search

lukemonahan / splunk_modinput_prometheus

A Splunk modular input for ingesting Prometheus metrics
Apache License 2.0
44 stars 23 forks source link

Splunk base update app #10

Open tioumen opened 5 years ago

tioumen commented 5 years ago

Hello,

Did you updated the splunk app on splunk base : https://splunkbase.splunk.com/app/4077/ Seems not based on the last update from github vs Splunk base.

Thanks.

lukemonahan commented 5 years ago

The Splunkbase version is behind Github. Please only use Github if you are able to deal with a few changes.

I will be changing some defaults and configuration keys and then releasing 1.0 to Splunkbase.

tioumen commented 5 years ago

Yeap, that's what I have done. I was needed the perfix and parser. I used your last update on github. What are the improvement you've in the pipe please ? :)

lukemonahan commented 5 years ago

The only further change I plan to make will be to make the dot-notation the default, with a new config key of "use_old_metric_name_conversion" to allow backwards compatibility.

tioumen commented 5 years ago

Hello, I compiled the code with the 2 new fields "MetricNamePrefix" & "MetricNameParse". It's works. But I'm getting 100% of CPU due to that parsing when ingesting metrics (that's not the case wtih previous version without these new fileds). I tried also to remove alternatively each option but I'm still getting 100% of CPU. I also tried to let the 2 new parameters empty and I'm getting the same amount of CPU load ... Do you have any advice please ?

Thanks.

ltmon commented 5 years ago

@tioumen Can you check if latest master fixes your issue?

tioumen commented 5 years ago

Sorry for the late answer. Still getting overload issue on the HF when I'm blacklisting / whitelisting metrics from Prometheus.