Smarts IP: Discovery timeout error; 'SNMP-ETIMEOUT-Timed out' seen in Smarts IP log

Products

• VMware

Issue/Introduction

Symptoms:




Smarts IP Discovery timeout error
Smarts IP log file displays the following error message:

[14-Jul-2008 04:50:42 PM+089ms BST] t@60 Discovery #4
SWFE-W-ETIMEOUT-GET_NEXT request timed out for Agent : x.x.x.x, OID:
   .1.3.6.1.2.1.2.2.1.3.10115
SNMP-ERESPONSE-No response from x.x.x.x, port 161
SNMP-ETIMEOUT-Timed out
                 

 

The snoop output shows a trap generated by the device's agent as a response to the get-next-request as in the following example:

get-next-request

Simple Network Management Protocol
      version: version-1 (0)
      community: ABC@2

trap

Simple Network Management Protocol
      version: version-1 (0)
      community: ABC
      data: trap (4)
          trap
                 enterprise: 1.3.6.1.4.1.9.1.217 (SNMPv2-SMI::enterprises.9.1.217)
                 agent-addr: x.x.x.x (x.x.x.x)
                 generic-trap: authenticationFailure (4)

Another Example for SNMP Timeout issue: 

Basically when a walk request is sent to the device SNMP agent process, the expectation is to get a response with the data or respond with "NoSuchInstance" and timing out is not a valid response:

The OID that the device was timing out was as follows:

pimInterfaceStatusOID
{".1.3.6.1.3.61.1.1.2.1.7."}

This is a multicast discovery OID and used to populate the "MCAST_TopologyCollection" class as part of the end device probe.

 After loading the MIB walk file in EMC lab machine; performed a manual SNMP walk on the above OID - .1.3.6.1.3.61.1.1.2.1.7 and below is the response: 

$ ./sm_snmp.exe -c public -s 2c -d <Device IP> -p <Domain Port> walk .1.3.6.1.3.61.1.1.2.1.7
SNMP Walk MIB starting at .1.3.6.1.3.61.1.1.2.1.7
.1.3.6.1.3.61.1.1.2.1.7.110 = 1
.1.3.6.1.3.61.1.1.2.1.7.111 = 1

With respect to the above output; there is an output and a valid response. Even "NoSuchInstance" response is considered a valid response.

Same command has been executed in user environment against the device <"Problamatic Device"> and observe the below output. 

[root@xxxx bin]# ./sm_snmp -c <community_string> -s 2c -d <Device IP> -p <Domain Port> walk .1.3.6.1.3.61.1.1.2.1.7
SNMP Walk MIB starting at .1.3.6.1.3.61.1.1.2.1.7
[July 11, 2016 2:47:07 PM GMT+01:00 +898ms] t@2097300336 <Primary Thread>
SNMP-E-ERESPONSE-No response from 169.35.198.221, port 51309
SNMP-ETIMEOUT-Timed out; in file "/work/tancurrent/DMT-9.2.0.0/19/smarts/snmp/lib/SNMP_RequestsSender.c" at line 1337

With respect to the above output; its timing out.  

Environment

VMware Smart Assurance - SMARTS

Cause

The SNMP-TIMEOUT is happening for the VLAN walks. The VLAN walk walks the dot1dBridge (OID ".1.3.6.1.2.1.17") appending the VLAN ID to the original community string. For each of the getnext requests with community string "ABC@2" (this is the walk for VLAN ID 2) the agent sends an authenticationFailure trap. Since there is no response for these requests, the time out occurs.

Resolution

Smarts is working as designed. This issue indicates a problem with the SNMP agent and must be addressed with the SNMP agent vendor.