Add EventBridge Inferred Spans

[nhulston]

What does this PR do?

• This PR fixes/adds EventBridge inferred spans for Lambda.
• Specifically, this PR supports inferred spans for Lambda -> EventBridge -> Lambda, Lambda -> EventBridge -> SQS -> Lambda, and Lambda -> EventBridge -> SNS -> Lambda. These are the most common use cases that customers want supported.

Motivation

SNS and SQS are already supported. There was some code in place for EventBridge, but it didn't work properly. Even the simplest case, Lambda -> EventBridge -> Lambda, didn't show a span for EventBridge. This PR aims to fix this, and handle other common cases like EventBridge -> SQS and EventBridge -> SNS.

Additional Notes

• I have a PR (here) to inject trace context into the Java tracer. At the time of writing this, the tracers for .NET and Go do not inject trace context into SQS/SNS/EventBridge, but that's my next project.

• The AWS API has no reliable way of getting the start time of the EventBridge span. It only gives time with precision to the second, which is not precise enough to create spans. Therefore, we try to get the start time from the tracer in the field SentTimestamp. If this field does not exist, we fall back to AWS API's start time.

• Also, the AWS API has no way of getting the EventBridge bus name, so we attempt to get this from the tracer in the field BusName for the span resource name. If this doesn't exist, we set the resource name to "EventBridge".

• This should not be a breaking change, but reviewers should double check my changes and keep this in mind. If the user is using an outdated tracer or .NET/Go that doesn't yet support inferred spans for EB/SQS/SNS, these changes shouldn't break anything, but it should show an "EventBridge" span on any Lambda function invoked by the EventBridge event:

Traces before these changes

Lambda --> EventBridge --> Lambda Two different traces. Second trace is missing an EventBridge span

Lambda --> EventBridge --> SQS --> Lambda Missing EventBridge span

Lambda --> EventBridge --> SNS --> Lambda Missing EventBridge span

Traces after these changes

Lambda --> EventBridge --> Lambda

Lambda --> EventBridge --> SQS --> Lambda

Lambda --> EventBridge --> SNS --> Lambda

Possible Drawbacks / Trade-offs

Describe how to test/QA your changes

Run local tests with invoke test --targets=./pkg/serverless from the root of the project

It's more complicated to test manually:

1. Create a serverless architecture that's supported, such as a Java Lambda function that publishes an event to EventBridge, an EventBridge event bus that has a target SQS queue or SNS notifier, and another Java Lambda function that subscribes to the queue/notifier.
2. Instrument the lambda functions using datadog-ci
3. Build the extension: ./scripts/publish_sandbox.sh
4. Build the new Java tracer from this PR: ./gradlew publishToMavenLocal && cd ~/.m2/repository/com/datadoghq/dd-java-agent/1.40.0-SNAPSHOT && zip layer.zip dd-java-agent-1.40.0-SNAPSHOT.jar
5. Create new layers on AWS by uploading those zip files, and replace your Lambda layers with your custom layers.
6. Set the environment variable JAVA_TOOL_OPTIONS: -javaagent:"/opt/dd-java-agent-1.40.0-SNAPSHOT.jar" on your lambda functions to use the new Java tracer
7. Invoke your first lambda function, and check Datadog for results. The spans should be combined into a single trace.

[bits-bot]

All committers have signed the CLA.

[pr-commenter[bot]]

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=44525616 --os-family=ubuntu

Note: This applies to commit 1d7799ef

[github-actions[bot]]

Serverless Benchmark Results

BenchmarkStartEndInvocation comparison between c7eb6e5fd9102291c697a89e7ae4150ae93ae56a and c23d241dbaaa05747cbe1477917d67406e089f70.

tl;dr

Use these benchmarks as an insight tool during development.

1. Skim down the vs base column 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.

2. 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                            84.74µ ± 6%    87.03µ ±  4%        ~ (p=0.165 n=10)
api-gateway-kong-appsec.json                       65.47µ ± 1%    68.28µ ±  1%   +4.28% (p=0.000 n=10)
api-gateway-kong.json                              64.24µ ± 1%    67.65µ ±  1%   +5.32% (p=0.000 n=10)
api-gateway-non-proxy-async.json                   102.2µ ± 1%    106.8µ ±  1%   +4.55% (p=0.000 n=10)
api-gateway-non-proxy.json                         102.2µ ± 3%    107.2µ ±  1%   +4.81% (p=0.001 n=10)
api-gateway-websocket-connect.json                 67.49µ ± 1%    71.99µ ±  2%   +6.66% (p=0.000 n=10)
api-gateway-websocket-default.json                 60.98µ ± 1%    64.97µ ±  1%   +6.55% (p=0.000 n=10)
api-gateway-websocket-disconnect.json              61.15µ ± 1%    65.05µ ±  1%   +6.39% (p=0.000 n=10)
api-gateway.json                                   112.8µ ± 0%    119.3µ ±  3%   +5.75% (p=0.000 n=10)
application-load-balancer.json                     61.83µ ± 1%    66.94µ ±  2%   +8.26% (p=0.000 n=10)
cloudfront.json                                    46.60µ ± 2%    50.23µ ±  2%   +7.81% (p=0.000 n=10)
cloudwatch-events.json                             37.80µ ± 2%    39.76µ ±  2%   +5.17% (p=0.000 n=10)
cloudwatch-logs.json                               63.15µ ± 3%    66.54µ ±  1%   +5.36% (p=0.000 n=10)
custom.json                                        30.55µ ± 1%    32.16µ ±  2%   +5.25% (p=0.000 n=10)
dynamodb.json                                      92.81µ ± 0%    96.36µ ±  2%   +3.83% (p=0.000 n=10)
empty.json                                         28.93µ ± 1%    30.35µ ±  2%   +4.90% (p=0.000 n=10)
eventbridge-custom.json                            42.06µ ± 2%    49.04µ ±  3%  +16.61% (p=0.000 n=10)
http-api.json                                      72.48µ ± 1%    75.17µ ±  1%   +3.71% (p=0.000 n=10)
kinesis-batch.json                                 69.61µ ± 1%    74.12µ ±  2%   +6.49% (p=0.000 n=10)
kinesis.json                                       53.62µ ± 1%    56.82µ ±  1%   +5.96% (p=0.000 n=10)
s3.json                                            59.38µ ± 1%    62.89µ ±  2%   +5.90% (p=0.000 n=10)
sns-batch.json                                     89.68µ ± 4%    93.85µ ±  2%   +4.65% (p=0.000 n=10)
sns.json                                           64.23µ ± 1%    69.99µ ±  1%   +8.96% (p=0.000 n=10)
snssqs.json                                        108.6µ ± 1%    113.1µ ±  2%   +4.15% (p=0.000 n=10)
snssqs_no_dd_context.json                          97.95µ ± 1%   104.58µ ±  2%   +6.77% (p=0.000 n=10)
sqs-aws-header.json                                55.08µ ± 2%    63.04µ ±  3%  +14.46% (p=0.000 n=10)
sqs-batch.json                                     93.94µ ± 2%   100.08µ ±  1%   +6.54% (p=0.000 n=10)
sqs.json                                           68.94µ ± 2%    75.93µ ±  2%  +10.14% (p=0.000 n=10)
sqs_no_dd_context.json                             63.18µ ± 5%    69.51µ ±  3%  +10.03% (p=0.000 n=10)
eventbridge-no-bus.json                                           48.54µ ±  3%
eventbridge-no-sent-timestamp.json                                48.16µ ±  2%
eventbridgesns.json                                               70.00µ ±  1%
eventbridgesqs.json                                               81.89µ ± 61%
geomean                                            65.89µ         68.97µ         +6.58%

                                      │ baseline/benchmark.log │        current/benchmark.log         │
                                      │          B/op          │     B/op      vs base                │
api-gateway-appsec.json                           37.25Ki ± 0%   37.33Ki ± 0%   +0.19% (p=0.000 n=10)
api-gateway-kong-appsec.json                      26.92Ki ± 0%   26.92Ki ± 0%        ~ (p=0.239 n=10)
api-gateway-kong.json                             24.41Ki ± 0%   24.42Ki ± 0%        ~ (p=0.306 n=10)
api-gateway-non-proxy-async.json                  48.03Ki ± 0%   48.10Ki ± 0%   +0.14% (p=0.000 n=10)
api-gateway-non-proxy.json                        47.26Ki ± 0%   47.32Ki ± 0%   +0.12% (p=0.000 n=10)
api-gateway-websocket-connect.json                25.45Ki ± 0%   25.50Ki ± 0%   +0.20% (p=0.000 n=10)
api-gateway-websocket-default.json                21.36Ki ± 0%   21.41Ki ± 0%   +0.25% (p=0.000 n=10)
api-gateway-websocket-disconnect.json             21.14Ki ± 0%   21.19Ki ± 0%   +0.23% (p=0.000 n=10)
api-gateway.json                                  49.54Ki ± 0%   49.58Ki ± 0%   +0.08% (p=0.000 n=10)
application-load-balancer.json                    22.33Ki ± 0%   23.29Ki ± 0%   +4.31% (p=0.000 n=10)
cloudfront.json                                   17.63Ki ± 0%   17.68Ki ± 0%   +0.27% (p=0.000 n=10)
cloudwatch-events.json                            11.67Ki ± 0%   11.73Ki ± 0%   +0.49% (p=0.000 n=10)
cloudwatch-logs.json                              53.35Ki ± 0%   53.37Ki ± 0%   +0.04% (p=0.005 n=10)
custom.json                                       9.720Ki ± 0%   9.763Ki ± 0%   +0.44% (p=0.000 n=10)
dynamodb.json                                     40.75Ki ± 0%   40.82Ki ± 0%   +0.15% (p=0.000 n=10)
empty.json                                        9.275Ki ± 0%   9.313Ki ± 0%   +0.42% (p=0.001 n=10)
eventbridge-custom.json                           13.40Ki ± 0%   15.68Ki ± 0%  +16.98% (p=0.000 n=10)
http-api.json                                     23.72Ki ± 0%   23.95Ki ± 0%   +0.96% (p=0.000 n=10)
kinesis-batch.json                                27.00Ki ± 0%   27.17Ki ± 0%   +0.64% (p=0.000 n=10)
kinesis.json                                      17.79Ki ± 1%   18.00Ki ± 0%   +1.21% (p=0.000 n=10)
s3.json                                           20.33Ki ± 0%   20.52Ki ± 0%   +0.92% (p=0.000 n=10)
sns-batch.json                                    38.65Ki ± 0%   39.42Ki ± 0%   +2.00% (p=0.000 n=10)
sns.json                                          23.96Ki ± 0%   24.62Ki ± 0%   +2.76% (p=0.000 n=10)
snssqs.json                                       50.52Ki ± 0%   50.84Ki ± 0%   +0.62% (p=0.000 n=10)
snssqs_no_dd_context.json                         44.77Ki ± 0%   45.28Ki ± 0%   +1.14% (p=0.000 n=10)
sqs-aws-header.json                               18.79Ki ± 0%   19.53Ki ± 1%   +3.94% (p=0.000 n=10)
sqs-batch.json                                    41.59Ki ± 0%   42.39Ki ± 1%   +1.93% (p=0.000 n=10)
sqs.json                                          25.54Ki ± 0%   26.25Ki ± 1%   +2.76% (p=0.000 n=10)
sqs_no_dd_context.json                            20.52Ki ± 1%   21.40Ki ± 1%   +4.27% (p=0.000 n=10)
eventbridge-no-bus.json                                          14.68Ki ± 0%
eventbridge-no-sent-timestamp.json                               14.70Ki ± 0%
eventbridgesns.json                                              23.41Ki ± 0%
eventbridgesqs.json                                              26.97Ki ± 2%
geomean                                           25.68Ki        25.14Ki        +1.59%

                                      │ baseline/benchmark.log │        current/benchmark.log        │
                                      │       allocs/op        │ allocs/op   vs base                 │
api-gateway-appsec.json                             629.0 ± 0%   630.0 ± 0%       ~ (p=0.370 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                    725.0 ± 0%   726.0 ± 0%       ~ (p=0.656 n=10)
api-gateway-non-proxy.json                          716.0 ± 0%   716.0 ± 0%       ~ (p=1.000 n=10)
api-gateway-websocket-connect.json                  453.0 ± 0%   453.0 ± 0%       ~ (p=0.211 n=10)
api-gateway-websocket-default.json                  379.0 ± 0%   379.0 ± 0%       ~ (p=0.303 n=10)
api-gateway-websocket-disconnect.json               369.5 ± 0%   370.0 ± 0%       ~ (p=0.141 n=10)
api-gateway.json                                    790.0 ± 0%   791.0 ± 0%  +0.13% (p=0.023 n=10)
application-load-balancer.json                      352.0 ± 0%   353.0 ± 0%  +0.28% (p=0.000 n=10)
cloudfront.json                                     283.5 ± 0%   284.0 ± 0%  +0.18% (p=0.033 n=10)
cloudwatch-events.json                              220.0 ± 0%   220.0 ± 0%       ~ (p=0.211 n=10)
cloudwatch-logs.json                                215.0 ± 0%   215.5 ± 0%       ~ (p=0.350 n=10)
custom.json                                         168.0 ± 0%   169.0 ± 0%  +0.60% (p=0.000 n=10)
dynamodb.json                                       589.0 ± 0%   589.0 ± 0%       ~ (p=0.124 n=10)
empty.json                                          160.0 ± 1%   160.0 ± 1%       ~ (p=0.628 n=10)
eventbridge-custom.json                             254.0 ± 0%   270.0 ± 0%  +6.30% (p=0.000 n=10)
http-api.json                                       432.0 ± 0%   435.0 ± 0%  +0.69% (p=0.000 n=10)
kinesis-batch.json                                  390.0 ± 0%   392.5 ± 0%  +0.64% (p=0.000 n=10)
kinesis.json                                        285.0 ± 1%   288.0 ± 0%  +1.05% (p=0.000 n=10)
s3.json                                             358.0 ± 0%   360.0 ± 0%  +0.56% (p=0.000 n=10)
sns-batch.json                                      454.5 ± 0%   468.0 ± 0%  +2.97% (p=0.000 n=10)
sns.json                                            323.0 ± 0%   335.0 ± 1%  +3.72% (p=0.000 n=10)
snssqs.json                                         437.5 ± 0%   440.5 ± 0%  +0.69% (p=0.000 n=10)
snssqs_no_dd_context.json                           399.0 ± 0%   409.0 ± 0%  +2.51% (p=0.000 n=10)
sqs-aws-header.json                                 273.5 ± 1%   288.5 ± 1%  +5.48% (p=0.000 n=10)
sqs-batch.json                                      503.0 ± 0%   518.5 ± 1%  +3.08% (p=0.000 n=10)
sqs.json                                            350.5 ± 0%   365.0 ± 1%  +4.14% (p=0.000 n=10)
sqs_no_dd_context.json                              322.5 ± 1%   338.5 ± 1%  +4.96% (p=0.000 n=10)
eventbridge-no-bus.json                                          261.0 ± 0%
eventbridge-no-sent-timestamp.json                               261.0 ± 0%
eventbridgesns.json                                              369.0 ± 0%
eventbridgesqs.json                                              387.5 ± 2%
geomean                                             376.1        372.3       +1.32%
¹ all samples are equal

[pr-commenter[bot]]

Regression Detector

Regression Detector Results

Run ID: b2ddef8d-7fac-4d4e-b9f7-9315f527e4ec Metrics dashboard Target profiles

Baseline: c7eb6e5fd9102291c697a89e7ae4150ae93ae56a Comparison: 1d7799ef0e67b80715f5c09580063d09cc267a0f

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
➖	pycheck_lots_of_tags	% cpu utilization	+1.52	[-1.24, +4.28]	1	Logs
➖	basic_py_check	% cpu utilization	+1.32	[-1.61, +4.25]	1	Logs
➖	idle	memory utilization	+0.04	[-0.00, +0.09]	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
➖	uds_dogstatsd_to_api_cpu	% cpu utilization	-0.03	[-0.78, +0.73]	1	Logs
➖	file_tree	memory utilization	-0.13	[-0.25, -0.01]	1	Logs
➖	otel_to_otel_logs	ingress throughput	-0.21	[-1.04, +0.61]	1	Logs
➖	tcp_syslog_to_blackhole	ingress throughput	-0.44	[-0.49, -0.39]	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:

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".