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

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



Hello guys,

 

Another problem happens and it looks wired. I can test the snmpget -i + <arg> to cisco2 from boss after I re-matched the ip address of boss and experiment switch : ) But in the experiment, it fails with such error message:

 

*** 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 believe boss is sending package through 192.168.1.70(boss) to 192.168.1.1(experiment switch). And during the experiment, I get the vlan set up from boss terminal, however, after the experiment fails, those setup was cleaned up.

%wap snmpit -l

VLAN     Project/Experiment VName     Members

--------------------------------------------------------------------------------

18       testbed/test11     link0-del pc:1 pc3:2

19       testbed/test11     link0-del pc2:1 pc3:1

 

My ns script is quite simple because there are only 3 free nodes for testing:

set link0 [$ns duplex-link $nodeB $nodeA 30Mb 50ms DropTail]

tb-set-link-loss $link0 0.01

 

I am stuck here as my knowledge is limited. Does anyone has an idea about this problem?

 

Best Regards,

 

Cheng Cui

578-5445 • 231 Johnston Hall • Baton Rouge, LA 70803

 

-----Original Message-----
From: Keith Sklower [mailto:sklower@vangogh.CS.Berkeley.EDU]

Subject: RE: [Testbed-admins] need advice regarding "SNMP GET failed"

   From ccui1@tigers.lsu.edu Sun Apr 12 15:56:39 2009

   When I check the "show running config", I can find the set community
string,
   as follows:

   ************************************
   cisco2#show running-config
   Building configuration...

   Current configuration : 3457 bytes
   !
   ! Last configuration change at 10:50:49 CDT Sun Apr 12 2009
   ! NVRAM config last updated at 10:51:57 CDT Sun Apr 12 2009
   !
   version 12.2
   no service pad
   service timestamps debug uptime
   service timestamps log uptime
   no service password-encryption
   !
   hostname cisco2
   ...
   ...
   interface Vlan1
    no ip address
   !
   interface Vlan2
    ip address 192.168.1.1 255.255.255.0
   !
   ip default-gateway 192.168.0.70
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> Note This <<<<<<<<<<<<<<<<<<

I'm a little confused here; in an earlier email, you said that
cisco2 was pingable from your boss, but I'm wondering how that
can be since there does not appear to be any way for boss to send
packets to vlan 2  (the ip default-gateway is not reachable from it).


   ip classless
   ip http server
   ip http secure-server
   !
   !
   access-list 1 permit 192.168.0.70
   access-list 2 permit 192.168.0.70
   no cdp run
   snmp-server community vlan2 RW 1
   snmp-server community cisco2 RW 2
   ...
   ************************************

   But I didn't see any error message like "C388 Unknown community name".

It was not an error message, it was the second or third line down
in the output of the "Show snmp" command.

   Meanwhile, as you mentioned, I can try snmpit -v 1 <other args> on both
   switch. It looks fine on control switch, but still time out on the
   experiment switch, with the same community string in the database. Here
is
   the output:

[snip]


Well, at least it appears that your database is appropriately
configure for both switches ... ;-)