Complex Condition for an Action

[jammann]

Hi everybody

I tried to figure out (with Zabbix 1.6.4) if I could make an Action depend on the following condition

(Trigger Description not like "Test") AND
(NOT (Host == "A" AND Time period in 1-7,22:45-23:15))

Reason: Host A automatically reboots every night at 23:00 and I want to suppress alerts. And we have some triggers defined with "Test" in the name, which we don't want alerts for either.

I did not succeed in specifying the above condition in one single action. As workaround, I created 2 action, so the problem is solved.

I just want to know if the above could have been done in one single action?

PS: Logic says I can rewrite the above as

(Trigger Description not like "Test") AND
(Host <> "A" OR Time period not in 1-7,22:45-23:15)

but I didn't succeed with this either :-/

[Andreas Bollhalder]

Hi

I use the zabbix-sender in the restart script to inform the ZABBIX server about the restart.

In my case, this will enable a trigger named "Host restart by script initiated" which will stay TRUE (by help of the nodata() function) for a while to allow the server to completely restart. The trigger has the severity of "Information" which is configured in the media to not start any actions. Then I make all needed dependencies to this trigger so ZABBIX won't send alerts in this case. With this, you never have to change a trigger if the restart time changes.

The only thing I haven't done is the use of escalation if some service/item didn't come up after the restart.

Andreas

[jammann]

Great idea, it works smoothly! I'd never have thought of this

Still many tricks to learn with Zabbix ...

[bashman]

(Trigger Description not like "Test") AND
(Host <> "A" OR Time period not in 1-7,22:45-23:15)

but I didn't succeed with this either :-/

It seems that Trigger Description refers to trigger comment field, so add Test to comment.

http://www.zabbix.com/forum/showthread.php?t=13672&highlight=trigger+description

[bashman]

Hi

I use the zabbix-sender in the restart script to inform the ZABBIX server about the restart.

In my case, this will enable a trigger named "Host restart by script initiated" which will stay TRUE (by help of the nodata() function) for a while to allow the server to completely restart. The trigger has the severity of "Information" which is configured in the media to not start any actions. Then I make all needed dependencies to this trigger so ZABBIX won't send alerts in this case. With this, you never have to change a trigger if the restart time changes.

The only thing I haven't done is the use of escalation if some service/item didn't come up after the restart.

Andreas

I have a complex scenario:

A Java application sending trappers to Zabbix Server from a tomcat.

In tomcat init script, I have zabbix_sender telling Zabbix Server that service has manually stopped.

I have defined dependencies like this: "Java application trapper" depends on "Tomcat manual stop", so when I stop the tomcat, the java trapper won't notify by email.

Well, this doesn't work, I can see that zabbix_sender was received before java trapper and I got both email notifications.

If I send the zabbix_sender manually, instead of using tomcat init script, and stop tomcat, it works great.

The curious thing is that I have the same results in the "Tomcat manual stop" and "Java application trapper", "lastest 500 values".

Does anybody know what's going on?.

[bashman]

I have a complex scenario:

A Java application sending trappers to Zabbix Server from a tomcat.

In tomcat init script, I have zabbix_sender telling Zabbix Server that service has manually stopped.

I have defined dependencies like this: "Java application trapper" depends on "Tomcat manual stop", so when I stop the tomcat, the java trapper won't notify by email.

Well, this doesn't work, I can see that zabbix_sender was received before java trapper and I got both email notifications.

If I send the zabbix_sender manually, instead of using tomcat init script, and stop tomcat, it works great.

The curious thing is that I have the same results in the "Tomcat manual stop" and "Java application trapper", "lastest 500 values".

Does anybody know what's going on?.

Reported as: https://support.zabbix.com/browse/ZBX-3164

[subba5678]

Hi Bashman,
I have a problem with my trigger , If i get one trigger stating " Zabbix Agent is Down " or " Ping Availability is Zero
" for that i am getting nearly 10 Resolve alerts for the same . Please find the details of Trigger description as well as Action Description please help me out

{Toko - cesl tot657:icmpping[100.100.3.90,5,300,32,100].min(300)}=0

Thanks,
Subbu

[bashman]

Try to modify your item to icmpping only.

[subba5678]

Hi Bashman,
Thanks for the reply. I tried with the ICMPING also but not able to fix this. Please help me how to stop the Recovery alerts i am getting more for 1 Severity Problem . Please help me out

Thanks,
Subbu

[bashman]

I don't know if I understood, are you saying that you're getting more than one alert per trigger (the one you showed) for one action (the one you showed)?.

[subba5678]

Hi,
Thanks for the quick reply , I will explain in clear way that i created a trigger to send an alert if the agent is down , whenever agent goes down i got 1 trigger saying that agent goes down . Whenever the agent start i need to get only one resolved alert but i am getting more than 5 resolved alerts . Also i checked the action was configured only one time Please find the attchment of my action configuration .Please help me out.

{Toko - cesl tot657:status.min(900)}=2


Thanks,
Subbu

[bashman]

That's strange, if you get one alert about zabbix agent is down, you should get only one recovery message about zabbix agent is down.

The action seems fine, but I don't understand your trigger:

{Toko - cesl tot657:status.min(900)}=2

[subba5678]

HI Bashman,
Thanks for the reply . I am monitoring totally 50 Servers in Zabbix . In that nearly about 10 servers I am getting multilple recover alerts . for all the servers i am using the same Trigger . please find the my trigger details . Please help me out , I am using this in the production

Template_Windows:svs- Rosse TAL765 Zabbix Agent is down {svs - Rosse TAL765:status.min(900)}=2



Thanks,
Subbu