Hi all,
I am using nagios 4.2 to monitor our local servers. But we face some problem on email notification function. if i set the host 'max_check_attempts' > 1, nagios will never send email notification to contact groups. If set the 'max_check_attempts' = 1, everything is normal. Email notification will send immediately if any service down. However, we don't want too many false alarm and want to set the max_check_attempts to 2. Can anyone help me? I am using amazon ec2 linux instance. Thanks. Below is the config of nagios.



host1.cfg

define hostgroup{
hostgroup_name hostgroup1
alias hostgroup1
}

define host {
use linux-server
host_name host1
hostgroups hostgroup1
alias host1
address xx.xx.xx.xx
register 1
}

define service{
host_name host1
service_description PING
check_command check_ping!150.0,20%!500.0,60%
max_check_attempts 2
check_interval 2
retry_interval 2
check_period 24x7
check_freshness 1
contact_groups admins
notification_interval 2
notification_period 24x7
notifications_enabled 1
register 1
}

define service{
use generic-service
host_name host1
service_description Server HTTP URL available check
check_command check_http! -H xxxxxxxxxxxx -u xxxxxxxxxxxxxx -w 8
max_check_attempts 2
check_interval 2
check_period 24x7
}



contacts.cfg

define contact{
contact_name nagiosadmin
use generic-contact
alias Nagios Admin

email xxxxxxxxxxxxxxxx
service_notification_commands notify-service-by-email
host_notification_commands notify-service-by-email
}



define contactgroup{
contactgroup_name admins
alias Nagios Administrators
members nagiosadmin
}


templates.cfg

define contact{
name generic-contact
service_notification_period 24x7
host_notification_period 24x7
service_notification_options w,u,c,r,f,s
host_notification_options d,u,r,f,s
service_notification_commands notify-service-by-email
host_notification_commands notify-host-by-email
register 0
}


Best Regards,
Kelvin

Hi all,
There are some additional information in nagios.log. I found that the status of service changed to 'HARD' but still no SERVICE NOTIFICATION. But the host check is normal. What am I doing wrong here?


[1470986558] SERVICE ALERT: xxxxxxxxx;PING;CRITICAL;SOFT;1;PING CRITICAL - Packet loss = 100%
[1470986588] HOST ALERT: xxxxxxxxx;DOWN;SOFT;1;PING CRITICAL - Packet loss = 100%
[1470986650] SERVICE ALERT: xxxxxxxxx;Server HTTP URL available check;CRITICAL;SOFT;1;CRITICAL - Socket timeout
[1470986678] SERVICE ALERT: xxxxxxxxx;PING;CRITICAL;HARD;2;PING CRITICAL - Packet loss = 100%
[1470986678] HOST ALERT: xxxxxxxxx;DOWN;SOFT;2;PING CRITICAL - Packet loss = 100%
[1470986768] HOST ALERT: xxxxxxxxx;DOWN;SOFT;3;PING CRITICAL - Packet loss = 100%
[1470986770] SERVICE ALERT: xxxxxxxxx;Server HTTP URL available check;CRITICAL;HARD;2;CRITICAL - Socket timeout
[1470986800] HOST ALERT: xxxxxxxxx;DOWN;SOFT;4;PING CRITICAL - Packet loss = 100%
[1470986890] HOST ALERT: xxxxxxxxx;DOWN;SOFT;5;PING CRITICAL - Packet loss = 100%
[1470986948] HOST ALERT: xxxxxxxxx;DOWN;SOFT;6;PING CRITICAL - Packet loss = 100%
[1470987038] HOST ALERT: xxxxxxxxx;DOWN;SOFT;7;PING CRITICAL - Packet loss = 100%
[1470987128] HOST ALERT: xxxxxxxxx;DOWN;SOFT;8;PING CRITICAL - Packet loss = 100%
[1470987150] SERVICE ALERT: xxxxxxxxx;Server HTTP URL available check;CRITICAL;SOFT;1;CRITICAL - Socket timeout
[1470987150] SERVICE ALERT: xxxxxxxxx;Server HTTP URL available check;CRITICAL;SOFT;1;CRITICAL - Socket timeout
[1470987160] HOST ALERT: xxxxxxxxx;DOWN;SOFT;9;PING CRITICAL - Packet loss = 100%
[1470987250] HOST ALERT: xxxxxxxxx;DOWN;HARD;10;PING CRITICAL - Packet loss = 100%
[1470987250] HOST NOTIFICATION: nagiosadmin;xxxxxxxxx;DOWN;notify-service-by-email;PING CRITICAL - Packet loss = 100%

But you get the Host DOWN notification?
It looks like its soft state blocks the service notifications.
If you stop http while ping still works then you should get the service notification.

1 members found this post helpful.

It seems that you are right. If only http down, nagios will send the service notification to us. So this is the nagios default function that will ignore the service notification if the host is down? Thanks.

Yes, I think the service alert suppression is a feature.
Normally you have a dozen service checks per host, and you want to only get the Host DOWN notification.
For example see this article

1 members found this post helpful.

Well, i think the host down notification is enough. Thanks MadeInGermany