[HA] add support for HA configuration in subsection and expose via API

[truthbk]

What does this PR do?

This PR adds an experimental HA config subsection and said configuration is exposed via an API endpoint for potential consumption by other processes.

Motivation

Ease of use of configuration of alternative endpoints for HA support.

Additional Notes

This is an initial and experimental approach to introduce HA support configuration semantics in the agent. Depending on future development decisions additional logic may be added to implement forwarding logic in the agent (or not).

Possible Drawbacks / Trade-offs

Describe how to test/QA your changes

The HA semantics are added in a new ha subsection to the configuration template. Please refer to the internal RFC available for a full description of the options available. Environment variables are indeed available to override configuration file definitions. To test the features we should test:

1. Config file definitions are correctly picked up.
2. Environment variable overrides (DD_HA_*) supersede config file definitions.
3. CLI commands available allow toggling HA endpoints at runtime.
4. HA configuration is correctly exposed over the API.

CLI changes can be triggered with the runtime config CLI facilities:

datadog-agent config set ha.failover true
datadog-agent config get ha.failover

API can be queried to collect the configuration (and may also be queried via the CLI as an API frontend):

datadog-agent config
datadog-agent config subsection ha

Reviewer's Checklist

• If known, an appropriate milestone has been selected; otherwise the Triage milestone is set.
• Use the major_change label if your change either has a major impact on the code base, is impacting multiple teams or is changing important well-established internals of the Agent. This label will be use during QA to make sure each team pay extra attention to the changed behavior. For any customer facing change use a releasenote.
• A release note has been added or the changelog/no-changelog label has been applied.
• Changed code has automated tests for its functionality.
• Adequate QA/testing plan information is provided. Except if the qa/skip-qa label, with required either qa/done or qa/no-code-change labels, are applied.
• At least one team/.. label has been applied, indicating the team(s) that should QA this change.
• If applicable, docs team has been notified or an issue has been opened on the documentation repo.
• If applicable, the need-change/operator and need-change/helm labels have been applied.
• If applicable, the k8s/<min-version> label, indicating the lowest Kubernetes version compatible with this feature.
• If applicable, the config template has been updated.

[bits-bot]

All committers have signed the CLA.

[pr-commenter]

Bloop Bleep... Dogbot Here

Regression Detector Results

Run ID: 364d6220-473f-4faa-a73d-c508fb6aaf36
Baseline: 5af28a5
Comparison: b21b693
Total CPUs: 7

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.

Declared stable experiments that are now erratic

An experiment is erratic (i.e., not stable) if its coefficient of variation is at least 0.10.

perf	experiment	goal	Δ mean %	Δ mean % CI	confidence
➖	otel_to_otel_logs	ingress throughput	+2.45	[+1.73, +3.18]	100.00%

Declared erratic experiments that are now stable

An experiment is stable (i.e., not erratic) if its coefficient of variation is less than 0.10.

perf	experiment	goal	Δ mean %	Δ mean % CI	confidence
➖	file_tree	memory utilization	+0.99	[+0.89, +1.09]	100.00%
➖	idle	memory utilization	-0.67	[-0.70, -0.65]	100.00%

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	confidence
➖	otel_to_otel_logs	ingress throughput	+2.45	[+1.73, +3.18]	100.00%
➖	tcp_syslog_to_blackhole	ingress throughput	+1.23	[+1.17, +1.30]	100.00%
➖	file_tree	memory utilization	+0.99	[+0.89, +1.09]	100.00%
➖	file_to_blackhole	% cpu utilization	+0.46	[-6.15, +7.07]	9.08%
➖	process_agent_standard_check_with_stats	memory utilization	+0.12	[+0.08, +0.16]	100.00%
➖	dogstatsd_string_interner_64MiB_100	ingress throughput	+0.02	[-0.02, +0.06]	51.93%
➖	dogstatsd_string_interner_128MiB_100	ingress throughput	+0.00	[-0.05, +0.05]	0.00%
➖	dogstatsd_string_interner_8MiB_100	ingress throughput	+0.00	[-0.04, +0.04]	0.00%
➖	dogstatsd_string_interner_64MiB_1k	ingress throughput	+0.00	[-0.04, +0.04]	0.00%
➖	uds_dogstatsd_to_api	ingress throughput	+0.00	[-0.04, +0.04]	0.00%
➖	dogstatsd_string_interner_128MiB_1k	ingress throughput	+0.00	[-0.06, +0.06]	0.00%
➖	dogstatsd_string_interner_8MiB_1k	ingress throughput	-0.00	[-0.04, +0.04]	0.00%
➖	dogstatsd_string_interner_8MiB_10k	ingress throughput	-0.00	[-0.05, +0.05]	0.00%
➖	tcp_dd_logs_filter_exclude	ingress throughput	-0.00	[-0.06, +0.06]	0.00%
➖	dogstatsd_string_interner_8MiB_50k	ingress throughput	-0.00	[-0.04, +0.04]	0.11%
➖	trace_agent_msgpack	ingress throughput	-0.02	[-0.04, +0.01]	68.15%
➖	trace_agent_json	ingress throughput	-0.03	[-0.07, -0.00]	90.30%
➖	dogstatsd_string_interner_8MiB_100k	ingress throughput	-0.05	[-0.07, -0.03]	99.98%
➖	process_agent_real_time_mode	memory utilization	-0.08	[-0.11, -0.05]	100.00%
➖	process_agent_standard_check	memory utilization	-0.36	[-0.42, -0.31]	100.00%
➖	idle	memory utilization	-0.67	[-0.70, -0.65]	100.00%

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".

[tobz]

👍🏻

[hush-hush]

We don't scrub data here ? This would mean exposing sensitive information locally to any process.

We started working on something similar with #21025. The idea was to have a dedicated API for sharing sensitive information between agent processes. A different API allows us to build more restrictive security system in the future. It also allows customer to bind the CLI API on a different addr that the "sensitive API" (which is useful for container setups).

We might want to move the ha endpoint to this new API ?