Hello,
In promtail I have set up the metric endpoint with a prometheus server set to scrap every 60 seconds. I have multiple metric stages in the config, one for example:
- match:
selector: '{job="lds"}'
stages:
- regex:
expression: "connect to spamd on (?P<spamhost>.*) failed, retrying"
- metrics:
retried_spamd_connection:
type: Counter
description: "Lines for spamc retrying the connection to spamd"
source: spamhost
config:
action: inc
This is a fairly rare line to see, so in most cases I wouldn’t expect it to change much through out the day. But, given the 60s scrape interval, I would expect to have a value every minute in prometheus, but that is not the case. In the last 24 hours, if I query for just the name of the metric, I have 2 values a few minutes apart from eachother. There have been no other issues with either prometheus scraping or promtail log injesting that I am aware of, but I would expect to have way more data for that metric. Is there something differant about metrics from promtail, even though they are in prometheus? Using promtail 3.1 and prometheus 2.53. Thanks