Is your feature request related to a problem? Please describe.
In the context of the Kubernetes Filter, I'd like to be able to suggest multiple parsers through the annotation. The idea is that logs could be formatted in different ways according to the context, and making one parser to handle the different cases can be hard.
Describe the solution you'd like
Each parser could be seperated by a comma like so :
fluentbit.io/parser_stderr: "parser-one,parser-two,..."
fluentbit.io/parser_stdout: "parser-three,parser-four,..."
Then we would iterate on the parsers, and use the first one that has a match.
Describe alternatives you've considered
I tried using the Parser Filter, but then we lose the Kubernetes Metadata.
Additional context
This feature was already suggested through #5526 and #8173. I am willing to work on the MR :)
Is your feature request related to a problem? Please describe.
In the context of the Kubernetes Filter, I'd like to be able to suggest multiple parsers through the annotation. The idea is that logs could be formatted in different ways according to the context, and making one parser to handle the different cases can be hard.
Describe the solution you'd like
Each parser could be seperated by a comma like so :
Then we would iterate on the parsers, and use the first one that has a match.
Describe alternatives you've considered
I tried using the
ParserFilter, but then we lose the Kubernetes Metadata.Additional context
This feature was already suggested through #5526 and #8173. I am willing to work on the MR :)