aiogram-prometheus 0.2.3

Aiogram`s exporter for Prometheus

aiogram Prometheus Exporter

Module for exporting monitoring values for Prometheus

Functionality

• Monitoring the status of bots and dispatchers
• Middleware for monitoring the bot's network activity
• Middleware for monitoring the event handler performance

Installation

pip install aiogram-prometheus

Quick start

• aiogram.BotAiogramCollector - base info abut started bot
• aiogram.PrometheusMetricRequestMiddleware - tracking requests from bot to server
• aiogram.DispatcherAiogramCollector - base info abut started app
• aiogram.PrometheusMetricMessageMiddleware - tracking event for app processing
• aiogram.StorageAiogramCollector - tracking actions with fsm storage
• aiogram.PushGatewayClient - easy way to push your metrics to pushgateway server

import asyncio
import logging

from aiogram import Bot, Dispatcher
from aiogram.fsm.storage.memory import MemoryStorage
from aiogram.types import Message
from decouple import config

from aiogram_prometheus import (
    BotAiogramCollector,
    DispatcherAiogramCollector,
    PrometheusMetricMessageMiddleware,
    PrometheusMetricStorageMixin,
    PrometheusMetricRequestMiddleware,
    PushGatewayClient,
    StorageAiogramCollector,
)

logging.basicConfig(level='DEBUG')

logger = logging.getLogger(__name__)

bot = Bot('TOKEN')

# Bot info metrics
BotAiogramCollector().add_bot(bot)

# Metric requests
# which are made by the target bot
bot.session.middleware(PrometheusMetricRequestMiddleware())

# Metric storage
# Change "MemoryStorage" to your storage
class _Storage(PrometheusMetricStorageMixin, MemoryStorage):
    pass

storage_collector = StorageAiogramCollector()
storage = _Storage(storage_collector)

dp = Dispatcher(storage=storage)

# Metric message
# which are processed by the dispatcher
dp.message.middleware(PrometheusMetricMessageMiddleware())


# Metric base info
DispatcherAiogramCollector(dp)

push_gateway_client = PushGatewayClient('http://localhost:9091/', 'job-name')

@dp.startup()
async def on_startup(bot: Bot):
    push_gateway_client.schedule_push(5)

@dp.message()
async def handle(message: Message) -> None:
    await message.reply('Ok')

asyncio.run(dp.start_polling(bot))

Functionality

aiogram.BotAiogramCollector

You should use this collector if you want to track information about running bots. The metrics include most of the available information about the bot, including its id, username and full_name

from aiogram import Bot
from aiogram_prometheus import BotAiogramCollector

bot = Bot(TOKEN)

BotAiogramCollector(bot)

aiogram.PrometheusMetricRequestMiddleware

This is an intermediate layer for requests that are sent to telegram servers on behalf of a specific bot. Use this middleware to track requests, their types, and their execution times.

from aiogram import Bot
from aiogram_prometheus import PrometheusMetricRequestMiddleware

bot = Bot(TOKEN)
bot.session.middleware(PrometheusMetricRequestMiddleware())

aiogram.DispatcherAiogramCollector

You should use this collector if you want to track general application information. This will be useful if you want to keep the aiogram_version and telegram_api up to date

from aiogram import Dispatcher
from aiogram_prometheus import DispatcherAiogramCollector

dp = Dispatcher()

DispatcherAiogramCollector(dp)

aiogram.PrometheusMetricMessageMiddleware

this intermediate layer is needed to track the events that the dispatcher processes. You will receive information about the event, the execution times (by your application), and the message (if the event is a message).

Note: if there is no handler for a message, then the message will not be tracked

from aiogram import Dispatcher
from aiogram_prometheus import PrometheusMetricMessageMiddleware

dp = Dispatcher()
dp.message.middleware(PrometheusMetricMessageMiddleware())

aiogram.StorageAiogramCollector

This collector is used inside a storage mixin. You should use it if you need transparency when working with storage. You will be able to see how often and how much data you save and read.

note. To use this collector you must use a mixin (aiogram.PrometheusMetricStorageMixin)

from aiogram import Dispatcher
from aiogram.fsm.storage.memory import MemoryStorage
from aiogram_prometheus import PrometheusMetricStorageMixin, StorageAiogramCollector

class _Storage(PrometheusMetricStorageMixin, MemoryStorage):
    pass

storage = _Storage(StorageAiogramCollector())

dp = Dispatcher(storage=storage)

aiogram.PushGatewayClient

Collecting metrics from application software is not an easy task. You can run a web application in parallel or use pushgateway. If everything is clear with the first, then there may be problems with the second. You can use the client implemented here, which starts when the application starts and sends metrics to the server every X seconds.

from aiogram import Dispatcher, Bot
from aiogram_prometheus import PushGatewayClient

dp = Dispatcher()

push_gateway_client = PushGatewayClient('http://localhost:9091/', 'job-name')

@dp.startup()
async def on_startup(bot: Bot):
    push_gateway_client.schedule_push(5)

Contribute

Issue Tracker: https://gitlab.com/rocshers/python/aiogram-prometheus/-/issues
Source Code: https://gitlab.com/rocshers/python/aiogram-prometheus

Before adding changes:

make install-dev

After changes:

make format test