fix(install): preserve user fleet data on re-seed + refresh active units (CRITICAL) (#632)

Co-authored-by: Jason Woltje <jason@diversecanvas.com>
Co-committed-by: Jason Woltje <jason@diversecanvas.com>

packages/mosaic/src/cli.ts

@@ -27,6 +27,7 @@ import {
   formatAllPackagesTable,
   getInstallAllCommand,
   runFrameworkReseed,
+  refreshActiveFleetUnits,
   readRosterAgentNames,
   buildRelaunchCommands,
   FRAMEWORK_RESEED_PACKAGE,
@@ -466,6 +467,12 @@ program
       const reseed = runFrameworkReseed();
       if (reseed.ok) {
         console.log('✔ Framework re-seeded.');
+        // Propagate shipped systemd unit fixes to the ACTIVE units (re-seed only
+        // touches ~/.config/mosaic/systemd/user; systemd runs ~/.config/systemd/user).
+        const units = refreshActiveFleetUnits();
+        if (units.refreshed.length > 0) {
+          console.log(`✔ Refreshed ${units.refreshed.length} active systemd unit(s).`);
+        }
         const agents = readRosterAgentNames();
         if (agents.length > 0) {
           if (opts.relaunch) {

packages/mosaic/src/config/file-adapter.test.ts

@@ -153,6 +153,30 @@ describe('FileConfigAdapter.syncFramework — defaults seeding', () => {
     expect(readFileSync(join(fixture.mosaicHome, 'AGENTS.md'), 'utf-8')).toBe('# AGENTS default\n');
   });
 
+  it('preserves user fleet data (roster.yaml, agents/, run/) through a keep-mode sync', async () => {
+    // Regression for the roster-loss bug (#631): user-authored fleet files must
+    // survive the framework re-seed that `mosaic update` runs.
+    mkdirSync(join(fixture.mosaicHome, 'fleet', 'run'), { recursive: true });
+    mkdirSync(join(fixture.mosaicHome, 'fleet', 'agents'), { recursive: true });
+    writeFileSync(join(fixture.mosaicHome, 'fleet', 'roster.yaml'), 'version: 1\nMINE\n');
+    writeFileSync(join(fixture.mosaicHome, 'fleet', 'run', 'a.hb'), 'ts=x\n');
+    writeFileSync(join(fixture.mosaicHome, 'fleet', 'agents', 'a.env'), 'X=1\n');
+    // The framework ships fleet/examples — it should still seed/refresh.
+    mkdirSync(join(fixture.sourceDir, 'fleet', 'examples'), { recursive: true });
+    writeFileSync(join(fixture.sourceDir, 'fleet', 'examples', 'general.yaml'), '# preset\n');
+
+    const adapter = new FileConfigAdapter(fixture.mosaicHome, fixture.sourceDir);
+    await adapter.syncFramework('keep');
+
+    expect(readFileSync(join(fixture.mosaicHome, 'fleet', 'roster.yaml'), 'utf-8')).toBe(
+      'version: 1\nMINE\n',
+    );
+    expect(existsSync(join(fixture.mosaicHome, 'fleet', 'run', 'a.hb'))).toBe(true);
+    expect(existsSync(join(fixture.mosaicHome, 'fleet', 'agents', 'a.env'))).toBe(true);
+    // framework-owned fleet/examples is seeded
+    expect(existsSync(join(fixture.mosaicHome, 'fleet', 'examples', 'general.yaml'))).toBe(true);
+  });
+
   it('is a no-op for seeding when defaults/ dir does not exist', async () => {
     rmSync(fixture.defaultsDir, { recursive: true });
 

packages/mosaic/src/config/file-adapter.ts

@@ -173,6 +173,13 @@ export class FileConfigAdapter implements ConfigService {
             'memory',
             'sources',
             'credentials',
+            // User-authored fleet data MUST survive `mosaic update`'s re-seed.
+            // The framework seeds only fleet/examples + fleet/roles +
+            // fleet/roster.schema.json; the operator's roster, per-agent env, and
+            // heartbeat run dir stay user-owned. (Mirror of install.sh PRESERVE_PATHS.)
+            'fleet/*.yaml',
+            'fleet/agents',
+            'fleet/run',
           ]
         : [];
 

packages/mosaic/src/runtime/update-checker.reseed.spec.ts

@@ -7,7 +7,9 @@ import {
   buildRelaunchCommands,
   readRosterAgentNames,
   runFrameworkReseed,
+  refreshActiveFleetUnits,
 } from './update-checker.js';
+import { existsSync, readFileSync } from 'node:fs';
 
 /**
  * F3-m3 / R13: `mosaic update` re-seeds the framework + (opt-in) relaunches
@@ -83,3 +85,41 @@ describe('runFrameworkReseed', () => {
     rmSync(missing, { recursive: true, force: true });
   });
 });
+
+describe('refreshActiveFleetUnits', () => {
+  let root: string;
+  let mosaicHome: string;
+  let configHome: string;
+
+  beforeEach(() => {
+    root = mkdtempSync(join(tmpdir(), 'mosaic-units-'));
+    mosaicHome = join(root, 'mosaic');
+    configHome = join(root, 'config');
+    mkdirSync(join(mosaicHome, 'systemd', 'user'), { recursive: true });
+    mkdirSync(join(configHome, 'systemd', 'user'), { recursive: true });
+    // Freshly re-seeded units (new content).
+    writeFileSync(join(mosaicHome, 'systemd', 'user', 'mosaic-agent@.service'), 'NEW\n');
+    writeFileSync(join(mosaicHome, 'systemd', 'user', 'mosaic-tmux-holder.service'), 'NEW\n');
+  });
+  afterEach(() => rmSync(root, { recursive: true, force: true }));
+
+  it('refreshes active units when a fleet is already installed', () => {
+    // Active dir already carries mosaic units (stale) → fleet is installed.
+    writeFileSync(join(configHome, 'systemd', 'user', 'mosaic-agent@.service'), 'OLD\n');
+    const res = refreshActiveFleetUnits(mosaicHome, {
+      XDG_CONFIG_HOME: configHome,
+    } as NodeJS.ProcessEnv);
+    expect(res.refreshed).toContain('mosaic-agent@.service');
+    expect(
+      readFileSync(join(configHome, 'systemd', 'user', 'mosaic-agent@.service'), 'utf-8'),
+    ).toBe('NEW\n');
+  });
+
+  it('is a no-op when no fleet is installed (active dir has no mosaic units)', () => {
+    const res = refreshActiveFleetUnits(mosaicHome, {
+      XDG_CONFIG_HOME: configHome,
+    } as NodeJS.ProcessEnv);
+    expect(res.refreshed).toEqual([]);
+    expect(existsSync(join(configHome, 'systemd', 'user', 'mosaic-agent@.service'))).toBe(false);
+  });
+});

packages/mosaic/src/runtime/update-checker.ts

@@ -14,7 +14,14 @@
  */
 
 import { execSync } from 'node:child_process';
-import { existsSync, mkdirSync, readFileSync, writeFileSync } from 'node:fs';
+import {
+  existsSync,
+  mkdirSync,
+  readFileSync,
+  writeFileSync,
+  readdirSync,
+  copyFileSync,
+} from 'node:fs';
 import { homedir } from 'node:os';
 import { dirname, join, resolve } from 'node:path';
 import { fileURLToPath } from 'node:url';
@@ -536,6 +543,47 @@ export function readRosterAgentNames(mosaicHome = join(homedir(), '.config', 'mo
   return names;
 }
 
+/**
+ * Refresh the ACTIVE systemd user units from the freshly re-seeded copies.
+ *
+ * The re-seed updates `~/.config/mosaic/systemd/user/*.service`, but the units
+ * systemd actually runs live at `~/.config/systemd/user/`. Without this copy,
+ * shipped unit fixes (e.g. the socket-env change) never take effect after
+ * `mosaic update` until `mosaic fleet install` is re-run. Best-effort + scoped:
+ * only refreshes when a fleet is already installed (the active dir already
+ * carries `mosaic-*` units), so non-fleet hosts are untouched.
+ */
+export function refreshActiveFleetUnits(
+  mosaicHome = join(homedir(), '.config', 'mosaic'),
+  env: NodeJS.ProcessEnv = process.env,
+): { refreshed: string[]; ok: boolean; reason?: string } {
+  const src = join(mosaicHome, 'systemd', 'user');
+  const configHome = env['XDG_CONFIG_HOME'] ?? join(homedir(), '.config');
+  const dest = join(configHome, 'systemd', 'user');
+  if (!existsSync(src)) return { refreshed: [], ok: true };
+  // Only refresh when a fleet is already installed (active dir has mosaic units).
+  const fleetInstalled =
+    existsSync(dest) &&
+    readdirSync(dest).some((f) => f.startsWith('mosaic-') && f.endsWith('.service'));
+  if (!fleetInstalled) return { refreshed: [], ok: true };
+  const units = readdirSync(src).filter((f) => f.startsWith('mosaic-') && f.endsWith('.service'));
+  const refreshed: string[] = [];
+  for (const unit of units) {
+    try {
+      copyFileSync(join(src, unit), join(dest, unit));
+      refreshed.push(unit);
+    } catch {
+      // best-effort per unit
+    }
+  }
+  try {
+    execSync('systemctl --user daemon-reload', { stdio: 'ignore', timeout: 15_000 });
+  } catch {
+    // non-systemd host or no session bus — non-fatal
+  }
+  return { refreshed, ok: true };
+}
+
 /** Build the per-agent systemd relaunch commands (drain+relaunch via restart). */
 export function buildRelaunchCommands(agentNames: string[]): string[][] {
   return agentNames.map((name) => [