Application Insight Logs - Accessing & Troubleshooting

The Atlas back-end infrastructure is housed in Azure within a dedicated Atlas Resource Group created during your Atlas Deployment, where a series of function apps and application insights are kept. These generate a set of logs during use, execution and provisioning of Atlas workspaces and related functionality.

These logs are crucial to detecting, diagnosing and troubleshooting any issues or exceptions encountered. Once fetched and sent to ClearPeople our tehcnical team will be able to investigate the issue in order to find a resolution, for example if a workspace is failing to provision after a retry, as well as Application Performance Issues. 

NOTE:  It is recommended that you be Global Administrator in the Tenant and Owner of the subscription and Resource Group. You will need sufficient permissions to be able to view, access and edit information.

NOTE: The article describes how to retrieve the wf logs, but depending on the error, you may be asked to retrieve a different set of logs, following the same process. Your CSA or ClearPeople contact will confirm which logs are required. 3 common examples are:

AppInsights _API logs	Parts of a Web part are not displayed correctly or an error doing an action
AppInsights _WF logs	Provisioning error
AppInsights _tpf logs	Template creation error from Connex studio

To obtain the logs:

Open your Microsoft Portal in your Microsoft Azure Tenant where your Atlas Resource Group was deployed (rg-atlas-xxxxxx-xxxx) in your browser Azure portal.  Once you have located the Atlas resource group, select it.

Locate the Application Insight Resources within the Resource Group.  These will be clearly identified by the Type Application Insights.

From the list of Application Insights select the one with the name that begins with:  appi-atlas-wf-XXXXXX-XXXX (has wf in the name).  Select this App Insight resource.

From the overview page of the App Insight, click on the Search.

We now need to narrow down the search criteria to locate the corresponding values that are required.

First we look at the time the error took place.

Access the Local Time filter (default value is last 24 hours).  If the issue you are experiencing occurred in the last 24 hours, leave this value.  If not, you can define your own time range or setup a custom time range.  You can also select to show the time in Local time or UTC.   When these settings are configured click on Apply.

Then we narrow down the event types.

We can filter by any Event type we require, however initially, we filter by Exceptions.  Once this filter has been applied, click on View in Logs.

The Log view will indicate the Kusto Query Language (KQL) with the filters that were applied.  These can be edited and re-run if you are familiar with KQL.

If the number of exceptions in the Time Period that we have set in the query, is greater than 100, we will need to modify this query so that ALL the exceptions are shown.  For this, delete the “| take 100 “ from the end of the query and re-run the query.

From the Export button at the top of the Query editor, select Export to CSV – All Columns.

This will generate a query_data.csv file that will be automatically downloaded to the Downloads Folder.  You can rename the file with Exceptions to help identify it.

IMPORTANT:  Follow the above steps and generate another csv file with ALL event types, for the same period of time.  Generate the csv file and rename it ALL Events. 

REMEMBER: that you will also have to delete the “| take 100 “ from the end of this query and re-run the instruction as the number of event with ALL will normally exceed 100.

Forward these both these logs to your ClearPeople representative.