Transport closes before logs are shipped

[yaron-idan]

I'm trying to use this transport in an AWS Lambda function and I'm finding that the function terminates before the logs are shipped to Loki.
I've noticed this issue in Winston's main repository and tried to implement this workaround in order to force the function to wait for the logger to finish shipping the logs but it does not wait for the requests to be sent before resolving the promise and closing the function.

I suspect it has something to do with this line which immediately emits a logged event to winston instead of waiting for the log record to be pushed, but honestly I feel I'm not familiar enough with Winston to be sure this is exactly what's happening.

Could emitting the logged event further down the line help with the scenario I've presented?

[JaniAnttonen]

Nice find. Might want to take a closer look at the events winston listens to. I'll come back at you, but you're welcome to submit a solution if you find one :)

[JaniAnttonen]

It is a bit weird though, if the logged event firing is the culprit. The index.js structure is based on https://github.com/winstonjs/winston/blob/bc6a363ec9164d03f266cab7c918d92208d2ec73/examples/custom-transport.js, that's why the logged event gets fired before the transport. Are you using batching or sending out logs as they come? Does changing that option change things?

[JaniAnttonen]

#51 might have fixed this?

[zabil]

Hi, We see this issue where the logs are not completely shipped.

For reference here is the setup.

  logger = winston.createLogger({
    level,
    transports: [
      new LokiTransport({
        host: lokiHost,
        batching: false,
        format: json(),
        onConnectionError: (err) => console.error(err),
      }),
    ]
  });

 export default logger;

Usage in lambda

import log from "logger";

export const handler = async (event: S3ObjectEvent) => {
  log.info("Received S3 Notification event");
  // And other log events
}

We have noticed that not all the log statements are being sent, especially the ones at the end. We think this happens because the handler shuts down after the response.

Version info

    "node_modules/winston-loki": {
      "version": "6.0.6",
      "resolved": "https://registry.npmjs.org/winston-loki/-/winston-loki-6.0.6.tgz",
      ...

[rlfrahm]

I'm encountering this issue as well. I'm using this for scripts run by ts-node, so my workaround has been to wait for 5 seconds at the end of the script before process.exit(0). I'm not sure what to do for an API endpoint for instance yet. I see there are some options to pass in to the LokiTransport constructor like timeout and gracefulShutdown, as well as flush() available from the class, but I've not tested a combo that solves this issue.