Prevent auto recovery


I have a problem when an incident is created, a recovery message is automatically sent 10 minutes later and the incident closes, while I do not want this to happen and the incident stays open until I close it manually.

Please Advice


Hey @iman1

Can you share a link to an example alert incident?

Hi Thank you for responding.
I have set 500 Error on VictorOps.I received three Notification After Incident Is created.

New Relic: Provider . Http Code Target 5 XX Count . Sum > 4 at least once in 5 minutes on 'arn:aws:elasticloadbalancing:us-west-2:xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

I received recovery notification from newrelic and the incident was closed.

Thanks @iman1

I’m hoping to take a look at the alert policy - do you have a direct link to that policy in your New Relic Account?

Hey @iman1

If you look at the time period for that alert violation you’ll see that very quickly the value you are alerting against comes below your threshold of 4. Meaning that the violation is then resolved, and the incident is closed.

You can configure auto-closing violation timelines, that is, auto-closing the violation after a certain time, even if the alert is still in violation.

You can not stop the violation from closing in cases where the threshold is no longer being breached.

See this query in your account to see the value coming below 4

Hi ,
My query is
SELECT average(provider.httpCodeTarget5XXCount.Sum) FROM LoadBalancerSample where provider.loadBalancerArn = ‘arn:aws:elasticloadbalancing:us-west-2:xxxxxxxxxxxx:loadbalancer/app/ce461192-xxxxxxxxxxxxx/xxxxxxxxxxxxxx’ FACET SINCE 30 MINUTES AGO TIMESERIES

would you help me for repair it


Hi @iman1

I don’t see an issue with that query - can you clarify what in there you are looking to repair?