nagios-charmers team mailing list archive

[Alvaro Uria <1829474@xxxxxxxxxxxxxxxxxx>]

Nagios usually has a burst of checks every 5 minutes. Those will make
the load avg go up for a brief period of time, and it will recover
pretty quickly once the 10-seconds checks finish running.

Nagios units may end up as "1.0, 7.0, 4.0" when they are recovering from
said burst.

-- 
You received this bug notification because you are a member of Nagios
Charm developers, which is subscribed to Nagios Charm.
https://bugs.launchpad.net/bugs/1829474

Title:
  load_monitor should allow an "auto" value

Status in Nagios Charm:
  New

Bug description:
  nagios-charm default config value for "load_monitor" is
  5.0!4.0!3.0!10.0!6.0!4.0 (which means 1/5/15 avg for warning+critical
  (6 values).

  I think load_monitor should allow a value "auto" by default, which would calculate the load average thresholds by running:
  * warning: nproc*2,int(nproc*2.5),nproc*2
  * critical: nproc*3,nproc*4,int(nproc*2.5)

  Those values could even be increased since the important part is to
  make nagios able to run checks (which is more related to max open
  files or memory available).

  If the host is a container, I think check_load should be deactivated
  (the host should be monitored).

To manage notifications about this bug go to:
https://bugs.launchpad.net/nagios-charm/+bug/1829474/+subscriptions