We are running Openshift Container Platform 3.4.1. It is using the RedHat provided EFK aggregrated logging solution to log to Elastic Search. We've also enabled to secure forwarder to Splunk. Splunk is our strategic logging solution so we only really want the logs to go there.
We also experience situations whereby ES has an issue, and the logs no longer get forwarded to Splunk.
I'd consider just building a standalone Fluentd solution (that sends to Splunk), but our users are now familiar with the way that aggregated logging enriches the data, and I don't think a standalone FluentD container would do that.
Does anyone know if it is possible to modify the aggregated logging so that it doesn't go to ES at all, and only uses the secure forwarder to send the data to Splunk? Or, drop the aggregated logging, and use a standard FluentD installation whilst maintaining the enrichment that aggregated logging provides?
BTW - it's only my second week with Openshift, so apologies if this is a vague or poor question :) Thanks M