Add ez-assistant and kerberos service folders

This commit is contained in:
kelin
2026-02-11 14:56:03 -05:00
parent e4e8ae1b87
commit 9ccfb36923
4471 changed files with 746463 additions and 0 deletions

View File

@@ -0,0 +1,8 @@
{
"id": "diagnostics-otel",
"configSchema": {
"type": "object",
"additionalProperties": false,
"properties": {}
}
}

View File

@@ -0,0 +1,16 @@
import type { MoltbotPluginApi } from "clawdbot/plugin-sdk";
import { emptyPluginConfigSchema } from "clawdbot/plugin-sdk";
import { createDiagnosticsOtelService } from "./src/service.js";
const plugin = {
id: "diagnostics-otel",
name: "Diagnostics OpenTelemetry",
description: "Export diagnostics events to OpenTelemetry",
configSchema: emptyPluginConfigSchema(),
register(api: MoltbotPluginApi) {
api.registerService(createDiagnosticsOtelService());
},
};
export default plugin;

View File

@@ -0,0 +1,24 @@
{
"name": "@moltbot/diagnostics-otel",
"version": "2026.1.26",
"type": "module",
"description": "Moltbot diagnostics OpenTelemetry exporter",
"moltbot": {
"extensions": [
"./index.ts"
]
},
"dependencies": {
"@opentelemetry/api": "^1.9.0",
"@opentelemetry/api-logs": "^0.211.0",
"@opentelemetry/exporter-logs-otlp-http": "^0.211.0",
"@opentelemetry/exporter-metrics-otlp-http": "^0.211.0",
"@opentelemetry/exporter-trace-otlp-http": "^0.211.0",
"@opentelemetry/resources": "^2.5.0",
"@opentelemetry/sdk-logs": "^0.211.0",
"@opentelemetry/sdk-metrics": "^2.5.0",
"@opentelemetry/sdk-node": "^0.211.0",
"@opentelemetry/sdk-trace-base": "^2.5.0",
"@opentelemetry/semantic-conventions": "^1.39.0"
}
}

View File

@@ -0,0 +1,220 @@
import { beforeEach, describe, expect, test, vi } from "vitest";
const registerLogTransportMock = vi.hoisted(() => vi.fn());
const telemetryState = vi.hoisted(() => {
const counters = new Map<string, { add: ReturnType<typeof vi.fn> }>();
const histograms = new Map<string, { record: ReturnType<typeof vi.fn> }>();
const tracer = {
startSpan: vi.fn((_name: string, _opts?: unknown) => ({
end: vi.fn(),
setStatus: vi.fn(),
})),
};
const meter = {
createCounter: vi.fn((name: string) => {
const counter = { add: vi.fn() };
counters.set(name, counter);
return counter;
}),
createHistogram: vi.fn((name: string) => {
const histogram = { record: vi.fn() };
histograms.set(name, histogram);
return histogram;
}),
};
return { counters, histograms, tracer, meter };
});
const sdkStart = vi.hoisted(() => vi.fn().mockResolvedValue(undefined));
const sdkShutdown = vi.hoisted(() => vi.fn().mockResolvedValue(undefined));
const logEmit = vi.hoisted(() => vi.fn());
const logShutdown = vi.hoisted(() => vi.fn().mockResolvedValue(undefined));
vi.mock("@opentelemetry/api", () => ({
metrics: {
getMeter: () => telemetryState.meter,
},
trace: {
getTracer: () => telemetryState.tracer,
},
SpanStatusCode: {
ERROR: 2,
},
}));
vi.mock("@opentelemetry/sdk-node", () => ({
NodeSDK: class {
start = sdkStart;
shutdown = sdkShutdown;
},
}));
vi.mock("@opentelemetry/exporter-metrics-otlp-http", () => ({
OTLPMetricExporter: class {},
}));
vi.mock("@opentelemetry/exporter-trace-otlp-http", () => ({
OTLPTraceExporter: class {},
}));
vi.mock("@opentelemetry/exporter-logs-otlp-http", () => ({
OTLPLogExporter: class {},
}));
vi.mock("@opentelemetry/sdk-logs", () => ({
BatchLogRecordProcessor: class {},
LoggerProvider: class {
addLogRecordProcessor = vi.fn();
getLogger = vi.fn(() => ({
emit: logEmit,
}));
shutdown = logShutdown;
},
}));
vi.mock("@opentelemetry/sdk-metrics", () => ({
PeriodicExportingMetricReader: class {},
}));
vi.mock("@opentelemetry/sdk-trace-base", () => ({
ParentBasedSampler: class {},
TraceIdRatioBasedSampler: class {},
}));
vi.mock("@opentelemetry/resources", () => ({
Resource: class {
// eslint-disable-next-line @typescript-eslint/no-useless-constructor
constructor(_value?: unknown) {}
},
}));
vi.mock("@opentelemetry/semantic-conventions", () => ({
SemanticResourceAttributes: {
SERVICE_NAME: "service.name",
},
}));
vi.mock("clawdbot/plugin-sdk", async () => {
const actual = await vi.importActual<typeof import("clawdbot/plugin-sdk")>("clawdbot/plugin-sdk");
return {
...actual,
registerLogTransport: registerLogTransportMock,
};
});
import { createDiagnosticsOtelService } from "./service.js";
import { emitDiagnosticEvent } from "clawdbot/plugin-sdk";
describe("diagnostics-otel service", () => {
beforeEach(() => {
telemetryState.counters.clear();
telemetryState.histograms.clear();
telemetryState.tracer.startSpan.mockClear();
telemetryState.meter.createCounter.mockClear();
telemetryState.meter.createHistogram.mockClear();
sdkStart.mockClear();
sdkShutdown.mockClear();
logEmit.mockClear();
logShutdown.mockClear();
registerLogTransportMock.mockReset();
});
test("records message-flow metrics and spans", async () => {
const registeredTransports: Array<(logObj: Record<string, unknown>) => void> = [];
const stopTransport = vi.fn();
registerLogTransportMock.mockImplementation((transport) => {
registeredTransports.push(transport);
return stopTransport;
});
const service = createDiagnosticsOtelService();
await service.start({
config: {
diagnostics: {
enabled: true,
otel: {
enabled: true,
endpoint: "http://otel-collector:4318",
protocol: "http/protobuf",
traces: true,
metrics: true,
logs: true,
},
},
},
logger: {
info: vi.fn(),
warn: vi.fn(),
error: vi.fn(),
debug: vi.fn(),
},
});
emitDiagnosticEvent({
type: "webhook.received",
channel: "telegram",
updateType: "telegram-post",
});
emitDiagnosticEvent({
type: "webhook.processed",
channel: "telegram",
updateType: "telegram-post",
durationMs: 120,
});
emitDiagnosticEvent({
type: "message.queued",
channel: "telegram",
source: "telegram",
queueDepth: 2,
});
emitDiagnosticEvent({
type: "message.processed",
channel: "telegram",
outcome: "completed",
durationMs: 55,
});
emitDiagnosticEvent({
type: "queue.lane.dequeue",
lane: "main",
queueSize: 3,
waitMs: 10,
});
emitDiagnosticEvent({
type: "session.stuck",
state: "processing",
ageMs: 125_000,
});
emitDiagnosticEvent({
type: "run.attempt",
runId: "run-1",
attempt: 2,
});
expect(telemetryState.counters.get("moltbot.webhook.received")?.add).toHaveBeenCalled();
expect(telemetryState.histograms.get("moltbot.webhook.duration_ms")?.record).toHaveBeenCalled();
expect(telemetryState.counters.get("moltbot.message.queued")?.add).toHaveBeenCalled();
expect(telemetryState.counters.get("moltbot.message.processed")?.add).toHaveBeenCalled();
expect(telemetryState.histograms.get("moltbot.message.duration_ms")?.record).toHaveBeenCalled();
expect(telemetryState.histograms.get("moltbot.queue.wait_ms")?.record).toHaveBeenCalled();
expect(telemetryState.counters.get("moltbot.session.stuck")?.add).toHaveBeenCalled();
expect(telemetryState.histograms.get("moltbot.session.stuck_age_ms")?.record).toHaveBeenCalled();
expect(telemetryState.counters.get("moltbot.run.attempt")?.add).toHaveBeenCalled();
const spanNames = telemetryState.tracer.startSpan.mock.calls.map((call) => call[0]);
expect(spanNames).toContain("moltbot.webhook.processed");
expect(spanNames).toContain("moltbot.message.processed");
expect(spanNames).toContain("moltbot.session.stuck");
expect(registerLogTransportMock).toHaveBeenCalledTimes(1);
expect(registeredTransports).toHaveLength(1);
registeredTransports[0]?.({
0: "{\"subsystem\":\"diagnostic\"}",
1: "hello",
_meta: { logLevelName: "INFO", date: new Date() },
});
expect(logEmit).toHaveBeenCalled();
await service.stop?.();
});
});

View File

@@ -0,0 +1,566 @@
import { metrics, trace, SpanStatusCode } from "@opentelemetry/api";
import type { SeverityNumber } from "@opentelemetry/api-logs";
import { OTLPLogExporter } from "@opentelemetry/exporter-logs-otlp-http";
import { OTLPMetricExporter } from "@opentelemetry/exporter-metrics-otlp-http";
import { OTLPTraceExporter } from "@opentelemetry/exporter-trace-otlp-http";
import { Resource } from "@opentelemetry/resources";
import { BatchLogRecordProcessor, LoggerProvider } from "@opentelemetry/sdk-logs";
import { PeriodicExportingMetricReader } from "@opentelemetry/sdk-metrics";
import { NodeSDK } from "@opentelemetry/sdk-node";
import { ParentBasedSampler, TraceIdRatioBasedSampler } from "@opentelemetry/sdk-trace-base";
import { SemanticResourceAttributes } from "@opentelemetry/semantic-conventions";
import type { MoltbotPluginService, DiagnosticEventPayload } from "clawdbot/plugin-sdk";
import { onDiagnosticEvent, registerLogTransport } from "clawdbot/plugin-sdk";
const DEFAULT_SERVICE_NAME = "moltbot";
function normalizeEndpoint(endpoint?: string): string | undefined {
const trimmed = endpoint?.trim();
return trimmed ? trimmed.replace(/\/+$/, "") : undefined;
}
function resolveOtelUrl(endpoint: string | undefined, path: string): string | undefined {
if (!endpoint) return undefined;
if (endpoint.includes("/v1/")) return endpoint;
return `${endpoint}/${path}`;
}
function resolveSampleRate(value: number | undefined): number | undefined {
if (typeof value !== "number" || !Number.isFinite(value)) return undefined;
if (value < 0 || value > 1) return undefined;
return value;
}
export function createDiagnosticsOtelService(): MoltbotPluginService {
let sdk: NodeSDK | null = null;
let logProvider: LoggerProvider | null = null;
let stopLogTransport: (() => void) | null = null;
let unsubscribe: (() => void) | null = null;
return {
id: "diagnostics-otel",
async start(ctx) {
const cfg = ctx.config.diagnostics;
const otel = cfg?.otel;
if (!cfg?.enabled || !otel?.enabled) return;
const protocol = otel.protocol ?? process.env.OTEL_EXPORTER_OTLP_PROTOCOL ?? "http/protobuf";
if (protocol !== "http/protobuf") {
ctx.logger.warn(`diagnostics-otel: unsupported protocol ${protocol}`);
return;
}
const endpoint = normalizeEndpoint(otel.endpoint ?? process.env.OTEL_EXPORTER_OTLP_ENDPOINT);
const headers = otel.headers ?? undefined;
const serviceName =
otel.serviceName?.trim() || process.env.OTEL_SERVICE_NAME || DEFAULT_SERVICE_NAME;
const sampleRate = resolveSampleRate(otel.sampleRate);
const tracesEnabled = otel.traces !== false;
const metricsEnabled = otel.metrics !== false;
const logsEnabled = otel.logs === true;
if (!tracesEnabled && !metricsEnabled && !logsEnabled) return;
const resource = new Resource({
[SemanticResourceAttributes.SERVICE_NAME]: serviceName,
});
const traceUrl = resolveOtelUrl(endpoint, "v1/traces");
const metricUrl = resolveOtelUrl(endpoint, "v1/metrics");
const logUrl = resolveOtelUrl(endpoint, "v1/logs");
const traceExporter = tracesEnabled
? new OTLPTraceExporter({
...(traceUrl ? { url: traceUrl } : {}),
...(headers ? { headers } : {}),
})
: undefined;
const metricExporter = metricsEnabled
? new OTLPMetricExporter({
...(metricUrl ? { url: metricUrl } : {}),
...(headers ? { headers } : {}),
})
: undefined;
const metricReader = metricExporter
? new PeriodicExportingMetricReader({
exporter: metricExporter,
...(typeof otel.flushIntervalMs === "number"
? { exportIntervalMillis: Math.max(1000, otel.flushIntervalMs) }
: {}),
})
: undefined;
if (tracesEnabled || metricsEnabled) {
sdk = new NodeSDK({
resource,
...(traceExporter ? { traceExporter } : {}),
...(metricReader ? { metricReader } : {}),
...(sampleRate !== undefined
? {
sampler: new ParentBasedSampler({
root: new TraceIdRatioBasedSampler(sampleRate),
}),
}
: {}),
});
await sdk.start();
}
const logSeverityMap: Record<string, SeverityNumber> = {
TRACE: 1 as SeverityNumber,
DEBUG: 5 as SeverityNumber,
INFO: 9 as SeverityNumber,
WARN: 13 as SeverityNumber,
ERROR: 17 as SeverityNumber,
FATAL: 21 as SeverityNumber,
};
const meter = metrics.getMeter("moltbot");
const tracer = trace.getTracer("moltbot");
const tokensCounter = meter.createCounter("moltbot.tokens", {
unit: "1",
description: "Token usage by type",
});
const costCounter = meter.createCounter("moltbot.cost.usd", {
unit: "1",
description: "Estimated model cost (USD)",
});
const durationHistogram = meter.createHistogram("moltbot.run.duration_ms", {
unit: "ms",
description: "Agent run duration",
});
const contextHistogram = meter.createHistogram("moltbot.context.tokens", {
unit: "1",
description: "Context window size and usage",
});
const webhookReceivedCounter = meter.createCounter("moltbot.webhook.received", {
unit: "1",
description: "Webhook requests received",
});
const webhookErrorCounter = meter.createCounter("moltbot.webhook.error", {
unit: "1",
description: "Webhook processing errors",
});
const webhookDurationHistogram = meter.createHistogram("moltbot.webhook.duration_ms", {
unit: "ms",
description: "Webhook processing duration",
});
const messageQueuedCounter = meter.createCounter("moltbot.message.queued", {
unit: "1",
description: "Messages queued for processing",
});
const messageProcessedCounter = meter.createCounter("moltbot.message.processed", {
unit: "1",
description: "Messages processed by outcome",
});
const messageDurationHistogram = meter.createHistogram("moltbot.message.duration_ms", {
unit: "ms",
description: "Message processing duration",
});
const queueDepthHistogram = meter.createHistogram("moltbot.queue.depth", {
unit: "1",
description: "Queue depth on enqueue/dequeue",
});
const queueWaitHistogram = meter.createHistogram("moltbot.queue.wait_ms", {
unit: "ms",
description: "Queue wait time before execution",
});
const laneEnqueueCounter = meter.createCounter("moltbot.queue.lane.enqueue", {
unit: "1",
description: "Command queue lane enqueue events",
});
const laneDequeueCounter = meter.createCounter("moltbot.queue.lane.dequeue", {
unit: "1",
description: "Command queue lane dequeue events",
});
const sessionStateCounter = meter.createCounter("moltbot.session.state", {
unit: "1",
description: "Session state transitions",
});
const sessionStuckCounter = meter.createCounter("moltbot.session.stuck", {
unit: "1",
description: "Sessions stuck in processing",
});
const sessionStuckAgeHistogram = meter.createHistogram("moltbot.session.stuck_age_ms", {
unit: "ms",
description: "Age of stuck sessions",
});
const runAttemptCounter = meter.createCounter("moltbot.run.attempt", {
unit: "1",
description: "Run attempts",
});
if (logsEnabled) {
const logExporter = new OTLPLogExporter({
...(logUrl ? { url: logUrl } : {}),
...(headers ? { headers } : {}),
});
logProvider = new LoggerProvider({ resource });
logProvider.addLogRecordProcessor(
new BatchLogRecordProcessor(logExporter, {
...(typeof otel.flushIntervalMs === "number"
? { scheduledDelayMillis: Math.max(1000, otel.flushIntervalMs) }
: {}),
}),
);
const otelLogger = logProvider.getLogger("moltbot");
stopLogTransport = registerLogTransport((logObj) => {
const safeStringify = (value: unknown) => {
try {
return JSON.stringify(value);
} catch {
return String(value);
}
};
const meta = (logObj as Record<string, unknown>)._meta as
| {
logLevelName?: string;
date?: Date;
name?: string;
parentNames?: string[];
path?: {
filePath?: string;
fileLine?: string;
fileColumn?: string;
filePathWithLine?: string;
method?: string;
};
}
| undefined;
const logLevelName = meta?.logLevelName ?? "INFO";
const severityNumber = logSeverityMap[logLevelName] ?? (9 as SeverityNumber);
const numericArgs = Object.entries(logObj)
.filter(([key]) => /^\d+$/.test(key))
.sort((a, b) => Number(a[0]) - Number(b[0]))
.map(([, value]) => value);
let bindings: Record<string, unknown> | undefined;
if (typeof numericArgs[0] === "string" && numericArgs[0].trim().startsWith("{")) {
try {
const parsed = JSON.parse(numericArgs[0]);
if (parsed && typeof parsed === "object" && !Array.isArray(parsed)) {
bindings = parsed as Record<string, unknown>;
numericArgs.shift();
}
} catch {
// ignore malformed json bindings
}
}
let message = "";
if (numericArgs.length > 0 && typeof numericArgs[numericArgs.length - 1] === "string") {
message = String(numericArgs.pop());
} else if (numericArgs.length === 1) {
message = safeStringify(numericArgs[0]);
numericArgs.length = 0;
}
if (!message) {
message = "log";
}
const attributes: Record<string, string | number | boolean> = {
"moltbot.log.level": logLevelName,
};
if (meta?.name) attributes["moltbot.logger"] = meta.name;
if (meta?.parentNames?.length) {
attributes["moltbot.logger.parents"] = meta.parentNames.join(".");
}
if (bindings) {
for (const [key, value] of Object.entries(bindings)) {
if (typeof value === "string" || typeof value === "number" || typeof value === "boolean") {
attributes[`moltbot.${key}`] = value;
} else if (value != null) {
attributes[`moltbot.${key}`] = safeStringify(value);
}
}
}
if (numericArgs.length > 0) {
attributes["moltbot.log.args"] = safeStringify(numericArgs);
}
if (meta?.path?.filePath) attributes["code.filepath"] = meta.path.filePath;
if (meta?.path?.fileLine) attributes["code.lineno"] = Number(meta.path.fileLine);
if (meta?.path?.method) attributes["code.function"] = meta.path.method;
if (meta?.path?.filePathWithLine) {
attributes["moltbot.code.location"] = meta.path.filePathWithLine;
}
otelLogger.emit({
body: message,
severityText: logLevelName,
severityNumber,
attributes,
timestamp: meta?.date ?? new Date(),
});
});
}
const spanWithDuration = (
name: string,
attributes: Record<string, string | number>,
durationMs?: number,
) => {
const startTime =
typeof durationMs === "number" ? Date.now() - Math.max(0, durationMs) : undefined;
const span = tracer.startSpan(name, {
attributes,
...(startTime ? { startTime } : {}),
});
return span;
};
const recordModelUsage = (evt: Extract<DiagnosticEventPayload, { type: "model.usage" }>) => {
const attrs = {
"moltbot.channel": evt.channel ?? "unknown",
"moltbot.provider": evt.provider ?? "unknown",
"moltbot.model": evt.model ?? "unknown",
};
const usage = evt.usage;
if (usage.input) tokensCounter.add(usage.input, { ...attrs, "moltbot.token": "input" });
if (usage.output) tokensCounter.add(usage.output, { ...attrs, "moltbot.token": "output" });
if (usage.cacheRead)
tokensCounter.add(usage.cacheRead, { ...attrs, "moltbot.token": "cache_read" });
if (usage.cacheWrite)
tokensCounter.add(usage.cacheWrite, { ...attrs, "moltbot.token": "cache_write" });
if (usage.promptTokens)
tokensCounter.add(usage.promptTokens, { ...attrs, "moltbot.token": "prompt" });
if (usage.total) tokensCounter.add(usage.total, { ...attrs, "moltbot.token": "total" });
if (evt.costUsd) costCounter.add(evt.costUsd, attrs);
if (evt.durationMs) durationHistogram.record(evt.durationMs, attrs);
if (evt.context?.limit)
contextHistogram.record(evt.context.limit, {
...attrs,
"moltbot.context": "limit",
});
if (evt.context?.used)
contextHistogram.record(evt.context.used, {
...attrs,
"moltbot.context": "used",
});
if (!tracesEnabled) return;
const spanAttrs: Record<string, string | number> = {
...attrs,
"moltbot.sessionKey": evt.sessionKey ?? "",
"moltbot.sessionId": evt.sessionId ?? "",
"moltbot.tokens.input": usage.input ?? 0,
"moltbot.tokens.output": usage.output ?? 0,
"moltbot.tokens.cache_read": usage.cacheRead ?? 0,
"moltbot.tokens.cache_write": usage.cacheWrite ?? 0,
"moltbot.tokens.total": usage.total ?? 0,
};
const span = spanWithDuration("moltbot.model.usage", spanAttrs, evt.durationMs);
span.end();
};
const recordWebhookReceived = (
evt: Extract<DiagnosticEventPayload, { type: "webhook.received" }>,
) => {
const attrs = {
"moltbot.channel": evt.channel ?? "unknown",
"moltbot.webhook": evt.updateType ?? "unknown",
};
webhookReceivedCounter.add(1, attrs);
};
const recordWebhookProcessed = (
evt: Extract<DiagnosticEventPayload, { type: "webhook.processed" }>,
) => {
const attrs = {
"moltbot.channel": evt.channel ?? "unknown",
"moltbot.webhook": evt.updateType ?? "unknown",
};
if (typeof evt.durationMs === "number") {
webhookDurationHistogram.record(evt.durationMs, attrs);
}
if (!tracesEnabled) return;
const spanAttrs: Record<string, string | number> = { ...attrs };
if (evt.chatId !== undefined) spanAttrs["moltbot.chatId"] = String(evt.chatId);
const span = spanWithDuration("moltbot.webhook.processed", spanAttrs, evt.durationMs);
span.end();
};
const recordWebhookError = (
evt: Extract<DiagnosticEventPayload, { type: "webhook.error" }>,
) => {
const attrs = {
"moltbot.channel": evt.channel ?? "unknown",
"moltbot.webhook": evt.updateType ?? "unknown",
};
webhookErrorCounter.add(1, attrs);
if (!tracesEnabled) return;
const spanAttrs: Record<string, string | number> = {
...attrs,
"moltbot.error": evt.error,
};
if (evt.chatId !== undefined) spanAttrs["moltbot.chatId"] = String(evt.chatId);
const span = tracer.startSpan("moltbot.webhook.error", {
attributes: spanAttrs,
});
span.setStatus({ code: SpanStatusCode.ERROR, message: evt.error });
span.end();
};
const recordMessageQueued = (
evt: Extract<DiagnosticEventPayload, { type: "message.queued" }>,
) => {
const attrs = {
"moltbot.channel": evt.channel ?? "unknown",
"moltbot.source": evt.source ?? "unknown",
};
messageQueuedCounter.add(1, attrs);
if (typeof evt.queueDepth === "number") {
queueDepthHistogram.record(evt.queueDepth, attrs);
}
};
const recordMessageProcessed = (
evt: Extract<DiagnosticEventPayload, { type: "message.processed" }>,
) => {
const attrs = {
"moltbot.channel": evt.channel ?? "unknown",
"moltbot.outcome": evt.outcome ?? "unknown",
};
messageProcessedCounter.add(1, attrs);
if (typeof evt.durationMs === "number") {
messageDurationHistogram.record(evt.durationMs, attrs);
}
if (!tracesEnabled) return;
const spanAttrs: Record<string, string | number> = { ...attrs };
if (evt.sessionKey) spanAttrs["moltbot.sessionKey"] = evt.sessionKey;
if (evt.sessionId) spanAttrs["moltbot.sessionId"] = evt.sessionId;
if (evt.chatId !== undefined) spanAttrs["moltbot.chatId"] = String(evt.chatId);
if (evt.messageId !== undefined) spanAttrs["moltbot.messageId"] = String(evt.messageId);
if (evt.reason) spanAttrs["moltbot.reason"] = evt.reason;
const span = spanWithDuration("moltbot.message.processed", spanAttrs, evt.durationMs);
if (evt.outcome === "error") {
span.setStatus({ code: SpanStatusCode.ERROR, message: evt.error });
}
span.end();
};
const recordLaneEnqueue = (
evt: Extract<DiagnosticEventPayload, { type: "queue.lane.enqueue" }>,
) => {
const attrs = { "moltbot.lane": evt.lane };
laneEnqueueCounter.add(1, attrs);
queueDepthHistogram.record(evt.queueSize, attrs);
};
const recordLaneDequeue = (
evt: Extract<DiagnosticEventPayload, { type: "queue.lane.dequeue" }>,
) => {
const attrs = { "moltbot.lane": evt.lane };
laneDequeueCounter.add(1, attrs);
queueDepthHistogram.record(evt.queueSize, attrs);
if (typeof evt.waitMs === "number") {
queueWaitHistogram.record(evt.waitMs, attrs);
}
};
const recordSessionState = (
evt: Extract<DiagnosticEventPayload, { type: "session.state" }>,
) => {
const attrs: Record<string, string> = { "moltbot.state": evt.state };
if (evt.reason) attrs["moltbot.reason"] = evt.reason;
sessionStateCounter.add(1, attrs);
};
const recordSessionStuck = (
evt: Extract<DiagnosticEventPayload, { type: "session.stuck" }>,
) => {
const attrs: Record<string, string> = { "moltbot.state": evt.state };
sessionStuckCounter.add(1, attrs);
if (typeof evt.ageMs === "number") {
sessionStuckAgeHistogram.record(evt.ageMs, attrs);
}
if (!tracesEnabled) return;
const spanAttrs: Record<string, string | number> = { ...attrs };
if (evt.sessionKey) spanAttrs["moltbot.sessionKey"] = evt.sessionKey;
if (evt.sessionId) spanAttrs["moltbot.sessionId"] = evt.sessionId;
spanAttrs["moltbot.queueDepth"] = evt.queueDepth ?? 0;
spanAttrs["moltbot.ageMs"] = evt.ageMs;
const span = tracer.startSpan("moltbot.session.stuck", { attributes: spanAttrs });
span.setStatus({ code: SpanStatusCode.ERROR, message: "session stuck" });
span.end();
};
const recordRunAttempt = (evt: Extract<DiagnosticEventPayload, { type: "run.attempt" }>) => {
runAttemptCounter.add(1, { "moltbot.attempt": evt.attempt });
};
const recordHeartbeat = (
evt: Extract<DiagnosticEventPayload, { type: "diagnostic.heartbeat" }>,
) => {
queueDepthHistogram.record(evt.queued, { "moltbot.channel": "heartbeat" });
};
unsubscribe = onDiagnosticEvent((evt: DiagnosticEventPayload) => {
switch (evt.type) {
case "model.usage":
recordModelUsage(evt);
return;
case "webhook.received":
recordWebhookReceived(evt);
return;
case "webhook.processed":
recordWebhookProcessed(evt);
return;
case "webhook.error":
recordWebhookError(evt);
return;
case "message.queued":
recordMessageQueued(evt);
return;
case "message.processed":
recordMessageProcessed(evt);
return;
case "queue.lane.enqueue":
recordLaneEnqueue(evt);
return;
case "queue.lane.dequeue":
recordLaneDequeue(evt);
return;
case "session.state":
recordSessionState(evt);
return;
case "session.stuck":
recordSessionStuck(evt);
return;
case "run.attempt":
recordRunAttempt(evt);
return;
case "diagnostic.heartbeat":
recordHeartbeat(evt);
return;
}
});
if (logsEnabled) {
ctx.logger.info("diagnostics-otel: logs exporter enabled (OTLP/HTTP)");
}
},
async stop() {
unsubscribe?.();
unsubscribe = null;
stopLogTransport?.();
stopLogTransport = null;
if (logProvider) {
await logProvider.shutdown().catch(() => undefined);
logProvider = null;
}
if (sdk) {
await sdk.shutdown().catch(() => undefined);
sdk = null;
}
},
} satisfies MoltbotPluginService;
}