bug: Prometheus unable to scrape /metrics

[astraldawn]

Describe the bug

Prometheus is unable to scrape metrics for latest (as of 90bb8f7), no issue with stable

What do you see?

In http://localhost:9090/targets, a single target with:

• State: DOWN
• Error: "INVALID" is not a valid start token

What do you expect to see?

In http://localhost:9090/targets, a single target with:

• State: UP

List the steps that must be taken to reproduce this issue

1. Run the example in https://github.com/TwiN/gatus/blob/master/.examples/docker-compose-grafana-prometheus
2. Go to http://localhost:9090/targets, the error message appears
3. Swap tag to stable
4. Error message disappears after a short while

Version

latest (90bb8f7)

Additional information

Go to localhost:8080/metrics

• latest: attempts to download metrics.gz
• stable: plaintext page with prometheus metrics

[TwiN]

Excellent catch.

This appears to be the root cause prometheus/client_golang#622

I've recently enabled gzip compression on the entire router, and it seems like because promhttp.Handler() uses its own compression, it causes an issue.