fluentforwardreceiver : logs cannot be received due to wrong timestamp type

[Bhogayata-Keval]

Bug Description
Fluent Forward Receiver cannot receive logs in case the timestamp follows uint64

Steps to reproduce

Case 1 : If you use "fluent-logger" npm package and try to send logs directly to fluentreceiver endpoint (Ex. localhost:8006)

receivers:
  fluentforward:
    endpoint: localhost:8006

const FluentClient = require("@fluent-org/logger").FluentClient;
const logger = new FluentClient("tag_prefix", {
  socket: {
    host: "localhost",
    port: 8006,
    timeout: 3000, // 3 seconds
  }
});

// send an event record with 'tag.label'
logger.emit('label', {record: 'this is a log'});

This setup does not work unless I add a uint64 case in conversion.go > timeFromTimestamp

func timeFromTimestamp(ts interface{}) (time.Time, error) {
	switch v := ts.(type) {
	case uint64:
		fmt.Println("---- this case needs to be added !")
		return time.Unix(int64(v), 0), nil
	case int64:
		return time.Unix(v, 0), nil
	case *eventTimeExt:
		return time.Time(*v), nil
	default:
		return time.Time{}, fmt.Errorf("unknown type of value: %v", ts)
	}
}

Case 2 : If you use "fluent" protocol for both input & output configuration

[SERVICE]
    Flush         5
    Daemon        off
    
[INPUT]
    Name    forward
    Host    localhost
    Port    24224
    
[OUTPUT]
    Name   stdout
    Match  *

[OUTPUT]
    Name    forward
    Host    localhost
    Port    8006

Same thing happens as case 1 : logs does not pass unless you add a case to handle uint64 timestamp !!

[github-actions]

Pinging code owners: @dmitryax

[benewen-brav]

I'm also experiencing this issue on otel/opentelemetry-collector-contrib:0.58.0

Can also confirm that the fix above works for me.

@Bhogayata-Keval I noticed your SO post describing the issue also: https://stackoverflow.com/questions/72709477/can-i-use-fluent-logger-with-otel-fluentforwardreceiver

Glad to see you found it possible to directly use the fluentforward receiver as the log driver for your container. However, I couldn't get this to work with older versions of docker, <=18 seem unsupported (a constraint of one of our current environments, rancher 1.6).

Therefore I am keen to see a fix to use fluentbit as an intermediary. I might read up the contribution steps for this repo when I have the time.

[dmitryax]

@Bhogayata-Keval @benewen-brav since we have the solution suggested already, does any of you want to submit a PR?

[github-actions]

This issue has been inactive for 60 days. It will be closed in 60 days if there is no activity. To ping code owners by adding a component label, see Adding Labels via Comments, or if you are unsure of which component this issue relates to, please ping @open-telemetry/collector-contrib-triagers. If this issue is still relevant, please ping the code owners or leave a comment explaining why it is still relevant. Otherwise, please close it.

Pinging code owners:

• receiver/fluentforward: @dmitryax

See Adding Labels via Comments if you do not have permissions to add labels yourself.