Monday 26 March 2018

Building an Edge Server Port Monitor with Azure Function Apps – Part 3

This blog is an expansion on the previous Part 1 and Part 2 posts found here and here. The process of setting up the Function App for this part 3 section is the same as was documented in Part 1 and 2.  I suggest you head over to the first two parts and give them a good read before moving onto this post.

In part 3 we will be adding to the Function App so it can save data over time that we can use to graph and manipulate in Power Bi. To do this we will expand the application to do the following:
  • Save all port test results to Azure Storage tables for analysis.
  • Use Power Bi to connect to Azure Storage Tables and create nice graphs showing the status of Edge servers over a period of time.

Just like in Part 2 we will use the Azure Storage Tables module for Powershell to allow our application to keep both a short term memory for logging errors as well as a long term memory for logging all port testing attempts.

Step 1
To begin with follow Steps 1 through 5 of Part 2.

Step 2 – Download a copy of the script
You can grab a copy of the script I wrote for part 3 from here:

Step 3 - Update variables

Update the Storage Account details in the Powershell Script.

IMPORTANT: These settings are slightly different than in Part 2. In this case we have 2 different storage tables: one of them stores the current state of edge servers (same as part 2) and the other one stores all attempts into a separate table (which we will use for analysis in Power Bi):
$subscriptionName = "Visual Studio Premium with MSDN"  #SUBSCRIPTION NAME
$resourceGroup = "EdgePortTester-Part003"      #RESOURCE GROUP
$storageAccount = "edgeporttesterp8dbf" #STORAGE ACCOUNT NAME
$tableName = "EdgeTesterTablePart3Current"      #CHOOSE A NAME 1
$tableName2 = "EdgeTesterTablePart3Results"      #CHOOSE A NAME 2
$partitionKey = "EdgeTesterStoragePart3Current"      #CHOOSE A NAME 1
$partitionKey2 = "EdgeTesterStoragePart3Results"      #CHOOSE A NAME 2
$storageAccountKey = "7asdkjhasd7KHDKJHAS0dsflasdnnlasd099asdpncsdlknclLJSDLjbadksdjbfa9su9duhoasivRqXA615jQ=="             #STORAGE ACCOUNT > ACCESS KEYS

Don’t forget to fill in your Mail Jet email account information (as you did in Part 1) and add your Skype for Business Edge server's details. See Part 1 for more details. Enter your Mail Jet API Key (Username) and Secret Key (Password) and paste them into the following section of the script:

$emailUsername = "kjh3k23h4kjhkj37573f8f020879dff7"     
$emailPassword = "9898f98fhdjkkdjh46cd418100075a3b"

Edit the Skype for Business Edge server details as required. These are entered as an array of hash tables. The sections highlighted in yellow can be changed. In this case the application is monitoring 2 Edge servers, one in Melbourne and one in Sydney.

Access Edge
Web Conferencing
AV Edge
Access Edge
Web Conferencing
AV Edge
Note: The script only supports testing TCP ports at this time.

$Records = @(@{"Location" = "Melbourne"; "ServerName" = ""; "ServerRole" = "Federation"; "DestinationPort" = "5061"; "Protocol" = "TCP"})
$Records += @(@{"Location" = "Melbourne"; "ServerName" = ""; "ServerRole" = "Access Edge"; "DestinationPort" = "443"; "Protocol" = "TCP"})
$Records += @(@{"Location" = "Melbourne"; "ServerName" = ""; "ServerRole" = "Web Conferencing"; "DestinationPort" = "443"; "Protocol" = "TCP"})
$Records += @(@{"Location" = "Melbourne"; "ServerName" = ""; "ServerRole" = "AV Edge"; "DestinationPort" = "443"; "Protocol" = "TCP"})

$Records += @(@{"Location" = "Sydney"; "ServerName" = ""; "ServerRole" = "Federation"; "DestinationPort" = "5061"; "Protocol" = "TCP"})
$Records += @(@{"Location" = "Sydney"; "ServerName" = ""; "ServerRole" = "Access Edge"; "DestinationPort" = "443"; "Protocol" = "TCP"})
$Records += @(@{"Location" = "Sydney"; "ServerName" = ""; "ServerRole" = "Web Conferencing"; "DestinationPort" = "443"; "Protocol" = "TCP"})
$Records += @(@{"Location" = "Sydney"; "ServerName" = ""; "ServerRole" = "AV Edge"; "DestinationPort" = "443"; "Protocol" = "TCP"})

Step 4 – Parameter Tweaking
This version of the script like part 2 has a few settings that you can tweak. These are how many failures on each port is required before an email gets sent ($RequiredNumberOfFailuresBeforeEmail). There is also a setting for consolidating multiple errors or recoveries into a single email ($consolidateEmailsOnError and $consolidateEmailsOnError). Set these as you like:

#This is the number of required port check failures before an email is sent out
$RequiredNumberOfFailuresBeforeEmail = 3

#Send 1 email rather than one per record
$consolidateEmailsOnError = $true
$consolidateEmailsOnRecover = $true

Step 5 - Download Azure Storage Explorer
Now let your Function Application run for a while and gather some data. At any time you can look into your Function Apps Table Storage using Azure Storage Explorer. This application will show you all of the rows in your storage tables and allow you to see and edit as you see fit. You can download your free copy from here:

Once you have logged into your Azure Account within Azure Storage Explorer you can dig into your storage tables by selecting Storage Accounts > (Storage Resource Group Name) > Tables to see your table data. Note, there will be no table or data until you actually start running the Function App.

Step 6 - Download Power Bi
Now download a copy of Power BI for desktop:

Install the downloaded Power Bi on your PC.

Step 7 - Open Power Bi
Open Power Bi Desktop and you will be greeted with a splash screen and dialog. Click on the “Get Data” button:

Step 8 - Import Data
The Get Data dialog will then be displayed. Select “Azure Table Storage” from the list and click the “Connect” button:

Step 8 - Account URL dialog
Power Bi will now request an Account Name or URL to connect to:

Step 9 - Find Account URL
The account name for the dialog above can be found in the Azure Portal under the storage account Overview > Tables section:

The “Table service endpoint” is the value you will need to fill in the dialog with:

Step 10 - Paste in URL
Enter the Table service endpoint into the “Azure Table Storage” dialog and click “OK”:

Step 11 - Enter Account Key
You will now be asked to enter your “Account Key”:

This can be found under the Storage Account > Access Keys section in the Azure Portal:

Enter the Account Key and click “Connect”:

Step 12 - Load Data
Power Bi will now connect to your Table Storage and list up all of the Tables in there. Select the Results table and click “Load”:

Step 13 - Not all data is displayed
Power Bi will now download your data into the application. You may notice though that all of the columns that you can see in Azure Storage Explorer will not be displayed:

Step 14 - Edit Query
To be able to see all of the columns you need to do a little extra work. On the right hand side of the screen, Right Click on Table name at the top of the top of the column names listed and select “Edit Query”:

Step 15 - Expand content column
You will now see an extended view of the data that includes a “Content” Column:

On the top right of the Column click on the double arrow “expand” button:

You will now get a full list of all of the additional columns available that are stored in Table Storage as a Json blob. Tick the Columns that you want to include in your graphs and data analysis and click OK:

Step 16 - All columns are now available
You will now see the extra data columns:

 Click the “Close and Apply” button from the Home Tab:

The full array of data is now available for you to do as you please with:

Step 18 - Make charts
Back on the "Report" tab you can now put together some nice looking graphs of your data. Here is an example of a Pie Chart and a Bar Chart showing information about the number of errors for each role:

To create these graphs you use the following settings:

From here you can play with the data in Power Bi and make whatever graph you like (I included location in the data so you can even plot your Edge servers on a map). This is what makes Power Bi so powerful!

The Wrap Up

This post ends my series on creating an Edge port monitor with Azure Function Apps. I hope that in addition to helping you monitor your edge servers, this has been informative and taught you some new skills that might help in the future when making your own Function Apps.

Read more →

Popular Posts