HTTP Monitor - how to retry on error?

[masterman]

Hello,

I created an HTTP monitor which runs every 5 minutes. The monitor also has a performance threshold and it will create a problem when 3 out of 5 recent executions passes the threshold.

However, this can potentially be 25 minutes before the problem is detected. As far as I know, there is no way to retry as with Browser monitor. Does anyone know a workaround 

[AntonPineiro]

Hi,

This looks like a product idea.

Best regards

[HannahM · Accepted solution]

The purpose of HTTP Monitors is to check the availability of API endpoints or perform simple HTTP checks for single-resource availability. As such, retry on error has been considered and decided against as it would mask network or other errors that teams need to know about and resolve. 

Now that we have workflows, you could create a workflow to run an on-demand execution if the HTTP Monitor failed

[marco_irmer]

Since this is an HTTP monitor that presumably checks availability of a single resource, might it make sense to create a problem after a single failure? Or are the scenarios where a single failure is considered tolerable?

[shilpi_mukherje]

I have a scenario where DNS resolution is intermittently not available for fractions of seconds. It is a tolerable condition, since we are monitoring 220+ application domains using HTTP monitor every 15 mins from 10 locations (average polling is too frequent) but due to monitor failed runs, synthetic monitoring is showing degraded availability of the application monitors (<99.9% where it could have been 100% if retry is allowed)