Which point? I can't reproduce your error message.

% fluentd -c multi_file_buffer.conf
2020-11-17 19:48:40 +0900 [info]: parsing config file is succeeded path="multi_file_buffer.conf"
2020-11-17 19:48:40 +0900 [info]: gem 'fluent-plugin-prometheus' version '1.8.1'
2020-11-17 19:48:40 +0900 [info]: gem 'fluent-plugin-record-modifier' version '2.0.1'
2020-11-17 19:48:40 +0900 [info]: gem 'fluent-plugin-rewrite-tag-filter' version '2.3.0'
2020-11-17 19:48:40 +0900 [info]: gem 'fluentd' version '1.11.4'
2020-11-17 19:48:40 +0900 [info]: [out_fwd] adding forwarding server 'localhost:24224' host="localhost" port=24224 weight=60 plugin_id="out_fwd"
2020-11-17 19:48:40 +0900 [info]: using configuration file: <ROOT>
  <system>
    workers 2
    root_dir "/tmp/fluentd"
  </system>
  <match pattern>
    @type forward
    @id out_fwd
    <service>
      host "localhost"
      port 24224
    </service>
    <buffer>
      @type "file"
    </buffer>
  </match>
</ROOT>
2020-11-17 19:48:40 +0900 [info]: starting fluentd-1.11.4 pid=94982 ruby="2.6.3"
2020-11-17 19:48:40 +0900 [info]: spawn command to main:  cmdline=["/path/to/ruby", "-Eascii-8bit:ascii-8bit", "/path/to/fluentd", "-c", "multi_file_buffer.conf", "--under-supervisor"]
2020-11-17 19:48:41 +0900 [info]: adding match pattern="pattern" type="forward"
2020-11-17 19:48:41 +0900 [info]: #0 [out_fwd] adding forwarding server 'localhost:24224' host="localhost" port=24224 weight=60 plugin_id="out_fwd"
2020-11-17 19:48:41 +0900 [info]: #1 [out_fwd] adding forwarding server 'localhost:24224' host="localhost" port=24224 weight=60 plugin_id="out_fwd"
2020-11-17 19:48:41 +0900 [info]: #0 starting fluentd worker pid=94997 ppid=94982 worker=0
2020-11-17 19:48:41 +0900 [info]: #1 starting fluentd worker pid=94998 ppid=94982 worker=1
2020-11-17 19:48:41 +0900 [info]: #0 fluentd worker is now running worker=0
2020-11-17 19:48:41 +0900 [info]: #1 fluentd worker is now running worker=1

This configuration leads to the above-mentioned error message with fluentd 1.11.5, but uncommenting the path parameter fixes it. Any clues?

<system>
      root_dir /var/log/fluentd
      workers 2
</system>

# Skipped

<match>
        @type copy
        @id internal_out

        <store>
          @type elasticsearch
          host "#{ENV['ES_HOST']}"
          port 9200
          scheme https
          user "#{ENV['ES_USER']}"
          password "#{ENV['ES_PASSWORD']}"
          logstash_format true
          logstash_prefix internal
          logstash_dateformat %Y%m%d
          include_tag_key true
          include_timestamp true
          time_key time
          type_name _doc
          reconnect_on_error true
          tag_key @log_name
          prefer_oj_serializer true
          reconnect_on_error true
          reload_on_failure true
          reload_connections false
          verify_es_version_at_startup false
          default_elasticsearch_version 7
          ssl_version TLSv1_2
          suppress_type_name true
          request_timeout 60s
          <buffer>
            @type file
            # path /var/log/fluentd/kubernetes.internal.system.buffer
            flush_mode interval
            retry_type exponential_backoff
            flush_thread_count 6
            flush_interval 5s
            retry_forever
            retry_max_interval 30
            retry_randomize true
            total_limit_size 5GB
            overflow_action "block"
          </buffer>
        </store>        
      </match>
    </label>

You forgot to add @id to elasticsearch setting.

root_dir/@id parameter
These parameters must be specified when you use the file buffer.

Are you saying I need @id in both <match> and <store> definitions?