Re: SNMP Traps

I've been using SNMP for a very long time, but I have never had a
particular reason to create my own custom MIBS or work with generating
custom trap's.  Most of my work with SNMP is for performance metrics
(cpu/mem), command definitions in the snmpd.conf file (custom programs
which i want to query exit codes in via snmpd), etc.

Scott


On Tue, Oct 6, 2009 at 12:18 PM, Matthew Lechleider
<[email protected]> wrote:
> SNMP is not as difficult as it may seem.  It can be your best friend and
> open a lot of doors.
>
> Once you learn to write SNMP commands you will see it is no different than
> learning Linux commands.
>
> Maybe try writing a bash or pearl script with some basic SNMP commands to
> help your issue?
>
>
> tcpdump is another cool tool to use to verify your SNMP data is arriving and
> behaving properly.
>
>
> Scott Behrens wrote:
>>
>> Interesting idea John, I will take a look.  SNMP is not nearly as
>> simple as it seems (after doing more research).
>>
>> On Tue, Oct 6, 2009 at 10:05 AM, John Kristoff <[email protected]> wrote:
>>
>>>
>>> On Tue, Oct 06, 2009 at 09:24:05AM -0500, Scott Behrens wrote:
>>>
>>>>
>>>> I was wondering if anyone had experience writing custom MIB's or
>>>> working with SNMP Traps.  I need to come up with a custom trap for
>>>> detecting core dumps on some systems I've been working with.  The
>>>> information I have found seems pretty confusing and complex, and I was
>>>> wondering if anyone had good tips.
>>>>
>>>
>>> I never run my own SNMP trap manager.  Never liked SNMP for that
>>> sort of thing.  Its been a long time since I looked, but most of
>>> the SNMP manager consoles I have seen where big bloated, buggy
>>> pieces of crap.  No kidding.  Plus, SNMP is not so simple as the
>>> name implies to work with.  I've had better luck using syslog
>>> for similar functionality.
>>>
>>> Something else I've used that I'd propose is something like a
>>> safetynet script.  You can do something basic in a start-up
>>> script like this:
>>>
>>>  STOP_RUN=/path/to/file/to/touch/for/normal/shutdown
>>>  while [ ! -f ${STOP_RUN} ]
>>>  do
>>>   /path/to/crappy/app
>>>   sleep 5
>>>   echo 'restarting' | mail -s '### crappy app alert' [email protected]
>>>  done
>>>
>>>
>>> If crapp appy crashes, the control script should start it up again,
>>> unless you've touched the file pointed to by STOP_RUN.  You want
>>> that in case you actually want to stop the app for real and not
>>> have it automatically restarted.
>>>
>>> John
>>>
>>>
>>
>> _______________________________________________
>> DLC mailing list
>> [email protected]
>> http://mailman.depaul.edu/mailman/listinfo/dlc
>> Use the Above Link to Unsubcribe!!
>> http://linux.depaul.edu/
>>
>
> _______________________________________________
> DLC mailing list
> [email protected]
> http://mailman.depaul.edu/mailman/listinfo/dlc
> Use the Above Link to Unsubcribe!!
> http://linux.depaul.edu/
>
_______________________________________________
DLC mailing list
[email protected]
http://mailman.depaul.edu/mailman/listinfo/dlc
Use the Above Link to Unsubcribe!!
http://linux.depaul.edu/