Comment on page
StackState Self-hosted v4.6.x
Hi! So, you've just installed StackState and you are ready to get started.
StackState visualizes components in the Topology Perspective by the layer and domain that they are placed in. These are logical groupings of components. Layers are displayed on the vertical axis. Domains are displayed on the horizontal axis.
A topology consists of components and relations combined with their health state. Because topologies can get very large, StackState automatically groups the components.
The health state of a component is indicated by two colors:
- The inner color of a component indicates the health state that is calculated for the component itself.
- The outer color indicates the potential impact of components that this component depends on.
The direction of a relation's arrow indicates dependency. For example,
app -> dbmeans:
db. Health propagates in the opposite direction. So if the
dbcomponent turns red, the
appcomponent will turn red too.
The timeline at the bottom of the screen gives you the ability to go to any point in time. All the information that you see (component details, metric streams, etc.) is relative to the topology that existed at the currently selected topology time. Normally, StackState is in live mode, this means that StackState automatically displays the latest state of the stack.
To see the details of a component or a relation, you can click on it. Click on SHOW ALL PROPERTIES to see all details of a component.
Both components and relations can have one or multiple telemetry streams. The most common type is a metric stream also known as time series. If you click on a metric stream, you can see the metric stream in a popup.
If you click on a log stream, you can see the log stream in a popup. Again, there are a number of drill-down capabilities available on the left-hand side of the popup.
To quickly find the cause of any deviating component, head to the right-hand pane in the selected view where you can find the Problems section of the sidebar. It reduces the cognitive flow and provides immediate understanding of ongoing problems. The Problems section pane will show you the cause of problems in your view. This summary is based on the components impacted in your current view combined with all (potential) causes. Problems are automatically clustered by their root cause.
Problems and issues are displayed in the following order order:
- StackState prioritizes problems in order of creation date/time, with the oldest problem at the top of the list and the most recently created at the bottom.
- Component-specific issues grouped inside problems are displayed from the most recent ones at the top to the oldest at the bottom of the list.
Not all components displayed in the problem pane are necessarily also visible in the current topology view. To make the root cause of a problem visible, hover a mouse pointer over the component and select
show -> root causefrom the
To show all events for the selected Topology, select the Events Perspective from the top of the screen. Examples of important events that may appear here are health state changes and changes to the components themselves, like version changes. With event handlers, you can configure StackState to react to any events, for example, by automatically creating a ticket or triggering some automation.
The Events Perspective is not the only place you can find events; you can find the latest events in the Events section of the View Details pane on the right of the screen.