Re: [Testbed-admins] update need advice regarding "SNMP SET failed"

[Robert P Ricci <ricci@cs.utah.edu>]

So, what happens if you try the command I suggested to Adit?

    snmpwalk -m all <switchname> <communitystring> portAdminSpeed

(this will help us determine if maybe we're just using the wrong method
of indexing, etc.)

Another question: Can you tell whether the ports on your experiment
switch are capable of doing speeds other than gigabit? Most can do
10/100/1000, but I've run into some that are gigabit-only, and that
could result in this error.

Thus spake Cheng Cui on Mon, Apr 13, 2009 at 07:40:56PM -0500:
> We use cisco WS-C3560G-24TS as our experiment switch. Just for the first simple step to check MIB, I go through online and find that:
> "MIBs supported by Catalyst 3560 series since IOS release 12.1(19)EA1b June 1995"
> => CISCO-STACK-MIB
> => CISCO-VLAN-MEMBERSHIP-MIB 
> => CISCO-VTP-MIB
> That means our experiment switch has such MIBs.
> 
> OK, then let's see further investigation on this MIB problem.
> 
> Best Regards,
> 
> Cheng Cui
> 578-5445 ??? 231 Johnston Hall ??? Baton Rouge, LA 70803
> 
> -----Original Message-----
> From: Ranadive, Adit U [mailto:adit262@cc.gatech.edu] 
> Sent: Monday, April 13, 2009 5:33 PM
> To: Keith Sklower
> Cc: azad@cct.lsu.edu; testbed-admins@flux.utah.edu; ccui1@tigers.lsu.edu; lahey@isi.edu; stoller@flux.utah.edu
> Subject: Re: [Testbed-admins] update need advice regarding "SNMP SET failed"
> 
> Yes, we're having the same problem with Cisco 4006/4506 switches.
> 
> The MIB being consulted is the CISCO-STACK-MIB. I had posted some
> mail earlier which discussed that. It would seem that these IOS switches
> do not have support for this MIB.
> To find which MIBs are supported by your switch you should look at the
> MIB Locator page on the cisco website.
> 
> Thanks,
> Adit
> 
> 
> ----- Original Message -----
> From: "Keith Sklower" <sklower@vangogh.cs.berkeley.edu>
> To: ccui1@tigers.lsu.edu, lahey@isi.edu, sklower@vangogh.cs.berkeley.edu, stoller@flux.utah.edu
> Cc: azad@cct.lsu.edu, testbed-admins@flux.utah.edu
> Sent: Monday, April 13, 2009 5:37:39 PM GMT -05:00 US/Canada Eastern
> Subject: Re: [Testbed-admins] update need advice regarding "SNMP SET failed"
> 
>     From: "Cheng Cui" <ccui1@tigers.lsu.edu>
>     Date: Mon, 13 Apr 2009 16:25:16 -0500
> 
>     Hello guys,
> 
>      
> 
>     I want to give more detail on our problem below. The previous problem" SNMP
>     GET failed - In /usr/testbed/bin/snmpit" was solved because the boss ip
>     (192.168.0.70) is mis-match with the cisco2(192.168.1.1). But now we have
>     another problem like "SNMP SET failed":
> 
>     *** WARNING: snmpit:
>     ***   SNMP SET failed - In /usr/testbed/bin/snmpit
>     ***   SNMPIT set failed for device cisco2 (try 10 of 10)
>     ***   Variable was [portDuplex,0.5,2,INTEGER]
>     ***   Returned (undefined), ErrorNum was 6
>     ***   Error string is: noAccess
> 
> I have the very strong impression that the Georgia Tech guys
> are having the same problem with a different cisco switch type
>      
> 
>     For simplicity, I altered our topology design as boss will not be directly
>     connected to the Experiment switch.
> 
> Since you can do an snmp "get" with the read-write string, I do not believe
> that it is a problem with topology or configuration.
>      
> 
> At Berkeley, we don't have any cisco switches; on Deter there is one
> large switch running CatOS.
> 
> I have limited experience dealing with cisco switches, but
> I might suggest the following -> if Cisco provides some remote
> configuration tool using snmp that you contrive some setup
> where you can run tcpdump and snoop on the packet dialog
> when you configure the duplex for the port.
> 
> For example, the configuration that HP(Procurve) provides is windows based,
> and I had to manually set up a bridging bsd node between
> two vlans inorder to discover what the magic sequence was for
> creating a vlan via SNMP in order to write snmpit_hp.pm.
> 
> It's then going to require some hunting through all the cisco mibs
> in order to find which one is being consulted.
> 
> Good Luck!
> 
> Keith Sklower
> _______________________________________________
> Testbed-admins mailing list
> Testbed-admins@flux.utah.edu
> http://www.flux.utah.edu/mailman/listinfo/testbed-admins
> 
> 
> _______________________________________________
> Testbed-admins mailing list
> Testbed-admins@flux.utah.edu
> http://www.flux.utah.edu/mailman/listinfo/testbed-admins

-- 
/-----------------------------------------------------------
| Robert P Ricci <ricci@cs.utah.edu> | <ricci@flux.utah.edu>
| Research Associate, University of Utah Flux Group
| www.flux.utah.edu | www.emulab.net
\-----------------------------------------------------------