datadog-agent
datadog-agent copied to clipboard
DataDog
usm: Refactor monitors to use common uprobe attacher
What does this PR do?
This PR applies the new uprobe attacher introduced in #27663 to the existing USM monitors, unifying features and behavior.
Motivation
Additional Notes
Possible Drawbacks / Trade-offs
Describe how to test/QA your changes
Several tests were performed in the load-test environment, using the manual QA checklist as a reference. The base version was the agent release 7.55. The commit under test for this branch was d87ae5175511826d94b6f623798ed8a1e33fcc3e, tests will be repeated before merge with the latest available commit of this branch.
Additionally, tests were performed with the process-load-test configuration to better test the performance impact of the changes, which should be more clear with a higher load of processes being created/stopped. The differences in CPU and memory usages were not significant. The profiling comparisons were also checked and the CPU usage was similar in the changed code. Details:
Go Package Import Differences
Baseline: e7715df43b089834037cb4e3e22fef36ec6cb1d3 Comparison: eba45f56323eb19b1c0aa51d53a26b66cb862ed6
| binary | os | arch | change |
|---|---|---|---|
| system-probe | linux | amd64 | +4, -0
+github.com/DataDog/datadog-agent/pkg/ebpf/uprobes
+github.com/DataDog/datadog-agent/pkg/network/protocols/http/testutil
+github.com/lorenzosaino/go-sysctl
+golang.org/x/net/netutil
|
| system-probe | linux | arm64 | +4, -0
+github.com/DataDog/datadog-agent/pkg/ebpf/uprobes
+github.com/DataDog/datadog-agent/pkg/network/protocols/http/testutil
+github.com/lorenzosaino/go-sysctl
+golang.org/x/net/netutil
|
![cit-pr-commenter[bot] avatar](https://avatars.githubusercontent.com/in/330445?v=4 "Published by a pub.dev verified publisher"){kind=small}
Regression Detector
Regression Detector Results
Run ID: 91280807-246a-4dfd-9217-e9347ac33071 Metrics dashboard Target profiles
Baseline: e7715df43b089834037cb4e3e22fef36ec6cb1d3 Comparison: eba45f56323eb19b1c0aa51d53a26b66cb862ed6
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
No significant changes in experiment optimization goals
Confidence level: 90.00% Effect size tolerance: |Δ mean %| ≥ 5.00%
There were no significant changes in experiment optimization goals at this confidence level and effect size tolerance.
Fine details of change detection per experiment
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | otel_to_otel_logs | ingress throughput | +0.38 | [-0.44, +1.21] | 1 | Logs |
| ➖ | idle | memory utilization | +0.35 | [+0.30, +0.39] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api | ingress throughput | -0.00 | [-0.00, +0.00] | 1 | Logs |
| ➖ | tcp_dd_logs_filter_exclude | ingress throughput | -0.00 | [-0.01, +0.01] | 1 | Logs |
| ➖ | tcp_syslog_to_blackhole | ingress throughput | -0.09 | [-0.14, -0.04] | 1 | Logs |
| ➖ | file_tree | memory utilization | -0.28 | [-0.39, -0.17] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api_cpu | % cpu utilization | -0.49 | [-1.25, +0.27] | 1 | Logs |
| ➖ | basic_py_check | % cpu utilization | -1.12 | [-3.91, +1.67] | 1 | Logs |
| ➖ | pycheck_lots_of_tags | % cpu utilization | -1.79 | [-4.41, +0.83] | 1 | Logs |
Bounds Checks
| perf | experiment | bounds_check_name | replicates_passed |
|---|---|---|---|
| ❌ | idle | memory_usage | 7/10 |
Explanation
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".
![pr-commenter[bot] avatar](https://avatars.githubusercontent.com/u/365230?v=4 "Published by a pub.dev verified publisher"){kind=small}
Test changes on VM
Use this command from test-infra-definitions to manually test this PR changes on a VM:
inv create-vm --pipeline-id=44304380 --os-family=ubuntu
Note: This applies to commit eba45f56
Serverless Benchmark Results
BenchmarkStartEndInvocation comparison between 74cdc875da220b7cf3886440aa48b0803680e5bb and 5990ff5009a3348c18f6226adea7a133336c6e6a.
tl;dr
Use these benchmarks as an insight tool during development.
-
Skim down the
vs basecolumn in each chart. If there is a~, then there was no statistically significant change to the benchmark. Otherwise, ensure the estimated percent change is either negative or very small. -
The last row of each chart is the
geomean. Ensure this percentage is either negative or very small.
What is this benchmarking?
The BenchmarkStartEndInvocation compares the amount of time it takes to call the start-invocation and end-invocation endpoints. For universal instrumentation languages (Dotnet, Golang, Java, Ruby), this represents the majority of the duration overhead added by our tracing layer.
The benchmark is run using a large variety of lambda request payloads. In the charts below, there is one row for each event payload type.
How do I interpret these charts?
The charts below comes from benchstat. They represent the statistical change in duration (sec/op), memory overhead (B/op), and allocations (allocs/op).
The benchstat docs explain how to interpret these charts.
Before the comparison table, we see common file-level configuration. If there are benchmarks with different configuration (for example, from different packages), benchstat will print separate tables for each configuration.
The table then compares the two input files for each benchmark. It shows the median and 95% confidence interval summaries for each benchmark before and after the change, and an A/B comparison under "vs base". ... The p-value measures how likely it is that any differences were due to random chance (i.e., noise). The "~" means benchstat did not detect a statistically significant difference between the two inputs. ...
Note that "statistically significant" is not the same as "large": with enough low-noise data, even very small changes can be distinguished from noise and considered statistically significant. It is, of course, generally easier to distinguish large changes from noise.
Finally, the last row of the table shows the geometric mean of each column, giving an overall picture of how the benchmarks changed. Proportional changes in the geomean reflect proportional changes in the benchmarks. For example, given n benchmarks, if sec/op for one of them increases by a factor of 2, then the sec/op geomean will increase by a factor of ⁿ√2.
I need more help
First off, do not worry if the benchmarks are failing. They are not tests. The intention is for them to be a tool for you to use during development.
If you would like a hand interpreting the results come chat with us in #serverless-agent in the internal DataDog slack or in #serverless in the public DataDog slack. We're happy to help!
Benchmark stats
goos: linux
goarch: amd64
pkg: github.com/DataDog/datadog-agent/pkg/serverless/daemon
cpu: AMD EPYC 7763 64-Core Processor
│ baseline/benchmark.log │ current/benchmark.log │
│ sec/op │ sec/op vs base │
api-gateway-appsec.json 87.24µ ± 3% 86.75µ ± 4% ~ (p=0.393 n=10)
api-gateway-kong-appsec.json 69.53µ ± 3% 65.54µ ± 1% -5.73% (p=0.000 n=10)
api-gateway-kong.json 67.30µ ± 2% 65.47µ ± 1% -2.71% (p=0.000 n=10)
api-gateway-non-proxy-async.json 108.9µ ± 3% 104.8µ ± 2% -3.75% (p=0.000 n=10)
api-gateway-non-proxy.json 108.5µ ± 1% 106.3µ ± 2% -2.02% (p=0.015 n=10)
api-gateway-websocket-connect.json 71.30µ ± 1% 70.14µ ± 1% -1.63% (p=0.015 n=10)
api-gateway-websocket-default.json 63.73µ ± 1% 63.68µ ± 1% ~ (p=0.971 n=10)
api-gateway-websocket-disconnect.json 63.66µ ± 1% 63.87µ ± 2% ~ (p=0.579 n=10)
api-gateway.json 116.9µ ± 1% 115.9µ ± 1% ~ (p=0.190 n=10)
application-load-balancer.json 64.69µ ± 4% 64.86µ ± 1% ~ (p=0.670 n=10)
cloudfront.json 47.38µ ± 4% 48.38µ ± 1% ~ (p=0.165 n=10)
cloudwatch-events.json 38.46µ ± 1% 38.91µ ± 2% +1.19% (p=0.011 n=10)
cloudwatch-logs.json 66.14µ ± 1% 65.71µ ± 2% ~ (p=0.105 n=10)
custom.json 31.55µ ± 1% 32.09µ ± 1% +1.72% (p=0.000 n=10)
dynamodb.json 96.22µ ± 2% 95.09µ ± 2% ~ (p=0.393 n=10)
empty.json 30.01µ ± 1% 30.27µ ± 2% ~ (p=0.218 n=10)
eventbridge-custom.json 42.83µ ± 2% 43.22µ ± 2% ~ (p=0.190 n=10)
http-api.json 73.43µ ± 1% 73.48µ ± 2% ~ (p=0.529 n=10)
kinesis-batch.json 72.68µ ± 2% 71.84µ ± 1% ~ (p=0.143 n=10)
kinesis.json 55.25µ ± 1% 55.41µ ± 1% ~ (p=0.353 n=10)
s3.json 61.04µ ± 2% 59.85µ ± 2% -1.94% (p=0.043 n=10)
sns-batch.json 92.64µ ± 2% 91.41µ ± 1% -1.33% (p=0.004 n=10)
sns.json 66.45µ ± 2% 65.83µ ± 2% ~ (p=0.123 n=10)
snssqs.json 113.3µ ± 2% 114.2µ ± 2% ~ (p=0.631 n=10)
snssqs_no_dd_context.json 100.3µ ± 2% 100.7µ ± 2% ~ (p=0.529 n=10)
sqs-aws-header.json 56.64µ ± 1% 57.31µ ± 2% ~ (p=0.089 n=10)
sqs-batch.json 96.19µ ± 4% 94.77µ ± 1% -1.47% (p=0.015 n=10)
sqs.json 70.70µ ± 1% 70.09µ ± 1% ~ (p=0.063 n=10)
sqs_no_dd_context.json 64.72µ ± 3% 64.81µ ± 4% ~ (p=0.796 n=10)
geomean 68.27µ 67.87µ -0.59%
│ baseline/benchmark.log │ current/benchmark.log │
│ B/op │ B/op vs base │
api-gateway-appsec.json 37.26Ki ± 0% 37.32Ki ± 0% +0.17% (p=0.000 n=10)
api-gateway-kong-appsec.json 26.92Ki ± 0% 26.92Ki ± 0% ~ (p=0.926 n=10)
api-gateway-kong.json 24.42Ki ± 0% 24.42Ki ± 0% ~ (p=0.753 n=10)
api-gateway-non-proxy-async.json 48.05Ki ± 0% 48.09Ki ± 0% +0.09% (p=0.000 n=10)
api-gateway-non-proxy.json 47.26Ki ± 0% 47.31Ki ± 0% +0.10% (p=0.000 n=10)
api-gateway-websocket-connect.json 25.47Ki ± 0% 25.50Ki ± 0% +0.14% (p=0.000 n=10)
api-gateway-websocket-default.json 21.37Ki ± 0% 21.41Ki ± 0% +0.19% (p=0.000 n=10)
api-gateway-websocket-disconnect.json 21.15Ki ± 0% 21.19Ki ± 0% +0.17% (p=0.000 n=10)
api-gateway.json 49.56Ki ± 0% 49.56Ki ± 0% ~ (p=0.093 n=10)
application-load-balancer.json 22.34Ki ± 0% 23.28Ki ± 0% +4.20% (p=0.000 n=10)
cloudfront.json 17.65Ki ± 0% 17.66Ki ± 0% +0.09% (p=0.000 n=10)
cloudwatch-events.json 11.69Ki ± 0% 11.72Ki ± 0% +0.25% (p=0.000 n=10)
cloudwatch-logs.json 53.37Ki ± 0% 53.36Ki ± 0% ~ (p=0.985 n=10)
custom.json 9.720Ki ± 0% 9.723Ki ± 0% ~ (p=0.469 n=10)
dynamodb.json 40.78Ki ± 0% 40.80Ki ± 0% ~ (p=0.110 n=10)
empty.json 9.277Ki ± 0% 9.295Ki ± 0% ~ (p=0.085 n=10)
eventbridge-custom.json 13.41Ki ± 0% 13.46Ki ± 0% +0.36% (p=0.009 n=10)
http-api.json 23.73Ki ± 0% 23.78Ki ± 0% +0.21% (p=0.000 n=10)
kinesis-batch.json 27.02Ki ± 0% 27.01Ki ± 0% ~ (p=0.271 n=10)
kinesis.json 17.78Ki ± 0% 17.84Ki ± 0% +0.34% (p=0.002 n=10)
s3.json 20.37Ki ± 0% 20.37Ki ± 0% ~ (p=0.671 n=10)
sns-batch.json 38.70Ki ± 0% 38.69Ki ± 0% ~ (p=0.810 n=10)
sns.json 24.00Ki ± 0% 23.97Ki ± 0% ~ (p=0.256 n=10)
snssqs.json 50.77Ki ± 0% 50.81Ki ± 0% ~ (p=0.218 n=10)
snssqs_no_dd_context.json 44.81Ki ± 0% 44.87Ki ± 0% +0.12% (p=0.007 n=10)
sqs-aws-header.json 18.81Ki ± 0% 18.90Ki ± 0% +0.51% (p=0.005 n=10)
sqs-batch.json 41.65Ki ± 0% 41.58Ki ± 0% ~ (p=0.075 n=10)
sqs.json 25.55Ki ± 1% 25.57Ki ± 0% ~ (p=0.481 n=10)
sqs_no_dd_context.json 20.69Ki ± 1% 20.73Ki ± 1% ~ (p=0.529 n=10)
geomean 25.71Ki 25.77Ki +0.25%
│ baseline/benchmark.log │ current/benchmark.log │
│ allocs/op │ allocs/op vs base │
api-gateway-appsec.json 629.5 ± 0% 629.5 ± 0% ~ (p=1.000 n=10)
api-gateway-kong-appsec.json 488.0 ± 0% 488.0 ± 0% ~ (p=1.000 n=10) ¹
api-gateway-kong.json 466.0 ± 0% 466.0 ± 0% ~ (p=1.000 n=10)
api-gateway-non-proxy-async.json 726.0 ± 0% 725.5 ± 0% ~ (p=0.650 n=10)
api-gateway-non-proxy.json 716.0 ± 0% 716.0 ± 0% ~ (p=1.000 n=10) ¹
api-gateway-websocket-connect.json 453.0 ± 0% 453.0 ± 0% ~ (p=1.000 n=10)
api-gateway-websocket-default.json 379.0 ± 0% 379.0 ± 0% ~ (p=0.474 n=10)
api-gateway-websocket-disconnect.json 370.0 ± 0% 370.0 ± 0% ~ (p=1.000 n=10)
api-gateway.json 791.0 ± 0% 791.0 ± 0% ~ (p=0.582 n=10)
application-load-balancer.json 352.0 ± 0% 353.0 ± 0% +0.28% (p=0.000 n=10)
cloudfront.json 284.0 ± 0% 284.0 ± 0% ~ (p=1.000 n=10) ¹
cloudwatch-events.json 220.0 ± 0% 220.0 ± 0% ~ (p=1.000 n=10)
cloudwatch-logs.json 215.5 ± 0% 215.0 ± 0% ~ (p=1.000 n=10)
custom.json 168.0 ± 0% 168.0 ± 1% ~ (p=1.000 n=10)
dynamodb.json 589.0 ± 0% 589.0 ± 0% ~ (p=0.837 n=10)
empty.json 160.0 ± 1% 160.0 ± 1% ~ (p=1.000 n=10)
eventbridge-custom.json 254.0 ± 0% 254.0 ± 0% ~ (p=0.582 n=10)
http-api.json 433.0 ± 0% 432.0 ± 0% ~ (p=0.656 n=10)
kinesis-batch.json 391.0 ± 0% 390.0 ± 0% -0.26% (p=0.000 n=10)
kinesis.json 285.0 ± 0% 285.0 ± 0% ~ (p=0.224 n=10)
s3.json 358.0 ± 0% 358.0 ± 0% ~ (p=0.667 n=10)
sns-batch.json 455.0 ± 0% 455.0 ± 0% ~ (p=0.423 n=10)
sns.json 323.0 ± 0% 323.0 ± 0% ~ (p=0.440 n=10)
snssqs.json 450.0 ± 0% 450.0 ± 0% ~ (p=0.732 n=10)
snssqs_no_dd_context.json 399.0 ± 0% 399.5 ± 0% ~ (p=0.386 n=10)
sqs-aws-header.json 274.0 ± 0% 275.0 ± 0% +0.36% (p=0.013 n=10)
sqs-batch.json 504.0 ± 0% 503.0 ± 0% ~ (p=0.117 n=10)
sqs.json 351.0 ± 1% 351.0 ± 0% ~ (p=0.578 n=10)
sqs_no_dd_context.json 324.5 ± 1% 324.5 ± 0% ~ (p=0.955 n=10)
geomean 376.8 376.8 -0.01%
¹ all samples are equal
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
[Fast Unit Tests Report]
On pipeline 44304380 (CI Visibility). The following jobs did not run any unit tests:
Jobs:
- tests_deb-arm64-py3
- tests_deb-x64-py3
- tests_flavor_dogstatsd_deb-x64
- tests_flavor_heroku_deb-x64
- tests_flavor_iot_deb-x64
- tests_rpm-arm64-py3
- tests_rpm-x64-py3
- tests_windows-x64
If you modified Go files and expected unit tests to run in these jobs, please double check the job logs. If you think tests should have been executed reach out to #agent-devx-help