Post-1.0: OTLP tracing, broker image + compose service, install rehearsal

- tc-telemetry: fmt subscriber always; with [telemetry] otlp_endpoint
  set, spans batch-export over OTLP/HTTP. Tested against a REAL OTLP
  receiver decoding the actual protobuf (official proto types): the
  emitted span and service.name arrive on the wire. No endpoint = no
  export = no network (air-gap stance). tower-http TraceLayer gives
  every API request a span
- The broker finally has its own image (images/broker.Dockerfile,
  9.5MB from scratch) — the Helm chart referenced one that never
  existed — and the compose deployment now RUNS the broker, sharing a
  socket volume with the server (the unix-socket equivalent of the K8s
  sidecar). Compose secret flows were silently dead before this
- server.Dockerfile fixes surfaced by the rehearsal: the workspace
  build needs tools/ (bundler joined the workspace) and
  images/seccomp/ (include_str! profile) in the build context
- scripts/rehearse-install.sh (plan: clean-VM rehearsal): assembles a
  REAL signed bundle from the built images (server/frontend/broker/
  postgres/socket-proxy), runs the customer path — offline verify,
  docker load, compose up — and asserts /healthz plus the served login
  page before teardown. Passing locally; wired as a release.yml step,
  which also builds/ships the broker + socket-proxy images now

161 Rust tests + 29 journeys; clean-room rehearsal green.

Co-Authored-By: Claude Fable 5 <[email protected]>
This commit is contained in:
Omar Sobh
2026-06-10 11:35:32 -05:00
co-authored by Claude Fable 5
parent 05e9612688
commit 8046853feb
17 changed files with 569 additions and 7 deletions
+25
View File
@@ -0,0 +1,25 @@
[package]
name = "tc-telemetry"
version = "0.1.0"
edition.workspace = true
rust-version.workspace = true
publish.workspace = true
license.workspace = true
[dependencies]
opentelemetry = "0.32"
opentelemetry-otlp = { version = "0.32", default-features = false, features = ["trace", "http-proto", "reqwest-blocking-client"] }
opentelemetry_sdk = "0.32"
thiserror = { workspace = true }
tracing = "0.1"
tracing-opentelemetry = "0.33"
tracing-subscriber = { version = "0.3", features = ["env-filter"] }
[dev-dependencies]
axum = "0.8"
opentelemetry-proto = { version = "0.32", default-features = false, features = ["trace", "gen-tonic-messages"] }
prost = "0.14"
tokio = { workspace = true }
[lints]
workspace = true
+92
View File
@@ -0,0 +1,92 @@
//! Tracing setup (§16 NFR: observability). Always installs a fmt
//! subscriber honoring `RUST_LOG`; with an OTLP endpoint configured,
//! spans also export over OTLP/HTTP — to a collector shipped in the
//! air-gapped bundle or any cloud backend. No endpoint, no export, no
//! network: telemetry never phones home uninvited.
use opentelemetry::trace::TracerProvider as _;
use opentelemetry::KeyValue;
use opentelemetry_otlp::WithExportConfig;
use opentelemetry_sdk::trace::SdkTracerProvider;
use opentelemetry_sdk::Resource;
use tracing_subscriber::layer::SubscriberExt;
use tracing_subscriber::util::SubscriberInitExt;
use tracing_subscriber::{EnvFilter, Layer};
/// Keeps the exporter alive; dropping it flushes pending spans.
pub struct TelemetryGuard {
provider: Option<SdkTracerProvider>,
}
impl TelemetryGuard {
/// Forces all batched spans out — call before process exit (tests
/// call it to assert delivery).
pub fn flush(&self) {
if let Some(provider) = &self.provider {
let _ = provider.force_flush();
}
}
}
impl Drop for TelemetryGuard {
fn drop(&mut self) {
if let Some(provider) = self.provider.take() {
let _ = provider.force_flush();
let _ = provider.shutdown();
}
}
}
#[derive(Debug, thiserror::Error)]
pub enum TelemetryError {
#[error("otlp exporter: {0}")]
Exporter(String),
#[error("subscriber already installed")]
AlreadyInstalled,
}
/// Installs the global subscriber. `otlp_endpoint` is the collector's
/// OTLP/HTTP base (e.g. `http://otel-collector:4318`).
pub fn init(
service_name: &str,
otlp_endpoint: Option<&str>,
) -> Result<TelemetryGuard, TelemetryError> {
let fmt = tracing_subscriber::fmt::layer().with_target(false);
let filter = EnvFilter::try_from_default_env()
.unwrap_or_else(|_| EnvFilter::new("info,sqlx=warn,hyper=warn"));
match otlp_endpoint {
Some(endpoint) => {
let exporter = opentelemetry_otlp::SpanExporter::builder()
.with_http()
.with_endpoint(format!("{}/v1/traces", endpoint.trim_end_matches('/')))
.build()
.map_err(|e| TelemetryError::Exporter(e.to_string()))?;
let provider = SdkTracerProvider::builder()
.with_batch_exporter(exporter)
.with_resource(
Resource::builder()
.with_attributes([KeyValue::new("service.name", service_name.to_owned())])
.build(),
)
.build();
let tracer = provider.tracer(service_name.to_owned());
let otel = tracing_opentelemetry::layer().with_tracer(tracer);
tracing_subscriber::registry()
.with(fmt.with_filter(filter))
.with(otel)
.try_init()
.map_err(|_| TelemetryError::AlreadyInstalled)?;
Ok(TelemetryGuard {
provider: Some(provider),
})
}
None => {
tracing_subscriber::registry()
.with(fmt.with_filter(filter))
.try_init()
.map_err(|_| TelemetryError::AlreadyInstalled)?;
Ok(TelemetryGuard { provider: None })
}
}
}
+81
View File
@@ -0,0 +1,81 @@
//! Spans reach a REAL OTLP/HTTP receiver: a live server accepting
//! protobuf at /v1/traces, decoded with the official proto types — the
//! exact wire contract any collector speaks.
use std::sync::Arc;
use opentelemetry_proto::tonic::collector::trace::v1::ExportTraceServiceRequest;
use prost::Message;
use tokio::sync::Mutex;
type Received = Arc<Mutex<Vec<ExportTraceServiceRequest>>>;
#[tokio::test(flavor = "multi_thread")]
async fn spans_export_to_a_real_otlp_receiver() {
let received: Received = Arc::new(Mutex::new(Vec::new()));
let state = received.clone();
let app = axum::Router::new()
.route(
"/v1/traces",
axum::routing::post(
|axum::extract::State(state): axum::extract::State<Received>,
body: axum::body::Bytes| async move {
let request =
ExportTraceServiceRequest::decode(body.as_ref()).expect("otlp protobuf");
state.lock().await.push(request);
// An empty ExportTraceServiceResponse.
(
[("content-type", "application/x-protobuf")],
Vec::<u8>::new(),
)
},
),
)
.with_state(state);
let listener = tokio::net::TcpListener::bind("127.0.0.1:0").await.unwrap();
let endpoint = format!("http://{}", listener.local_addr().unwrap());
tokio::spawn(async move {
axum::serve(listener, app).await.unwrap();
});
let guard = tc_telemetry::init("teamclaw-test", Some(&endpoint)).expect("init");
{
let span = tracing::info_span!("approval_decision", approval_id = "ap_123");
let _entered = span.enter();
tracing::info!("decided");
}
guard.flush();
// Find our span among everything exported.
let requests = received.lock().await;
assert!(!requests.is_empty(), "no OTLP export arrived");
let mut span_names = Vec::new();
let mut service_names = Vec::new();
for request in requests.iter() {
for resource in &request.resource_spans {
if let Some(res) = &resource.resource {
for attr in &res.attributes {
if attr.key == "service.name" {
if let Some(value) = &attr.value {
service_names.push(format!("{value:?}"));
}
}
}
}
for scope in &resource.scope_spans {
for span in &scope.spans {
span_names.push(span.name.clone());
}
}
}
}
assert!(
span_names.iter().any(|name| name == "approval_decision"),
"spans: {span_names:?}"
);
assert!(
service_names.iter().any(|s| s.contains("teamclaw-test")),
"services: {service_names:?}"
);
}