Http.max_content_length update on cluster

Hello, Alex.
We pick this value because i find Encountered a retryable error (will retry with exponential backoff) {:code=>413,

You suggest to leave this value by default. And look in the direction of reducing the input data from logstash.
Because in the future it will affect the performance of the cluster?

Also set this value. Because the data did not go to Elasticsearch. After restarting logstash, the data was uploaded correctly.

And this setting «http.max_content_length: 200mb» should be specified on all nodes of the cluster (master, data) or only on those where logstash output looks?