From 9fb569bdb057e5db63b4dbf2d81f704c16a37f8b Mon Sep 17 00:00:00 2001
From: Karstein Phobic Nyvold Kvistad <karstein.kvistad@maritimerobotics.com>
Date: Sun, 26 Apr 2026 01:20:39 +0200
Subject: [PATCH] shutdown_codesys: kill orphan CODESYS.exe when launcher has
 no tracked PID

Real bug surfaced after a VSC restart left a CODESYS.exe alive that
the new MCP server didn't spawn (state=stopped/error, this.pid=null).
The previous shutdown() early-returned at the top -- the launcher
was 'stopped' so it considered itself done -- which:

  1. Left the orphan CODESYS.exe alive (couldn't run the project).
  2. Made the refuse-on-duplicate guard (95a884b) block every
     subsequent launch_codesys with 'CODESYS already running'.

The launcher was effectively bricked: shutdown said success but
did nothing, launch refused. Hit during today's release-pipeline
test loop.

Fix: when shutdown() is called with state=stopped/error AND this.pid
is null AND findRunningCodesysPids() returns non-empty, taskkill the
orphans before the early-return. Graceful WM_CLOSE first, then 2s
grace, then -F force-kill anything still alive.

Doesn't change the happy-path (launcher tracks its own PID, state
ready -> stopping -> stopped) -- that flow is untouched. The new
code only runs when the launcher would otherwise have been a no-op
on something the OS still has active.
---
 src/launcher.ts | 30 +++++++++++++++++++++++++++++-
 1 file changed, 29 insertions(+), 1 deletion(-)

diff --git a/src/launcher.ts b/src/launcher.ts
index 52f7730..afe93ac 100644
--- a/src/launcher.ts
+++ b/src/launcher.ts
@@ -188,7 +188,35 @@ export class CodesysLauncher implements ScriptExecutor {
 
   /** Graceful shutdown */
   async shutdown(): Promise<void> {
-    if (this.state === 'stopped' || this.state === 'stopping') return;
+    // Orphan-killing fallback: if the launcher itself has no tracked PID
+    // (state stopped/error after a fresh MCP server start) but a CODESYS.exe
+    // is alive on the box from a previous session, the previous early-return
+    // would say "shutdown_codesys success" and do nothing. This left the
+    // launcher's refuse-on-duplicate guard permanently blocking new spawns.
+    // Now we taskkill any orphans we can find before the early-return so the
+    // tool actually does something useful in this state.
+    if (this.state === 'stopped' || this.state === 'stopping') {
+      if (this.pid === null) {
+        const orphans = findRunningCodesysPids();
+        if (orphans.length > 0) {
+          launcherLog.info(`shutdown_codesys: launcher has no tracked PID but found ${orphans.length} orphan CODESYS.exe (PIDs: ${orphans.join(', ')}). Force-killing.`);
+          for (const pid of orphans) {
+            try {
+              execSync(`taskkill /PID ${pid}`, { timeout: 5000, stdio: 'ignore' });
+            } catch { /* ignore graceful failures, force-kill below */ }
+          }
+          // Give them a moment to close gracefully
+          await this.sleep(2_000);
+          const stillAlive = findRunningCodesysPids();
+          for (const pid of stillAlive) {
+            try {
+              execSync(`taskkill /F /PID ${pid}`, { timeout: 5000, stdio: 'ignore' });
+            } catch { /* nothing else to try */ }
+          }
+        }
+      }
+      return;
+    }
 
     this.setState('stopping');
     this.stopHealthMonitor();