Log_Level info reports error too

[ismailyenigul]

I use fluent-bit on EKS cluster.

even if default log_level is info,
I am getting error logs in the pod output too.

[SERVICE]
  Flush 5
  Grace 30
  Log_Level info
  Daemon off
  Parsers_File parsers.conf
  HTTP_Server On
  HTTP_Listen 0.0.0.0
  HTTP_Port 2020
  storage.path /var/fluent-bit/state/flb-storage/
  

[2023/10/30 19:48:01] [error] [filter:kubernetes:kubernetes.0] kubelet upstream connection error
[2023/10/31 10:02:16] [error] [http_client] broken connection to logs.eu-west-1.amazonaws.com:443 ?
[2023/10/31 10:03:46] [error] [output:cloudwatch_logs:cloudwatch_logs.0] Failed to send events

Is this expected behavior or bug? I will open a issue if you think that is a bug.

[patrick-stephens]

This is expected, it's normal behaviour - log level controls the minimum level that is reported, i.e. anything above info will be reported.