I develop an IT environment management system and considering using splunk for event analysis. I'd like to be able to use the network topology as an additional input for root cause analysis. For example, I'm interested to search for events from devices connected to the certain switch.
Initially, I was thinking about using a workflow and extending all events by information from the external topology repository. However, my concerns are about performance degradation and significant data volume increase in this case.
Is there is a way to to keep the topology data in splunk and to use it in the search ?
Depending on your network topo, there are a few options and ways to approach this.
One of the simplest would be to keep a cmdb that has connection mapping. In Splunk, this would be easiest in a csv (or database.) And then you can enrich your data as its searched by adding fields based on hostnames and ports in the lookup file and events.
There are other options here, depending on your current network stack. There are existing apps for Nagios, Snort etc.