logstash-filter-csv
logstash-filter-csv copied to clipboard
limit workers to 1 when autodetect_column_names is enabled
As fixing https://github.com/logstash-plugins/logstash-filter-csv/issues/65 is a big issue requiring fundamental work on logstash to synchronize between threads, would it be possible for the interim to
- Document that when
autodetect_column_names
is truepipeline.workers
must be set to 1 - Have a startup check such that if
autodetect_column_names
is true andpipeline.workers
is >1 then startup will fail (similar to the way the aggregate filter does).
the new csv codec should be more appropriate for this, when paired with the file input a new codec instance will be used per file.