[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [hobbit] Alert problems



Hi Leonardo,
	If my memory is working as expected :) -- when you use "%" it
means your using regular expression, and in this case it is doing
correctly. 

	Try change it in a way it reads as:
 HOST=%^servername$ SERVICE=conn COLOR=red RECOVERED
         MAIL    id (at) email.com.br               FORMAT=sms REPEAT=1h


	Note the "$" after "servername".

	-wm


----------------------- Original Message -----------------------
From:    Leonardo Bozi <leobozi (at) yahoo.com>
To:      hobbit (at) hswn.dk

Date:    Thu, 29 May 2008 15:00:21 -0700 (PDT)
Subject: [hobbit] Alert problems
----

> Hi,
> 
> I upgraded Hobbit version 4.1.2p1 to 4.2.0 (CentOS 4 x86_64) with allinone.patch but I'm having problems to set alerts with clusters which have nodes with similar names. For exemple:
> servername -> head node name;
> servername001 - batch node name;
> servername002 - batch node name;
> ...
> Before the upgrade, the alerts was working well and th hobbit-alerts.cfg is configured like this:
> 
> HOST=%^servername SERVICE=conn COLOR=red RECOVERED
>         MAIL    id (at) email.com.br               FORMAT=sms REPEAT=1h
> 
> The alerts of the nodes are working after the upgrade, but the alert of the head node are appearing multiplied by the number of the nodes of the cluster. I attached a picture that show the problem.
> 
> Did anybody see a problem like that?
> 
> Thanks in advance.
> 
> Leonardo
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> ------------------------------------------------------------
> Abra sua conta no <http://br.rd.yahoo.com/mail/taglines/mail/*http://br.mail.yahoo.com/>Yahoo! Mail, o único sem limite de espaço para armazenamento!