My Splunk forwarder is forwarding ~5 events per second (per my Deployment Monitor on my search head). The maxKBps is set to 20,000 and Splunk is transferring at a rate of 16,000 KBps. Why is the number of events so low? I have 27TB of data (and then some) that need to be indexed, these are historical log files.
I do have indexer acknowledgement enabled in my environment as well.
By adding "maxQueueSize=500MB" to the outputs.conf file, under the [tcpout] stanza.
Change the 500MB to whatever you wish. It defaults to 500kb I believe. But be careful, because whatever you set this number to, it will consume 3x more RAM. So if you have it set to 500MB, it'll consume 1500MB of RAM.
Please this for more info (search for maxQueueSize):
http://docs.splunk.com/Documentation/Splunk/5.0.2/Admin/Outputsconf
By adding "maxQueueSize=500MB" to the outputs.conf file, under the [tcpout] stanza.
Change the 500MB to whatever you wish. It defaults to 500kb I believe. But be careful, because whatever you set this number to, it will consume 3x more RAM. So if you have it set to 500MB, it'll consume 1500MB of RAM.
Please this for more info (search for maxQueueSize):
http://docs.splunk.com/Documentation/Splunk/5.0.2/Admin/Outputsconf