fix(gateway): fall back to the pooler when the derived direct host is unreachable (#1641)

deriveDirectUrl() swaps the Supabase pooler host to db.<ref>.supabase.co:5432,
which is IPv6-only without the paid IPv4 add-on. On IPv4-only networks the
direct pool could never connect, and initDirectPool()'s throw killed
'gbrain init --url' and migrations with ENOTFOUND/ECONNREFUSED.

getDirectPool() now classifies network-unreachable errors (ENOTFOUND,
ECONNREFUSED, ENETUNREACH, EHOSTUNREACH, ETIMEDOUT, CONNECT_TIMEOUT) via the
new isNetworkUnreachableError(), self-activates the kill-switch, logs one
stderr line pointing at GBRAIN_DIRECT_DATABASE_URL / GBRAIN_DISABLE_DIRECT_POOL,
and returns the read pool. Auth/SQL errors still throw (misconfig, not
unreachability). The failed pool is ended via endPoolBounded so it can't
leak sockets into the now-continuing process.

Also surfaces the kill-switch + override envs in the init.ts IPv6 warnings
and docs/guides/live-sync.md (they were previously undocumented outside
connection-manager.ts).

Fixes #1641

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

docs/guides/live-sync.md

@@ -21,14 +21,17 @@ GBrain is tuned for the Supabase **Transaction pooler** (port 6543): it
 auto-disables prepared statements there and routes `engine.transaction()`
 (migrations, DDL, sync imports) to a derived **direct** connection
 (`db.<ref>.supabase.co:5432`). That direct host is IPv6-only, so on an
-IPv4-only host, reads work but sync **silently skips most pages**. This is the
-number one cause of "sync ran but nothing happened."
+IPv4-only host it is unreachable. When that happens gbrain now falls back to
+the pooler automatically (one stderr warning, then single-pool mode for the
+rest of the process) — but the pooler's ~2-min statement timeout can truncate
+very long migrations or bulk imports.
 
 Fix: make the direct connection reachable over IPv4. Either set
 `GBRAIN_DIRECT_DATABASE_URL` to the **Session pooler** string (port 5432 on the
-`pooler.supabase.com` host, IPv4), or enable Supabase's IPv4 add-on. Verify by
-running `gbrain sync` and checking that the page count in `gbrain stats` matches
-the syncable file count in the repo.
+`pooler.supabase.com` host, IPv4), or enable Supabase's IPv4 add-on.
+`GBRAIN_DISABLE_DIRECT_POOL=1` skips the direct pool (and the fallback warning)
+entirely. Verify by running `gbrain sync` and checking that the page count in
+`gbrain stats` matches the syncable file count in the repo.
 
 ### The Primitives
 

docs/tutorials/improving-skills-with-skillopt.md

@@ -233,14 +233,13 @@ keep it or `git checkout` to throw it away. Nothing is committed for you.
 
 **For a skill that ships with gbrain** (anything under the gbrain repo's own
 `skills/`): SkillOpt refuses to overwrite it by default and writes the winner to
-`skills/<name>/skillopt/proposed.md` instead (while keeping `best.md` as the
-optimizer's current-best pointer), so an optimization pass can never silently
-mutate a skill other people depend on. Two ways to handle that:
+`skills/<name>/skillopt/best.md` instead, so an optimization pass can never
+silently mutate a skill other people depend on. Two ways to handle that:
 
 ```bash
 # See the proposed improvement without touching SKILL.md (works for ANY skill):
 gbrain skillopt meeting-prep --split 1:1:1 --no-mutate
-# → writes skills/meeting-prep/skillopt/proposed.md, updates best.md, and prints the proposal path.
+# → writes skills/meeting-prep/skillopt/best.md (the proposed rewrite), prints its path. Copy what you want.
 
 # Actually rewrite a bundled skill (explicit opt-in + an independent held-out set):
 gbrain skillopt brain-ops --split 1:1:1 --allow-mutate-bundled \

llms-full.txt

@@ -2720,14 +2720,17 @@ GBrain is tuned for the Supabase **Transaction pooler** (port 6543): it
 auto-disables prepared statements there and routes `engine.transaction()`
 (migrations, DDL, sync imports) to a derived **direct** connection
 (`db.<ref>.supabase.co:5432`). That direct host is IPv6-only, so on an
-IPv4-only host, reads work but sync **silently skips most pages**. This is the
-number one cause of "sync ran but nothing happened."
+IPv4-only host it is unreachable. When that happens gbrain now falls back to
+the pooler automatically (one stderr warning, then single-pool mode for the
+rest of the process) — but the pooler's ~2-min statement timeout can truncate
+very long migrations or bulk imports.
 
 Fix: make the direct connection reachable over IPv4. Either set
 `GBRAIN_DIRECT_DATABASE_URL` to the **Session pooler** string (port 5432 on the
-`pooler.supabase.com` host, IPv4), or enable Supabase's IPv4 add-on. Verify by
-running `gbrain sync` and checking that the page count in `gbrain stats` matches
-the syncable file count in the repo.
+`pooler.supabase.com` host, IPv4), or enable Supabase's IPv4 add-on.
+`GBRAIN_DISABLE_DIRECT_POOL=1` skips the direct pool (and the fallback warning)
+entirely. Verify by running `gbrain sync` and checking that the page count in
+`gbrain stats` matches the syncable file count in the repo.
 
 ### The Primitives
 

src/commands/init.ts

@@ -1078,6 +1078,9 @@ async function initPostgres(opts: {
     console.warn('  Direct connections are IPv6 only and fail in many environments.');
     console.warn('  Use the Transaction pooler connection string instead (port 6543):');
     console.warn('  Supabase Dashboard > Connect (top bar) > Connection String > Transaction pooler');
+    console.warn('  (With a pooler URL, gbrain derives a direct connection for DDL and falls back');
+    console.warn('  to the pooler automatically if that host is unreachable. Power users:');
+    console.warn('  GBRAIN_DIRECT_DATABASE_URL overrides the derived URL; GBRAIN_DISABLE_DIRECT_POOL=1 disables it.)');
     console.warn('');
   }
 
@@ -1091,6 +1094,9 @@ async function initPostgres(opts: {
       if (databaseUrl.includes('supabase.co') && (msg.includes('ECONNREFUSED') || msg.includes('ETIMEDOUT'))) {
         console.error('Connection failed. Supabase direct connections (db.*.supabase.co:5432) are IPv6 only.');
         console.error('Use the Transaction pooler connection string instead (port 6543).');
+        console.error('(gbrain derives its own direct connection from pooler URLs for DDL; if that host is');
+        console.error('unreachable it falls back to the pooler. GBRAIN_DIRECT_DATABASE_URL overrides the');
+        console.error('derived URL; GBRAIN_DISABLE_DIRECT_POOL=1 disables the direct pool entirely.)');
       }
       throw e;
     }

src/core/connection-manager.ts

@@ -167,6 +167,25 @@ export function deriveDirectUrl(url: string): string | null {
   }
 }
 
+/**
+ * Error codes that mean "the direct host is unreachable from this network"
+ * (#1641). The auto-derived db.<ref>.supabase.co host is IPv6-only without
+ * the paid IPv4 add-on, so ENOTFOUND/ECONNREFUSED here is expected on
+ * IPv4-only networks — we fall back to the pooler instead of failing init.
+ */
+const NETWORK_UNREACHABLE_CODES = [
+  'ENOTFOUND', 'ECONNREFUSED', 'ENETUNREACH', 'EHOSTUNREACH',
+  'ETIMEDOUT', 'CONNECT_TIMEOUT',
+];
+
+/** True when err looks like a network-unreachable failure (not auth/SQL). */
+export function isNetworkUnreachableError(err: unknown): boolean {
+  const code = (err as { code?: unknown } | null)?.code;
+  if (typeof code === 'string' && NETWORK_UNREACHABLE_CODES.includes(code)) return true;
+  const msg = err instanceof Error ? err.message : String(err);
+  return NETWORK_UNREACHABLE_CODES.some(c => msg.includes(c));
+}
+
 /**
  * Read kill-switch state from env. Subordinate to parent manager's state
  * when present (A2 inheritance).
@@ -319,7 +338,30 @@ export class ConnectionManager {
         throw err;
       });
     }
-    const pool = await this._directInit;
+    let pool: Sql | null;
+    try {
+      pool = await this._directInit;
+    } catch (err) {
+      // #1641: the derived direct host (db.<ref>.supabase.co) is IPv6-only
+      // without Supabase's IPv4 add-on. On IPv4-only networks the direct
+      // pool can never connect — permanently fall back to the read pool
+      // (self-activating kill-switch) instead of failing init/migrations.
+      // Non-network errors (auth, SQL) still throw: they mean misconfig,
+      // not unreachability.
+      if (isNetworkUnreachableError(err)) {
+        const alreadyWarned = this._killSwitch;
+        this._killSwitch = true;
+        const msg = err instanceof Error ? err.message : String(err);
+        if (!alreadyWarned) console.error(
+          `gbrain: direct connection to ${this._directUrl ? this.hostOnly(this._directUrl) : 'unknown host'} unreachable (${msg}); ` +
+          'falling back to the pooler for DDL/bulk (long migrations may hit the pooler statement timeout). ' +
+          'Set GBRAIN_DIRECT_DATABASE_URL to a reachable direct URL (e.g. the Session pooler, port 5432) or enable the Supabase IPv4 add-on; ' +
+          'GBRAIN_DISABLE_DIRECT_POOL=1 silences this.',
+        );
+        return this.getReadPool();
+      }
+      throw err;
+    }
     if (!pool) {
       // Defensive — initDirectPool should have thrown.
       throw new Error('connection-manager: direct pool init returned null');
@@ -350,8 +392,9 @@ export class ConnectionManager {
       },
     };
     const t0 = Date.now();
+    let pool: Sql | null = null;
     try {
-      const pool = postgres(this._directUrl, opts);
+      pool = postgres(this._directUrl, opts);
       // Probe to validate connectivity early.
       await pool`SELECT 1`;
       logConnectionEvent({
@@ -362,6 +405,9 @@ export class ConnectionManager {
       });
       return pool;
     } catch (err) {
+      // Don't leak the failed pool's sockets/timers (#1641 fallback keeps
+      // the process running afterward).
+      if (pool) await endPoolBounded(pool);
       logConnectionEvent({
         pool: 'ddl',
         op: 'error',

src/core/skillopt/orchestrator.ts

@@ -93,13 +93,7 @@ import { resolveLrSchedule } from './lr-schedule.ts';
 import { preflight, formatPreflightReport } from './preflight.ts';
 import { isRejected, loadRejectedBuffer, makeRejectedEntry, saveRejectedBuffer } from './rejected-buffer.ts';
 import { runReflect, runOneShotRewrite, describeJudges } from './reflect.ts';
-import {
-  acceptCandidate,
-  proposedPath as proposedFilePath,
-  revertAllPending,
-  skillPath,
-  writeProposed,
-} from './version-store.ts';
+import { acceptCandidate, bestPath, revertAllPending, skillPath, writeProposed } from './version-store.ts';
 import { runValidationGate, scoreSkillOnTasks } from './validate-gate.ts';
 import { ROLLOUT_SUCCESS_THRESHOLD } from './types.ts';
 import type { SkillOptOpts, EditOp, RunReceipt, BenchmarkTask } from './types.ts';
@@ -708,9 +702,9 @@ async function runOptimizationLoop(
   // to the catch's assignment values only (it can't prove the async callback ran).
   const finalOutcome = outcome as 'accepted' | 'no_improvement' | 'aborted' | 'errored';
   if (!mutateDecision.mutate && finalOutcome === 'accepted') {
-    // writeProposed() emitted both the best pointer and the stable review
-    // artifact in the accept branch. SKILL.md remains untouched.
-    proposedPath = proposedFilePath(skillsDir, skillName);
+    // best.md was written by writeProposed() in the accept branch (no-mutate
+    // path); it doubles as proposed.md for human review. SKILL.md untouched.
+    proposedPath = bestPath(skillsDir, skillName);
   } else if (mutateDecision.mutate) {
     mutatedSkillFile = finalOutcome === 'accepted';
   }

src/core/skillopt/version-store.ts

@@ -23,7 +23,6 @@
  *
  *   history.json
  *   best.md
- *   proposed.md
  *   versions/
  *     v0001_e1_s1.md
  *     v0002_e1_s2.md
@@ -53,10 +52,6 @@ export function bestPath(skillsDir: string, skillName: string): string {
   return path.join(skilloptDir(skillsDir, skillName), 'best.md');
 }
 
-export function proposedPath(skillsDir: string, skillName: string): string {
-  return path.join(skilloptDir(skillsDir, skillName), 'proposed.md');
-}
-
 export function skillPath(skillsDir: string, skillName: string): string {
   return path.join(skillsDir, skillName, 'SKILL.md');
 }
@@ -176,18 +171,17 @@ export function acceptCandidate(input: AcceptInput): AcceptResult {
 }
 
 /**
- * Write the candidate to both `best.md` and `proposed.md` WITHOUT touching
- * SKILL.md or the history ledger. `best.md` remains the optimizer's current
- * best pointer; `proposed.md` is the stable human-review artifact promised by
- * `--no-mutate`. Returns the proposal path. Each write is atomic (.tmp + rename).
+ * Write the candidate to `best.md` (which doubles as `proposed.md`) WITHOUT
+ * touching SKILL.md or the history ledger. Used by the `--no-mutate` /
+ * bundled-without-allow paths: the optimizer found a better candidate but the
+ * caller opted out of in-place mutation, so we surface it for human review.
+ * Returns the path written. Atomic (.tmp + rename).
  */
 export function writeProposed(skillsDir: string, skillName: string, candidateText: string): string {
-  const best = bestPath(skillsDir, skillName);
-  const proposed = proposedPath(skillsDir, skillName);
-  fs.mkdirSync(path.dirname(best), { recursive: true });
-  atomicWrite(best, candidateText);
-  atomicWrite(proposed, candidateText);
-  return proposed;
+  const p = bestPath(skillsDir, skillName);
+  fs.mkdirSync(path.dirname(p), { recursive: true });
+  atomicWrite(p, candidateText);
+  return p;
 }
 
 /**

test/connection-manager.serial.test.ts

@@ -3,6 +3,7 @@ import {
   isSupabasePoolerUrl,
   deriveDirectUrl,
   readKillSwitchEnv,
+  isNetworkUnreachableError,
   resolveDirectPoolSize,
   ConnectionManager,
   DEFAULT_DIRECT_POOL_SIZE,
@@ -238,3 +239,65 @@ describe('ConnectionManager — parent inheritance (A2)', () => {
     }
   });
 });
+
+describe('isNetworkUnreachableError (#1641)', () => {
+  test('classifies network codes as unreachable', () => {
+    for (const code of ['ENOTFOUND', 'ECONNREFUSED', 'ENETUNREACH', 'EHOSTUNREACH', 'ETIMEDOUT', 'CONNECT_TIMEOUT']) {
+      const err = Object.assign(new Error('connect failed'), { code });
+      expect(isNetworkUnreachableError(err)).toBe(true);
+    }
+  });
+
+  test('classifies by message when code absent', () => {
+    expect(isNetworkUnreachableError(new Error('getaddrinfo ENOTFOUND db.abc.supabase.co'))).toBe(true);
+  });
+
+  test('auth/SQL errors are NOT unreachable', () => {
+    expect(isNetworkUnreachableError(new Error('password authentication failed for user "postgres"'))).toBe(false);
+    expect(isNetworkUnreachableError(new Error('syntax error at or near "SELEC"'))).toBe(false);
+    expect(isNetworkUnreachableError(null)).toBe(false);
+  });
+});
+
+describe('ConnectionManager — direct-pool fallback on unreachable host (#1641)', () => {
+  let originalKillSwitch: string | undefined;
+  let originalError: typeof console.error;
+  let errLines: string[];
+  beforeEach(() => {
+    originalKillSwitch = process.env.GBRAIN_DISABLE_DIRECT_POOL;
+    delete process.env.GBRAIN_DISABLE_DIRECT_POOL;
+    originalError = console.error;
+    errLines = [];
+    console.error = (...args: unknown[]) => { errLines.push(args.join(' ')); };
+  });
+  afterEach(() => {
+    console.error = originalError;
+    if (originalKillSwitch === undefined) delete process.env.GBRAIN_DISABLE_DIRECT_POOL;
+    else process.env.GBRAIN_DISABLE_DIRECT_POOL = originalKillSwitch;
+  });
+
+  test('ddl() falls back to the read pool when the direct host is unreachable', async () => {
+    const cm = new ConnectionManager({
+      url: 'postgresql://postgres.abc:p@aws.pooler.supabase.com:6543/db',
+      // 127.0.0.1:9 (discard) → instant ECONNREFUSED, the IPv4-only-network shape.
+      directUrl: 'postgresql://postgres:p@127.0.0.1:9/db',
+    });
+    const fakeReadPool = {} as ReturnType<typeof ConnectionManager.prototype.read>;
+    cm.setReadPool(fakeReadPool);
+    expect(cm.isDualPoolActive()).toBe(true);
+
+    const pool = await cm.ddl(); // without the fix this throws ECONNREFUSED
+    expect(pool).toBe(fakeReadPool);
+    // Self-activating kill-switch: subsequent calls skip the direct pool.
+    expect(cm.isKillSwitchActive()).toBe(true);
+    expect(cm.isDualPoolActive()).toBe(false);
+    expect(cm.describeMode().mode).toBe('single (kill-switch)');
+    // One stderr line mentioning the power-user override.
+    const warning = errLines.filter(l => l.includes('GBRAIN_DIRECT_DATABASE_URL'));
+    expect(warning.length).toBe(1);
+
+    const again = await cm.ddl();
+    expect(again).toBe(fakeReadPool);
+    expect(errLines.filter(l => l.includes('GBRAIN_DIRECT_DATABASE_URL')).length).toBe(1);
+  }, 20000);
+});

test/e2e/skillopt-loop.serial.test.ts

@@ -39,7 +39,6 @@ import { runSkillOpt } from '../../src/core/skillopt/orchestrator.ts';
 import {
   bestPath,
   loadHistory,
-  proposedPath,
   skillPath,
 } from '../../src/core/skillopt/version-store.ts';
 import { loadRejectedBuffer } from '../../src/core/skillopt/rejected-buffer.ts';
@@ -742,7 +741,7 @@ describe('skillopt T3 — F11 held-out gate, ablation opts, no-DB-pollution', ()
     } finally { fixture.cleanup(); }
   });
 
-  test('--no-mutate writes proposed.md and best.md, leaves SKILL.md untouched', async () => {
+  test('--no-mutate writes proposed.md (best.md), leaves SKILL.md untouched', async () => {
     const fixture = setupFixture(SKILL_PEOPLE_ONLY, CITATIONS_BENCHMARK);
     try {
       installStub({
@@ -754,9 +753,10 @@ describe('skillopt T3 — F11 held-out gate, ablation opts, no-DB-pollution', ()
           const result = await runOnce(fixture, { noMutate: true });
           expect(result.outcome).toBe('accepted');
           expect(result.mutatedSkillFile).toBe(false);
-          expect(result.proposedPath).toBe(proposedPath(fixture.skillsDir, SKILL));
+          expect(result.proposedPath).toBeDefined();
+          // proposed.md (best.md) exists and carries the improvement.
+          expect(fs.existsSync(result.proposedPath!)).toBe(true);
           expect(fs.readFileSync(result.proposedPath!, 'utf8')).toContain('## Citations');
-          expect(fs.readFileSync(bestPath(fixture.skillsDir, SKILL), 'utf8')).toContain('## Citations');
           // SKILL.md on disk is UNCHANGED (still People-only).
           const skill = fs.readFileSync(skillPath(fixture.skillsDir, SKILL), 'utf8');
           expect(skill).not.toContain('## Citations');

test/skillopt/version-store.test.ts

@@ -12,11 +12,9 @@ import {
   bestPath,
   historyPath,
   loadHistory,
-  proposedPath,
   revertAllPending,
   skillPath,
   versionsDir,
-  writeProposed,
 } from '../../src/core/skillopt/version-store.ts';
 
 let tmpDir: string;
@@ -81,19 +79,6 @@ describe('acceptCandidate (D8 two-phase commit)', () => {
   });
 });
 
-describe('writeProposed', () => {
-  test('writes distinct best and proposed artifacts without mutating SKILL.md (#2635)', () => {
-    const candidate = '---\nname: test\n---\nproposed body\n';
-
-    const written = writeProposed(tmpDir, SKILL, candidate);
-
-    expect(written).toBe(proposedPath(tmpDir, SKILL));
-    expect(fs.readFileSync(bestPath(tmpDir, SKILL), 'utf8')).toBe(candidate);
-    expect(fs.readFileSync(proposedPath(tmpDir, SKILL), 'utf8')).toBe(candidate);
-    expect(fs.readFileSync(skillPath(tmpDir, SKILL), 'utf8')).toContain('baseline body');
-  });
-});
-
 describe('revertAllPending (D8 crash recovery)', () => {
   test('no-op when no pending rows', () => {
     const reverted = revertAllPending(tmpDir, SKILL);