Provide Prometheus export functionality for ACS instance monitoring data.

[leo79901]

The required feature described as a wish

To more effectively monitor the health status of private cloud instances, we hope the ACS management console can add a feature page or interface to directly obtain real-time operational data of instances (such as disk read/write volumes, network transmission volumes, etc.).

For this purpose, we have a specific suggestion: Could the ACS team implement the exposure of this monitoring data in the standard Prometheus format? This would allow us to eliminate the current two-step process of data collection via "Python + API" and data conversion via "Python + Prometheus exporter." Instead, we could directly configure our existing Prometheus service to pull the data and utilize Grafana for visualization. This would significantly enhance the efficiency and reliability of our monitoring system.

[NuxRo]

Hello,

I have asked for the same a couple of years ago and the response was that it would put too much stress on the management server and/or DB and it's best handled at the hypervisor level.

As such, I went ahead and done it with a libvirt exporter which gives me all the data I need and since the libvirt VM names are the CloudStack "instance names" then it's very easy to correlate the two.

Hope this helps.

[leo79901]

Thank you very much. This is indeed an excellent solution. I have found some materials, and I will give it a try.

[leo79901]

https://github.com/Tinkoff/libvirt-exporter
Excellent. I found this project and got it running. It provides a wealth of performance metrics, and I'll try running it in a test environment for a while to see how it performs.

[DaanHoogland]

please do @leo79901 if it is worthwhile we can move this to show and tell.