Currently working on deploying Splunk on AWS to work in conjunction with our current on-prem solution and I have 2 questions.
Can I configure our AWS Search heads to function as normal Search Heads AND as search peers for our on-prem solution? Or would I need dedicated search peers?
And would I be able to place the Search peers behind a NLB and point the on-prem distconf file to that NLB? Or would I have to hardcode the instances in the distconf file?
The AWS search heads can service the on-prem system, not as search peers, but as Federated Search (FS) providers. FS allows one Splunk environment (on-prem, in this example) to query another (AWS) and include those results as part of a local search. You can read more about FS at https://docs.splunk.com/Documentation/Splunk/latest/FederatedSearch/fsoptions
Never put a load balancer in a network path that uses the Splunk-to-Splunk protocol. LBs don't know that protocol and can't be relied on to manage the connections correctly. Put all of the search peers in the servers= line of distsearch.conf or use Indexer Discovery.
Thank you!
The AWS search heads can service the on-prem system, not as search peers, but as Federated Search (FS) providers. FS allows one Splunk environment (on-prem, in this example) to query another (AWS) and include those results as part of a local search. You can read more about FS at https://docs.splunk.com/Documentation/Splunk/latest/FederatedSearch/fsoptions
Never put a load balancer in a network path that uses the Splunk-to-Splunk protocol. LBs don't know that protocol and can't be relied on to manage the connections correctly. Put all of the search peers in the servers= line of distsearch.conf or use Indexer Discovery.