discovery: use libdd_discovery CGO bindings for service discovery#46876
Draft
martavicentenavarro wants to merge 3 commits intomainfrom
Draft
discovery: use libdd_discovery CGO bindings for service discovery#46876martavicentenavarro wants to merge 3 commits intomainfrom
martavicentenavarro wants to merge 3 commits intomainfrom
Conversation
martavicentenavarro
added
changelog/no-changelog
martavicentenavarro
force-pushed
the
marta/load_library
branch
from
cc5d45a to
ded76b0
Compare
martavicentenavarro
force-pushed
the
marta/load_library
branch
2 times, most recently
from
bbcce02 to
b587da9
Compare
Contributor
Files inventory check summaryFile checks results against ancestor d4c34bbf: Results for datadog-agent_7.78.0~devel.git.681.0ca2870.pipeline.103318857-1_amd64.deb:No change detected |
Contributor
Static quality checks✅ Please find below the results from static quality gates Successful checksInfo
16 successful checks with minimal change (< 2 KiB)
On-wire sizes (compressed)
|
Regression DetectorRegression Detector ResultsMetrics dashboard Baseline: d98e355 Optimization Goals: ✅ No significant changes detected
|
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | docker_containers_cpu | % cpu utilization | -0.41 | [-3.47, +2.64] | 1 | Logs |
Fine details of change detection per experiment
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | tcp_syslog_to_blackhole | ingress throughput | +0.99 | [+0.86, +1.13] | 1 | Logs |
| ➖ | quality_gate_metrics_logs | memory utilization | +0.95 | [+0.72, +1.19] | 1 | Logs bounds checks dashboard |
| ➖ | file_tree | memory utilization | +0.64 | [+0.59, +0.69] | 1 | Logs |
| ➖ | docker_containers_memory | memory utilization | +0.58 | [+0.51, +0.66] | 1 | Logs |
| ➖ | quality_gate_idle_all_features | memory utilization | +0.58 | [+0.54, +0.61] | 1 | Logs bounds checks dashboard |
| ➖ | quality_gate_idle | memory utilization | +0.16 | [+0.11, +0.21] | 1 | Logs bounds checks dashboard |
| ➖ | uds_dogstatsd_20mb_12k_contexts_20_senders | memory utilization | +0.09 | [+0.03, +0.15] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api | ingress throughput | +0.01 | [-0.19, +0.21] | 1 | Logs |
| ➖ | file_to_blackhole_100ms_latency | egress throughput | -0.00 | [-0.08, +0.08] | 1 | Logs |
| ➖ | tcp_dd_logs_filter_exclude | ingress throughput | -0.00 | [-0.11, +0.11] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api_v3 | ingress throughput | -0.01 | [-0.21, +0.19] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency | egress throughput | -0.01 | [-0.49, +0.47] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency | egress throughput | -0.01 | [-0.44, +0.42] | 1 | Logs |
| ➖ | ddot_metrics_sum_cumulative | memory utilization | -0.02 | [-0.16, +0.12] | 1 | Logs |
| ➖ | ddot_logs | memory utilization | -0.08 | [-0.13, -0.02] | 1 | Logs |
| ➖ | file_to_blackhole_500ms_latency | egress throughput | -0.11 | [-0.51, +0.29] | 1 | Logs |
| ➖ | ddot_metrics_sum_delta | memory utilization | -0.12 | [-0.29, +0.04] | 1 | Logs |
| ➖ | ddot_metrics_sum_cumulativetodelta_exporter | memory utilization | -0.17 | [-0.39, +0.06] | 1 | Logs |
| ➖ | otlp_ingest_metrics | memory utilization | -0.19 | [-0.35, -0.04] | 1 | Logs |
| ➖ | otlp_ingest_logs | memory utilization | -0.29 | [-0.39, -0.20] | 1 | Logs |
| ➖ | docker_containers_cpu | % cpu utilization | -0.41 | [-3.47, +2.64] | 1 | Logs |
| ➖ | ddot_metrics | memory utilization | -0.51 | [-0.69, -0.32] | 1 | Logs |
| ➖ | quality_gate_logs | % cpu utilization | -1.37 | [-2.96, +0.21] | 1 | Logs bounds checks dashboard |
Bounds Checks: ❌ Failed
| perf | experiment | bounds_check_name | replicates_passed | observed_value | links |
|---|---|---|---|---|---|
| ✅ | docker_containers_cpu | simple_check_run | 10/10 | 722 ≥ 26 | |
| ✅ | docker_containers_memory | memory_usage | 10/10 | 276.94MiB ≤ 370MiB | |
| ✅ | docker_containers_memory | simple_check_run | 10/10 | 721 ≥ 26 | |
| ✅ | file_to_blackhole_0ms_latency | memory_usage | 10/10 | 0.19GiB ≤ 1.20GiB | |
| ✅ | file_to_blackhole_0ms_latency | missed_bytes | 10/10 | 0B = 0B | |
| ✅ | file_to_blackhole_1000ms_latency | memory_usage | 10/10 | 0.23GiB ≤ 1.20GiB | |
| ✅ | file_to_blackhole_1000ms_latency | missed_bytes | 10/10 | 0B = 0B | |
| ✅ | file_to_blackhole_100ms_latency | memory_usage | 10/10 | 0.19GiB ≤ 1.20GiB | |
| ✅ | file_to_blackhole_100ms_latency | missed_bytes | 10/10 | 0B = 0B | |
| ✅ | file_to_blackhole_500ms_latency | memory_usage | 10/10 | 0.21GiB ≤ 1.20GiB | |
| ✅ | file_to_blackhole_500ms_latency | missed_bytes | 10/10 | 0B = 0B | |
| ✅ | quality_gate_idle | intake_connections | 10/10 | 3 = 3 | bounds checks dashboard |
| ❌ | quality_gate_idle | memory_usage | 9/10 | 175.13MiB > 175MiB | bounds checks dashboard |
| ✅ | quality_gate_idle_all_features | intake_connections | 10/10 | 2 ≤ 3 | bounds checks dashboard |
| ✅ | quality_gate_idle_all_features | memory_usage | 10/10 | 490.46MiB ≤ 550MiB | bounds checks dashboard |
| ✅ | quality_gate_logs | intake_connections | 10/10 | 4 ≤ 6 | bounds checks dashboard |
| ✅ | quality_gate_logs | memory_usage | 10/10 | 201.63MiB ≤ 220MiB | bounds checks dashboard |
| ✅ | quality_gate_logs | missed_bytes | 10/10 | 0B = 0B | bounds checks dashboard |
| ✅ | quality_gate_metrics_logs | cpu_usage | 10/10 | 340.11 ≤ 2000 | bounds checks dashboard |
| ✅ | quality_gate_metrics_logs | intake_connections | 10/10 | 4 ≤ 6 | bounds checks dashboard |
| ✅ | quality_gate_metrics_logs | memory_usage | 10/10 | 408.38MiB ≤ 475MiB | bounds checks dashboard |
| ✅ | quality_gate_metrics_logs | missed_bytes | 10/10 | 0B = 0B | 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
❌ Failed. Some Quality Gates were violated.
- quality_gate_idle, bounds check memory_usage: 9/10 replicas passed. Failed 1 which is > 0. Gate FAILED.
- quality_gate_idle, 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 missed_bytes: 10/10 replicas passed. Gate passed.
- quality_gate_logs, 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 memory_usage: 10/10 replicas passed. Gate passed.
- quality_gate_metrics_logs, bounds check intake_connections: 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 cpu_usage: 10/10 replicas passed. Gate passed.
martavicentenavarro
force-pushed
the
marta/load_library
branch
4 times, most recently
from
8934a6f to
aa30b5f
Compare
Contributor
Author
|
@codex review |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: aa30b5f346
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".
tasks/system_probe.py
Outdated
Comment on lines
+843
to
+845
| if not is_windows and not is_macos: | ||
| build_rust_cgo_libs(ctx, arch=arch_obj) | ||
| build_tags.append("dd_discovery_cgo") |
There was a problem hiding this comment.
build_sysprobe_binary now always enables dd_discovery_cgo on Linux, which forces linking against libdd_discovery even when --static is requested. In the static path, get_build_flags adds -extldflags=-static (tasks/libs/common/utils.py), so the linker expects static archives, but this change wires in a shared-library workflow (libdd_discovery.so), causing static system-probe builds to fail. This regresses users/CI paths that rely on dda inv system-probe.build-sysprobe-binary --static.
Useful? React with 👍 / 👎.
martavicentenavarro
force-pushed
the
marta/load_library
branch
2 times, most recently
from
bce5f79 to
57c6b45
Compare
Yumasi
requested changes
Mar 10, 2026
martavicentenavarro
force-pushed
the
marta/load_library
branch
2 times, most recently
from
b8cc610 to
a922591
Compare
Yumasi
reviewed
Mar 17, 2026
tasks/build_tags.py
Outdated
| "zstd", | ||
| "cel", | ||
| "cws_instrumentation_injector_only", # used for building cws-instrumentation with only the injector code | ||
| "dd_discovery_rust", # used to enable the Rust-backed libdd_discovery implementation in system-probe |
Member
There was a problem hiding this comment.
Nit: appart from the two last entries, that set seems sorted. I would try to keep it this way, it makes it easier to see which build tag is used when building as they are said in this set's order.
tasks/system_probe.py
Outdated
| platform_flag = f"--platforms={platform_map[arch.kmt_arch]}" | ||
|
|
||
| for source_path in RUST_LIBS: | ||
| ctx.run(f"bazelisk run {platform_flag} -- @//{source_path}:install_libs --destdir={Path(source_path)}") |
Member
There was a problem hiding this comment.
Suggested change
| ctx.run(f"bazelisk run {platform_flag} -- @//{source_path}:install_libs --destdir={Path(source_path)}") | |
| ctx.run(f"bazelisk run --config=release {platform_flag} -- @//{source_path}:install_libs --destdir={Path(source_path)}") |
tasks/system_probe.py
Outdated
| platform_flag = f"--platforms={platform_map[arch.kmt_arch]}" | ||
|
|
||
| for source_path in RUST_LIBS: | ||
| ctx.run(f"bazelisk run {platform_flag} -- @//{source_path}:install_libs --destdir={Path(source_path)}") |
Member
There was a problem hiding this comment.
Suggested change
| ctx.run(f"bazelisk run {platform_flag} -- @//{source_path}:install_libs --destdir={Path(source_path)}") | |
| ctx.run(f"bazelisk run --config=release {platform_flag} -- @//{source_path}:install_libs --destdir={Path(source_path)}") |
martavicentenavarro
force-pushed
the
marta/load_library
branch
3 times, most recently
from
e9eae15 to
ed91e0a
Compare
Improves discovery: use dd_discovery_cgo build tag instead of built-in cgo fix omnibus build for CGo library integration Implement some improvements
martavicentenavarro
force-pushed
the
marta/load_library
branch
from
ed91e0a to
1cb3124
Compare
- Rename build tag dd_discovery_cgo -> dd_discovery_rust - Link against static library (add -lpthread -ldl -lm linker flags) - Remove libdd_discovery.so copy step from omnibus (no longer needed) - Add --config=release to bazelisk build command - Fix test build tags to avoid duplicate definitions with !dd_discovery_rust
martavicentenavarro
force-pushed
the
marta/load_library
branch
from
43041c1 to
0ca2870
Compare
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
2 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 does this PR do?
Replaces the pure-Go service discovery implementation with a CGO-backed one that delegates to the Rust libdd_discovery shared library. The original Go implementation is preserved as a fallback when the dd_discovery_cgo build tag is not set.
impl_module_linux.go and impl_test_helpers_linux_test.go mainly contain logic that previously lived in impl_linux.go and the Linux test files, reorganized so CGO and non-CGO paths can share it without intended behavior changes.
Motivation
DSCVR-369
Describe how you validated your changes
I validated the change by building the Rust shared library (libdd_discovery.so) and running discovery module tests in both execution modes: the CGO path (dd_discovery_cgo) and the non-CGO path (!dd_discovery_cgo).
Additional Notes
The new path delegates all service discovery logic to the Rust
libdd_discoverylibrary. As a result, there are known intentional differences with the non-CGO fallback path that will be resolved before the fallback is removed:GPUPIDslist.privilegedDetector: theprivilegedDetectorfield in the shareddiscoverystruct is only initialized and used in the previous path. It will be removed oncelibdd_discoverycovers full feature parity with the Go implementation.