Your data. Anywhere you go.

New Relic for iOS or Android


Download on the App Store    Android App on Google play


New Relic Insights App for iOS


Download on the App Store


Learn more

Close icon

Feature request: number of times synthetics monitor fails before alert triggered

feature-request
synthetics
legacy-alerts

#1

This request is a variant on some others I have seen posted.

I would like to be able to set my alerts for synthetics monitors to only trigger after X number of failures.

We have many synthetics api-scripted monitors checking our system. Generally they will fail once during a deployment and then return to passing. There is no way to set the alert to only trigger after 2 failures though, so I often receive emails for false positives that immediately resolve themselves. This becomes a problem since our developers and testers can deploy into their environments at any time, which leads to me receiving a dozen dud emails on a Saturday.


#2

Hi @megan13: By default the synthetic check will run 3 times when it receives its first failure. Then the alert will be raised.

Have you looked at adding a script into the deployment process to turn off the synthetics for the duration of the deployment and turn them back on when the deployment is finished?

Take a look at the Synthetics API documentation which would enable you to do this using your scripting/deployment tooling?


#3

By default the synthetic check will run 3 times when it receives its first failure. Then the alert will be raised.

In what duration gap does the 3 checks happen ?


#4

@arvind.subrmanian - The 3 checks happen immediately.
When the first check fails, 2 subsequent checks are scheduled immediately. There is no documented timeframe for these checks. The amount of time each check takes to run is different depending on the length of the script, how quickly the site typically loads, along with network speeds at that time the check runs. So there is no way to say exactly what that duration is between all of those checks.

That said, they are scheduled to run immediately after a failure is detected.