Best practice and configuration when using Amazon Managed Prometheus

[Deleted user]

Afternoon all I hope you can help?

We have subscribed to the Premium Jira Service Management plan.

Ive setup a basic monitoring platform using Amazon Managed Prometheus and Amazon Managed Grafana.  In Amazon Managed Prometheus we have AlertManager outputting alerts to SNS as this is the only receiver currently supported by Amazon Managed Prometheus.

Can you advise on a sensible way to configure Jira Service Management and OpsGenie? Im struggling with finding documentation that can advise a process/flow.

We are a little confused as to how it is laid out/configured.

We want to be able to create issues into Jira based on severity level labels passed from AlertManager. Any alerts that are P1 or P2 are automatically sent to OpsGenie to alert the on call team. If the alert is P3 or P4 its to create a Jira ticket and alert the relevant team between office hours. P5 informational can just go directly to slack for the team to be aware of a small issue thats not service impacting.

We also have a Jira portal where issues can be created manually. Again if its a P1/P2 its sent to OpsGenie to alert the relevant teams.

Do we need to create the project in Jira then in OpsGenie create our teams.

Under each OpsGenie team do we then configure the integration with Jira Service Management  and configure rules which send traffic to destinations based on alert level?

Also under each team in OpsGenie I believe we should subscribe to the SNS topic that AlertManager writes into allowing OpsGenie to see these topic events.

So the rules/flow would be something like this I think?

Web Service Portal > Jira Service Management > OpsGenie integration > P1/P2 alert via OpsGenie, create Jira incident ticket, slack channel etc

Web Service Portal > Jira Service Management > OpsGenie Integration > P3/P4 > create Jira Incident Ticket for investigation during office hours. No on call alerts raised.

AlertManager > SNS Topic > OpsGenie > Jira Service Management Integration > P1/P2 alert via OpsGenie, create incident Jira ticket and slack channel.

AlertManager > SNS Topic >  OpsGenie  > Jira Service Management Integration > P3/P4 create Jira ticket and send to relevant team to investigate during office hours. No on call alerts raised.

AlertManager > SNS Topic > OpsGenie > Jira Service Management integration > P5 information sent to team via slack to investigate during office hours only. No on call alerts raised.

Were just a bit confused at the moment and would appreciate some guidance - especially if you have used Jira Service Management with Amazon Managed Prometheus in the past.

Cheers

Andrew