Dependency Graph in Kosal - Root Cause Analysis¶
Overview¶
In order to calculate root cause accurately, CloudAEye leverages application information such as dependency graph
. You may set up the dependency graph
one time for each application for the root cause analysis to work correctly.
Prerequisites¶
- Quick Start Guide is completed
Setting Up Dependency Graph¶
From the left navigation, select Dependency Graph
. You will see a page that allows you to configure the dependency graph
.
Define Graph in Editor¶
You may provide the dependency graph for your application manually by typing in the JSON on the left side of the screen. You may see the dependency graph on the right side when you click on the Preview
button. Click Submit
to save the dependency graph definition when everything looks correct.
Collect from Datadog¶
You may start by collecting the dependency graph from Datadog. After that, you may edit the JSON to correct any mistakes before saving the JSON for your application.
Collect via AI Analysis of Uploaded Application Code¶
If you prefer to setup the dependency graph
from your code directly, you may do so by uploading one or more zip file of your application code. Click on Analyze Code
button to begin analysis.
After the code analysis, you will see a detailed sequence diagram of your application.
If you see any mistakes or missing information, you may type that in the chat prompt and ask for generating the diagram again. Please provide specific instructions on what is missing for AI analysis to correct the mistakes. After few iterations, when the sequence diagram looks correct, you may click on the Confirm Graph
button to convert the Plant UML
code to JSON format. Click Submit
to save the dependency graph
definition.
Collect from Istio + Kiali & Upload¶
If you are using Istio and Kiali, you may provide us the graph JSON file using the following steps.
Step 1:¶
From the Kiali UI, click the help icon at the top in the masthead (the circle with the question mark in it) to get a dropdown menu. Click "View Debug Info" found in that dropdown menu.
Step 2:¶
After that a popup window will appear. In that popup window, select the "Additional State" tab. Click the "Download" button to download the JSON file.
Step 3:¶
Locate the JSON file you downloaded from Kiali UI and upload to the CloudAEye screen. Click on Upload & Analyze
to convert the JSON to CloudAEye format. Click Submit
to save the dependency graph
definition.