[hobbit] trap column not showing up

[FARRIOR, Andy]

The trap column will only display the last alert it receives.  If a trap
for a particular host hasn't been generated within Hobbit's "no response
from client" window, the trap column will be set to "green".  This is to
keep the column from going "purple".

If three traps for a given host are processed, it will only display the
color of the last trap.  If you are just using Hobbit, it will log any
color changes.  If you are using the SQL logging in SNMPTT, you'll have
a history of SNMP traps.

So if you receive these traps in sequence:
  green
  red
  green

The trap column will update with the first "green" status.  It will turn
"red" and send an alert when it receives the second status.  The trap
column will turn back to green when it receives the last status.

It's very possible that when you go to check on Hobbit after receiving
an alert for the "red" status, that the column will have changed back to
"green" assuming the last trap happened shortly after the second trap.


There's no intelligence in the trap.pl client that checks the status of
the previous alert or the type of previous alert (you can receive two
different traps in sequence that are "red" or CRITICAL) to determine if
it should update the status or not.  You have to rely on the SQL logging
if you want to see that since it logs every alert individually.



Andy





-----Original Message-----
From: Joshua Krause [mailto:joshuakrause at triad.rr.com] 
Sent: Monday, September 17, 2007 12:04 PM
To: hobbit at hswn.dk
Subject: Re: [hobbit] trap column not showing up

I have my traps working now.  But my question is where to set the time  
period of how long they stay a certain color?


On Sep 14, 2007, at 7:28 PM, FARRIOR, Andy wrote:

>
> After you convert you MIBs using "snmpttconvertmib", you have to edit
> the various lines in the snmptt.conf.* files that start with "EVENT"  
> and
> change the status from "Normal" to something more useful.  For  
> instance,
> by default, all of Cisco's traps have a status of "Normal".
>
> Here's a list of common event status settings and their corresponding
> Hobbit alert color:
>
> Normal - green
> INFORMATIONAL - green
> MINOR - yellow
> WARNING - yellow
> SEVERE - red
> MAJOR - red
> CRITICAL - red
>
>
> Example:
>
> The Cisco "linkDown" traps:
>
> EVENT linkDown .1.3.6.1.2.1.11.0.2 "Status Events" Normal
> EVENT linkDown .1.3.6.1.6.3.1.1.5.0.2 "Status Events" Normal
>
> If you want this to generate an alert, change "Normal" to  
> "WARNING" (for
> yellow) or "CRITICAL" (for red)
>
> The Dell and APC traps are usually predefined with the various status
> settings.
>
> Hope that helps,
> Andy
>
>
>
>
> -----Original Message-----
> From: Joshua Krause [mailto:joshuakrause at triad.rr.com]
> Sent: Friday, September 14, 2007 12:21 PM
> To: hobbit at hswn.dk
> Subject: Re: [hobbit] trap column not showing up
>
> Where do i need to tell hobbit/snmptt what traps i want to alarm out
> on?  I am a little fuzzy on where i need to configure that stuff at.
> And is there a way to customize the message that is sent to hobbit.
> Say someone logs in and does a config change.  Can I just have "Config
> Change" sent to hobbit?
>
> Thanks
> Josh
>
> **** trap column is there***
>
> On Sep 14, 2007, at 12:26 PM, Ralph Mitchell wrote:
>
>> On 9/14/07, Joshua Krause <joshuakrause at triad.rr.com> wrote:
>>> I think i know why i may not be getting the trap column but needed
>>> some  insight on fixing it.  The trap is coming in with the
>>> hostname of:
>>>
>>> gsonc-spg-ups-1.eng.gbo.twcable.com
>>>
>>> But the hostname in the bb-hosts file is just:
>>>
>>> gsonc-spg-ups-1
>>>
>>> How could i go about parsing out the " .eng.gbo.twcable.com " from
>>> the
>>> hostname coming across in the trap?
>>>
>>> Thanks
>>> Josh
>>>
>>> On Sep 14, 2007, at 11:03 AM, Joshua Krause wrote:
>>>
>>>> I am having an issue with getting the trap column to show up on a
>>>> host.  I have everything setup the way the documentation said to.
>>>> And all of my log files are logging except for the trap.log, which
>>>> looks to be generated when the hobbitlaunch.cfg is ran.  Anyone  
>>>> have
>>>> anything that might be able to help me?
>>
>> Try the "CLIENT:hostname" tag in the bb-hosts file:
>>
>>   x.x.x.x   gsonc-spg-ups-1   # CLIENT:gsonc-spg-
>> ups-1.eng.gbo.twcable.com
>>
>> I think that's supposed to take care of this situation.
>>
>> Ralph Mitchell
>>
>> To unsubscribe from the hobbit list, send an e-mail to
>> hobbit-unsubscribe at hswn.dk
>>
>
>
> To unsubscribe from the hobbit list, send an e-mail to
> hobbit-unsubscribe at hswn.dk
>
>
>
> To unsubscribe from the hobbit list, send an e-mail to
> hobbit-unsubscribe at hswn.dk
>


To unsubscribe from the hobbit list, send an e-mail to
hobbit-unsubscribe at hswn.dk