feat(codec,sources): influxdb line protcol decoder

[MichaHoffmann]

address #11337

[pront]

Hello @MichaHoffmann, thank you for this PR! I know this is a late response but this fell through the cracks. Left a few comments.

[pront]

Can we add a test per metric type?

[sebinsunny]

Based on this comment, it should be okay to use the gauge data type for Influx's 'measurement,' right? Since we don't have encoding for InfluxDB output.

[pront]

Should we treat other metric types as an error?

[sebinsunny]

@pront Sorry to get back to you. Do we need that? The format for Influx line protocol is like this and doesn't specify any metric types: cpu,host=A,region=west usage_system=64i,usage_user=10i 1590488773254420000. We're just parsing it into the gauge metric type. Since the input is always like this, does the metric type matter? I fixed the formatting issue. In the upcoming PR, I can add support for InfluxDB encoding so we can specify the metric type tag and support all metric types. Is it okay to add encoding for InfluxDB line protocol?

[pront]

Also, please add a changelog fragment here: https://github.com/vectordotdev/vector/tree/master/changelog.d

[RedForza]

Hello! Just stumbled onto this PR. I am very interested into Influx line protocol support for vector and looking at the issues list others are too.

@MichaHoffmann Do you have a timeline when the outstanding comments could be resolved? Appreciate the work done so far.

[MichaHoffmann]

Hello! Just stumbled onto this PR. I am very interested into Influx line protocol support for vector and looking at the issues list others are too.

@MichaHoffmann Do you have a timeline when the outstanding comments could be resolved? Appreciate the work done so far.

Ah I didnt continue to work on it because I had the impression that we cannot use kafka as datasource since it forces the type to be "log". Ever since I have changed employers and since this was work related, I didnt think about it much anymore. I dont want to squat this issue either, if someone wants to pick it up, thats would be perfect.

[sebinsunny]

Also, please add a changelog fragment here: https://github.com/vectordotdev/vector/tree/master/changelog.d

@pront added the changelog, please review this PR, thanks

[jszwedko]

Thanks for this contribution!

[jszwedko]

@sebinsunny thanks for iterating on this. A small request to try to avoid force pushing. It makes viewing the change history more difficult.

[jszwedko]

(the commits will all be squashed during merge)

[sebinsunny]

A small request to try to avoid force pushing. It makes viewing the change history more difficult.

@jszwedko Sorry about this. The pipeline was failing due to missing docs, but it has been fixed now. Hope pipeline will be green now.

[pront]

A small request to try to avoid force pushing. It makes viewing the change history more difficult.

@jszwedko Sorry about this. The pipeline was failing due to missing docs, but it has been fixed now. Hope pipeline will be green now.

👍 re-running CI checks

[estherk15]

Small nit: Decodes the raw bytes as an [Influxdb Line Protocol][influxdb] message.

[estherk15]

Suggested change

	Decodes the raw bytes as a [Influxdb Line Protocol][influxdb] message.
	Decodes the raw bytes as an [Influxdb Line Protocol][influxdb] message.

[sebinsunny]

fixed the typo

[sebinsunny]

forced merge to resolve conflict

[sebinsunny]

Fixed the conflicts correctly. Earlier, the old reference for DeserializerConfig::Gelf(_) was added in DeserializerConfig::Bytes, which caused the test failure. I have now corrected the conflict.

[github-actions]

Regression Detector Results

Run ID: f0ecfe7a-d839-4b8d-ade3-2b580a904964 Metrics dashboard

Baseline: 787e1ec
Comparison: edbe7ca

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	links
❌	splunk_hec_route_s3	ingress throughput	-5.57	[-5.88, -5.26]

Experiments ignored for regressions

Regressions in experiments with settings containing erratic: true are ignored.

perf	experiment	goal	Δ mean %	Δ mean % CI	links
❌	file_to_blackhole	egress throughput	-7.73	[-14.39, -1.08]

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	links
➖	otlp_http_to_blackhole	ingress throughput	+2.90	[+2.72, +3.07]
➖	socket_to_socket_blackhole	ingress throughput	+1.78	[+1.71, +1.85]
➖	datadog_agent_remap_datadog_logs_acks	ingress throughput	+1.02	[+0.80, +1.23]
➖	otlp_grpc_to_blackhole	ingress throughput	+0.93	[+0.81, +1.05]
➖	http_to_s3	ingress throughput	+0.26	[-0.00, +0.53]
➖	datadog_agent_remap_datadog_logs	ingress throughput	+0.25	[+0.05, +0.45]
➖	syslog_splunk_hec_logs	ingress throughput	+0.06	[-0.03, +0.15]
➖	splunk_hec_to_splunk_hec_logs_acks	ingress throughput	+0.00	[-0.09, +0.10]
➖	splunk_hec_to_splunk_hec_logs_noack	ingress throughput	-0.02	[-0.12, +0.08]
➖	http_to_http_noack	ingress throughput	-0.02	[-0.14, +0.10]
➖	splunk_hec_indexer_ack_blackhole	ingress throughput	-0.03	[-0.12, +0.06]
➖	http_to_http_json	ingress throughput	-0.12	[-0.19, -0.06]
➖	datadog_agent_remap_blackhole	ingress throughput	-0.39	[-0.51, -0.28]
➖	syslog_loki	ingress throughput	-0.74	[-0.82, -0.67]
➖	http_elasticsearch	ingress throughput	-0.96	[-1.19, -0.73]
➖	http_to_http_acks	ingress throughput	-1.22	[-2.53, +0.10]
➖	syslog_regex_logs2metric_ddmetrics	ingress throughput	-1.33	[-1.51, -1.16]
➖	syslog_log2metric_tag_cardinality_limit_blackhole	ingress throughput	-1.51	[-1.61, -1.42]
➖	fluent_elasticsearch	ingress throughput	-1.63	[-2.14, -1.13]
➖	http_text_to_http_json	ingress throughput	-2.02	[-2.17, -1.88]
➖	syslog_log2metric_humio_metrics	ingress throughput	-2.05	[-2.18, -1.91]
➖	datadog_agent_remap_blackhole_acks	ingress throughput	-2.94	[-3.05, -2.84]
➖	syslog_humio_logs	ingress throughput	-3.06	[-3.18, -2.94]
➖	syslog_log2metric_splunk_hec_metrics	ingress throughput	-3.06	[-3.17, -2.95]
❌	splunk_hec_route_s3	ingress throughput	-5.57	[-5.88, -5.26]
❌	file_to_blackhole	egress throughput	-7.73	[-14.39, -1.08]

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: 3d125c8b-0cd2-4ec8-b31f-7357020832ba Metrics dashboard

Baseline: 56bd0dd
Comparison: 3fbb66c

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.

Experiments ignored for regressions

Regressions in experiments with settings containing erratic: true are ignored.

perf	experiment	goal	Δ mean %	Δ mean % CI	links
❌	file_to_blackhole	egress throughput	-14.31	[-20.24, -8.38]

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	links
➖	splunk_hec_route_s3	ingress throughput	+3.06	[+2.75, +3.37]
➖	syslog_splunk_hec_logs	ingress throughput	+3.05	[+2.92, +3.19]
➖	syslog_log2metric_splunk_hec_metrics	ingress throughput	+1.57	[+1.46, +1.67]
➖	syslog_log2metric_tag_cardinality_limit_blackhole	ingress throughput	+1.25	[+1.15, +1.34]
➖	datadog_agent_remap_datadog_logs	ingress throughput	+1.15	[+0.92, +1.38]
➖	http_elasticsearch	ingress throughput	+0.92	[+0.73, +1.12]
➖	http_text_to_http_json	ingress throughput	+0.79	[+0.67, +0.91]
➖	http_to_s3	ingress throughput	+0.79	[+0.53, +1.05]
➖	fluent_elasticsearch	ingress throughput	+0.78	[+0.28, +1.29]
➖	splunk_hec_indexer_ack_blackhole	ingress throughput	+0.06	[-0.02, +0.14]
➖	splunk_hec_to_splunk_hec_logs_noack	ingress throughput	+0.01	[-0.09, +0.11]
➖	splunk_hec_to_splunk_hec_logs_acks	ingress throughput	+0.01	[-0.12, +0.13]
➖	http_to_http_noack	ingress throughput	-0.08	[-0.19, +0.04]
➖	syslog_log2metric_humio_metrics	ingress throughput	-0.08	[-0.26, +0.10]
➖	http_to_http_json	ingress throughput	-0.14	[-0.22, -0.07]
➖	otlp_grpc_to_blackhole	ingress throughput	-0.69	[-0.81, -0.56]
➖	http_to_http_acks	ingress throughput	-0.75	[-2.07, +0.56]
➖	otlp_http_to_blackhole	ingress throughput	-0.81	[-0.97, -0.65]
➖	datadog_agent_remap_blackhole_acks	ingress throughput	-0.90	[-1.00, -0.79]
➖	datadog_agent_remap_blackhole	ingress throughput	-0.97	[-1.07, -0.87]
➖	syslog_loki	ingress throughput	-1.35	[-1.43, -1.28]
➖	datadog_agent_remap_datadog_logs_acks	ingress throughput	-1.38	[-1.55, -1.20]
➖	syslog_humio_logs	ingress throughput	-2.23	[-2.35, -2.11]
➖	socket_to_socket_blackhole	ingress throughput	-2.73	[-2.80, -2.66]
➖	syslog_regex_logs2metric_ddmetrics	ingress throughput	-3.20	[-3.37, -3.03]
❌	file_to_blackhole	egress throughput	-14.31	[-20.24, -8.38]

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".