fix(websocket sink): send encoded message as binary frame

[zhongchen]

fixes: #18037

[bits-bot]

All committers have signed the CLA.

[netlify]

✅ Deploy Preview for vrl-playground canceled.

Name	Link
🔨 Latest commit	fcb8d26
🔍 Latest deploy log	https://app.netlify.com/sites/vrl-playground/deploys/64c03229cc3cb80008a1b301

[netlify]

✅ Deploy Preview for vector-project canceled.

Name	Link
🔨 Latest commit	fcb8d26
🔍 Latest deploy log	https://app.netlify.com/sites/vector-project/deploys/64c032292da9e70008d98a63

[jszwedko]

@zhongchen thanks for contributing this! We've been wondering about use-cases Vector users have for the websocket sink. Would you be able to share yours? That'll help us understand how it is being used in the wild.

[zhongchen]

@jszwedko We have an existing data pipeline where clients open WebSocket connections to the server to send proto-encoded data. We want to introduce Vector to the data flow because Vector can automatically persist buffering data to disk so that we don't need to write custom code on the client side to achieve that. Using the Websocket sink also means we don't need to rewrite anything on the server side.

[jszwedko]

@zhongchen gotcha, thanks! So it sounds like the server you want to send data to only exposes a websocket interface for doing so. Is this a custom server? Or some off-the-shelf software?

[zhongchen]

@jszwedko The server is an In-house solution. We want to stream data from clients in real-time, so we chose to use websocket connections.

[jszwedko]

@jszwedko The server is an In-house solution. We want to stream data from clients in real-time, so we chose to use websocket connections.

Thanks for the context! Just curious, did you see advantages to websockets over HTTP/2 which also supports streaming?

[zhongchen]

@jszwedko The Websocket solution was a legacy solution we inherited and it works fine for us, so we continue using it.

changes made

[neuronull]

Thanks for making the suggested changes @zhongchen! I've triggered CI on the PR.

[zhongchen]

@neuronull Just made an additional to fix the format. cargo fmt -- --check is passing locally. don't know whether the CI needs to be triggered again or not.

[github-actions]

Regression Detector Results

Run ID: 0c0f67c9-84a3-4ec9-b865-f2e1a5ba13a0
Baseline: c592cb1
Comparison: b85f4f9
Total vector CPUs: 7

Explanation

A regression test is an integrated performance test for vector in a repeatable rig, with varying configuration for vector. What follows is a statistical summary of a brief vector run for each configuration across SHAs given above. The goal of these tests are to determine quickly if vector performance is changed and to what degree by a pull request.

Because a target's optimization goal performance in each experiment will vary somewhat each time it is run, we can only estimate mean differences in optimization goal relative to the baseline target. We express these differences as a percentage change relative to the baseline target, denoted "Δ mean %". These estimates are made to a precision that balances accuracy and cost control. We represent this precision as a 90.00% confidence interval denoted "Δ mean % CI": there is a 90.00% chance that the true value of "Δ mean %" is in that interval.

We decide whether a change in performance is a "regression" -- a change worth investigating further -- if both of the following two criteria are true:

1. The estimated |Δ mean %| ≥ 5.00%. This criterion intends to answer the question "Does the estimated change in mean optimization goal performance have a meaningful impact on your customers?". We assume that when |Δ mean %| < 5.00%, the impact on your customers is not meaningful. We also assume that a performance change in optimization goal is worth investigating whether it is an increase or decrease, so long as the magnitude of the change is sufficiently large.

2. Zero is not in the 90.00% confidence interval "Δ mean % CI" about "Δ mean %". This statement is equivalent to saying that there is at least a 90.00% chance that the mean difference in optimization goal is not zero. This criterion intends to answer the question, "Is there a statistically significant difference in mean optimization goal performance?". It also means there is no more than a 10.00% chance this criterion reports a statistically significant difference when the true difference in mean optimization goal is zero -- a "false positive". We assume you are willing to accept a 10.00% chance of inaccurately detecting a change in performance when no true difference exists.

The table below, if present, lists those experiments that have experienced a statistically significant change in mean optimization goal performance between baseline and comparison SHAs with 90.00% confidence OR have been detected as newly erratic. Negative values of "Δ mean %" mean that baseline is faster, whereas positive values of "Δ mean %" mean that comparison is faster. Results that do not exhibit more than a ±5.00% change in their mean optimization goal are discarded. An experiment is erratic if its coefficient of variation is greater than 0.1. The abbreviated table will be omitted if no interesting change is observed.

No interesting changes in experiment optimization goals with confidence ≥ 90.00% and |Δ mean %| ≥ 5.00%.

Fine details of change detection per experiment.

experiment	goal	Δ mean %	Δ mean % CI	confidence
datadog_agent_remap_blackhole_acks	ingress throughput	+3.23	[+3.05, +3.40]	100.00%
otlp_http_to_blackhole	ingress throughput	+1.91	[+1.74, +2.07]	100.00%
datadog_agent_remap_blackhole	ingress throughput	+1.90	[+1.79, +2.01]	100.00%
file_to_blackhole	egress throughput	+1.22	[-2.49, +4.93]	32.61%
syslog_splunk_hec_logs	ingress throughput	+1.07	[+0.99, +1.15]	100.00%
otlp_grpc_to_blackhole	ingress throughput	+0.62	[+0.51, +0.72]	100.00%
http_to_s3	ingress throughput	+0.26	[+0.06, +0.46]	90.53%
syslog_log2metric_humio_metrics	ingress throughput	+0.25	[+0.16, +0.34]	99.97%
syslog_log2metric_splunk_hec_metrics	ingress throughput	+0.20	[+0.09, +0.30]	98.56%
syslog_humio_logs	ingress throughput	+0.16	[+0.08, +0.23]	99.10%
enterprise_http_to_http	ingress throughput	+0.06	[+0.02, +0.10]	95.70%
fluent_elasticsearch	ingress throughput	+0.00	[-0.00, +0.00]	20.10%
http_to_http_noack	ingress throughput	+0.00	[-0.06, +0.06]	0.23%
splunk_hec_to_splunk_hec_logs_acks	ingress throughput	-0.00	[-0.07, +0.06]	1.72%
splunk_hec_indexer_ack_blackhole	ingress throughput	-0.01	[-0.05, +0.04]	14.83%
splunk_hec_to_splunk_hec_logs_noack	ingress throughput	-0.01	[-0.06, +0.03]	34.15%
datadog_agent_remap_datadog_logs	ingress throughput	-0.10	[-0.20, +0.00]	78.62%
http_to_http_acks	ingress throughput	-0.12	[-1.35, +1.11]	9.84%
http_to_http_json	ingress throughput	-1.01	[-1.07, -0.94]	100.00%
splunk_hec_route_s3	ingress throughput	-1.15	[-1.28, -1.02]	100.00%
socket_to_socket_blackhole	ingress throughput	-1.29	[-1.36, -1.23]	100.00%
syslog_loki	ingress throughput	-2.23	[-2.31, -2.14]	100.00%
http_text_to_http_json	ingress throughput	-2.47	[-2.53, -2.41]	100.00%
syslog_regex_logs2metric_ddmetrics	ingress throughput	-2.58	[-2.90, -2.27]	100.00%
datadog_agent_remap_datadog_logs_acks	ingress throughput	-2.86	[-2.98, -2.74]	100.00%