apps/metricator
2
0
Fork 0
Code Issues 5 Pull Requests Packages Projects Releases Wiki Activity
Prometheus metrics proxy for your classic NMS.
11 Commits 1 Branch 5 Tags 134 KiB
Go 91.4%
Shell 3.6%
Makefile 3.1%
Dockerfile 1.9%
529113a41a
Go to file
Stanislav N. aka pztrn 529113a41a
Metrics data is stored as structure, HTTP requests logging, got rid of context. 
Metrics data now stored as structures. This is a first step for autodiscovery
helping for NMSes.

HTTP requests now logged.

Got rid of context.Context for getting metric data in applications
because context.Context is useless here.
2020-12-23 13:27:17 +05:00
cmd/metricatord A very naive, simple and not so robust HTTP API responder implementation. 2020-11-29 06:09:35 +05:00
docs Metrics data is stored as structure, HTTP requests logging, got rid of context. 2020-12-23 13:27:17 +05:00
internal Metrics data is stored as structure, HTTP requests logging, got rid of context. 2020-12-23 13:27:17 +05:00
scripts Correctly set version info when building, fixed build script and added headers for apps endpoints. 2020-11-28 23:46:24 +05:00
.gitignore Rethink how metricator will work, golangci-lint and gitlab ci configs. 2020-11-29 03:22:39 +05:00
.gitlab-ci.yml Use proxified Docker image for linting. 2020-11-29 03:24:12 +05:00
.golangci.yml Linter config tuning and first version of Prometheus metrics parser. 2020-11-29 05:44:21 +05:00
go.mod Rethink how metricator will work, golangci-lint and gitlab ci configs. 2020-11-29 03:22:39 +05:00
go.sum Rethink how metricator will work, golangci-lint and gitlab ci configs. 2020-11-29 03:22:39 +05:00
Makefile Rethink how metricator will work, golangci-lint and gitlab ci configs. 2020-11-29 03:22:39 +05:00
metricator.example.yaml Rethink how metricator will work, golangci-lint and gitlab ci configs. 2020-11-29 03:22:39 +05:00
README.md Rethink how metricator will work, golangci-lint and gitlab ci configs. 2020-11-29 03:22:39 +05:00

README.md

Metricator

Simple proxy between prometheus-powered application and your NMS.

Why

I'm using NetXMS to monitor all of my systems. Many things I use exports metrics in prometheus format which can be utilized by custom parsing script. But I've encounter a performance problem when I need to monitor 30 metrics - parsing script will make 30 requests to prometheus endpoint which might affect performance.

Metricator will issue only one request and cache data in memory between them. Also it will expose HTTP API to get single metric which can be easily utilized with any NMS.

Caveats

  • No authorization. DO NOT expose Metricator to wild world!

Installation

TBW

Configuration

TBW

Documentation

Check docs directory.