short review PR is simple enough to be reviewed quickly label Feb 25, 2026

Enzu83 approved these changes Feb 25, 2026

View reviewed changes

rdesgroppes approved these changes Feb 25, 2026

View reviewed changes

KevinFairise2 merged commit 7ae91dc into main Feb 25, 2026
330 checks passed

KevinFairise2 deleted the pgimalac/skip-test-illegal-seek-port branch February 25, 2026 17:04

github-actions Bot added this to the 7.78.0 milestone Feb 25, 2026

chouetz added backport/7.76.x Automatically create a backport PR to the 7.76.x branch once the PR is merged backport/7.77.x Automatically create a backport PR to the 7.77.x branch once the PR is merged labels Feb 25, 2026

dd-octo-sts Bot mentioned this pull request Feb 25, 2026

[Backport 7.77.x] [AGENTRUN-1108] Skip portlist unit test when kernel return ESPIPE #46937

Merged

dd-octo-sts Bot mentioned this pull request Feb 25, 2026

[Backport 7.76.x] [AGENTRUN-1108] Skip portlist unit test when kernel return ESPIPE #46939

Merged

rdesgroppes added a commit that referenced this pull request Feb 25, 2026

Revert "[AGENTRUN-1108] Skip portlist unit test when kernel return ES…

ced3387

…PIPE (#46909)" This reverts commit 7ae91dc.

rdesgroppes mentioned this pull request Feb 25, 2026

Rework portlist poller to open /proc/net/* fresh on each poll #46927

Merged

dd-octo-sts Bot mentioned this pull request Feb 26, 2026

[Backport 7.77.x] Rework portlist poller to open /proc/net/* fresh on each poll #46975

Merged

dd-octo-sts Bot mentioned this pull request Feb 27, 2026

[Backport 7.76.x] Rework portlist poller to open /proc/net/* fresh on each poll #47028

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[AGENTRUN-1108] Skip portlist unit test when kernel return ESPIPE#46909

[AGENTRUN-1108] Skip portlist unit test when kernel return ESPIPE#46909
KevinFairise2 merged 1 commit intomainfrom
pgimalac/skip-test-illegal-seek-port

pgimalac commented Feb 25, 2026

Uh oh!

agent-platform-auto-pr Bot commented Feb 25, 2026

Uh oh!

cit-pr-commenter-54b7da Bot commented Feb 25, 2026

Experiments ignored for regressions

Fine details of change detection per experiment

Bounds Checks: ✅ Passed

Explanation

Uh oh!

agent-platform-auto-pr Bot commented Feb 25, 2026

Uh oh!

rdesgroppes left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

Conversation

pgimalac commented Feb 25, 2026

What does this PR do?

Motivation

Describe how you validated your changes

Additional Notes

Uh oh!

agent-platform-auto-pr Bot commented Feb 25, 2026

Files inventory check summary

Results for datadog-agent_7.78.0~devel.git.102.877c165.pipeline.98854485-1_amd64.deb:

Uh oh!

cit-pr-commenter-54b7da Bot commented Feb 25, 2026

Regression Detector

Regression Detector Results

Optimization Goals: ✅ No significant changes detected

Experiments ignored for regressions

Fine details of change detection per experiment

Bounds Checks: ✅ Passed

Explanation

CI Pass/Fail Decision

Uh oh!

agent-platform-auto-pr Bot commented Feb 25, 2026

Static quality checks

Uh oh!

rdesgroppes left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants