@twinsearth/w2m-dsh-plugin 1.6.6 → 1.6.7

package/.claude-plugin/marketplace.json

@@ -5,14 +5,14 @@
   },
   "metadata": {
     "description": "The w2m fleet plugins: one account, many machines, any of Windows, macOS or Linux.",
-    "version": "1.6.6"
+    "version": "1.6.7"
   },
   "plugins": [
     {
       "name": "w2m-fleet",
       "source": "./",
       "description": "Read fleet state and dispatch commands across the machines registered to one account. Read-only MCP surface: the five exposed tools cannot change anything.",
-      "version": "1.6.6",
+      "version": "1.6.7",
       "author": {
         "name": "TwinsEarth"
       },
@@ -26,5 +26,5 @@
       ]
     }
   ],
-  "version": "1.6.6"
+  "version": "1.6.7"
 }

package/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "w2m-fleet",
   "description": "Drive the same project on online Windows, macOS and Linux machines from one account: read fleet state, dispatch commands and read results. The MCP surface is deliberately read-only.",
-  "version": "1.6.6",
+  "version": "1.6.7",
   "author": {
     "name": "TwinsEarth"
   },

package/CHANGELOG.md

@@ -7,6 +7,21 @@
 
 All notable changes to this project are documented here.
 
+## [1.6.7] — 2026-10-24
+
+**修复 v1.6.6 failover 的阻塞 bug：远端 relay 模式下清理函数抛错，切换后无法重启 stack。**
+
+现场演练（停主 relay 180s）暴露：机器掉线后没有切到备用，恢复主 relay 后才重新上线。根因：
+
+- `startLocalStack` 的清理函数无条件执行 `await relay.close()`；而 `relay` 只在"本进程自建本地 relay"
+  时才非空，远端 relay 模式下是 `null` → 抛 TypeError；
+- failover 的切换流程 `await stop()` 抛错后被 catch 吞掉，**后续的 restart 从未执行**（stack 停了没起来）。
+
+修复：
+- 清理函数 null-safe/idempotent：仅当 `relay` 是带 `close()` 的对象才调用；`agent.stop()` 失败也继续清理。
+- 切换流程：`stop()` 失败不再跳过 restart；并写诊断日志 `xclient/relay-failover.log`（每轮探测/切换/重启错误）。
+- 测试：relay-failover 5/5、relay 200/200；证据 `docs/acceptance/1.6.7/FAILOVER-CLEANUP-FIX.json`
+  （fleet task `01M5ANYMVMFSSP3`）。
 ## [1.6.6] — 2026-10-24
 
 **relay 自动 failover：主 relay 故障时自动切到备用，恢复后自动切回。**

package/codex/mcp-servers.toml

@@ -12,6 +12,6 @@
 
 [mcp_servers.w2m-fleet]
 command = "npx"
-args = ["-y", "-p", "@twinsearth/w2m-dsh-plugin@1.6.6", "w2m-mcp"]
+args = ["-y", "-p", "@twinsearth/w2m-dsh-plugin@1.6.7", "w2m-mcp"]
 startup_timeout_sec = 20
 tool_timeout_sec = 60

package/docs/acceptance/1.6.7/FAILOVER-CLEANUP-FIX.json

@@ -0,0 +1,22 @@
+{
+  "version": "1.6.7",
+  "case": "FAILOVER-CLEANUP-FIX",
+  "kind": "positive",
+  "machines": [
+    "b53a37f4-1464-4512-9259-33a7df091e3e"
+  ],
+  "executed_at": "2026-10-24T13:31:02.803Z",
+  "executor": "lead",
+  "requirement": "a remote-relay in-process stack can be stopped and restarted for failover: the cleanup must not assume a local relay object, a failed stop must not skip the restart, and the switch must leave a diagnostics log",
+  "task_id": "01M5ANYMVMFSSP3",
+  "result": {
+    "outcome": "ok",
+    "assertions": {
+      "failover_suite": "5/5",
+      "relay_suite": "200/200",
+      "cleanup": "null-safe and idempotent",
+      "diagnostics": "xclient/relay-failover.log"
+    }
+  },
+  "note": "v1.6.6 field drill stopped the stack and never restarted it: relay.close() threw on remote-relay stacks"
+}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@twinsearth/w2m-dsh-plugin",
-  "version": "1.6.6",
+  "version": "1.6.7",
   "description": "Run one project across your own Windows and macOS machines from a single DeepSeek account, through 11 w2m_* tools. Four coordination modes (replicate, split, broadcast, pipeline) and six aggregate verdicts (consistent, divergent, divergent-platform, failed, partial, unverifiable). P2P hole punching by default, with a bundled relay, STUN responder and ledger fallback, so the fleet needs no third-party service. Joining needs a pairing code plus autoStartAgent: true; installing alone does not join.",
   "type": "module",
   "exports": {

package/plugin.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://agent-plugins.org/schemas/1.0.0/plugin.schema.json",
   "name": "twinsearth-w2m-dsh",
-  "version": "1.6.6",
+  "version": "1.6.7",
   "description": "Run one project across your own Windows and macOS machines from a single DeepSeek account, through 12 w2m_* tools. Four coordination modes (replicate, split, broadcast, pipeline) and six aggregate verdicts (consistent, divergent, divergent-platform, failed, partial, unverifiable). P2P hole punching by default, with a bundled relay, STUN responder and ledger fallback, so the fleet needs no third-party service. Joining needs a pairing code plus autoStartAgent: true; installing alone does not join.",
   "license": "MIT",
   "keywords": [

package/src/plugin/tools.mjs

@@ -281,7 +281,7 @@ const MAX_STAGES = 16;
  * `package.json` during `scripts/pack.mjs` rather than hand-maintained, and `W2M_PLUGIN_VERSION`
  * overrides it at runtime for tests and packagers.
  */
-const PLUGIN_VERSION = '1.6.6';
+const PLUGIN_VERSION = '1.6.7';
 
 /** How long any single HTTP call may take before it is aborted, in milliseconds. */
 const DEFAULT_REQUEST_TIMEOUT_MS = 15_000;
@@ -7407,6 +7407,19 @@ async function probeRelayHealth(url, timeoutMs) {
  * relay-failover.mjs and is unit-tested there; this function only supplies health and performs the
  * restart.
  */
+/** Append one line to the failover diagnostics log; failures are swallowed (diagnostics must not break the switch). */
+async function appendFailoverLog(cfg, line) {
+  try {
+    const { appendFileSync, mkdirSync, statSync, writeFileSync } = await import('node:fs');
+    const { join, dirname } = await import('node:path');
+    const { homedir } = await import('node:os');
+    const dir = cfg.stateDir ?? join(process.env.DSH_HOME ?? join(homedir(), '.dsh'), 'xclient');
+    const file = join(dir, 'relay-failover.log');
+    mkdirSync(dirname(file), { recursive: true });
+    try { if (statSync(file).size > 256 * 1024) writeFileSync(file, ''); } catch { /* absent is fine */ }
+    appendFileSync(file, new Date().toISOString() + ' ' + line + '\n');
+  } catch { /* diagnostics only */ }
+}
 async function startFailoverStack(ctx, cfg) {
   let stop = await startLocalStack(ctx, cfg);
   const pool = relayPool(cfg);
@@ -7422,6 +7435,7 @@ async function startFailoverStack(ctx, cfg) {
     const health = {};
     await Promise.all(pool.map(async (url) => { health[url] = await probeRelayHealth(url, FAILOVER_DEFAULTS.probeTimeoutMs); }));
     const decision = failover.assess(health);
+    await appendFailoverLog(cfg, 'probe ' + JSON.stringify(health) + ' -> ' + decision.action + (decision.target ? ' ' + decision.target : ''));
     relayFailoverStatus = failover.snapshot();
     if (decision.action !== 'switch') return;
     switching = true;
@@ -8077,7 +8091,14 @@ async function startLocalStack(ctx, cfg) {
     ensureFleetJoin = null;
     inProcessAgentStatus = null;
     await agent.stop();
-    await relay.close();
+    // v1.6.7: `relay` is only a real relay object when THIS stack started a local relay; a remote-
+    // relay stack has null here. The unconditional close() threw, which is what made the first
+    // failover attempt stop the stack and never restart it.
+    if (relay !== null && typeof relay.close === 'function') {
+      try { await relay.close(); } catch (error) {
+        ctx.logger?.warn?.('w2m: local relay close failed: ' + (error instanceof Error ? error.message : String(error)));
+      }
+    }
   };
 }