Harden lock backend config validation and fallback behavior

This commit is contained in:
thePR0M3TH3AN committed 2026-02-14 18:48:21 -05:00
1 parent 9a64554d87
commit 84d3e7eded
7 files changed
+467 -293

No files matched your search

+2
View File
@@ -8,5 +8,7 @@ export const DEFAULT_RELAYS = [
export const DEFAULT_TTL = 7200;
export const DEFAULT_NAMESPACE = 'torch';
export const DEFAULT_QUERY_TIMEOUT_MS = 15_000;
export const DEFAULT_PUBLISH_TIMEOUT_MS = 10_000;
export const DEFAULT_MIN_SUCCESSFUL_PUBLISHES = 1;
export const VALID_CADENCES = new Set(['daily', 'weekly']);
export const KIND_APP_DATA = 30078;
+24 -1
View File
@@ -32,6 +32,9 @@ node bin/torch-lock.mjs list
- `NOSTR_LOCK_RELAYS`
- `NOSTR_LOCK_TTL`
- `NOSTR_LOCK_QUERY_TIMEOUT_MS`
- `NOSTR_LOCK_PUBLISH_TIMEOUT_MS`
- `NOSTR_LOCK_MIN_SUCCESSFUL_PUBLISHES`
- `NOSTR_LOCK_RELAY_FALLBACKS`
- `NOSTR_LOCK_DAILY_ROSTER`
- `NOSTR_LOCK_WEEKLY_ROSTER`
- `TORCH_CONFIG_PATH`
@@ -50,7 +53,10 @@ Common settings:
- `nostrLock.namespace` — namespace prefix used in d-tags and hashtags.
- `nostrLock.relays` — relay list for check/lock/list operations.
- `nostrLock.ttlSeconds` — default lock TTL.
- `nostrLock.queryTimeoutMs` — relay query timeout.
- `nostrLock.queryTimeoutMs` — relay query timeout (ms, valid range: 100..120000).
- `nostrLock.publishTimeoutMs` — per-relay publish timeout (ms, valid range: 100..120000).
- `nostrLock.minSuccessfulRelayPublishes` — minimum successful publishes required before lock acquisition continues (default: `1`).
- `nostrLock.relayFallbacks` — optional fallback relay URLs used when primary query/publish attempts fail quorum.
- `nostrLock.dailyRoster` / `nostrLock.weeklyRoster` — optional per-project roster overrides.
- `dashboard.defaultCadenceView` — default dashboard view (`daily`, `weekly`, `all`).
- `dashboard.defaultStatusView` — default dashboard status filter (`active`, `all`).
@@ -64,6 +70,23 @@ For weekly repository-fit maintenance, TORCH also includes `src/prompts/weekly/r
Operational note: scheduler handoff commands are treated as required execution steps. A non-zero exit code (or missing command) is a hard failure: the scheduler writes a `_failed.md` task log, exits immediately, and does not publish `lock:complete` for that run.
## Lock backend production defaults
Recommended baseline for production scheduler runs:
- `nostrLock.relays`: 3+ geographically-diverse primary relays.
- `nostrLock.relayFallbacks`: 2 additional relays not present in primary list.
- `nostrLock.queryTimeoutMs`: `10000`
- `nostrLock.publishTimeoutMs`: `8000`
- `nostrLock.minSuccessfulRelayPublishes`: `2`
Validation behavior:
- Relay URLs must be absolute `ws://` or `wss://` URLs.
- Invalid relay URLs or invalid timeout/count ranges are fatal startup errors.
- Lock backend errors include phase (`query:primary`, `query:fallback`, `publish:primary`, `publish:fallback`), relay endpoint, and timeout value used.
## Scheduler lock reliability reporting
Run the lock reliability summary to aggregate recent scheduler outcomes by platform, cadence, backend error category, and relay endpoint:
+128 -38
View File
@@ -1,5 +1,10 @@
import { SimplePool } from 'nostr-tools/pool';
import { getQueryTimeoutMs } from './torch-config.mjs';
import {
getQueryTimeoutMs,
getPublishTimeoutMs,
getMinSuccessfulRelayPublishes,
getRelayFallbacks,
} from './torch-config.mjs';
import { KIND_APP_DATA } from './constants.mjs';
function nowUnix() {
@@ -42,56 +47,141 @@ function filterActiveLocks(locks) {
return locks.filter((lock) => !lock.expiresAt || lock.expiresAt > now);
}
export async function queryLocks(relays, cadence, dateStr, namespace) {
const pool = new SimplePool();
const tagFilter = `${namespace}-lock-${cadence}-${dateStr}`;
const queryTimeoutMs = getQueryTimeoutMs();
function withTimeout(promise, timeoutMs, timeoutMessage) {
let timeoutHandle;
const timeoutPromise = new Promise((_, reject) => {
timeoutHandle = setTimeout(() => reject(new Error(timeoutMessage)), timeoutMs);
});
return Promise.race([promise, timeoutPromise]).finally(() => {
if (timeoutHandle) clearTimeout(timeoutHandle);
});
}
function relayListLabel(relays) {
return relays.join(', ');
}
export async function queryLocks(relays, cadence, dateStr, namespace, deps = {}) {
const {
poolFactory = () => new SimplePool(),
getQueryTimeoutMsFn = getQueryTimeoutMs,
getRelayFallbacksFn = getRelayFallbacks,
errorLogger = console.error,
} = deps;
const pool = poolFactory();
const tagFilter = `${namespace}-lock-${cadence}-${dateStr}`;
const queryTimeoutMs = getQueryTimeoutMsFn();
const fallbackRelays = getRelayFallbacksFn();
const runQuery = async (relaySet, phase) => {
try {
const events = await withTimeout(
pool.querySync(relaySet, {
kinds: [KIND_APP_DATA],
'#t': [tagFilter],
}),
queryTimeoutMs,
`[${phase}] Relay query timed out after ${queryTimeoutMs}ms (relays: ${relayListLabel(relaySet)})`,
);
return filterActiveLocks(events.map(parseLockEvent));
} catch (err) {
const message = err instanceof Error ? err.message : String(err);
throw new Error(
`[${phase}] Relay query failed (timeout=${queryTimeoutMs}ms, relays=${relayListLabel(relaySet)}): ${message}`,
{ cause: err },
);
}
};
try {
const events = await Promise.race([
pool.querySync(relays, {
kinds: [KIND_APP_DATA],
'#t': [tagFilter],
}),
new Promise((_, reject) => {
timeoutHandle = setTimeout(() => reject(new Error('Relay query timed out')), queryTimeoutMs);
}),
]);
return filterActiveLocks(events.map(parseLockEvent));
} finally {
if (timeoutHandle) {
clearTimeout(timeoutHandle);
try {
return await runQuery(relays, 'query:primary');
} catch (primaryErr) {
if (!fallbackRelays.length) {
throw primaryErr;
}
errorLogger(`WARN: ${primaryErr.message}`);
errorLogger(`WARN: retrying query with fallback relays (${relayListLabel(fallbackRelays)})`);
return await runQuery(fallbackRelays, 'query:fallback');
}
pool.close(relays);
} finally {
const allRelays = [...new Set([...relays, ...fallbackRelays])];
pool.close(allRelays);
}
}
export async function publishLock(relays, event) {
const pool = new SimplePool();
async function publishToRelays(pool, relays, event, publishTimeoutMs, phase) {
const publishPromises = pool.publish(relays, event);
const settled = await Promise.allSettled(
publishPromises.map((publishPromise, index) => withTimeout(
publishPromise,
publishTimeoutMs,
`[${phase}] Publish timed out after ${publishTimeoutMs}ms (relay=${relays[index]})`,
)),
);
return settled.map((result, index) => {
if (result.status === 'fulfilled') {
return { relay: relays[index], success: true, phase };
}
const reason = result.reason;
const message = reason instanceof Error ? reason.message : String(reason ?? 'unknown');
return { relay: relays[index], success: false, phase, message };
});
}
export async function publishLock(relays, event, deps = {}) {
const {
poolFactory = () => new SimplePool(),
getPublishTimeoutMsFn = getPublishTimeoutMs,
getMinSuccessfulRelayPublishesFn = getMinSuccessfulRelayPublishes,
getRelayFallbacksFn = getRelayFallbacks,
} = deps;
const pool = poolFactory();
const publishTimeoutMs = getPublishTimeoutMsFn();
const minSuccesses = getMinSuccessfulRelayPublishesFn();
const fallbackRelays = getRelayFallbacksFn().filter((relay) => !relays.includes(relay));
const attempted = new Set();
const publishResults = [];
const attemptPhase = async (phaseRelays, phaseName) => {
if (!phaseRelays.length) return;
for (const relay of phaseRelays) {
attempted.add(relay);
}
const phaseResults = await publishToRelays(pool, phaseRelays, event, publishTimeoutMs, phaseName);
publishResults.push(...phaseResults);
};
try {
const publishPromises = pool.publish(relays, event);
const results = await Promise.allSettled(publishPromises);
const successes = results.filter((r) => r.status === 'fulfilled');
await attemptPhase(relays, 'publish:primary');
if (successes.length === 0) {
const errors = results.map((r, i) => {
if (r.status === 'rejected') {
const reason = r.reason;
const message = reason instanceof Error ? reason.message : String(reason ?? 'unknown');
return `${relays[i]}: ${message}`;
}
return `${relays[i]}: unknown`;
});
throw new Error(`Failed to publish to any relay:\n ${errors.join('\n ')}`);
let successCount = publishResults.filter((result) => result.success).length;
if (successCount < minSuccesses && fallbackRelays.length > 0) {
await attemptPhase(fallbackRelays, 'publish:fallback');
successCount = publishResults.filter((result) => result.success).length;
}
console.error(` Published to ${successes.length}/${relays.length} relays`);
if (successCount < minSuccesses) {
const failureLines = publishResults
.filter((result) => !result.success)
.map((result) => `${result.relay} (${result.phase}): ${result.message}`);
throw new Error(
`Failed relay publish quorum in publish phase: ${successCount}/${attempted.size} successful ` +
`(required=${minSuccesses}, timeout=${publishTimeoutMs}ms)\n ${failureLines.join('\n ')}`,
);
}
console.error(
` Published to ${successCount}/${attempted.size} relays ` +
`(required=${minSuccesses}, timeout=${publishTimeoutMs}ms)`,
);
return event;
} finally {
pool.close(relays);
pool.close([...attempted]);
}
}
+129 -7
View File
@@ -5,11 +5,52 @@ import {
DEFAULT_TTL,
DEFAULT_NAMESPACE,
DEFAULT_QUERY_TIMEOUT_MS,
DEFAULT_PUBLISH_TIMEOUT_MS,
DEFAULT_MIN_SUCCESSFUL_PUBLISHES,
} from './constants.mjs';
const DEFAULT_CONFIG_PATH = 'torch-config.json';
let cachedConfig = null;
const MIN_TIMEOUT_MS = 100;
const MAX_TIMEOUT_MS = 120_000;
function parsePositiveInteger(value) {
if (!Number.isFinite(value) || value <= 0) return null;
return Math.floor(value);
}
function parseRelayList(value) {
if (!Array.isArray(value)) return null;
const parsed = value.map((relay) => String(relay).trim()).filter(Boolean);
return parsed.length > 0 ? parsed : null;
}
function assertValidRelayUrl(relay, sourceLabel) {
let parsed;
try {
parsed = new URL(relay);
} catch {
throw new Error(`Invalid relay URL in ${sourceLabel}: "${relay}" (must be an absolute ws:// or wss:// URL)`);
}
if (parsed.protocol !== 'ws:' && parsed.protocol !== 'wss:') {
throw new Error(`Invalid relay URL in ${sourceLabel}: "${relay}" (protocol must be ws:// or wss://)`);
}
}
function assertTimeoutInRange(value, sourceLabel) {
if (!Number.isInteger(value) || value < MIN_TIMEOUT_MS || value > MAX_TIMEOUT_MS) {
throw new Error(
`Invalid ${sourceLabel}: ${value} (must be an integer between ${MIN_TIMEOUT_MS} and ${MAX_TIMEOUT_MS} ms)`,
);
}
}
function assertPositiveCount(value, sourceLabel) {
if (!Number.isInteger(value) || value <= 0) {
throw new Error(`Invalid ${sourceLabel}: ${value} (must be a positive integer)`);
}
}
function normalizeCadence(value, fallback) {
const normalized = String(value || '').trim().toLowerCase();
@@ -48,12 +89,13 @@ export function parseTorchConfig(raw, configPath = null) {
relays: Array.isArray(nostrLock.relays)
? nostrLock.relays.map((relay) => String(relay).trim()).filter(Boolean)
: null,
relayFallbacks: parseRelayList(nostrLock.relayFallbacks),
ttlSeconds: Number.isFinite(nostrLock.ttlSeconds) && nostrLock.ttlSeconds > 0
? Math.floor(nostrLock.ttlSeconds)
: null,
queryTimeoutMs: Number.isFinite(nostrLock.queryTimeoutMs) && nostrLock.queryTimeoutMs > 0
? Math.floor(nostrLock.queryTimeoutMs)
: null,
queryTimeoutMs: parsePositiveInteger(nostrLock.queryTimeoutMs),
publishTimeoutMs: parsePositiveInteger(nostrLock.publishTimeoutMs),
minSuccessfulRelayPublishes: parsePositiveInteger(nostrLock.minSuccessfulRelayPublishes),
dailyRoster: parseRoster(nostrLock.dailyRoster),
weeklyRoster: parseRoster(nostrLock.weeklyRoster),
},
@@ -99,21 +141,72 @@ export function loadTorchConfig(fileSystem = fs) {
}
cachedConfig = parseTorchConfig(raw, configPath);
validateLockBackendConfig(cachedConfig);
return cachedConfig;
}
function validateLockBackendConfig(config) {
const relays = config.nostrLock.relays || [];
const relayFallbacks = config.nostrLock.relayFallbacks || [];
for (const relay of relays) {
assertValidRelayUrl(relay, 'nostrLock.relays');
}
for (const relay of relayFallbacks) {
assertValidRelayUrl(relay, 'nostrLock.relayFallbacks');
}
if (config.nostrLock.queryTimeoutMs !== null) {
assertTimeoutInRange(config.nostrLock.queryTimeoutMs, 'nostrLock.queryTimeoutMs');
}
if (config.nostrLock.publishTimeoutMs !== null) {
assertTimeoutInRange(config.nostrLock.publishTimeoutMs, 'nostrLock.publishTimeoutMs');
}
if (config.nostrLock.minSuccessfulRelayPublishes !== null) {
assertPositiveCount(config.nostrLock.minSuccessfulRelayPublishes, 'nostrLock.minSuccessfulRelayPublishes');
}
}
function parseEnvRelayList(envValue, envName) {
const relays = envValue.split(',').map((r) => r.trim()).filter(Boolean);
for (const relay of relays) {
assertValidRelayUrl(relay, envName);
}
return relays;
}
function parseEnvInteger(envValue, envName) {
const parsed = parseInt(envValue, 10);
if (Number.isNaN(parsed)) {
throw new Error(`Invalid ${envName}: "${envValue}" (must be an integer)`);
}
return parsed;
}
export function getRelays() {
const config = loadTorchConfig();
const envRelays = process.env.NOSTR_LOCK_RELAYS;
if (envRelays) {
return envRelays.split(',').map((r) => r.trim()).filter(Boolean);
return parseEnvRelayList(envRelays, 'NOSTR_LOCK_RELAYS');
}
if (config.nostrLock.relays?.length) {
return config.nostrLock.relays;
}
for (const relay of DEFAULT_RELAYS) {
assertValidRelayUrl(relay, 'DEFAULT_RELAYS');
}
return DEFAULT_RELAYS;
}
export function getRelayFallbacks() {
const config = loadTorchConfig();
const envRelays = process.env.NOSTR_LOCK_RELAY_FALLBACKS;
if (envRelays) {
return parseEnvRelayList(envRelays, 'NOSTR_LOCK_RELAY_FALLBACKS');
}
return config.nostrLock.relayFallbacks || [];
}
export function getNamespace() {
const config = loadTorchConfig();
const namespace = (process.env.NOSTR_LOCK_NAMESPACE || config.nostrLock.namespace || DEFAULT_NAMESPACE).trim();
@@ -137,8 +230,37 @@ export function getQueryTimeoutMs() {
const config = loadTorchConfig();
const envValue = process.env.NOSTR_LOCK_QUERY_TIMEOUT_MS;
if (envValue) {
const parsed = parseInt(envValue, 10);
if (!Number.isNaN(parsed) && parsed > 0) return parsed;
const parsed = parseEnvInteger(envValue, 'NOSTR_LOCK_QUERY_TIMEOUT_MS');
assertTimeoutInRange(parsed, 'NOSTR_LOCK_QUERY_TIMEOUT_MS');
return parsed;
}
return config.nostrLock.queryTimeoutMs || DEFAULT_QUERY_TIMEOUT_MS;
const value = config.nostrLock.queryTimeoutMs || DEFAULT_QUERY_TIMEOUT_MS;
assertTimeoutInRange(value, 'effective query timeout');
return value;
}
export function getPublishTimeoutMs() {
const config = loadTorchConfig();
const envValue = process.env.NOSTR_LOCK_PUBLISH_TIMEOUT_MS;
if (envValue) {
const parsed = parseEnvInteger(envValue, 'NOSTR_LOCK_PUBLISH_TIMEOUT_MS');
assertTimeoutInRange(parsed, 'NOSTR_LOCK_PUBLISH_TIMEOUT_MS');
return parsed;
}
const value = config.nostrLock.publishTimeoutMs || DEFAULT_PUBLISH_TIMEOUT_MS;
assertTimeoutInRange(value, 'effective publish timeout');
return value;
}
export function getMinSuccessfulRelayPublishes() {
const config = loadTorchConfig();
const envValue = process.env.NOSTR_LOCK_MIN_SUCCESSFUL_PUBLISHES;
if (envValue) {
const parsed = parseEnvInteger(envValue, 'NOSTR_LOCK_MIN_SUCCESSFUL_PUBLISHES');
assertPositiveCount(parsed, 'NOSTR_LOCK_MIN_SUCCESSFUL_PUBLISHES');
return parsed;
}
const value = config.nostrLock.minSuccessfulRelayPublishes || DEFAULT_MIN_SUCCESSFUL_PUBLISHES;
assertPositiveCount(value, 'effective min successful relay publishes');
return value;
}