Collection all process stats using telegraf

How to collect stats for all running processes on a host using telegraf.

can we really use “*” in procstat

like
[[inputs.procstat]]
exe = ""
prefix = “

Any help appreciated.

We wish to use telegraf to monitor all processes on the influx host.

You can use pattern = ".*", but you will also need pid_tag = true in order to make the series unique.

This works but there is a big problem, this will quickly create many series so you will need a plan to deal with the series cardinality. I recently wrote some ways you can do this in the passenger plugin docs under series cardinality warning, and much of that would apply here.

1 Like

well we are getting messages in telegraf log

2018-04-11T19:12:12Z E! Error in plugin [inputs.procstat]: took longer to collect than collection interval (1s)