|
From: Jon D. <jd...@us...> - 2010-05-19 20:59:08
|
> 1. Re: create new template question (Buchan Milne)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Fri, 5 Mar 2010 22:44:35 +0100
> From: Buchan Milne <bg...@st...>
> Subject: Re: [Devmon] create new template question
> To: dev...@li...
> Cc: Jon Dustin <jd...@us...>
> Message-ID: <201...@st...>
> Content-Type: Text/Plain; charset="iso-8859-1"
>
> On Sunday, 28 February 2010 00:18:19 Jon Dustin wrote:
>> Greetings -
>>
>> I am trying to develop a new template for Cisco Wireless Controllers, and
>> running into a snag with devmon's features. Here is a quick snippet of the
>> SNMP conversations I am trying to analyze:
>>
>> (names of wireless access points connected to controller)
>> snmpwalk -v2c -cpublic 10.13.1.4 .1.3.6.1.4.1.14179.2.2.1.1.3
>> SNMPv2-SMI::enterprises.14179.2.2.1.1.3.0.35.235.10.79.64 = STRING:
>> "wap-1st-2" SNMPv2-SMI::enterprises.14179.2.2.1.1.3.0.37.132.253.194.144 =
>> STRING: "wap-2nd-5"
>> SNMPv2-SMI::enterprises.14179.2.2.1.1.3.0.37.132.253.200.192 = STRING:
>> "wap-2nd-4"
>>
>> (number of clients connected to each WAP - by radio)
>> snmpwalk -v2c -cpublic 10.13.1.4 .1.3.6.1.4.1.14179.2.2.2.1.15
>> SNMPv2-SMI::enterprises.14179.2.2.2.1.15.0.35.235.10.79.64.0 = Counter32: 4
>> SNMPv2-SMI::enterprises.14179.2.2.2.1.15.0.35.235.10.79.64.1 = Counter32: 1
>> SNMPv2-SMI::enterprises.14179.2.2.2.1.15.0.37.132.253.194.144.0 =
>> Counter32: 7
>> SNMPv2-SMI::enterprises.14179.2.2.2.1.15.0.37.132.253.194.144.1 =
>> Counter32: 3
>> SNMPv2-SMI::enterprises.14179.2.2.2.1.15.0.37.132.253.200.192.0 =
>> Counter32: 1
>> SNMPv2-SMI::enterprises.14179.2.2.2.1.15.0.37.132.253.200.192.1 =
>> Counter32: 1
>>
>> In the above examples, 0.35.235.10.79.64 represents a single WAP (the
>> repeater part of the oid identifies the MAC address in decimal). In the
>> above example, the WAP labeled as wap-1st-2 has a "repeater OID" of
>> 0.35.235.10.79.64.0, which corresponds to a MAC address of
>> 00:23:eb:0a:4f:40 (not particularly germane to this question, but I
>> thought I would include the example).
>>
>> Each WAP has two radios, which lead to two different results (.0 and .1).
>>
>> What I would *like* to have is a nice devmon table similar to "regular"
>> switches, with the WAP name in the 1st column, the .0 client count in
>> column 2, and .1 count in column 3:
>>
>> WAP Name 2.4GHz clients 5 GHz clients
>> wap-1st-2 4 1
>> wap-2nd-5 7 3
>> wap-2nd-4 1 1
>>
>> But I believe I have found a limitation with the devmon template
>> capabilities. I can easily get the names of the WAPs, and even the
>> repeater section of the OID (via a transform of INDEX). But I have tried
>> the CHAIN, INDEX, and REGSUB transform, and nothing is able to produce the
>> results I seek.
>>
>> I need to have a "devmon transforms" that can give me a new oid like
>> follows:
>>
>> wapClients24 = {.1.3.6.1.4.1.14179.2.2.2.1.15.}{wapRepeaterOid}.0
>> wapClients24 = {.1.3.6.1.4.1.14179.2.2.2.1.15.}{wapRepeaterOid}.1
>>
>> In all my reading of the devmon documentation, I cannot think of any scheme
>> to do this type of operation.
>>
>> Please accept my apologies for such a verbose post, I hope the details
>> might help you understand my situation. Any help would be appreciated.
>>
>
> I think it should be possible, and we just installed a Cisco Wireless
> Controller before I went on leave, so I will look at this as soon as I am
> back
> in the office ...
Hello Buchan -
Have you made any progress at supporting the Cisco WLC with devmon? I have written a perl script to poll the required SNMP mibs, and combine the results, but I'm running into a wall trying to learn how Xymon expects to see "devmon-like data" submitted. I like how you are sending an entire repeater-mib worth of results, and I want to accomplish the same thing.
So - IF you have not been able to get devmon working for the above request, can you send me a pointer about how to send data like devmon? I'm hoping to use the "auto-rrd" tools that devmon/xymon seem to provide, rather than being forced into another NCV script.
Thanks for anything you can pass along.
--
Jon Dustin - Network Specialist
University of Southern Maine
Portland, ME 207-780-4152
|