The MENU > TOPOLOGY page of the Backyards web interface displays the topology of services and workloads inside the mesh, and annotates it with real-time information about latency, throughput, or HTTP request failures. The topology page serves as a starting point of diagnosing problems within the mesh. Backyards is integrated with Grafana and Jaeger for easy access to in-depth monitoring and distributed traces of various services.
On the TOPOLOGY tab, you can see the service mesh control panel.
The nodes in the graph are services or workloads, while the arrows represent network connections between different services. This is based on Istio metrics retrieved from Prometheus.
The graph serves as a visual monitoring tool, as it displays various errors and metrics in the system. Click the ? icon on the left to show a legend of the graph to better understand what the icons mean in the graph.
If your application hasn’t received any traffic yet, there will be no metrics in the system so you won’t see any visualization yet. To send some traffic to your services as a test, see [Generate test load](/docs/backyards/dashboard-overview/generate-load/).
Select the data displayed 🔗︎
You can select and configure what is displayed on the graph and how often the data is updated in the top bar of the screen.
Display data only for the selected namespaces.
Set the frequency of automatic polling to show real-time metrics, or disable polling.
You can select the type of resources you want to display in the graph. The following resources can be displayed in a cluster: clusters, namespaces, apps, services, and workloads.
Workloads are always shown, they cannot be disabled.
Here’s an example when only apps, services and workloads are shown:
Showing clusters is important in multi-cloud and hybrid-cloud environments. For details, see Multi cluster.
Edge labels 🔗︎
The labels on the edges of the graph can display various real-time information about the traffic between services. You can display the following information:
- the protocol used in the communication (
- actual P95 latency,
- the current request rate (or throughput for TCP connections), or
- whether the connection is using mTLS or not.
Drill-down to the pods and nodes 🔗︎
Starting with Backyards version 1.2, you can drill-down from the MENU > TOPOLOGY page by selecting a service or a workload in the Istio service mesh. You can trace back an issue from the top-level service mesh layer by navigating deeper in the stack, and see the status and most important metrics of your Kubernetes controllers, pods, and nodes.
See Drill-down for details.