fix(observability): propagate tracing span context in stream sink request building

[neuronull]

Currently any vector logs or internal metrics that arise during the request building of stream sinks are missing automatic component tags.

This was discovered while implementing a sink validator for the component validation framework, for the internal metric component_discarded_events_total (#16847)

before:

2024-01-25T21:59:37.955554Z ERROR vector::internal_events::codecs: Failed serializing frame. error=LogEvent does not contain required field: "host" error_code="encoder_serialize" error_type="encoder_failed" stage="sending" internal_log_rate_limit=true
2024-01-25T21:59:37.955678Z ERROR vector_common::internal_event::component_events_dropped: Events dropped intentional=false count=1 reason="Failed serializing frame." internal_log_rate_limit=true
2024-01-25T21:59:37.956047Z ERROR sink{component_kind="sink" component_id=test_sink component_type=http}: vector::internal_events::common: Failed to build request. error=unable to encode event error_type="encoder_failed" stage="processing" internal_log_rate_limit=true

after:

2024-01-25T22:18:24.972116Z ERROR sink{component_kind="sink" component_id=test_sink component_type=http}: vector::internal_events::codecs: Failed serializing frame. error=LogEvent does not contain required field: "host" error_code="encoder_serialize" error_type="encoder_failed" stage="sending" internal_log_rate_limit=true
2024-01-25T22:18:24.972213Z ERROR sink{component_kind="sink" component_id=test_sink component_type=http}: vector_common::internal_event::component_events_dropped: Events dropped intentional=false count=1 reason="Failed serializing frame." internal_log_rate_limit=true
2024-01-25T22:18:24.972405Z ERROR sink{component_kind="sink" component_id=test_sink component_type=http}: vector::internal_events::common: Failed to build request. error=unable to encode event error_type="encoder_failed" stage="processing" internal_log_rate_limit=true

[dsmith3197]

nice find!

[datadog-vectordotdev]

Datadog Report

Branch report: neuronull/fix_missing_component_automatic_tags_sink_request_builder
Commit report: daae643
Test service: vector

✅ 0 Failed, 2291 Passed, 0 Skipped, 28m 51.89s Wall Time

[jszwedko]

Marking as request changes just to avoid queuing until my comment is resolved.

Good find though!

[jszwedko]

Perfect, thank you for updating the changelog!

[github-actions]

Regression Detector Results

Run ID: ad2328d5-e78c-4845-b3a9-a3b4378177b6
Baseline: cacb44f
Comparison: 88adcb9
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.

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI
➖	syslog_log2metric_humio_metrics	ingress throughput	+2.40	[+2.23, +2.57]
➖	syslog_splunk_hec_logs	ingress throughput	+2.29	[+2.20, +2.39]
➖	syslog_log2metric_tag_cardinality_limit_blackhole	ingress throughput	+1.84	[+1.73, +1.95]
➖	otlp_http_to_blackhole	ingress throughput	+1.54	[+1.38, +1.70]
➖	http_to_http_acks	ingress throughput	+0.77	[-0.55, +2.08]
➖	syslog_log2metric_splunk_hec_metrics	ingress throughput	+0.48	[+0.34, +0.63]
➖	file_to_blackhole	egress throughput	+0.46	[-2.03, +2.95]
➖	http_text_to_http_json	ingress throughput	+0.42	[+0.28, +0.56]
➖	http_to_http_noack	ingress throughput	+0.22	[+0.12, +0.31]
➖	splunk_hec_route_s3	ingress throughput	+0.21	[-0.28, +0.71]
➖	otlp_grpc_to_blackhole	ingress throughput	+0.05	[-0.04, +0.13]
➖	http_to_http_json	ingress throughput	+0.04	[-0.04, +0.12]
➖	syslog_loki	ingress throughput	+0.02	[-0.06, +0.10]
➖	splunk_hec_to_splunk_hec_logs_acks	ingress throughput	-0.00	[-0.16, +0.16]
➖	splunk_hec_indexer_ack_blackhole	ingress throughput	-0.00	[-0.15, +0.14]
➖	splunk_hec_to_splunk_hec_logs_noack	ingress throughput	-0.04	[-0.15, +0.08]
➖	http_to_s3	ingress throughput	-0.07	[-0.35, +0.21]
➖	enterprise_http_to_http	ingress throughput	-0.11	[-0.17, -0.04]
➖	datadog_agent_remap_blackhole	ingress throughput	-0.23	[-0.32, -0.14]
➖	datadog_agent_remap_blackhole_acks	ingress throughput	-0.27	[-0.38, -0.15]
➖	http_elasticsearch	ingress throughput	-0.41	[-0.48, -0.34]
➖	fluent_elasticsearch	ingress throughput	-0.55	[-1.02, -0.08]
➖	datadog_agent_remap_datadog_logs	ingress throughput	-0.65	[-0.74, -0.56]
➖	datadog_agent_remap_datadog_logs_acks	ingress throughput	-0.86	[-0.95, -0.78]
➖	syslog_humio_logs	ingress throughput	-1.64	[-1.75, -1.53]
➖	syslog_regex_logs2metric_ddmetrics	ingress throughput	-2.89	[-3.02, -2.76]
➖	socket_to_socket_blackhole	ingress throughput	-3.50	[-3.56, -3.43]

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".

[github-actions]

Regression Detector Results

Run ID: e9fd5905-8f00-4386-a981-10c90e70daa9
Baseline: acc9bbf
Comparison: 8aa4022
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

Significant changes in experiment optimization goals

Confidence level: 90.00%
Effect size tolerance: |Δ mean %| ≥ 5.00%

perf	experiment	goal	Δ mean %	Δ mean % CI
✅	http_elasticsearch	ingress throughput	+5.20	[+5.12, +5.27]

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI
✅	http_elasticsearch	ingress throughput	+5.20	[+5.12, +5.27]
➖	otlp_http_to_blackhole	ingress throughput	+4.30	[+4.15, +4.45]
➖	syslog_log2metric_humio_metrics	ingress throughput	+2.73	[+2.59, +2.87]
➖	http_to_s3	ingress throughput	+1.40	[+1.11, +1.68]
➖	otlp_grpc_to_blackhole	ingress throughput	+0.77	[+0.68, +0.85]
➖	syslog_regex_logs2metric_ddmetrics	ingress throughput	+0.68	[+0.57, +0.80]
➖	http_to_http_acks	ingress throughput	+0.56	[-0.75, +1.87]
➖	syslog_log2metric_splunk_hec_metrics	ingress throughput	+0.52	[+0.37, +0.66]
➖	splunk_hec_route_s3	ingress throughput	+0.41	[-0.09, +0.91]
➖	fluent_elasticsearch	ingress throughput	+0.38	[-0.10, +0.86]
➖	syslog_loki	ingress throughput	+0.36	[+0.31, +0.41]
➖	syslog_splunk_hec_logs	ingress throughput	+0.29	[+0.22, +0.35]
➖	socket_to_socket_blackhole	ingress throughput	+0.22	[+0.13, +0.31]
➖	syslog_log2metric_tag_cardinality_limit_blackhole	ingress throughput	+0.17	[+0.04, +0.30]
➖	http_to_http_noack	ingress throughput	+0.12	[+0.01, +0.23]
➖	http_to_http_json	ingress throughput	+0.04	[-0.04, +0.11]
➖	splunk_hec_to_splunk_hec_logs_acks	ingress throughput	-0.00	[-0.15, +0.15]
➖	splunk_hec_indexer_ack_blackhole	ingress throughput	-0.00	[-0.15, +0.14]
➖	datadog_agent_remap_blackhole	ingress throughput	-0.01	[-0.11, +0.09]
➖	splunk_hec_to_splunk_hec_logs_noack	ingress throughput	-0.05	[-0.17, +0.06]
➖	http_text_to_http_json	ingress throughput	-0.07	[-0.20, +0.06]
➖	enterprise_http_to_http	ingress throughput	-0.10	[-0.18, -0.03]
➖	datadog_agent_remap_datadog_logs_acks	ingress throughput	-0.43	[-0.51, -0.35]
➖	file_to_blackhole	egress throughput	-0.55	[-3.08, +1.98]
➖	datadog_agent_remap_datadog_logs	ingress throughput	-0.85	[-0.95, -0.75]
➖	syslog_humio_logs	ingress throughput	-0.90	[-1.00, -0.80]
➖	datadog_agent_remap_blackhole_acks	ingress throughput	-1.63	[-1.75, -1.52]

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".

[github-actions]

Regression Detector Results

Run ID: f8a62ece-29dc-4503-b224-ed448d52324d
Baseline: cacb44f
Comparison: 4195071
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.

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI
➖	syslog_log2metric_humio_metrics	ingress throughput	+2.31	[+2.19, +2.44]
➖	syslog_regex_logs2metric_ddmetrics	ingress throughput	+1.89	[+1.76, +2.01]
➖	syslog_log2metric_splunk_hec_metrics	ingress throughput	+1.54	[+1.40, +1.69]
➖	syslog_splunk_hec_logs	ingress throughput	+1.17	[+1.09, +1.24]
➖	splunk_hec_route_s3	ingress throughput	+0.98	[+0.48, +1.47]
➖	syslog_log2metric_tag_cardinality_limit_blackhole	ingress throughput	+0.92	[+0.79, +1.04]
➖	datadog_agent_remap_blackhole_acks	ingress throughput	+0.63	[+0.54, +0.73]
➖	syslog_loki	ingress throughput	+0.44	[+0.38, +0.51]
➖	http_to_http_acks	ingress throughput	+0.43	[-0.90, +1.76]
➖	http_to_http_noack	ingress throughput	+0.13	[+0.04, +0.23]
➖	http_to_http_json	ingress throughput	+0.06	[-0.02, +0.14]
➖	otlp_http_to_blackhole	ingress throughput	+0.03	[-0.12, +0.19]
➖	datadog_agent_remap_datadog_logs	ingress throughput	+0.01	[-0.07, +0.09]
➖	splunk_hec_to_splunk_hec_logs_acks	ingress throughput	+0.00	[-0.15, +0.16]
➖	splunk_hec_indexer_ack_blackhole	ingress throughput	-0.00	[-0.14, +0.14]
➖	splunk_hec_to_splunk_hec_logs_noack	ingress throughput	-0.02	[-0.14, +0.10]
➖	fluent_elasticsearch	ingress throughput	-0.07	[-0.54, +0.40]
➖	http_to_s3	ingress throughput	-0.16	[-0.44, +0.12]
➖	enterprise_http_to_http	ingress throughput	-0.17	[-0.25, -0.09]
➖	syslog_humio_logs	ingress throughput	-0.28	[-0.39, -0.17]
➖	otlp_grpc_to_blackhole	ingress throughput	-0.31	[-0.39, -0.22]
➖	datadog_agent_remap_blackhole	ingress throughput	-0.58	[-0.68, -0.48]
➖	datadog_agent_remap_datadog_logs_acks	ingress throughput	-0.62	[-0.70, -0.54]
➖	http_elasticsearch	ingress throughput	-0.64	[-0.71, -0.57]
➖	file_to_blackhole	egress throughput	-0.97	[-3.43, +1.50]
➖	http_text_to_http_json	ingress throughput	-1.73	[-1.85, -1.61]
➖	socket_to_socket_blackhole	ingress throughput	-1.94	[-2.02, -1.86]

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".