为什么Prometheus Python客户机和Prometheus节点客户机之间存在设计差异？

Question

普罗米修斯Python clientvs普罗米修斯Node.js client

我将Prometheus与基于烧瓶的应用程序以及Node.js/快递基于应用程序。下面是它们的代码。你知道吗

我对这里的普罗米修斯建筑有点困惑。为什么普罗米修斯在与Flask集成时又启动了一个web服务器，而在与Flask集成时却重用了express服务器Node.js/快递应用程序。请查找下面两个集成的代码。你知道吗

基于烧瓶的应用：

import graphene
from   apis.graphql.schema.queries \
                                import Query
from   flask                    import Flask
from   flask_graphql            import GraphQLView

import time
from prometheus_client import Counter, Histogram, start_http_server


FLASK_REQUEST_LATENCY = Histogram('flask_request_latency_seconds', 'Flask Request Latency', ['method', 'endpoint'])
FLASK_REQUEST_COUNT = Counter('flask_request_count', 'Flask Request Count', ['method', 'endpoint', 'http_status'])


def get_options():
    """Function to get the options passed via command line."""
    from cmdline.parse import OptionParser
    parser = OptionParser()

    # Add the port option with default value 5000
    parser.add_option(
        '-port',
        required=False,
        type=int,
        default=5002,
        help='Port on which to listen for incoming requests. '
             'Defaults to 5000.'
    )

    opts, args = parser.parse_args()
    return opts


def before_request():
    request.start_time = time.time()


def after_request(response):
    request_latency = time.time() - request.start_time
    FLASK_REQUEST_LATENCY.labels(request.method, request.path).observe(request_latency)
    FLASK_REQUEST_COUNT.labels(request.method, request.path, response.status_code).inc()
    return response

def monitor(app, port, addr=''):
    app.before_request(before_request)
    app.after_request(after_request)
    start_http_server(port, addr)


if __name__ == '__main__':
    # Get the command line options.
    options = get_options()

    # Create an instance of the Flask class for the web app.
    app = Flask(__name__)

    # Adds the endpoint /graphql.
    app.add_url_rule(
        '/graphql',
        view_func=GraphQLView.as_view(
            'graphql', schema=graphene.Schema(Query), graphiql=True
        )
    )

    monitor(app, options.port)

    app.run(
        host = '0.0.0.0',
        port = 5000,
        threaded=True
        )

基于节点的应用程序：

const express = require('express')
const bodyParser = require('body-parser')
const redis = require("redis")
const process = require('process')

const version = require('./version.json')
const redis_url = process.env.REDIS_URL || 'redis://localhost:6379'
const client = redis.createClient(redis_url)
const app = module.exports = express()

const Prometheus = require('prom-client')
const metricsInterval = Prometheus.collectDefaultMetrics()

const metrics = (req, res) => {
  res.set('Content-Type', Prometheus.register.contentType)
  res.end(Prometheus.register.metrics())
}

const gamingRequests = new Prometheus.Counter({
  name: 'gaming_requests_total',
  help: 'number of requests',
  labelNames: ['route', 'method']
})

const middleware = (req, res, next) => {
  gamingRequests.inc({route: req.path, method: req.method}, 1, Date.now())
  next()
}

app.use(bodyParser.json())
app.use(middleware)

app.get('/metrics', metrics)

app.get('/version', (req, res) => {
    res.setHeader('content-type', 'application/javascript')
    res.send(JSON.stringify({version: version.version}))
})

app.get('/score/:userId', (req, res) => {
  client.get(
    `score:${req.params.userId}`,
    (err, result) => {
      res.setHeader('content-type', 'application/javascript')
      if (err || !result) {
        return res.send(JSON.stringify({score: 0}))
      }
      return res.send(result)
    }
  )
})

app.listen(8080, '0.0.0.0' , () => {
    console.log(`Gaming App started on 0.0.0.0:8080`)
})

您是否注意到“启动http服务器（端口，地址）”以及应用程序运行（）如果是烧瓶？但我们没有两个网络服务器以防Node.js/快递应用程序。你知道吗