Any way to threshold the Health Checks?

We just enabled native Cloudflare Health Checks yesterday (not load balancing checks, just regular health checks). They work great! We’ve set up the emails to send to PagerDuty to create an alert every time a site goes offline. That works perfectly for prod/qa.

For develop, we actually expect a minor (~20 sec) outage when we deploy because of how things are configured. Is there any way to set a threshold of failures instead of just one failure before sending a notification? Basically, I want logic like if (health check failures > 3 in 1 minute) send notification. We’d have to pay a lot more to be able to do this on PagerDuty, so I’d like a native solution on Cloudflare, otherwise we’ll have to continue using Datadog :frowning:

Thanks!

1 Like

Thanks for the feedback. You could always change the interval of the health check to adjust this for now. But thats a good feature that could be added to future iterations of the product.

2 Likes

Unfortunately, not quite. If the server is entirely down (which is the case for ~20 sec as we swap deploys), the network response is an immediate 502 or 0. That caues retries/timeouts to be pretty useless here.

To fix this we have added two fields to the api. “consecutive_successes”: and “consecutive_fails”: . You could edit your health check with these fields. The api documentation hasn’t been updated yet but this is available right now.

Awesome!! Thanks for the help!