Checking the application status
In this demo, we will see how the seamless visibility from the network to the application enables faster root cause analysis of a simple ecommerce web application that consists of several tiers running in an infrastructure on top of the ACI fabric and APIC Controller.
With a quick glance, any application administrator can see that the application is healthy and the network conductivity between these services is also healthy
Checking the application and network status in APIC controller
On the APIC side, this view is reinforced for the network administrators who can visualize the application services mapped to the relevant ACI constructs; in this case endpoints, end point groups and application profiles along with the contract information.
As you can see from the health status, all the applications seem to work as expected.
In the next step, we will see how a contract change may result in the application failure.
Making the scheduled change
Now lets us consider a scenario when the network administrator makes a scheduled change to a contract.
Errors start to appear on the dashboard
Soon after the contract changes, there are increased errors on the application dashboard.
Network issue or an application issue?
The application administrator's first step is to determine if the issue is rooted in the application or the network. By navigating into the AppDynamics network dashboard, the application adminstrator can see if there are any network related errors impacting the application. They can also see the links between application services running on the ACI fabric turning red.
Digging a bit deeper
Inspecting the details at the failure node, the app administrator has sufficient information to determine something at the network level is impacting the performance between the order tier and the payment tier. The app admin can then cross launch into the ACI troubleshooting wizard to debug the issue further.
Visibility and troubleshooting data
The application admin can now download the PDF report detailing the relevant endpoints and associate network faults metrics and send over to the network administrator to help debug the issue.
Exporting data for offline diagnosis
A detailed report can be downloaded easily to further root cause the issue. The report consists of but not limited to end points, statistics, contracts, faults, events and audit logs with scope.
Finding the application profile with
The network administrator can now navigate to the desired Application Profile. The health score seems to be critical. With the report generated available, it will now be easy to debug the issue.
Inspecting the logs and making the necessary change
As expected and captured in the report, the contract between the respective EPGs are deleted which resulted in the application failure. The network administrator can take the necessary action in the next step.
Application back on online - APIC view
As recommended in the reprot, the network administrator rectifies the missing contract and the application is back online as indicated in the APIC controller (health score/health indicator turns green)
Application back up - AppDynamics view
The health score as seen in the AppDynamic Dashboard is green as well, indicating there are no more faults in the application and everything is back to normal.