Help with nested json parser for fluentd + kubernetes collector

Hi,

I'm quite new to using sumologic collector with kubernetes. I'm trying to unpack a log attribute that contains a json message.

I used the code below to configure my elasticsearch with kibana, however I'm not sure how to make similar changes for sumologic.

       <filter **>
          @type parser
          <parse>
            @type json
            json_parser json
          </parse>
          replace_invalid_sequence true
          emit_invalid_record_to_error false
          key_name message
        </filter>

        <filter **>
          @type parser
          <parse>
            @type json
            json_parser json
            keep_time_key true
          </parse>
          replace_invalid_sequence true
          emit_invalid_record_to_error false
          key_name log
        </filter>

This is a sample of the message column displayed in sumologic

{
timestamp: 1661346107025,
log : "2022-08-24T13:01:47.001331339Z stdout F {\"time\":\"2022-08-24T09:01:46.989-04:00\",\"msg\":\"An error has happened.\",\"X-REQ-SESSION-ID\":\"d8ed7bd4-5bfc-4eb4-b1da-4c4d35acbcd2\",\"traceId\":\"-\",\"spanId\":\"-\",\"parentSpanId\":\"-\",\"clientId\":\"-\",\"reqSessionId\":\"d8ed7bd4-5bfc-4eb4-b1da-4c4d35acbcd2\",\"userSessionId\":\"-\"}",
stream: null,
time: null
}

The only information that I care about is what is inside {\"time ... }

Is there a simple way to configure sumologic to unwrap that portion as the real message instead of the standard one above?

I appreciate any help.

Best,
Eduardo