I have deployed the log shipper using the guidelines here. However, the only logs being “shipped” are from the log shipper service itself.
The service does not seem to be picking up logs from any other service in my org even though I have set the ORG env var. I do have multiple orgs within my account. When I run fly logs for my other services, I do see logs coming through the console.
I also tried just forwarding all raw logs to stderr by deploying my own version of the shipper as follows:
Still no dice - the only logs coming through the shipper are from the shipper service itself.
Is there something I am missing, or a step I can take to debug? I have enabled the public graphql port for the shipper, though I’m not sure what to look for (vector top connects but only gives a blank screen as output).
My endpoint (HTTP_URL) was returning a 400 status code. I did not realize, however, that this failure would prevent vector from sending additional logs. I’d assumed that vector would send logs in spite of the failure condition. Once I returned a 200 status code, logs started to flow through again. The key for me was to better understand vector healthchecks.