multiple actions from trigger

[td3201]

Hello,

I have a vfs.fs trigger that looks like this:
{foo_Template_Windows:vfs.fs.size[c:,free].last(0)}<7415529472

I have a host that is generating an event for this trigger. That looks good and is working as expected. However, the action is acting funny. I am getting two notifications, one that says the {STATUS} is OK, and the other says the {STATUS} is PROBLEM. Not sure how this can be. This tells me that it is flapping but it isn't because I don't have recovery message checked. I dropped the period on the action to 60 seconds so I can troubleshoot and every 60 seconds I got two messages.

The action has escalations enabled and 1 step that sends me an email with the following parameters:
From: 1
To: 0
Period: 0
Default Message: checked.

I have a feeling that the problem lies in the escalation steps but not sure how. I would expect that this type of escalation will notify every 60 seconds forever until the status changes.

What am I missing?

[td3201]

Just for reference, here's the action and the two emails:

ACTION
=========================================
{TRIGGER.NAME}: {STATUS}
Latest Value: {{HOSTNAME}:{TRIGGER.KEY}.last(0)}

EMAIL1
=========================================
Low free disk space on 10.51.1.13 volume c:: OK
Latest Value: 4.76 GB

EMAIL2
=========================================
Low free disk space on 10.51.1.13 volume c:: PROBLEM
Latest Value: 4.76 GB

[jroberson]

It sounds like you need to add a condition match the value of the trigger before it sends you an e-mail. I had the same issue. Add this to the conditions for the action:

Trigger value = "PROBLEM"

Otherwise it will send you an email for each status change (OK & Problem). If you keep getting emails, but they all say "Problem" after that, then you have a flapping trigger.

[td3201]

I don't understand. The item is not recovering. These notifications are all happening with 1 event that is a problem event.

Besides, what is the recovery checkbox for then?

[jroberson]

I guess I misread your post. I was having a problem that I received two notifications for when the trigger sends "PROBLEM" and "OK", but I didn't want to get the "OK" notification. The solution was to only send when there was a "PROBLEM" status. That sounded like what you were experiencing. I don't use the recovery options yet, but you are correct that it sounds like you should receive the recovery message only when you get the "OK" status change. I don't use escalations myself, so I might not be able to help you with this one. Sorry.

BUT you might find this useful --> http://www.zabbix.com/forum/showthread.php?t=11433

[tekknokrat]

I have a similar issue.

I am looking for regex in a logfile via this trigger expression:

{scg_logging:log[/var/log/10.4.190.16/syslog,unsuccessful].count(600)}>1

When I have recovery checked I get also a PROBLEM & OK Message but the OK message is only a further regex check. Whats causing this behaviour?