crashlabs/huskies
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 37 Wiki Activity

Fixed exec bug

Browse Source
This commit is contained in:
Timmy
2026-07-15 14:55:27 +01:00
parent 549e0349d7
commit 2a7db16a77
3 changed files with 40 additions and 60 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docker/entrypoint.sh
+8
View File
@@ -75,4 +75,12 @@ if [ -d /workspace/frontend ] && [ -f /workspace/frontend/package.json ]; then
cd /workspace
fi
# If a rebuilt binary exists (from rebuild_and_restart or upgrade), prefer it
# over the image-baked /usr/local/bin/huskies.
REBUILT="/app/target/release/huskies"
if [ -x "$REBUILT" ] && [ "$1" = "huskies" ]; then
shift
exec "$REBUILT" "$@"
fi
exec "$@"
server/src/rebuild.rs
+26 -38
View File
@@ -74,17 +74,39 @@ impl BotShutdownNotifier {
}
}
// ── Shared shutdown sequence ─────────────────────────────────────────────
/// Flush CRDT persistence, remove port files, and exit so Docker restarts
/// the container with the new binary. Used by both `rebuild_and_restart`
/// and `upgrade_and_reexec`.
pub async fn drain_and_exit(project_root: &Path, label: &str) -> ! {
crate::crdt_state::flush_persistence(std::time::Duration::from_secs(5)).await;
let port_file = project_root.join(".huskies_port");
if port_file.exists() {
let _ = std::fs::remove_file(&port_file);
}
let cwd_port_file = std::path::Path::new(".huskies_port");
if cwd_port_file.exists() {
let _ = std::fs::remove_file(cwd_port_file);
}
slog!("[{label}] Exiting for Docker restart with new binary");
std::process::exit(0)
}
// ── Rebuild ──────────────────────────────────────────────────────────────
/// Rebuild the server binary and re-exec.
/// Rebuild the server binary and exit for Docker restart.
///
/// 1. Gracefully stops all running agents (kills PTY children).
/// 2. Runs `cargo build [-p huskies]` from the workspace root, matching
/// the current build profile (debug or release).
/// 3. If the build fails, returns the build error (server stays up).
/// 4. If the build succeeds, sends a best-effort shutdown notification (if a
/// [`BotShutdownNotifier`] is provided), then re-execs the process with
/// the new binary via `std::os::unix::process::CommandExt::exec()`.
/// [`BotShutdownNotifier`] is provided), then calls [`drain_and_exit`] to
/// flush persistence and exit. Docker's restart policy brings the
/// container back up with the new binary.
pub async fn rebuild_and_restart(
agents: &AgentPool,
project_root: &Path,
@@ -204,41 +226,7 @@ pub async fn rebuild_and_restart(
n.notify(ShutdownReason::Rebuild).await;
}
// 5b. Drain the persistence channel so no queued ops are lost when exec()
// replaces this process. Times out after 5 s with a logged warning
// naming the queue depth so any regression is visible in logs.
crate::crdt_state::flush_persistence(std::time::Duration::from_secs(5)).await;
// 6. Re-exec with the new binary.
// Use the cargo output path rather than current_exe() so that rebuilds
// inside Docker work correctly — the running binary may be installed at
// /usr/local/bin/huskies (read-only) while cargo writes the new binary
// to /app/target/release/huskies (a writable volume).
let new_exe = if cfg!(debug_assertions) {
workspace_root.join("target/debug/huskies")
} else {
workspace_root.join("target/release/huskies")
};
let args: Vec<String> = std::env::args().collect();
// Remove the port file before re-exec so the new process can write its own.
let port_file = project_root.join(".huskies_port");
if port_file.exists() {
let _ = std::fs::remove_file(&port_file);
}
// Also check cwd for port file.
let cwd_port_file = std::path::Path::new(".huskies_port");
if cwd_port_file.exists() {
let _ = std::fs::remove_file(cwd_port_file);
}
// Use exec() to replace the current process.
// This never returns on success.
use std::os::unix::process::CommandExt;
let err = std::process::Command::new(&new_exe).args(&args[1..]).exec();
// If we get here, exec() failed.
Err(format!("Failed to exec new binary: {err}"))
drain_and_exit(project_root, "rebuild").await
}
// ── Tests ────────────────────────────────────────────────────────────────
server/src/upgrade.rs
+6 -22
View File
@@ -60,9 +60,12 @@ pub async fn fetch_and_replace_binary(source_url: &str, target_path: &Path) -> R
// ── Full server upgrade (called from the running process) ─────────────────
/// Fetch a new binary, atomically replace the current executable, drain CRDT
/// persistence, and re-exec the running server process with its original args.
/// persistence, and exit so Docker restarts the container with the new binary.
///
/// The entrypoint script checks for a rebuilt binary at
/// /app/target/release/huskies and prefers it over the image-baked
/// /usr/local/bin/huskies.
///
/// This function never returns on success — `exec()` replaces the process.
/// On failure it returns `Err(message)` so the caller can report the error
/// while keeping the original server running.
pub async fn upgrade_and_reexec(source_url: &str, project_root: &Path) -> Result<String, String> {
@@ -70,26 +73,7 @@ pub async fn upgrade_and_reexec(source_url: &str, project_root: &Path) -> Result
fetch_and_replace_binary(source_url, &target).await?;
// Drain queued CRDT ops so nothing is lost when exec() replaces the process.
crate::crdt_state::flush_persistence(std::time::Duration::from_secs(5)).await;
// Clean up the port file so the new process can write a fresh one.
let port_file = project_root.join(".huskies_port");
if port_file.exists() {
let _ = std::fs::remove_file(&port_file);
}
let args: Vec<String> = std::env::args().collect();
slog!("[upgrade] Re-execing with new binary: {}", target.display());
use std::os::unix::process::CommandExt;
let err = std::process::Command::new(&target).args(&args[1..]).exec();
// exec() only returns on failure.
Err(format!(
"Failed to exec new binary at {}: {err}",
target.display()
))
crate::rebuild::drain_and_exit(project_root, "upgrade").await
}
// ── CLI upgrade (no re-exec) ─────────────────────────────────────────────