[Pkg-nagios-devel] Bug#857905: icinga2-common: Default downtimes config for load on locahost noisy and not generic enough

[Olivier Berger]

Package: icinga2-common
Version: 2.1.1-1
Severity: normal

Dear Maintainer,

I've just installed icinga2 on a system, and wondered why I got some
notifications sent by email reporting every night (excerpt of mail pile) :
Today 03:00 nagios at ...  DOWNTIMEEND - localhost - load is OK
Today 02:00 nagios at ...  DOWNTIMESTART - localhost - load is OK

It seems this is due to config by default in
/etc/icinga2/conf.d/downtimes.conf :

/**
 * The example downtime apply rule.
 */

apply ScheduledDowntime "backup-downtime" to Service {
  import "backup-downtime"

  assign where host.name == "localhost" && service.name == "load"
}

I think this is meant to adjust to higher load supposedly happening at
backup times at night.

Not everyone may notice this by default, as one needs to have activated
notification pushes by mail for instance, to detect such repetitive
messages. But I would expect that most installations would be notifying
these...

However, such a default config installed by default on Debian systems
seems dubious to me. I can see no clear hint that every Debian system
could be subject to load issues at night during that perdiod.

Worse, it's not exactly obvious, by reading the notification emails.
Nothing indicates where the rule lies, in the config files, for that
"alert".

I'd propose to remove such default in Debian. Or to indicate whether
there's a way to not be notified of such downtimes, in the default
config.

Thanks in advance.

Best regards,

-- System Information:
Debian Release: 9.0
  APT prefers testing-proposed-updates
  APT policy: (500, 'testing-proposed-updates'), (500, 'stable-updates'), (500, 'testing')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-2-amd64 (SMP w/8 CPU cores)
Locale: LANG=fr_FR.utf8, LC_CTYPE=fr_FR.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)