prometheus metrics format

[pavel987]

This PR adds an alternate display format for data. It's a lot of code that I modified after copy & pasting. I'm not a C developer so I wouldn't be surprised if there's a lot of things wrong with it. Let me know what needs to be fixed/changed to get this merged. Thanks!

[ghost]

Hi, Any news about this?

[SuperQ]

Have you tested this with promtool check-metrics mode?

[ghost]

error while linting: text format parsing error in line 37: second HELP line for metric name "nginx_server_requests"
error while linting: text format parsing error in line 37: second TYPE line for metric name "nginx_server_requests", or TYPE reported after samples
...

# HELP nginx_server_requests requests counter
# TYPE nginx_server_requests counter
nginx_server_requests{code="1xx",host="domain.com"} 0
nginx_server_requests{code="2xx",host="domain.com"} 17
nginx_server_requests{code="3xx",host="domain.com"} 2
nginx_server_requests{code="4xx",host="domain.com"} 3
nginx_server_requests{code="5xx",host="domain.com"} 0
# HELP nginx_server_requests requests counter
# TYPE nginx_server_requests counter
nginx_server_requests{code="1xx",host="_"} 0
nginx_server_requests{code="2xx",host="_"} 1812
nginx_server_requests{code="3xx",host="_"} 0
nginx_server_requests{code="4xx",host="_"} 8
nginx_server_requests{code="5xx",host="_"} 0
# HELP nginx_server_requests requests counter
# TYPE nginx_server_requests counter
nginx_server_requests{code="1xx",host="*"} 0
nginx_server_requests{code="2xx",host="*"} 1829
nginx_server_requests{code="3xx",host="*"} 2
nginx_server_requests{code="4xx",host="*"} 11
nginx_server_requests{code="5xx",host="*"} 0

It looks like a small patch would come in handy, but the gun works.
The data appears in the prometheus.

[vozlt]

Thanks for the pull request.
I'll check it out soon.

[vozlt]

That feature has added.
I have rewritten but it's not much different from yours.
One of the differences is that all mertric names are started with nginx_vts_(*).
I can not guarantee that it will work well because I know little about prometheus.
Please test it.
Thanks.

Latest commit: 2f9434f

[pavel987]

Did a bit of testing and it seems to work well. Do you plan on making a release with these changes anytime soon?

[vozlt]

I will release it after checking the histogram in #62 (comment).
Thanks.

[vozlt]

The metric type of histogram in format/prometheus has added.
See the directive vhost_traffic_status_histogram_buckets for details.
I quickly added the feature and did not test it much.
So it is necessary to verify that I have implemented the function correctly according to prometheus histogram metric type.
Please test it.
Thanks.

Latest commit: 2fd3ae4

[vozlt]

I have tagged the latest commit as v0.1.17.
Thanks.

[etlfg]

Don't know if this is a configuration problem from my side but I have many upstreams behind nginx and I only got one from the /status/format/prometheus. They appear when I use the /status/ or /status/format/json. Any explanation here ?

I didn't create a new issue as It may be an error in your code as you mentioned it.

Used version :
NGINX_VERSION=1.13.12
NGINX_VTS_VERSION=0.1.18

[SuperQ]

I think this is obsolete.

[vozlt]

@etlfg If you still encounter such a problem, please create a new issue.