Hello everyone
I need help to send information to 2 indexers.
By request of the client I need to send information from a heavy forwarder to an indexer A, if indexer A goes down the information must reach indexer B and when indexer A is back online the heavy forwarder must send the information back to the indexer A.
The priority must be in indexer A
I need to know how to tell the heavy forwarder that when indexer A is online it sends the data to indexer A.
Thank you
Hi Edgar.
The Splunk have some features to balance or distribute data indexing over multiples indexers and Datacenters, but Splunk dont have a way to make what you need natively.
I think that you can provide this feature using software development, based on type of data ingestion that you need.
If you need to create a copy of data in indexer B without a Indexer Cluster, you must to configure it on outputs.conf file in your Heavy Forwarder,
[tcpout]
defaultGroup = indexerA, indexerB
[tcpout:indexerA]
server = indexerAIP:9997
[tcpout:indexerB]
server = indexerBIP:9997
With this configuration, all data sent to IndexerA, will be sent to IndexerB too.
You're kind of replicating the main feature of indexer clustering. If you put indexer A and indexer B into a cluster then they'll each have a copy of all data regardless of where the data was sent. Also, a search will find the data in either indexer.
It's possible to tell a HF to send data to two indexers, but it's difficult (impossible?) to tell it to only use B if A is unavailable. Indeed, if both indexers are not available a forwarder will refuse to send to either one.