Nagios Support Forum

Hello,

I' setting up a different notification alert setting for my dev servers to only aletr during business hours. I have the host configured to only alert on a work hour time period and service checks with disabled notification alerts, however we are still getting alerts outside of the host notification alerts settings.
- Can you see any incorrect setting?
- Is it required to have 2 contacts? (1 for 24/7 and another for desired time)

HOST:
define host {
host_name host1
address xxx.xx.xx.xx
max_check_attempts 5
check_interval 5
retry_interval 1
check_period workhours_linux
contacts nagiosadmin
contact_groups linuxadmin
notification_interval 60
notification_period workhours_linux
first_notification_delay 0
notification_options d,r,s,
notifications_enabled 1
icon_image hp_ux.png
statusmap_image hp_ux.png
_xiwizard linuxsnmp
register 1
}


Checks: notification alert is off in Alert settings tab)

define service {
service_description tmp Disk Usage
hostgroup_name hostgroup1 (host1 is a member of this group)
check_command check_nrpe!check_tmp!!!!!!!
max_check_attempts 3
check_interval 3
retry_interval 1
check_period 24x7
notification_interval 30
first_notification_delay 0
notifications_enabled 0
_xiwizard nrpe
register 1
}

CONTACt:
define contact {
contact_name linuxadmin
alias admin1
host_notifications_enabled 1
service_notifications_enabled 1
host_notification_period 24/7
service_notification_period 24/7
host_notification_options d,u,r,f,s,
service_notification_options c,r,s,
can_submit_commands 1
email email_address
use xi_contact_generic (service notifiaction is skipped on this template)
}

In your host definition, you have:

contact_groups linuxadmin

Is the linuxadmin" contact a member of the "linuxadmin" contactgroup? This is not very clear. It's not a good practice to name the contact and the contactgroup with exactly the same names.

define contact {
contact_name linuxadmin

Also, can you post the "workhours_linux" timeperiod's definition and show us the actual email notification that you received?

HI,

I was making the post as generic as I can, we have names configured differently in our set up and just changed to linuxadmin on this post.

Here is the email we got (changed the ips/hostname)

From: nagios01@zzzz [mailto:nagios01@zzzzz]
Sent: Friday, March 20, 2015 7:38 PM
To: linuxadmin
Subject: PROBLEM (CRITICAL) host1 HP-UX /tmp
Importance: High

Nagios has detected a problem with this service.
CHECK_NRPE: Socket timeout after 30 seconds.
(host1) IP: xxx.xx.xx.xx
2015-03-20 19:38:01


and here is the timeperiod definition:

define timeperiod {
timeperiod_name workhours_linux
alias Normal Linux Work Hours
friday 05:00-19:00
thursday 05:00-19:00
wednesday 05:00-19:00
tuesday 05:00-19:00
monday 05:00-19:00
}


Thanks,

I was making the post as generic as I can...

It's difficult to try to guess which names are "real" and which are "generic".
Is the "host1 HP-UX" same as "host1"? Is "tmp Disk Usage" same as "/tmp"? Notifications are disabled for the " tmp Disk Usage" service, so you shouldn't be getting alerts. Can you check if you have any multiple instances of nagios running?

1. Is the "host1 HP-UX" same as "host1" == yes this is correct.
2. /tmmp Disk Usage" same as "/tmp"? Notifications are disabled for the " tmp Disk Usage" service === when the Notification Alerts tab is set to off, this will not inherit the setting for the HOST NOTIFICATION ALERT setting?
- does this mean that I need to define 2 service check for /tmp? One for 24/7 notification alerts and another /tmp service check but for "work hours" notification setting?
3. Can you check if you have any multiple instances of nagios running === Just 1 instance.

When you have:
in the service definition, the notifications will be disabled. See how this directive is used:
Can you email us your profile at [email protected]?

Admin->System Profile->Download Profile->save and email the "porfile.zip".

We will need to move this to our email ticketing system to further troubleshoot the issue. Thanks!