I have an indexer cluster composed of 2 indexers with a master node. I have seen that my master node is actually indexing data as well as both indexers. As I don't have a lot of disk space free on my master node, I don't want it to index data. How can I prevent him from doing that ?
THis link should give you all the configuration changes required to Forwarder all data from Master node to Indexer layer (no indexing at master node)
You need to have outputs.conf file on your Indexer Master. In that file, turn off indexing and configure it to forward to your indexers. In other words, outputs.conf should be similar to the one you have on your forwarders.
Following is a sample configuration:
index = false
forwardedindex.filter.disable = true
indexAndForward = false
autoLB = true
compressed = true
server = 10.37.20.38:9997,10.37.20.28:9997
Just turn off indexing in outputs.conf
[indexAndForward] index = [true|false] * If set to true, data is indexed. * If set to false, data is not indexed. * Default depends on whether the Splunk instance is configured as a forwarder, modified by any value configured for the indexAndForward attribute in [tcpout].
And obviously, add the configurations to forward the data to indexers
do you have apps that might be setting it at the app context
find $SPLUNK_HOME/etc/ -name outputs.conf|xargs grep --color index
Check using Btool to see if this is applied in any other configuration files
./splunk btool outputs list --debug
This is Splunk's internal configuration validation tool. The debug option will show you which file the configuration is applied in.
Thanks a lot for your answer. I have checked the configuration files but I don't have any outputs.conf in my local directory and the value is already set up to false on the same file in the default directory. However, I can say that it is still indexing data as I have checked the time of warm buckets.
Do you have another idea how could I do it ?