Can someone update on where the collected data from input plugin is saved when we use Prometheus Output Plugin?
@telegraf_vmware The Prometheus output plugin exposes a port on your Telegraf host and publishes Prometheus-formatted metrics on that port at the /metrics
path (by default). You can scrape those metrics and store them somewhere else, but the plugin itself doesn’t send Prometheus-formatted metrics to a remote endpoint.
Thanks for the update, but let me rephrase my question.
- Where is the data stored when the metrics is collected by the input plugin?
- Also, will there not be a time difference as the input metrics will be collected in the interval mentioned in the configuration and the time it is scrapped by the prometheus output plugin.
Example :
Input plugin collects the metrics every 5 min i.e 10:00 AM , 10:05 AM … and son on
and this is redirected to the Prometheus Output plugin.
I use Prometheus to scrap the metrics which is at relative interval from the time it starts. So if I start Prometheus at 10:02 at 300 seconds interval, any metrics collected by input plugin at 10:05 will be scraped only at 10:07 which is totally wrong.
Any prometheus output plugin should be an on-demand one which when called, should be calling the input plugin.