RunicGateway/link
3
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 8 Wiki Activity

feat(sidecar): start as a real Windows service #29

Merged
whitlocktech merged 1 commits from feat/windows-service into main 2026-08-07 18:53:16 +00:00
Conversation 0 Commits 1 Files Changed 7 +768 -300
7 changed files with 768 additions and 300 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
README.md
View File

@@ -48,6 +48,29 @@ to read the token back; it is not meant to be scraped from the log.
uo-link-sidecar --print-config --config /etc/runicgateway/sidecar.toml uo-link-sidecar --print-config --config /etc/runicgateway/sidecar.toml
``` ```
### Running as a service
The same binary runs in the foreground and as a system service — there is no `--service` flag to
remember, because the process can tell how it was started.
- **Linux/systemd** supervises any foreground process, so the unit just runs the binary. `SIGTERM`
(what `systemctl stop` sends) and `SIGINT` both unwind it cleanly; logs go to the journal.
- **Windows** cannot. The service control manager only supervises a process that connects back to
it within ~30 seconds via `StartServiceCtrlDispatcher`; a plain console program registered with
`sc.exe create` is killed with **error 1053** despite running perfectly. So on Windows the sidecar
speaks that handshake: started by the SCM it runs as a service, started from a shell the connect
fails with `ERROR_FAILED_SERVICE_CONTROLLER_CONNECT` and it falls through to an ordinary
foreground run. It reports `Running` only once the shard port is bound and the store is open, and
— having no console — logs to `uo-link-sidecar.<date>.log` beside its config, rolled daily.
Only the starting and stopping is platform-specific: `src/app.rs` is the entire sidecar and is
shared, while `src/windows.rs` and `src/unix.rs` do nothing but start it and tell it when to stop.
The Windows crates are declared under `[target.'cfg(windows)'.dependencies]`, so Cargo neither
resolves nor builds them for a Linux target.
Registering the service is the installer's job; to do it by hand see
[INSTALL.md Appendix A4](https://gitea.whitlocktech.com/RunicGateway/docs/src/branch/main/installer/INSTALL.md).
`.gitea/workflows/release.yml` cross-compiles Linux + Windows binaries and cuts a Gitea release on `.gitea/workflows/release.yml` cross-compiles Linux + Windows binaries and cuts a Gitea release on
every merge to `main` (conventional-commit versioning). See [`sidecar/README.md`](sidecar/README.md) every merge to `main` (conventional-commit versioning). See [`sidecar/README.md`](sidecar/README.md)
for configuration and the wire protocol. for configuration and the wire protocol.

95
sidecar/Cargo.lock generated
View File

@@ -242,6 +242,15 @@ version = "2.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "217698eaf96b4a3f0bc4f3662aaa55bdf913cd54d7204591faa790070c6d0853" checksum = "217698eaf96b4a3f0bc4f3662aaa55bdf913cd54d7204591faa790070c6d0853"
[[package]]
name = "crossbeam-channel"
version = "0.5.16"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d85363c37faeca707aef026efa9f3b34d077bce547e48f770770625c6013679e"
dependencies = [
"crossbeam-utils",
]
[[package]] [[package]]
name = "crossbeam-queue" name = "crossbeam-queue"
version = "0.3.13" version = "0.3.13"
@@ -284,6 +293,12 @@ dependencies = [
"zeroize", "zeroize",
] ]
[[package]]
name = "deranged"
version = "0.5.8"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7cd812cc2bc1d69d4764bd80df88b4317eaef9e773c75226407d9bc0876b211c"
[[package]] [[package]]
name = "digest" name = "digest"
version = "0.10.7" version = "0.10.7"
@@ -921,6 +936,12 @@ dependencies = [
"zeroize", "zeroize",
] ]
[[package]]
name = "num-conv"
version = "0.2.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "521739c6d2bac4aa25192232afe6841231376b2b26d4d9fae5ecf8ca5772e441"
[[package]] [[package]]
name = "num-integer" name = "num-integer"
version = "0.1.46" version = "0.1.46"
@@ -1048,6 +1069,12 @@ dependencies = [
"zerovec", "zerovec",
] ]
[[package]]
name = "powerfmt"
version = "0.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "439ee305def115ba05938db6eb1644ff94165c5ab5e9420d1c1bcedbba909391"
[[package]] [[package]]
name = "ppv-lite86" name = "ppv-lite86"
version = "0.2.21" version = "0.2.21"
@@ -1565,6 +1592,12 @@ version = "2.6.1"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "13c2bddecc57b384dee18652358fb23172facb8a2c51ccc10d74c157bdea3292" checksum = "13c2bddecc57b384dee18652358fb23172facb8a2c51ccc10d74c157bdea3292"
[[package]]
name = "symlink"
version = "0.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a7973cce6668464ea31f176d85b13c7ab3bba2cb3b77a2ed26abd7801688010a"
[[package]] [[package]]
name = "syn" name = "syn"
version = "2.0.118" version = "2.0.118"
@@ -1642,6 +1675,36 @@ dependencies = [
"cfg-if", "cfg-if",
] ]
[[package]]
name = "time"
version = "0.3.55"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cdb87b95ec50ddfa440816d227a17b2ccbdda963a316a727fda0fc4334f7d134"
dependencies = [
"deranged",
"num-conv",
"powerfmt",
"serde_core",
"time-core",
"time-macros",
]
[[package]]
name = "time-core"
version = "0.1.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9e1c906769ad99c88eaa54e728060edef082f8e358ff32030cb7c7d315e81109"
[[package]]
name = "time-macros"
version = "0.2.32"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7e689342a48d2ea927c87ea50cabf8594854bf940e9310208848d680d668ed85"
dependencies = [
"num-conv",
"time-core",
]
[[package]] [[package]]
name = "tinystr" name = "tinystr"
version = "0.8.3" version = "0.8.3"
@@ -1798,6 +1861,19 @@ dependencies = [
"tracing-core", "tracing-core",
] ]
[[package]]
name = "tracing-appender"
version = "0.2.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "050686193eb999b4bb3bc2acfa891a13da00f79734704c4b8b4ef1a10b368a3c"
dependencies = [
"crossbeam-channel",
"symlink",
"thiserror 2.0.18",
"time",
"tracing-subscriber",
]
[[package]] [[package]]
name = "tracing-attributes" name = "tracing-attributes"
version = "0.1.31" version = "0.1.31"
@@ -1913,7 +1989,9 @@ dependencies = [
"tokio", "tokio",
"toml", "toml",
"tracing", "tracing",
"tracing-appender",
"tracing-subscriber", "tracing-subscriber",
"windows-service",
] ]
[[package]] [[package]]
@@ -2025,6 +2103,12 @@ dependencies = [
"wasite", "wasite",
] ]
[[package]]
name = "widestring"
version = "1.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "72069c3113ab32ab29e5584db3c6ec55d416895e60715417b5b883a357c3e471"
[[package]] [[package]]
name = "windows-core" name = "windows-core"
version = "0.62.2" version = "0.62.2"
@@ -2075,6 +2159,17 @@ dependencies = [
"windows-link", "windows-link",
] ]
[[package]]
name = "windows-service"
version = "0.8.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "857224b3b211c6f3616921f081ee54721ee3ad2ace2fac6a6337e032f7b4dcf2"
dependencies = [
"bitflags",
"widestring",
"windows-sys 0.61.2",
]
[[package]] [[package]]
name = "windows-strings" name = "windows-strings"
version = "0.5.1" version = "0.5.1"

7
sidecar/Cargo.toml
View File

@@ -17,5 +17,12 @@ toml = "0.8"
getrandom = "0.2" getrandom = "0.2"
chrono = { version = "0.4", default-features = false, features = ["std", "clock"] } chrono = { version = "0.4", default-features = false, features = ["std", "clock"] }
# Speaking the Windows Service Control Manager's startup handshake, and logging somewhere other
# than the stdout a service does not have. Declared per target so Cargo neither resolves nor builds
# either crate for Linux — the Linux binary is byte-for-byte unaffected by Windows service support.
[target.'cfg(windows)'.dependencies]
windows-service = "0.8"
tracing-appender = "0.2"
[profile.release] [profile.release]
opt-level = 2 opt-level = 2

329
sidecar/src/app.rs Normal file
View File

@@ -0,0 +1,329 @@
//! The sidecar itself: everything that happens between "we have a config path" and "we were told
//! to stop". Identical on every platform.
//!
//! This module exists so that *how the process is started and stopped* — a bare `main` under
//! systemd, or a `ServiceMain` under the Windows SCM — is the only thing that differs between
//! hosts. The shard listener, the config, the store, the web server and the event loop are shared
//! code with no `#[cfg]` in sight.
//!
//! [`run`] is parameterised on the two things a supervisor cares about:
//!
//! - `ready` is called once the sidecar is actually up (listener bound, store open). The Windows
//! service reports `Running` to the SCM there, so a config or bind failure surfaces as a *start*
//! failure rather than a service that reports Running and then dies.
//! - `shutdown` is whatever "stop" means on this host: Ctrl-C and `SIGTERM` on Unix, the SCM's
//! `Stop` control on Windows.
use std::future::Future;
use std::sync::atomic::AtomicI64;
use std::sync::Arc;
use std::time::Instant;
use tokio::sync::{broadcast, mpsc};
use tracing::info;
use crate::{config, rpc, shard, store, web};
/// Runs the sidecar until `shutdown` resolves.
///
/// `config_path` is the `--config` argument, or `None` to resolve `$UOLINK_CONFIG` and the default
/// as usual.
pub async fn run<R, S>(config_path: Option<&str>, ready: R, shutdown: S) -> anyhow::Result<()>
where
R: FnOnce(),
S: Future<Output = ()>,
{
info!("uo-link sidecar starting");
let loaded = config::Config::load(config_path)?;
let cfg = loaded.cfg;
info!(
config = %loaded.path.display(),
shard = %cfg.shard.bind,
web = %cfg.web.bind,
db = %cfg.store.path,
auth = cfg.auth_required(),
"configuration loaded"
);
// Shard link: events in, commands out.
let (event_tx, mut event_rx) = mpsc::unbounded_channel::<shard::ShardEvent>();
let handle = shard::serve(&cfg.shard.bind, event_tx).await?;
// Live feed: every shard event fans out to all connected website WebSocket clients.
let (bcast_tx, _) = broadcast::channel::<String>(1024);
// Request/reply correlation for REST queries.
let rpc = rpc::Rpc::new();
// Durable store: event history, economy series, cached profiles, link map.
let store = store::Store::open(&cfg.store.path).await?;
// Health/observability state.
let started = Instant::now();
let last_event = Arc::new(AtomicI64::new(0));
// Website-facing HTTP server.
let web_state = web::AppState {
events: bcast_tx.clone(),
shard: handle.clone(),
rpc: rpc.clone(),
store: store.clone(),
token: Arc::new(cfg.web.auth_token.clone()),
started,
last_event: last_event.clone(),
};
let web_bind = cfg.web.bind.clone();
tokio::spawn(async move {
if let Err(e) = web::serve(&web_bind, web_state).await {
tracing::error!(error = %e, "web server exited");
}
});
// Event loop: a line that correlates to a pending REST call is a reply — route it to the
// waiting caller and stop. Everything else is a live event: log it, persist it, broadcast it.
let feed_tx = bcast_tx.clone();
let route_rpc = rpc.clone();
let event_store = store.clone();
let last_event_ts = last_event.clone();
let replay_handle = handle.clone(); // re-push external news to the shard on (re)connect
let mut total: u64 = 0;
tokio::spawn(async move {
while let Some(ev) = event_rx.recv().await {
// Any line from the shard — including pong heartbeats — is a sign of life.
last_event_ts.store(now_ms(), std::sync::atomic::Ordering::Relaxed);
if route_rpc.try_route(&ev.value).await {
continue; // consumed as a reply
}
total += 1;
match ev.kind.as_str() {
"server.hello" | "mob.login" | "mob.logout" | "player.death" | "vendor.sale"
| "house.decay" | "link.request" => {
info!(kind = %ev.kind, n = total, "{}", ev.value);
}
_ => tracing::debug!(kind = %ev.kind, n = total, "{}", ev.value),
}
// Persist, then broadcast. `pong` and `ws.hello` are ephemeral chatter, not history.
if ev.kind != "pong" {
let t = ev
.value
.get("t")
.and_then(|v| v.as_i64())
.unwrap_or_else(now_ms);
let text = ev.value.to_string();
if let Err(e) = event_store.insert_event(t, &ev.kind, &text).await {
tracing::warn!(error = %e, "failed to persist event");
}
// The champ board is a live projection: champ.update folds in the latest state (one
// row per spawn), champ.remove drops a spawn that despawned or was slain.
match ev.kind.as_str() {
"champ.update" => {
if let Some(serial) = ev.value.get("serial").and_then(|s| s.as_str()) {
if let Err(e) = event_store
.upsert_champ(
serial,
ev.value.get("status").and_then(|s| s.as_str()),
ev.value.get("name").and_then(|n| n.as_str()),
&text,
t,
)
.await
{
tracing::warn!(error = %e, "failed to upsert champ board");
}
}
}
"champ.remove" => {
if let Some(serial) = ev.value.get("serial").and_then(|s| s.as_str()) {
if let Err(e) = event_store.delete_champ(serial).await {
tracing::warn!(error = %e, "failed to remove champ board row");
}
}
}
// Guild board (Protocol 2.0): guild.update folds in the latest roster (one row
// per guild id); guild.remove drops a disbanded guild.
"guild.update" => {
if let Some(id) = ev.value.get("id").and_then(|v| v.as_i64()) {
if let Err(e) = event_store
.upsert_guild(
id,
ev.value.get("name").and_then(|n| n.as_str()),
&text,
t,
)
.await
{
tracing::warn!(error = %e, "failed to upsert guild board");
}
}
}
"guild.remove" => {
if let Some(id) = ev.value.get("id").and_then(|v| v.as_i64()) {
if let Err(e) = event_store.delete_guild(id).await {
tracing::warn!(error = %e, "failed to remove guild board row");
}
}
}
// Governor board (Protocol 2.0): city.update folds in each city's latest
// governance state (one row per city).
"city.update" => {
if let Some(city) = ev.value.get("city").and_then(|c| c.as_str()) {
if let Err(e) = event_store.upsert_governor(city, &text, t).await {
tracing::warn!(error = %e, "failed to upsert governor board");
}
}
}
// House registry (Protocol 2.0): house.update folds in each house's latest state
// (one row per serial); house.remove drops a demolished/traded house.
"house.update" => {
if let Some(serial) = ev.value.get("serial").and_then(|s| s.as_str()) {
if let Err(e) = event_store
.upsert_house(
serial,
ev.value.get("name").and_then(|n| n.as_str()),
&text,
t,
)
.await
{
tracing::warn!(error = %e, "failed to upsert house registry");
}
}
}
"house.remove" => {
if let Some(serial) = ev.value.get("serial").and_then(|s| s.as_str()) {
if let Err(e) = event_store.delete_house(serial).await {
tracing::warn!(error = %e, "failed to remove house registry row");
}
}
}
// Points/loyalty boards (Protocol 3.0): one row per point system, keyed by
// the shard's own PointsType name. The plugin only emits a system whose top N
// actually moved, so this is a sparse stream of overwrites — and there is no
// `points.remove` to handle, because the shard's set of systems is fixed at
// startup and cannot shrink.
"points.board" => {
if let Some(system) = ev.value.get("system").and_then(|s| s.as_str()) {
if let Err(e) = event_store
.upsert_points_board(
system,
ev.value.get("nameString").and_then(|n| n.as_str()),
&text,
t,
)
.await
{
tracing::warn!(error = %e, "failed to upsert points board");
}
}
}
// Player-vendor market index (Protocol 3.0). Each frame is authoritative for
// one vendor — the shard's round-robin sweep only emits a shop whose contents,
// prices or location actually moved — so this is a whole-row overwrite.
//
// Unlike the boards above there IS a remove: a vendor is dismissed, expires, or
// its owner switches off the in-game Vendor Search flag, and any of those must
// take the shop off the site. The last of the three is a privacy control, so
// dropping the row promptly is the point rather than housekeeping.
"vendor.listing" => {
if let Some(serial) = ev.value.get("serial").and_then(|s| s.as_str()) {
let loc = ev.value.get("location");
let field = |k: &str| loc.and_then(|l| l.get(k));
if let Err(e) = event_store
.upsert_vendor(
serial,
ev.value.get("shopName").and_then(|v| v.as_str()),
ev.value.get("ownerName").and_then(|v| v.as_str()),
field("map").and_then(|v| v.as_str()),
field("x").and_then(|v| v.as_i64()),
field("y").and_then(|v| v.as_i64()),
field("region").and_then(|v| v.as_str()),
ev.value.get("count").and_then(|v| v.as_i64()),
&text,
t,
)
.await
{
tracing::warn!(error = %e, "failed to upsert vendor listing");
}
}
}
"vendor.listing.remove" => {
if let Some(serial) = ev.value.get("serial").and_then(|s| s.as_str()) {
if let Err(e) = event_store.delete_vendor(serial).await {
tracing::warn!(error = %e, "failed to remove vendor listing");
}
}
}
// Shard ruleset (Protocol 3.0): a singleton projection. The shard re-emits
// world.ruleset on every connect, so this row is simply overwritten; `rev`
// lets a reader tell a re-send from an actual config change.
"world.ruleset" => {
if let Err(e) = event_store
.upsert_ruleset(ev.value.get("rev").and_then(|r| r.as_str()), &text, t)
.await
{
tracing::warn!(error = %e, "failed to upsert ruleset");
}
}
_ => {}
}
}
// On a shard (re)connect, re-push the stored external news: the shard rebuilds
// TownCryerSystem.NewsEntries from scratch each boot and does not persist ours. Replay
// with announce=false so a restart does not re-proclaim every article at once. news.add
// is idempotent by id, so replaying to a still-populated shard is harmless.
if ev.kind == "server.hello" {
match event_store.news_all().await {
Ok(items) => {
for mut item in items {
if let Some(obj) = item.as_object_mut() {
obj.insert("announce".to_string(), serde_json::json!(false));
}
if !replay_handle.send(item.to_string()).await {
break; // shard went away mid-replay
}
}
}
Err(e) => tracing::warn!(error = %e, "news replay: could not read stored news"),
}
}
let _ = feed_tx.send(ev.value.to_string());
}
});
// Heartbeat to the shard, exercising the command path.
let ping_handle = handle.clone();
tokio::spawn(async move {
loop {
tokio::time::sleep(std::time::Duration::from_secs(15)).await;
if ping_handle.is_connected().await {
let _ = ping_handle
.send(r#"{"kind":"ping","id":"sidecar-heartbeat"}"#.to_string())
.await;
}
}
});
// Everything that can fail at startup has now either failed or succeeded: the shard port is
// bound and the store is open. A supervisor may call this "started".
ready();
shutdown.await;
info!("shutting down");
Ok(())
}
pub fn now_ms() -> i64 {
use std::time::{SystemTime, UNIX_EPOCH};
SystemTime::now()
.duration_since(UNIX_EPOCH)
.map(|d| d.as_millis() as i64)
.unwrap_or(0)
}

334
sidecar/src/main.rs
View File

@@ -1,21 +1,37 @@
//! uo-link sidecar. //! uo-link sidecar.
//! //!
//! Terminates the loopback link to the ServUO shard and exposes a website-facing HTTP surface. So //! Terminates the loopback link to the ServUO shard and exposes a website-facing HTTP surface: the
//! far: the shard link (bidirectional) and a WebSocket live feed. REST queries and SQLite come next. //! shard link (bidirectional), a WebSocket live feed, and REST queries backed by SQLite.
//!
//! # Layout
//!
//! `main` does argument handling and nothing else; the sidecar proper lives in [`app`] and is the
//! same code on every platform. Only *how the process is started and stopped* is
//! platform-specific:
//!
//! ```text
//! systemd ──▶ main ──▶ unix::run ─────────────────────────┐
//! ├──▶ app::run
//! SCM ─────▶ main ──▶ windows::run ──▶ ServiceMain ───────┘
//! └─▶ console fallback ──┘
//! ```
//!
//! The Windows half is not optional politeness: the SCM refuses to supervise a program that does
//! not speak its startup handshake (see [`windows`]). The platform modules are gated with `#[cfg]`
//! and their dependencies are declared per target, so none of it reaches a Linux build.
mod app;
mod cli; mod cli;
mod config; mod config;
mod rpc; mod rpc;
mod shard; mod shard;
mod store; mod store;
#[cfg(unix)]
mod unix;
mod web; mod web;
#[cfg(windows)]
mod windows;
use std::sync::atomic::AtomicI64;
use std::sync::Arc;
use std::time::Instant;
use tokio::sync::{broadcast, mpsc};
use tracing::info;
use tracing_subscriber::EnvFilter; use tracing_subscriber::EnvFilter;
/// Wire-protocol version between the website and the sidecar. Bump this whenever an event or /// Wire-protocol version between the website and the sidecar. Bump this whenever an event or
@@ -32,8 +48,10 @@ use tracing_subscriber::EnvFilter;
/// there is deliberately no feature-negotiation array: v3 implies all three kinds. /// there is deliberately no feature-negotiation array: v3 implies all three kinds.
pub const PROTOCOL_VERSION: u32 = 3; pub const PROTOCOL_VERSION: u32 = 3;
#[tokio::main] // Not `#[tokio::main]`: on Windows the SCM dispatcher takes over this thread and starts the runtime
async fn main() -> anyhow::Result<()> { // itself, on its own thread, once the service actually begins. The runtime is built by whichever
// platform module ends up running.
fn main() -> anyhow::Result<()> {
let args = match cli::parse(std::env::args().skip(1)) { let args = match cli::parse(std::env::args().skip(1)) {
Ok(args) => args, Ok(args) => args,
Err(msg) => { Err(msg) => {
@@ -66,299 +84,15 @@ async fn main() -> anyhow::Result<()> {
cli::Mode::Run => {} cli::Mode::Run => {}
} }
init_tracing(); #[cfg(windows)]
info!("uo-link sidecar starting"); return windows::run(args.config.as_deref());
let loaded = config::Config::load(args.config.as_deref())?; #[cfg(unix)]
let cfg = loaded.cfg; return unix::run(args.config.as_deref());
info!(
config = %loaded.path.display(),
shard = %cfg.shard.bind,
web = %cfg.web.bind,
db = %cfg.store.path,
auth = cfg.auth_required(),
"configuration loaded"
);
// Shard link: events in, commands out.
let (event_tx, mut event_rx) = mpsc::unbounded_channel::<shard::ShardEvent>();
let handle = shard::serve(&cfg.shard.bind, event_tx).await?;
// Live feed: every shard event fans out to all connected website WebSocket clients.
let (bcast_tx, _) = broadcast::channel::<String>(1024);
// Request/reply correlation for REST queries.
let rpc = rpc::Rpc::new();
// Durable store: event history, economy series, cached profiles, link map.
let store = store::Store::open(&cfg.store.path).await?;
// Health/observability state.
let started = Instant::now();
let last_event = Arc::new(AtomicI64::new(0));
// Website-facing HTTP server.
let web_state = web::AppState {
events: bcast_tx.clone(),
shard: handle.clone(),
rpc: rpc.clone(),
store: store.clone(),
token: Arc::new(cfg.web.auth_token.clone()),
started,
last_event: last_event.clone(),
};
let web_bind = cfg.web.bind.clone();
tokio::spawn(async move {
if let Err(e) = web::serve(&web_bind, web_state).await {
tracing::error!(error = %e, "web server exited");
}
});
// Event loop: a line that correlates to a pending REST call is a reply — route it to the
// waiting caller and stop. Everything else is a live event: log it, persist it, broadcast it.
let feed_tx = bcast_tx.clone();
let route_rpc = rpc.clone();
let event_store = store.clone();
let last_event_ts = last_event.clone();
let replay_handle = handle.clone(); // re-push external news to the shard on (re)connect
let mut total: u64 = 0;
tokio::spawn(async move {
while let Some(ev) = event_rx.recv().await {
// Any line from the shard — including pong heartbeats — is a sign of life.
last_event_ts.store(now_ms(), std::sync::atomic::Ordering::Relaxed);
if route_rpc.try_route(&ev.value).await {
continue; // consumed as a reply
} }
total += 1; /// Logging for a foreground run: human-readable, on stdout.
match ev.kind.as_str() { pub fn init_console_tracing() {
"server.hello" | "mob.login" | "mob.logout" | "player.death" | "vendor.sale"
| "house.decay" | "link.request" => {
info!(kind = %ev.kind, n = total, "{}", ev.value);
}
_ => tracing::debug!(kind = %ev.kind, n = total, "{}", ev.value),
}
// Persist, then broadcast. `pong` and `ws.hello` are ephemeral chatter, not history.
if ev.kind != "pong" {
let t = ev
.value
.get("t")
.and_then(|v| v.as_i64())
.unwrap_or_else(now_ms);
let text = ev.value.to_string();
if let Err(e) = event_store.insert_event(t, &ev.kind, &text).await {
tracing::warn!(error = %e, "failed to persist event");
}
// The champ board is a live projection: champ.update folds in the latest state (one
// row per spawn), champ.remove drops a spawn that despawned or was slain.
match ev.kind.as_str() {
"champ.update" => {
if let Some(serial) = ev.value.get("serial").and_then(|s| s.as_str()) {
if let Err(e) = event_store
.upsert_champ(
serial,
ev.value.get("status").and_then(|s| s.as_str()),
ev.value.get("name").and_then(|n| n.as_str()),
&text,
t,
)
.await
{
tracing::warn!(error = %e, "failed to upsert champ board");
}
}
}
"champ.remove" => {
if let Some(serial) = ev.value.get("serial").and_then(|s| s.as_str()) {
if let Err(e) = event_store.delete_champ(serial).await {
tracing::warn!(error = %e, "failed to remove champ board row");
}
}
}
// Guild board (Protocol 2.0): guild.update folds in the latest roster (one row
// per guild id); guild.remove drops a disbanded guild.
"guild.update" => {
if let Some(id) = ev.value.get("id").and_then(|v| v.as_i64()) {
if let Err(e) = event_store
.upsert_guild(
id,
ev.value.get("name").and_then(|n| n.as_str()),
&text,
t,
)
.await
{
tracing::warn!(error = %e, "failed to upsert guild board");
}
}
}
"guild.remove" => {
if let Some(id) = ev.value.get("id").and_then(|v| v.as_i64()) {
if let Err(e) = event_store.delete_guild(id).await {
tracing::warn!(error = %e, "failed to remove guild board row");
}
}
}
// Governor board (Protocol 2.0): city.update folds in each city's latest
// governance state (one row per city).
"city.update" => {
if let Some(city) = ev.value.get("city").and_then(|c| c.as_str()) {
if let Err(e) = event_store.upsert_governor(city, &text, t).await {
tracing::warn!(error = %e, "failed to upsert governor board");
}
}
}
// House registry (Protocol 2.0): house.update folds in each house's latest state
// (one row per serial); house.remove drops a demolished/traded house.
"house.update" => {
if let Some(serial) = ev.value.get("serial").and_then(|s| s.as_str()) {
if let Err(e) = event_store
.upsert_house(
serial,
ev.value.get("name").and_then(|n| n.as_str()),
&text,
t,
)
.await
{
tracing::warn!(error = %e, "failed to upsert house registry");
}
}
}
"house.remove" => {
if let Some(serial) = ev.value.get("serial").and_then(|s| s.as_str()) {
if let Err(e) = event_store.delete_house(serial).await {
tracing::warn!(error = %e, "failed to remove house registry row");
}
}
}
// Points/loyalty boards (Protocol 3.0): one row per point system, keyed by
// the shard's own PointsType name. The plugin only emits a system whose top N
// actually moved, so this is a sparse stream of overwrites — and there is no
// `points.remove` to handle, because the shard's set of systems is fixed at
// startup and cannot shrink.
"points.board" => {
if let Some(system) = ev.value.get("system").and_then(|s| s.as_str()) {
if let Err(e) = event_store
.upsert_points_board(
system,
ev.value.get("nameString").and_then(|n| n.as_str()),
&text,
t,
)
.await
{
tracing::warn!(error = %e, "failed to upsert points board");
}
}
}
// Player-vendor market index (Protocol 3.0). Each frame is authoritative for
// one vendor — the shard's round-robin sweep only emits a shop whose contents,
// prices or location actually moved — so this is a whole-row overwrite.
//
// Unlike the boards above there IS a remove: a vendor is dismissed, expires, or
// its owner switches off the in-game Vendor Search flag, and any of those must
// take the shop off the site. The last of the three is a privacy control, so
// dropping the row promptly is the point rather than housekeeping.
"vendor.listing" => {
if let Some(serial) = ev.value.get("serial").and_then(|s| s.as_str()) {
let loc = ev.value.get("location");
let field = |k: &str| loc.and_then(|l| l.get(k));
if let Err(e) = event_store
.upsert_vendor(
serial,
ev.value.get("shopName").and_then(|v| v.as_str()),
ev.value.get("ownerName").and_then(|v| v.as_str()),
field("map").and_then(|v| v.as_str()),
field("x").and_then(|v| v.as_i64()),
field("y").and_then(|v| v.as_i64()),
field("region").and_then(|v| v.as_str()),
ev.value.get("count").and_then(|v| v.as_i64()),
&text,
t,
)
.await
{
tracing::warn!(error = %e, "failed to upsert vendor listing");
}
}
}
"vendor.listing.remove" => {
if let Some(serial) = ev.value.get("serial").and_then(|s| s.as_str()) {
if let Err(e) = event_store.delete_vendor(serial).await {
tracing::warn!(error = %e, "failed to remove vendor listing");
}
}
}
// Shard ruleset (Protocol 3.0): a singleton projection. The shard re-emits
// world.ruleset on every connect, so this row is simply overwritten; `rev`
// lets a reader tell a re-send from an actual config change.
"world.ruleset" => {
if let Err(e) = event_store
.upsert_ruleset(ev.value.get("rev").and_then(|r| r.as_str()), &text, t)
.await
{
tracing::warn!(error = %e, "failed to upsert ruleset");
}
}
_ => {}
}
}
// On a shard (re)connect, re-push the stored external news: the shard rebuilds
// TownCryerSystem.NewsEntries from scratch each boot and does not persist ours. Replay
// with announce=false so a restart does not re-proclaim every article at once. news.add
// is idempotent by id, so replaying to a still-populated shard is harmless.
if ev.kind == "server.hello" {
match event_store.news_all().await {
Ok(items) => {
for mut item in items {
if let Some(obj) = item.as_object_mut() {
obj.insert("announce".to_string(), serde_json::json!(false));
}
if !replay_handle.send(item.to_string()).await {
break; // shard went away mid-replay
}
}
}
Err(e) => tracing::warn!(error = %e, "news replay: could not read stored news"),
}
}
let _ = feed_tx.send(ev.value.to_string());
}
});
// Heartbeat to the shard, exercising the command path.
let ping_handle = handle.clone();
tokio::spawn(async move {
loop {
tokio::time::sleep(std::time::Duration::from_secs(15)).await;
if ping_handle.is_connected().await {
let _ = ping_handle
.send(r#"{"kind":"ping","id":"sidecar-heartbeat"}"#.to_string())
.await;
}
}
});
tokio::signal::ctrl_c().await?;
info!("shutting down");
Ok(())
}
fn now_ms() -> i64 {
use std::time::{SystemTime, UNIX_EPOCH};
SystemTime::now()
.duration_since(UNIX_EPOCH)
.map(|d| d.as_millis() as i64)
.unwrap_or(0)
}
fn init_tracing() {
tracing_subscriber::fmt() tracing_subscriber::fmt()
.with_env_filter( .with_env_filter(
EnvFilter::try_from_default_env().unwrap_or_else(|_| EnvFilter::new("info")), EnvFilter::try_from_default_env().unwrap_or_else(|_| EnvFilter::new("info")),

41
sidecar/src/unix.rs Normal file
View File

@@ -0,0 +1,41 @@
//! Unix startup and shutdown.
//!
//! There is no supervisor protocol to speak: systemd starts the process, and stops it by sending
//! `SIGTERM`. All this module does is translate the two signals that mean "stop" into the future
//! [`crate::app::run`] waits on, so a `systemctl stop` unwinds the same way a Ctrl-C does instead
//! of being killed by the default `SIGTERM` disposition mid-write.
use tokio::signal::unix::{signal, SignalKind};
pub fn run(config_path: Option<&str>) -> anyhow::Result<()> {
crate::init_console_tracing();
tokio::runtime::Runtime::new()?.block_on(crate::app::run(config_path, || {}, shutdown_signal()))
}
/// Resolves on the first `SIGINT` or `SIGTERM`.
async fn shutdown_signal() {
// A failure to install a handler is not worth aborting a running sidecar for: fall back to
// pending, which leaves that signal's default disposition (terminate) in place.
let mut term = match signal(SignalKind::terminate()) {
Ok(s) => s,
Err(e) => {
tracing::warn!(error = %e, "could not listen for SIGTERM");
std::future::pending::<()>().await;
unreachable!()
}
};
let mut int = match signal(SignalKind::interrupt()) {
Ok(s) => s,
Err(e) => {
tracing::warn!(error = %e, "could not listen for SIGINT");
term.recv().await;
return;
}
};
tokio::select! {
_ = term.recv() => tracing::info!("SIGTERM received"),
_ = int.recv() => tracing::info!("SIGINT received"),
}
}

239
sidecar/src/windows.rs Normal file
View File

@@ -0,0 +1,239 @@
//! Windows startup and shutdown: the SCM handshake.
//!
//! Unlike systemd, the Windows Service Control Manager cannot supervise an arbitrary console
//! program. A binary registered with `sc.exe create` has ~30 seconds to call
//! `StartServiceCtrlDispatcher` and connect back to the SCM; one that never does is killed with
//! **error 1053, "the service did not respond to the start request in a timely fashion"** — even
//! though the process itself started perfectly and is sitting there serving traffic. That is the
//! entire reason this module exists.
//!
//! ## One binary, two ways in
//!
//! The dispatcher is tried first and *failing is expected*: when the process was started from a
//! shell rather than by the SCM, the connect fails with `ERROR_FAILED_SERVICE_CONTROLLER_CONNECT`
//! (1063), and that — and only that — falls through to a normal foreground run. So
//! `uo-link-sidecar.exe --config ...` stays an ordinary console app you can Ctrl-C, `cargo run`
//! still works, and the same binary can be registered as a service with no `--service` flag for an
//! operator to forget. Any other dispatcher error is a real failure and is reported.
//!
//! ## Logging goes to a file, because a service has no stdout
//!
//! Under the SCM there is no console attached, so the normal stdout subscriber writes into the
//! void. In service mode the sidecar logs to a daily-rolled file next to its config instead
//! (`uo-link-sidecar.YYYY-MM-DD.log`, seven kept). A service whose start fails leaves a reason
//! behind rather than only an SCM error code.
use std::ffi::OsString;
use std::path::PathBuf;
use std::sync::Arc;
use std::sync::OnceLock;
use std::time::Duration;
use tokio::sync::Notify;
use tracing_subscriber::EnvFilter;
use windows_service::service::{
ServiceControl, ServiceControlAccept, ServiceExitCode, ServiceState, ServiceStatus, ServiceType,
};
use windows_service::service_control_handler::{self, ServiceControlHandlerResult};
use windows_service::{define_windows_service, service_dispatcher};
/// Must match the name the installer registers (`installer/src/service.rs::WINDOWS_SERVICE`). For
/// an own-process service the SCM ignores it, but a mismatch would be a trap for whoever converts
/// this to a shared-process service later.
pub const SERVICE_NAME: &str = "RunicGatewayLink";
const SERVICE_TYPE: ServiceType = ServiceType::OWN_PROCESS;
/// `ERROR_FAILED_SERVICE_CONTROLLER_CONNECT` — "this process was not started by the SCM", which is
/// the normal answer when a human runs the binary.
const ERROR_FAILED_SERVICE_CONTROLLER_CONNECT: i32 = 1063;
/// `service_main` is called through an `extern "system"` trampoline and so can capture nothing.
/// The parsed `--config` is handed over here instead of being re-parsed, so the service and a
/// console run resolve their configuration through exactly the same code path.
static CONFIG_PATH: OnceLock<Option<String>> = OnceLock::new();
pub fn run(config_path: Option<&str>) -> anyhow::Result<()> {
let _ = CONFIG_PATH.set(config_path.map(str::to_string));
match service_dispatcher::start(SERVICE_NAME, ffi_service_main) {
Ok(()) => Ok(()),
// Not started by the SCM: this is a foreground run, which is not an error.
Err(windows_service::Error::Winapi(e))
if e.raw_os_error() == Some(ERROR_FAILED_SERVICE_CONTROLLER_CONNECT) =>
{
console_run(config_path)
}
Err(e) => Err(anyhow::Error::new(e)
.context("could not connect to the Windows service control manager")),
}
}
/// A normal foreground run: stdout logging, Ctrl-C to stop.
fn console_run(config_path: Option<&str>) -> anyhow::Result<()> {
crate::init_console_tracing();
tokio::runtime::Runtime::new()?.block_on(crate::app::run(config_path, || {}, async {
let _ = tokio::signal::ctrl_c().await;
}))
}
define_windows_service!(ffi_service_main, service_main);
fn service_main(_arguments: Vec<OsString>) {
// Arguments are deliberately ignored: for an own-process service the `binPath=` arguments
// arrive on the process command line and have already been parsed in `main`. What lands here
// is whatever was typed after `sc start`, which nothing in this deployment uses.
if let Err(e) = serve() {
// Nowhere left to report to but the log: the status handle is gone or was never obtained.
tracing::error!(error = %e, "service exited with an error");
}
}
fn serve() -> anyhow::Result<()> {
let config_path = CONFIG_PATH.get().cloned().flatten();
// Held for the life of the service: dropping the guard stops the background log writer.
let _log_guard = init_service_tracing(config_path.as_deref());
// The SCM calls the control handler on its own thread, so the stop signal crosses a thread
// boundary into the async world. `notify_one` stores a permit if nothing is waiting yet, so a
// stop that arrives during startup is not lost.
let stop = Arc::new(Notify::new());
let handler_stop = stop.clone();
let status_handle =
service_control_handler::register(SERVICE_NAME, move |control| match control {
ServiceControl::Interrogate => ServiceControlHandlerResult::NoError,
ServiceControl::Stop | ServiceControl::Shutdown => {
handler_stop.notify_one();
ServiceControlHandlerResult::NoError
}
_ => ServiceControlHandlerResult::NotImplemented,
})?;
// Registering the handler is the handshake 1053 was about. Everything after this point gets to
// take as long as it credibly needs, as long as the state keeps being reported.
status_handle.set_service_status(ServiceStatus {
service_type: SERVICE_TYPE,
current_state: ServiceState::StartPending,
controls_accepted: ServiceControlAccept::empty(),
exit_code: ServiceExitCode::Win32(0),
checkpoint: 0,
wait_hint: Duration::from_secs(30),
process_id: None,
})?;
let ready_handle = status_handle;
let result = tokio::runtime::Runtime::new()?.block_on(crate::app::run(
config_path.as_deref(),
// Reported only once the shard port is bound and the store is open, so a bad config or a
// taken port fails the *start* instead of flapping Running → Stopped a moment later.
move || {
let _ = ready_handle.set_service_status(ServiceStatus {
service_type: SERVICE_TYPE,
current_state: ServiceState::Running,
controls_accepted: ServiceControlAccept::STOP | ServiceControlAccept::SHUTDOWN,
exit_code: ServiceExitCode::Win32(0),
checkpoint: 0,
wait_hint: Duration::default(),
process_id: None,
});
},
async move { stop.notified().await },
));
// A failed run must leave a nonzero SERVICE_EXIT_CODE behind: `sc query` reporting STOPPED with
// exit code 0 is what made the original failure look like a clean stop.
let exit_code = match &result {
Ok(()) => ServiceExitCode::Win32(0),
Err(e) => {
tracing::error!(error = %e, "sidecar failed");
ServiceExitCode::ServiceSpecific(1)
}
};
status_handle.set_service_status(ServiceStatus {
service_type: SERVICE_TYPE,
current_state: ServiceState::Stopped,
controls_accepted: ServiceControlAccept::empty(),
exit_code,
checkpoint: 0,
wait_hint: Duration::default(),
process_id: None,
})?;
result
}
/// Where the service writes its log: beside the config it was pointed at, which is the directory
/// the installer already provisions and grants the service account write access to.
fn log_dir(config_path: Option<&str>) -> PathBuf {
if let Some(parent) = config_path
.map(PathBuf::from)
.as_deref()
.and_then(|p| p.parent())
.filter(|p| !p.as_os_str().is_empty())
{
return parent.to_path_buf();
}
match std::env::var_os("ProgramData") {
Some(program_data) => PathBuf::from(program_data).join("RunicGateway"),
None => std::env::temp_dir(),
}
}
/// Returns `None` if the log file could not be opened — a service that cannot write a log is still
/// a service worth running, and the SCM start must not fail over it.
fn init_service_tracing(
config_path: Option<&str>,
) -> Option<tracing_appender::non_blocking::WorkerGuard> {
let appender = tracing_appender::rolling::Builder::new()
.rotation(tracing_appender::rolling::Rotation::DAILY)
.filename_prefix("uo-link-sidecar")
.filename_suffix("log")
.max_log_files(7)
.build(log_dir(config_path))
.ok()?;
let (writer, guard) = tracing_appender::non_blocking(appender);
tracing_subscriber::fmt()
.with_env_filter(
EnvFilter::try_from_default_env().unwrap_or_else(|_| EnvFilter::new("info")),
)
.with_ansi(false) // a log file is not a terminal
.with_writer(writer)
.init();
Some(guard)
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn log_dir_follows_the_config_file() {
assert_eq!(
log_dir(Some(r"C:\ProgramData\RunicGateway\sidecar.toml")),
PathBuf::from(r"C:\ProgramData\RunicGateway")
);
}
#[test]
fn a_bare_filename_does_not_become_the_filesystem_root() {
// `--config sidecar.toml` has a parent of "", which as a path means the root of the current
// drive — somewhere a service account cannot write. Fall back instead.
let dir = log_dir(Some("sidecar.toml"));
assert_ne!(dir, PathBuf::from(""));
assert!(dir.is_absolute(), "{}", dir.display());
}
#[test]
fn no_config_falls_back_to_program_data() {
let dir = log_dir(None);
assert!(dir.is_absolute(), "{}", dir.display());
}
#[test]
fn service_name_matches_the_installer() {
// installer/src/service.rs::WINDOWS_SERVICE. Kept as a literal on both sides — the two
// repos are released independently and do not share a crate.
assert_eq!(SERVICE_NAME, "RunicGatewayLink");
}
}