Hi,
I've a problem with an NTP daemon dying now and again.
An action has been written to restart it, but it only triggers for some servers, not others.
For my linux servers I have a Linux-servers-Action that normally just reports problems by email. A new ntp-restart-action has been added to restart the NTP automatically and email.
The Linux-servers-Action was modified to ignore NTP errors:
Trigger value = "PROBLEM"
Trigger <> "NTP daemon"
Host group = "Linux servers"
To be caught by ntp-restart-action:
Trigger value = "PROBLEM"
Trigger = "NTP daemon"
For most systems I get the email generated by Linux-servers-Action, and just one single systems triggers ntp-restart-action.
Now the trigger used is identical in both cases and all servers concerned belong to "Linux servers", I'm mystified about this.
Where could one start tracing the problem?
Is there a way to force the order of actions being matched?
Zabbix 1.8.2.
Thanks in advance...
I've a problem with an NTP daemon dying now and again.
An action has been written to restart it, but it only triggers for some servers, not others.
For my linux servers I have a Linux-servers-Action that normally just reports problems by email. A new ntp-restart-action has been added to restart the NTP automatically and email.
The Linux-servers-Action was modified to ignore NTP errors:
Trigger value = "PROBLEM"
Trigger <> "NTP daemon"
Host group = "Linux servers"
To be caught by ntp-restart-action:
Trigger value = "PROBLEM"
Trigger = "NTP daemon"
For most systems I get the email generated by Linux-servers-Action, and just one single systems triggers ntp-restart-action.
Now the trigger used is identical in both cases and all servers concerned belong to "Linux servers", I'm mystified about this.
Where could one start tracing the problem?
Is there a way to force the order of actions being matched?
Zabbix 1.8.2.
Thanks in advance...
roc.num[ntpd].last(0)}=0
Comment