RSA SecurID Appliance application for Splunk

 

This application was designed to give users usable data surrounding the activity taking place on their RSA SecurID appliances. This application will work with both the RSA SecurID Appliance 130 and 230 models. It is freely available on Splunkbase here: http://splunk-base.splunk.com/apps/33495/splunk-for-rsa-securid-appliances

Pre-deployment Assumptions:

  1. The RSA appliances are configured to send SNMP traps and allow SNMP read access using SNMPv2.
  2. The Splunk server is accepting SNMP traps and logging them to /var/log/snmptraps.log or the SNMP traps are being absorbed by Splunk in some manner and given a sourcetype name "snmptrap".
  3. The Splunk server has SNMP access to the RSA appliance.
  4. The snmpget command is installed and in your $PATH

Application Configuration:

Scripted Inputs:

For the "Network Activity" view to properly work there is a scripted input that needs to be configured. This scripted input uses the snmpget command to retrieve specific values from the device. If you have multiple devices then you need to configure multiple scripted inputs.

Follow these steps:

  1. Copy the sample inputs.conf file from $SPLUNK_HOME/etc/apps/RSASecurID/default/inputs.conf to your local folder, just so no changes are overwritten if the application is updated.
  2. Edit the inputs.conf file and change the script stanza to reflect your device configuration: [script://$SPLUNK_HOME/etc/apps/RSASecurID/bin/getSnmpData.sh public 1.1.1.1] disabled = 1 Change "public" to be the community name configured on your appliance that has read access. Change "1.1.1.1" to be the IP Address of your appliance. Change "disabled = 1" to "disabled = 0" to enable the scripted input.
  3. If you have multiple appliances, just copy/paste the [script://] stanza for as many appliances as you have and configure the appropriate values as mentioned above.

Monitored Inputs:

There is an example [monitor://] stanza in the inputs.conf file. Configure this for the proper location of the file that your SNMP traps are being logged to. If the SNMP traps are already being indexed by Splunk then this can be ignored.

Reports in this Application:

Summary View:

  • All Users Accessing the Device(s) Count of Events (5min spans)
  • Total Failed/Successful Logins (5min spans)
  • Top Ten Connecting Hosts
  • Top Ten Actions User

Activity View:

  • Successful Actions Failed Actions
  • Successful Action Reasons
  • Failed Action Reasons
  • Login Failures by User After Hours (<9am and >5pm)
  • Admin Events
  • System Level Actions
  • Runtime Level Actions
  • Admin Level Actions

Network Activity View:

  • Received KBytes by Interface
  • Transferred KBytes by Interface
  • Total Inbound Packets by Interface
  • Total Outbound Packets by Interface
  • Total TCP In/Out Segments
  • Total UDP In/Out Segments
  • Total TCP Active/Passive Connections Opened
  • Total TCP and UDP Error Counts
  • ICMP In/Out Messages
  • ICMP Inbound Echos
  • ICMP In/Out Destination Unreachables

TODO:

  1. Making the Event Search form prettier
  2. Add a correlation view to detect abnormalities in the events