lfg2025/archy
1
0
Fork 0
Code Issues 25 Pull Requests 4 Actions Packages Projects Releases 74 Wiki Activity

perf(async): remove blocking std::process::Command from async paths

Browse Source 
Every production process spawn reachable from a tokio worker now uses
tokio::process: the install path's podman-port probe, the dependencies
disk check, factory-reset restart, config host-IP detection, the
orchestrator's host-facts helpers (resolve_dynamic_env and its call
sites made async to carry it through), and AutoRuntime's podman/docker
probes.

The FIPS transport probe is the special case: is_available() is a sync
trait method called from async route(), so instead of blocking ~50ms
on systemctl per stale-cache hit it now serves the cached value and
refreshes on a background thread (stale-while-revalidate) — bounded
staleness, zero stalled workers.

§C of the 1.8.0 hardening plan; container/transport/config/package
suites green.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
archipelago
2026-07-04 09:00:50 -04:00
co-authored by Claude Fable 5
parent 01cbec27ed
commit 4c75bb3d38
8 changed files with 132 additions and 73 deletions
Download Patch File Download Diff File Expand all files Collapse all files
core/archipelago/src/container/prod_orchestrator.rs
+42 -30
View File
@@ -1440,7 +1440,7 @@ impl ProdContainerOrchestrator {
.collect()
};
let mut report = ReconcileReport::default();
let disk_gb = self.disk_gb();
let disk_gb = self.disk_gb().await;
// Register every candidate before the (sequential, possibly slow)
// pass so the scanner overlays queued-but-down apps as Restarting
// instead of Stopped. Each app is deregistered as its turn finishes,
@@ -1620,7 +1620,7 @@ impl ProdContainerOrchestrator {
}
let mut resolved_manifest = lm.manifest.clone();
self.resolve_dynamic_env(&mut resolved_manifest)?;
self.resolve_dynamic_env(&mut resolved_manifest).await?;
let name = compute_container_name(&lm.manifest);
// An explicitly user-stopped app MUST stay stopped. The reconcile filter
@@ -1970,7 +1970,7 @@ impl ProdContainerOrchestrator {
async fn install_fresh(&self, lm: &LoadedManifest) -> Result<()> {
self.ensure_app_secrets(&lm.manifest.app.id).await?;
let mut resolved_manifest = lm.manifest.clone();
self.resolve_dynamic_env(&mut resolved_manifest)?;
self.resolve_dynamic_env(&mut resolved_manifest).await?;
resolve_catalog_image(&mut resolved_manifest);
let resolved = resolved_manifest.app.container.resolve().ok_or_else(|| {
@@ -2263,7 +2263,7 @@ impl ProdContainerOrchestrator {
// Re-render the manifest with dynamic env baked in, then go
// through the same install path a fresh install would.
let mut resolved = lm.manifest.clone();
self.resolve_dynamic_env(&mut resolved)?;
self.resolve_dynamic_env(&mut resolved).await?;
self.install_via_quadlet(&resolved, name)
.await
.with_context(|| format!("Phase 3.3: re-install {name} via Quadlet"))?;
@@ -2329,7 +2329,7 @@ impl ProdContainerOrchestrator {
let restart_required = quadlet::contains_stale_health_gate(&old_body);
let mut resolved = lm.manifest.clone();
self.resolve_dynamic_env(&mut resolved)?;
self.resolve_dynamic_env(&mut resolved).await?;
// Same catalog/pinned-version image resolution the installer applies, so
// the drift re-render doesn't revert a pinned version back to the
// manifest's shipped `:latest` tag on the next reconcile tick.
@@ -2404,7 +2404,7 @@ impl ProdContainerOrchestrator {
for dep in dependencies {
let mut resolved = dep.manifest.clone();
self.resolve_dynamic_env(&mut resolved)?;
self.resolve_dynamic_env(&mut resolved).await?;
let name = compute_container_name(&dep.manifest);
if self.runtime.get_container_status(&name).await.is_err() {
continue;
@@ -2588,7 +2588,7 @@ impl ProdContainerOrchestrator {
}
.read("bitcoin-rpc-password")
.context("lnd pre-start: read bitcoin RPC password")?;
let bitcoin_host = self.bitcoin_host();
let bitcoin_host = self.bitcoin_host().await;
let outcome = lnd::ensure_config(&self.lnd_paths, &rpc_pass, &bitcoin_host)
.await
.context("lnd pre-start: ensure lnd.conf")?;
@@ -2706,10 +2706,12 @@ impl ProdContainerOrchestrator {
tokio::time::sleep(std::time::Duration::from_secs(1)).await;
}
fn detect_host_facts(&self) -> HostFacts {
let host_ip = Self::detect_host_ip().unwrap_or_else(|| "127.0.0.1".to_string());
let host_mdns = Self::detect_host_mdns();
let disk_gb = self.disk_gb();
async fn detect_host_facts(&self) -> HostFacts {
let host_ip = Self::detect_host_ip()
.await
.unwrap_or_else(|| "127.0.0.1".to_string());
let host_mdns = Self::detect_host_mdns().await;
let disk_gb = self.disk_gb().await;
HostFacts {
host_ip,
host_mdns,
@@ -2723,9 +2725,8 @@ impl ProdContainerOrchestrator {
/// Container name of the running Bitcoin node (`bitcoin-knots` or
/// `bitcoin-core`) for the `{{BITCOIN_HOST}}` derived-env placeholder.
/// Synchronous `podman ps` to match the surrounding host-fact detection;
/// defaults to `bitcoin-knots` when none is running (B12).
fn bitcoin_host(&self) -> String {
/// Defaults to `bitcoin-knots` when none is running (B12).
async fn bitcoin_host(&self) -> String {
#[cfg(test)]
if let Some(host) = &self.test_bitcoin_host {
return host.clone();
@@ -2733,9 +2734,10 @@ impl ProdContainerOrchestrator {
// Mirrors api::rpc::package::dependencies (the legacy install path);
// both Bitcoin node variants are reachable on archy-net by name.
const BITCOIN_NAMES: &[&str] = &["bitcoin-knots", "bitcoin-core", "bitcoin"];
let names = Command::new("podman")
let names = tokio::process::Command::new("podman")
.args(["ps", "--format", "{{.Names}}"])
.output()
.await
.ok()
.filter(|o| o.status.success())
.map(|o| String::from_utf8_lossy(&o.stdout).into_owned())
@@ -2753,8 +2755,12 @@ impl ProdContainerOrchestrator {
self.test_bitcoin_host = Some(host.to_string());
}
fn detect_host_ip() -> Option<String> {
let output = Command::new("hostname").arg("-I").output().ok()?;
async fn detect_host_ip() -> Option<String> {
let output = tokio::process::Command::new("hostname")
.arg("-I")
.output()
.await
.ok()?;
if !output.status.success() {
return None;
}
@@ -2762,9 +2768,10 @@ impl ProdContainerOrchestrator {
stdout.split_whitespace().next().map(|s| s.to_string())
}
fn detect_host_mdns() -> String {
let hostname = Command::new("hostname")
async fn detect_host_mdns() -> String {
let hostname = tokio::process::Command::new("hostname")
.output()
.await
.ok()
.and_then(|o| {
if o.status.success() {
@@ -2782,13 +2789,18 @@ impl ProdContainerOrchestrator {
}
}
fn detect_disk_gb() -> u64 {
async fn detect_disk_gb() -> u64 {
let target = if Path::new("/var/lib/archipelago").exists() {
"/var/lib/archipelago"
} else {
"/"
};
let output = match Command::new("df").arg("-k").arg(target).output() {
let output = match tokio::process::Command::new("df")
.arg("-k")
.arg(target)
.output()
.await
{
Ok(o) if o.status.success() => o,
_ => return 0,
};
@@ -2808,12 +2820,12 @@ impl ProdContainerOrchestrator {
kb / 1_000_000
}
fn disk_gb(&self) -> u64 {
async fn disk_gb(&self) -> u64 {
#[cfg(test)]
if let Some(disk_gb) = self.test_disk_gb {
return disk_gb;
}
Self::detect_disk_gb()
Self::detect_disk_gb().await
}
/// Ensure app-specific secrets exist *before* env resolution. The Bitcoin
@@ -2838,14 +2850,14 @@ impl ProdContainerOrchestrator {
Ok(())
}
fn resolve_dynamic_env(&self, manifest: &mut AppManifest) -> Result<()> {
async fn resolve_dynamic_env(&self, manifest: &mut AppManifest) -> Result<()> {
// Materialise any manifest-declared generated secrets before they're
// read below. This is the single chokepoint every install/reconcile
// path funnels through, so an app's secrets exist by the time its
// `secret_env` resolves — no per-app code, no host provisioning.
crate::container::secrets::ensure_generated_secrets(&self.secrets_dir, manifest)?;
let mut facts = self.detect_host_facts();
let mut facts = self.detect_host_facts().await;
// Only pay the podman cost to detect Knots-vs-Core when this manifest
// actually templates the Bitcoin node into its env (mempool — B12).
if manifest
@@ -2855,7 +2867,7 @@ impl ProdContainerOrchestrator {
.iter()
.any(|e| e.template.contains("{{BITCOIN_HOST}}"))
{
facts.bitcoin_host = self.bitcoin_host();
facts.bitcoin_host = self.bitcoin_host().await;
}
let mut env = manifest.app.environment.clone();
env.extend(manifest.app.container.resolve_derived_env(&facts));
@@ -3208,7 +3220,7 @@ impl ProdContainerOrchestrator {
) -> Result<String> {
let mut out = content.to_string();
if out.contains("{{HOST_IP}}") || out.contains("{{HOST_MDNS}}") {
let facts = self.detect_host_facts();
let facts = self.detect_host_facts().await;
out = out
.replace("{{HOST_IP}}", &facts.host_ip)
.replace("{{HOST_MDNS}}", &facts.host_mdns);
@@ -3297,7 +3309,7 @@ impl ProdContainerOrchestrator {
/// however the box is reached locally. (Generalised from the old per-app
/// netbird TLS helper, deleted in #20 ph4: rsa:2048, 10-year, no per-app Rust.)
async fn ensure_manifest_certs(&self, manifest: &AppManifest) -> Result<()> {
let facts = self.detect_host_facts();
let facts = self.detect_host_facts().await;
let render = |s: &str| {
s.replace("{{HOST_IP}}", &facts.host_ip)
.replace("{{HOST_MDNS}}", &facts.host_mdns)
@@ -3594,7 +3606,7 @@ impl ContainerOrchestrator for ProdContainerOrchestrator {
self.ensure_app_secrets(app_id).await?;
let name = compute_container_name(&lm.manifest);
let mut resolved_manifest = lm.manifest.clone();
self.resolve_dynamic_env(&mut resolved_manifest)?;
self.resolve_dynamic_env(&mut resolved_manifest).await?;
let service = format!("{name}.service");
if self.quadlet_unit_exists(&name).await? {
@@ -4328,7 +4340,7 @@ app:
orch.set_bitcoin_host_for_test(node);
let mut manifest = AppManifest::parse(yaml).unwrap();
orch.resolve_dynamic_env(&mut manifest).unwrap();
orch.resolve_dynamic_env(&mut manifest).await.unwrap();
assert!(
manifest