Improve monitoring of Transit Gateway attachments

[dms1981]

User Story

As a Modernisation Platform Engineer I want to receive alerts when Transit Gateway events occur So that I can resolve issues that could impact Modernisation Platform Customers

User Type(s)

Value

With active monitoring of Transit Gateway events through AWS Network Manager, we can take action to resolve issues before they are reported to us by our customers.

Questions / Assumptions / Hypothesis

Questions

Where would we send the alerts? Pagerduty? Slack

Hypothesis

At present we make use of metric-based alerts, which are only appropriate when VPC attachments are in active use. By using Network Firewall events we can monitor what we're really interested in, without the overhead of needing to generate traffic for VPCs without active customers using them.

Proposal

Look into AWS Network Manager as a way of monitoring Transit Gateway events, as these events are not - so far as I know - monitor-able directly against Transit Gateway

Definition of done

• [ ] Investigate if this can be done with Terraform
• [ ] Identify the level at which this could be implemented (eg, root account level?)
• [ ] another team member has reviewed
• [ ] Alerts potentially implemented

Reference

How to write good user stories https://docs.aws.amazon.com/vpc/latest/tgwnm/monitoring-events.html#network-topology-events

[davidkelliott]

This may also improve the transit gateway monitoring - https://aws.amazon.com/about-aws/whats-new/2022/07/amazon-vpc-flow-logs-transit-gateway-improved-visibility-monitoring/

[dms1981]

This looks to only be partially available through Terraform. I can see that it's possible to create a Networkmanager site through terraform, but I don't see a way to onboard that into CloudWatch Insights through Terraform.

• https://docs.aws.amazon.com/vpc/latest/tgwnm/monitoring-events.html

[github-actions[bot]]

This issue is stale because it has been open 90 days with no activity.

[github-actions[bot]]

This issue is stale because it has been open 90 days with no activity.

[github-actions[bot]]

This issue is stale because it has been open 90 days with no activity.