1. Support
  2. Knowledge base
  3. 100026646

How to check SNMP trap configuration on 50xx appliances

Article: 100026646
Last Published: 2012-03-20
Ratings: 0 1
Product(s): Appliances

Problem

How to set up and check SNMP functionality in 50xx appliances.

Solution

1) Test SNMP connectivity via script /opt/pdconfigure/scripts/support/send_SNMPtrap:

If possible start an snmpd listener on some host in the network. Alternatively download open source snmp software for PC and use the desktop as a target.

Run the following command on the appliance:

 /opt/pdconfigure/scripts/support/send_SNMPtrap darkstar public 'test message' 200

(replace 'darkstar' with suitable snmp listener, described above)

-----------------
Calculating description size based on desired trap size (200 bytes) ...
  - Description size = 105 bytes
  - Actual Trap size = 200 bytes (95 bytes overhead)
-----------------
Building SNMP trap...
  - Uptime = 41287983
  - Warning level = 4
  - Description = test message (trapsize=200) test message (trapsize=200) test message (trapsize=200) test message (trapsiz
-----------------
Sending SNMP trap (size = 200 bytes) ...
  - RequestId = 679347917
  - Community = public
  - SNMP server = darkstar
Done.
-----------------
 

You should be able to observe the following entry where snmpd logs on target snmp listener:

Mar 20 08:24:37 darkstar snmptrapd[16014]: [ID 702911 daemon.warning] rbeyvm08.min.veritas.com [10.82.90.68]: Trap SNMPv2-MIB::sysUpTime = Timeticks: (41287983) 4 days, 18:41:19.83, SNMPv2-MIB::snmpTrapOID.0 = OID: SNMPv2-SMI::enterprises.1302.3.12.10.2.4, SNMPv2-SMI::enterprises.1302.3.12.10.1.3 = STRING: "test message (trapsize=200) test message (trapsize=200) test message (trapsize=200) test message (trapsiz"
 

2) Configure SNMP on the 50xx appliance via CLISH.

Show the current configuration with the 'system/snmp show' command:

rbeyvm08.system> snmp show
        SNMP PORT      : 162
        SNMP SERVER    : darkstar
        SNMP ENABLE    : 1
        SNMP COMMUNITY : public
rbeyvm08.system>
 

Change SNMP SERVER with 'system/snmp add' command:

rbeyvm08.system> snmp add somesnmptarget
Successfully updated SNMP SERVER to somesnmptarget
rbeyvm08.system> snmp show
        SNMP PORT      : 162
        SNMP SERVER    : somesnmptarget
        SNMP ENABLE    : 1
        SNMP COMMUNITY : public
 

 

3) Verify trap is generated and sent via tcpdump

Start by running tcpdump on the appliance and referencing the snmp target server. You will need to know the ethernet interface used to reach the snmp target server. This is typically eth1.

tcpdump -i eth1 host darkstar
 

On 50xx appliances, snmp traps are NOT generated in real time. Every 15 minutes cron executes the 'callhome.pl' script, which checks for issues and sends appropriate traps if found. You can simulate a failure by unplugging ONE of the two power cords from the rear of the appliance. Wait at least 15 minutes before plugging power cord back in to allow sufficient time for cron to run callhome script.

For the send_SMPtrap command, you will see output similar to this from tcpdump:

 # tcpdump -i eth1 host darkstar
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on eth2, link-type EN10MB (Ethernet), capture size 96 bytes
08:33:19.406592 IP rbeyvm08.min.veritas.com.32806 > darkstar.min.veritas.com.snmptrap:  V2Trap(183)  system.sysUpTime=41340172 [|snmp]

If all of this has been completed successfully, the trap is being generated by the appliance and sent onto the network. From there any further responsibility is in the hands of the network administrator.

 

 

Was this content helpful?

Article Languages