The pm2-cluster-prometheus from jakeh

Returning JSON Instead of Prometheus Format

I've been trying to implement metrics for a cluster program, and this seems to be the tool to go for. Thanks a lot for the same <3

I'm facing an issue, and I would like to know if you can help me with the same. I've used the following statements in my code along with appropriate imports.

const aggregatorRegistry = new AggregatorRegistry();
// register for prometheus
app.get('/metrics', async (_, res) => {
    const metrics = await getAggregateMetrics();
    res.set('Content-Type', aggregatorRegistry.contentType);
    res.send(metrics);
});

The /metrics endpoint is queried by Prometheus and since it is returning a JSON (attached at bottom), Prometheus is giving the following error.

I've searched the web for a solution, however, my knowledge with clusters and metrics is very less and I would be really delighted if you can guide me on the right path.

{"_metrics":{"graphql_queries_parsed":{"name":"graphql_queries_parsed","labelNames":["operationName","operation"],"hashMap":{},"help":"The amount of GraphQL queries that have been parsed.","aggregator":"sum"},"graphql_queries_validation_started":{"name":"graphql_queries_validation_started","labelNames":["operationName","operation"],"hashMap":{},"help":"The amount of GraphQL queries that have started validation.","aggregator":"sum"},"graphql_queries_resolved":{"name":"graphql_queries_resolved","labelNames":["operationName","operation"],"hashMap":{},"help":"The amount of GraphQL queries that have had their operation resolved.","aggregator":"sum"},"graphql_queries_execution_started":{"name":"graphql_queries_execution_started","labelNames":["operationName","operation"],"hashMap":{},"help":"The amount of GraphQL queries that have started executing.","aggregator":"sum"},"graphql_queries_errored":{"name":"graphql_queries_errored","labelNames":["operationName","operation"],"hashMap":{},"help":"The amount of GraphQL queries that have encountered errors.","aggregator":"sum"},"graphql_queries_responded":{"name":"graphql_queries_responded","labelNames":["operationName","operation"],"hashMap":{},"help":"The amount of GraphQL queries that have been executed and been attempted to send to the client. This includes requests with errors.","aggregator":"sum"},"graphql_resolver_time":{"name":"graphql_resolver_time","help":"The time to resolve a GraphQL field.","aggregator":"sum","upperBounds":[0.005,0.01,0.025,0.05,0.1,0.25,0.5,1,2.5,5,10],"bucketValues":{"1":0,"5":0,"10":0,"0.005":0,"0.01":0,"0.025":0,"0.05":0,"0.1":0,"0.25":0,"0.5":0,"2.5":0},"sum":0,"count":0,"hashMap":{},"labelNames":["parentType","fieldName","returnType"]},"graphql_total_request_time":{"name":"graphql_total_request_time","help":"The time to complete a GraphQL query.","aggregator":"sum","upperBounds":[0.005,0.01,0.025,0.05,0.1,0.25,0.5,1,2.5,5,10],"bucketValues":{"1":0,"5":0,"10":0,"0.005":0,"0.01":0,"0.025":0,"0.05":0,"0.1":0,"0.25":0,"0.5":0,"2.5":0},"sum":0,"count":0,"hashMap":{},"labelNames":["operationName","operation"]},"http_request_duration_seconds":{"name":"http_request_duration_seconds","help":"duration histogram of http responses labeled with: status_code, method, path","aggregator":"sum","upperBounds":[0.003,0.03,0.1,0.3,1.5,10],"bucketValues":{"10":0,"0.003":0,"0.03":0,"0.1":0,"0.3":0,"1.5":0},"sum":0,"count":0,"hashMap":{},"labelNames":["status_code","method","path"]},"up":{"name":"up","labelNames":[],"hashMap":{"":{"value":1,"labels":{}}},"help":"1 = up, 0 = not up","aggregator":"sum"}},"_collectors":[],"_defaultLabels":{}}

PM2 Dependency Outdated

pm2-cluster-prometheus/package.json

Line 29 in ecffcc6

"pm2": "^3.5.1"

I was looking at resolving an issue in one of our projects (specifically GHSA-vh95-rmgr-6w4m). And this pm2 version relies on an outdated version of mkdirp that relies on minimist 0.0.8.

What's the feasibility of bumping this dependency to the latest pm2?

PM2 running more than 1 clustered process

First off, just wanted to say I have been looking for something like this for a while! Nice work!

It looks like as part of your metric gathering process that you send a message to all PM2 controlled process to report their stats and this works great if you only have 1 clustered process running in PM2. If you have multiple, it sends a request to all and waits for a reply. If the other process is not running your module, it does not respond causing a timeout. If it is running your module, it will inadvertently report stats for a process a different process. I was wondering if you could add a filter to only send a message to process with the same process name "process.env.name". I figure a quick filter under getProcList would do the trick. :)

jakeh / pm2-cluster-prometheus Goto Github PK

pm2-cluster-prometheus's People

Contributors

Stargazers

Watchers

Forkers

pm2-cluster-prometheus's Issues

Returning JSON Instead of Prometheus Format

PM2 Dependency Outdated

PM2 running more than 1 clustered process

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent