Hi,
I set up EFK Cluster in my Kubernetes. The fluent version is 0.12. And I have a question regarding configuring log filters/matches to parse the logs.
I wish to parse logs that I collect from my services in Kuebernetes before fluentd sends them to the elasticsearch. I want to parse only logs that come from my services that are in a specific Kubernetes namespaces and not others. I found that I can apply filter only on tags and that there are Wild Cards that I can use, but, after some research I found that the tag that my logs get looks like this:
"kubernetes.var.log.containers.<name_of_pod>_<name_of_namespace>_... .log"
I wish to write a filter that will choose tags of specific namespace for example:
<fliter my_namespace>
...
</filter>
Is this possible?