Definitely not an issue with your app, it's an issue with the V2 runtime, unfortunately I do not have an update besides the one I gave in the thread you linked.

You may switch back to the legacy runtime in your service settings until this is resolved.

We're tracking this ticket internally and hope to have the issue resolved soon. Thanks both

Thanks @brody and @melissa. I have switched to the legacy runtime and it started showing more logs but still missing quite a bit of logs.

For example, I have a medusa scheduled job that runs every 1 hour. It prints number of items it processed in that run. For the last 1 day it has correctly run 24 times and I can see that based on the db updates. But it printed the logs only 3 times (screenshot attached).

Is there some scenarios where the entire log will not be captured? Can we force it to capture all the logs?

The legacy runtime has no known issues with logs, as long as you are printing to stdout/stderr unbuffered they will print.

That's strange. The scheduled job in the attached screenshot prints using console.log. The job runs every hour but as you see in the screenshot I get the log 3-4 times a day.

We would recommend you definitely use the cron for anything like this. Jobs that run using application level stuff, we can't debug. You'll want to check with the Template author to see if there's any issues here

Sorry for the late reply, but all known logging issues on the v2 runtime have been fixed.