Abhith Rajan — everything

Cost Optimization: Auto Shutdown and Startup of Tagged VMs with Azure Automation

Abhith Rajan — Tue, 30 May 2023 09:26:00 GMT

In today’s cloud computing era, optimizing costs has become crucial for effective resource management. One area where organizations can achieve significant cost savings is by automating the shutdown and startup processes of virtual machines (VMs) in their test environments. Microsoft Azure offers various solutions to help accomplish this goal, and one powerful approach is through Azure Automation.

The Significance of Cost Optimization

Test environments play a vital role in software development and testing processes. However, these environments often remain idle during non-working hours, such as nights and weekends. Continuously running VMs during these periods can lead to unnecessary expenses. By implementing an automated solution that shuts down and starts up VMs based on predefined schedules, organizations can save a substantial amount of money.

Azure Automation: Streamlining Cost Optimization

Azure Automation is a cloud-based service that enables you to automate the management of your Azure resources. It offers a wide range of capabilities for process automation, configuration management, and update management. Through the use of Runbooks, you can schedule the shutdown and startup of VMs based on specific criteria, optimizing costs in test environments.

Runbooks

Implementing VM Shutdown with Azure Automation

Stop VM Runbook

To implement the VM shutdown process, you can utilize PowerShell scripts within an Azure Automation runbook. The following code snippet illustrates how to shut down VMs during weekdays (Monday to Friday) from 10 PM to 6 AM:

# Get the current time in the time zone of your choice
$TimeZone = [System.TimeZoneInfo]::FindSystemTimeZoneById("Arabian Standard Time")
$ScheduledTime = [System.TimeZoneInfo]::ConvertTimeFromUtc([DateTime]::UtcNow, $TimeZone)

# Connect to Azure Account using Managed Identity
$Conn = Connect-AzAccount -Identity

# Weekday Shutdown (Mon-Fri, 10pm-6am)
if (($ScheduledTime.DayOfWeek -ge "Monday") -and ($ScheduledTime.DayOfWeek -le "Friday") -and ($ScheduledTime.ToString("HH:mm") -ge "22:00") -or ($ScheduledTime.ToString("HH:mm") -lt "06:00")) {
   $VMs = Get-AzResource -Tag @{ Availability = "Available 6am-10pm Mon-Fri, Unavailable 10pm-6am Mon-Fri and all day Sat-Sun" } -ResourceType "Microsoft.Compute/virtualMachines"
   foreach ($VM in $VMs) {
       Stop-AzVM -ResourceGroupName $VM.ResourceGroupName -Name $VM.Name -Force
   }
}

This code snippet retrieves the current time in the specified time zone, connects to your Azure account using Managed Identity, and identifies the VMs based on a specified tag “Availability = Available 6am-10pm Mon-Fri, Unavailable 10pm-6am Mon-Fri and all day Sat-Sun”. It checks if the current time falls within the desired shutdown window and proceeds to stop each VM using the Stop-AzVM command.

Schedule for VM Shutdown:

Stop VM Schedule

Implementing VM Startup with Azure Automation

Similarly, you can implement the VM startup process using another PowerShell script within a separate Azure Automation runbook. The script can be scheduled to run during working hours or any other predefined schedule. The code snippet below demonstrates how to start VMs during weekdays:

# Get the current time in the time zone of your choice
$TimeZone = [System.TimeZoneInfo]::FindSystemTimeZoneById("Arabian Standard Time")
$ScheduledTime = [System.TimeZoneInfo]::ConvertTimeFromUtc([DateTime]::UtcNow, $TimeZone)

# Connect to Azure Account using Managed Identity
$Conn = Connect-AzAccount -Identity

# Weekday Start (Mon-Fri, 6am-10pm)
if (($ScheduledTime.DayOfWeek -ge "Monday") -and ($ScheduledTime.DayOfWeek -le "Friday") -and ($ScheduledTime.ToString("HH:mm") -ge "06:00") -and ($ScheduledTime.ToString("HH:mm") -lt "22:00")) {
   $VMs = Get-AzResource -Tag @{ Availability = "Available 6am-10pm Mon-Fri, Unavailable 10pm-6am Mon-Fri and all day Sat-Sun" } -ResourceType "Microsoft.Compute/virtualMachines"
   foreach ($VM in $VMs) {
       Start-AzVM -ResourceGroupName $VM.ResourceGroupName -Name $VM.Name
   }
}

The above code snippet retrieves the current time, connects to your Azure account using Managed Identity, and starts VMs based on the specified criteria. It identifies VMs tagged with “Availability = Available 6am-10pm Mon-Fri, Unavailable 10pm-6am Mon-Fri and all day Sat-Sun” and initiates the startup process using the Start-AzVM command.

Schedule for VM Startup:

Start VM Schedule

Additional Considerations

When implementing Azure Automation for VM shutdown and startup, keep the following points in mind:

Create an Azure Automation account to host your runbooks.
Enable Managed Identity for Azure Automation to securely connect to your Azure resources.
Grant the necessary permissions to the Managed Identity. In this case, assigning the Virtual Machine Contributor role at the resource group level is sufficient for performing the shutdown and startup actions.
Ensure relevant VMs are appropriately tagged for efficient automation.

By following these guidelines and leveraging Azure Automation, you can effectively optimize costs by automating the shutdown and startup of tagged VMs in your test environments. This approach ensures that VMs are active only when needed,thereby reducing unnecessary expenses. Additionally, it streamlines resource management and ensures that resources are available and ready for use during working hours.

Moreover, Azure Automation provides flexibility in scheduling the shutdown and startup processes based on specific criteria. You can customize the schedules according to your organization’s requirements, taking into account factors such as different time zones or varying working hours across teams.

Implementing VM shutdown and startup through Azure Automation offers several benefits:

Cost savings: By automatically shutting down VMs during non-working hours, organizations can significantly reduce costs associated with running idle resources. This optimization can lead to substantial savings over time.
Resource availability: VMs can be automatically started up before working hours, ensuring that the necessary resources are available when the team begins their tasks. This eliminates delays caused by manual startup processes and improves productivity.
Scalability and agility: Azure Automation allows for easy scaling of the solution to accommodate changing needs. As your organization grows or experiences fluctuations in demand, you can adjust the automation processes accordingly, ensuring efficient resource utilization.
Centralized management: Azure Automation provides a centralized platform for managing and monitoring the automated shutdown and startup processes. You can view and track the status of the runbooks, monitor resource utilization, and make adjustments as needed.
Compliance and security: By leveraging Managed Identity and assigning the appropriate permissions, Azure Automation ensures secure access to Azure resources without the need for storing credentials within the automation scripts. This improves compliance with security best practices and reduces potential risks.

In conclusion, automating the shutdown and startup processes of tagged VMs using Azure Automation is a powerful solution for cost optimization and resource management in test environments. By implementing predefined schedules and leveraging the flexibility of Azure Automation, organizations can achieve significant cost savings, improve resource availability, and streamline operations.

Remember, cost optimization is an ongoing process, and regularly reviewing and adjusting your automation schedules based on changing requirements can further enhance your cost-saving efforts.

If you have any questions or need clarification on any part of the post, feel free to leave a comment, and I’ll be glad to assist you.

Discover the power of Azure Automation for cost optimization in Azure environments. Cut unnecessary expenses, optimize resource management, and efficiently manage tagged VMs with automated VM shutdown and startup. https://t.co/bVUXhZ2cwT #Azure #CostOptimization #Automation
— Abhith Rajan (@AbhithRajan) May 30, 2023

Decode JWT Token

Abhith Rajan — Mon, 29 Aug 2022 09:26:00 GMT

Method

function decodeJwt(token) {
  var base64Payload = token.split(".")[1];
  var payloadBuffer = Buffer.from(base64Payload, "base64");
  return JSON.parse(payloadBuffer.toString());
}

Usage

var payload = decodeJwt(
  "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiIxMjM0NTY3ODkwIiwibmFtZSI6IkpvaG4gRG9lIiwiaWF0IjoxNTE2MjM5MDIyfQ.SflKxwRJSMeKKF2QT4fwpMeJf36POk6yJV_adQssw5c"
);
console.log(payload);

Output Environment Variable

Abhith Rajan — Mon, 25 Jul 2022 09:26:00 GMT

If you want to print the value of an environment variable, you can use the $env:variable_name syntax.

$env:MY_VARIABLE

Replace MY_VARIABLE with the name of the environment variable you want to print.

Delete All Local/Remote Git Tags

Abhith Rajan — Thu, 02 Jun 2022 09:26:00 GMT

On rare occasions, you may want to remove all local and remote git tags from your repository. For that, you can follow the below Recommended Steps.

In short, first, we are replacing the local tags with the remote tags, then removing all remote tags with reference to the local tags, and finally, removing all local tags.

Recommended Steps

1. Delete all local tags

git tag -d $(git tag -l)

2. Fetch all remote tags

git fetch

Retrieves all remote tags giving you a complete list of remote tags.

3. Delete All remote tags

git push origin --delete $(git tag -l)

Deletes the remote tags with reference to the local list.

4. Delete All local tags

git tag -d $(git tag -l)

Once again, deletes all local tags.

Unnecessary Details

One of the rare occasions to delete all tags in my case was, that we were maintaining a monorepo in Azure DevOps, and wanted to migrate it to GitHub, but in a repo per component manner. So,

First I did import the monorepo by following 👉 Code Migration From Azure DevOps to GitHub,
Then removed all other component’s code from the repo (git history is still dirty with other component’s old commits)
Re-configured the existing Azure pipeline triggers, i.e switched to a tag-based trigger. So, on each new release in GitHub, the connected Azure pipelines will run.

Now, the monorepo had so many tags including the tags of the other components. So, I wanted to remove all the tags from the repo before proceeding with the GitHub releases, and I did use the Recommended Steps above.

How We Use Jira Automation

Abhith Rajan — Sun, 10 Apr 2022 09:26:00 GMT

We use Jira for our issue/task management at work. As a DevOps guy, one of my favorite features of Jira is Jira automation.

For automation starters, the good thing is there are many predefined templates available at the Jira automation template library, which we can use as a reference to create our own with adaptations as per our needs.

Before getting into the automation rules which we use, to get a better picture, better to know about our current workflow.

Workflow

Our current workflow is,

flowchart LR
Backlog --> Todo
Todo --> InDev[In Dev]
InDev --> ReadyForQA[Ready for QA]
ReadyForQA --> InTest[In Test]
InTest --> IsQAPassed{Is QA passed?}
IsQAPassed --> |Yes| ReadyToDeploy[Ready to Deploy]
IsQAPassed -. Yes .-> Done
IsQAPassed --> |No| Todo
ReadyToDeploy --> Done

And, we configured our Jira to have the following two custom fields for issue types, which are,

Developer
Tester

Both of them are of the type User Picker (single user).

This is to retain the developer, and tester information on the issue since the assignee may change during state transition as per the workflow.

Custom fields

Now, let’s jump into the automation rules which we use. They are,

Automation rules

Auto-assign to whoever created the issue
Close duplicate issues
Link issues that are mentioned in the comments
When a commit is created and the status is ‘To do’ → then transition this issue to ‘In Dev’
On Pull Request created → comment the details
On Pull Request merged → comment the details
Comment on Ready for QA items if not assigned to a QA
Assign issue back to Developer once Ready To Deploy
Assign issue back to Developer once Done
When issue is resolved → comment on related issues
Label issues as stale that have not had recent interaction

Automation

I hope the names are self-explanatory. Now let’s dig into each and will try to explain the use case for each,

1. Auto-assign to whoever created the issue

This is very simple automation. When an issue is created without an assignee, automatically assign it to the person who created it.

This ensures that someone is always accountable for an issue by making sure it is always assigned to whoever created it.

Auto-assign to whoever created the issue

And the result will be,

Automation auto assigned to reporter

There’s a template for the same in the Jira automation template library.

👉 Smart auto-assign

2. Close duplicate issues

Sometimes, we have multiple similar issues. We can close the second issue and keep the first issue open just by adding a link to the second issue as,

The second issue duplicates the first issue.

Link issue as duplicates

The automation rule for closing such duplicate is,

Trigger - Close duplicate issues

That will do,

Duplicate issue closed by automation

3. Link issues that are mentioned in the comments

This will help us to link issues that are mentioned in the comments. And, along with the automation rule When issue is resolved → comment on related issues, this is a great way of notifying the related issues and the watchers of the same.

The result is,

Automation links the commented issue

4. When a commit is created and the status is ‘To do’ → then transition this issue to ‘In Dev’

We already connected our GitHub repos to our JIRA cloud. And, this automation will help us to transition the issue to ‘In Dev’ when a commit is created in Github.

You can use the smart commit message to trigger this automation.

👉 Process issues with smart commits | Jira Software Cloud | Atlassian Support

Trigger - When a commit is created and the status is 'To do' → then transition this issue to 'In Dev'

That’s it, automation will take care of the issue transition,

Issue transition

5. On Pull Request created → comment the details

This is to comment on the issue about the details of the pull request so the watchers are informed that there’s some progress on the work item.

Comment - On Pull Request created → comment the details

Then automation will do,

Comments PR details when PR created

6. On Pull Request merged → comment the details

Similar to On Pull Request created → comment the details, this is to comment the details of the pull request on the issue so the watchers are informed that there is more progress on the work item.

Comment - On Pull Request merged → comment the details

Then automation will do,

Comments PR details when PR merged

7. Comment on Ready for QA items if not assigned to a QA

As per our workflow, the dev should assign the task to a QA person when the task is ready for QA. But, if the dev doesn’t assign the task to a QA person, then automation comments on the issue to alert the dev person.

That’s it, automation will notify the developer in case they forgot to assign the task to a QA person.

Automation comment

8. Assign issue back to Developer once Ready To Deploy

Once the QA passed, the QA person should assign the task back to the developer of that work item along with changing the status to Ready to Deploy. But, we observed that most time the QA person forgets to assign the task back to the dev. To avoid this, we are using this automation rule to assign the task back to the dev once the task is ready to deploy.

Change assignee

Now automation will take care of the assignee switching,

Assignee changed

9. Assign issue back to Developer once Done

This is similar to Assign issue back to Developer once Ready To Deploy, but this is to assign the task back to the dev once the task is done.

That is because, once QA passed, QA can either mark it as Done or Ready to Deploy.

As I mentioned in the automation rule 3. Link issues that are mentioned in the comments, this is a great way of notifying the related issues and the watchers of the same.

Comment on related issues

This rule does,

Commented on related issue

11. Label issues as stale that have not had recent interaction

After a period of inactivity (now set to 60d), a stale label will be applied to mark an issue with Todo status as stale and posts a comment to notify contributors that the issue will be closed.

Label issues as stale that have not had recent interaction

Before adding this rule to the automation, I had around ~250 issues assigned to me in the Backlog. Most of them were created by myself since I have a habit of making notes of all work items in the Jira after any work-related discussions. But all of those were sitting there for so long whereas our priorities were wholly different and some were too old and stale of course. And, we recently started to work based on the backlog items so having a long backlog with stale work items is not a good thing and it was hard to prioritize the work too.

So, I decided to add this rule to the automation and it labeled the stale issues.

Stale issues

Then I went around all my stale issues, closed the ones which were not relevant anymore, and the ones which are not stale simply removed the stale label. In the end, I ended up having around ~50 items in the backlog. So, picking up items for the next sprint was easy and I was able to prioritize the backlog items.

Is closing stale issues really a good idea?

In any successful software project, there’s always more work to do than people to do it. As more and more work piles up, it becomes paralyzing. Just making decisions about what work should and shouldn’t get done can exhaust all available resources.

Focusing on issues that are actively affecting humans is an effective method for prioritizing work.

To some, a robot trying to close stale issues may seem inhospitable or offensive to contributors. But the alternative is to disrespect them by setting false expectations and implicitly ignoring their work. This automation makes it explicit: if work is not progressing, then it’s stale. A comment is all it takes to keep the conversation alive.

Planned Improvements

Rule to auto-assign to the tester when transitioned to Ready for QA
Rule to auto-assign to the developer when transitioned to Todo from In Test (QA not passed case)
Close issues that are marked as stale, after a period of inactivity.

Final Thoughts

The Jira automation helps us to manage the work items more efficiently and we are loving it. If you have any questions related to any of the Jira automation rules, or if you have any suggestions, please feel free to post them in the comments 👇

Additional Resources

Most popular Jira automation templates | Atlassian

This is how we use Jira Automation.
https://t.co/uZkdcDGtH2

#jira #projectmanagement
— Abhith Rajan (@AbhithRajan) April 11, 2022

GitHub vs Azure DevOps - For Code

Abhith Rajan — Tue, 01 Feb 2022 09:26:00 GMT

I have been using Azure DevOps as well as GitHub for many years now. All open-source stuff is on GitHub (some private repositories too.), and all work stuff is on Azure DevOps. I love Azure DevOps, but for code management, I prefer using GitHub instead.

Since both are owned by Microsoft, and seeing the recent developments and features all happening on the GitHub side compared to Azure DevOps, my mind is pretty clear on which to choose for the future. It’s not just me, but many people agree on the same. Jump to the 40:20 on the below video and hear yourself.

youtube: https://www.youtube.com/watch?v=K5aO97x2208&start=2420s

So, I would like to see this change (move all our repo to GitHub) in my organization. Currently, we use Azure DevOps for both code & CI/CD pipelines.

Since it is an organization-wide change, I cannot just go and tell my manager about this change. Instead, I have to present what benefits we are getting through this change.

For that, I thought of creating an internal presentation and side-by-side this public blog.

So in this post, I will try to list out things that attract me towards GitHub for code management.

Dependabot

Automated dependency updates built into GitHub.

One of the main attracting factors towards GitHub. Since we have many projects and tracking all the dependencies manually is not possible at all.

👉 Keep all your packages up to date with Dependabot

Dependabot’s automated PRs with updated packages/dependencies help a lot.

Dependabot PRs

And the security alerts,

Dependabot Alerts

Code Scanning

Code scanning is GitHub’s native SAST (Static Application Security Testing) tool, a developer-first approach to SAST that enables vulnerabilities to be found and remediated effortlessly before they reach production.

Here’s what GitHub code scanning can help you do:

Find, triage, and prioritize vulnerabilities in your code.
See security issues in your pull requests as part of your review process.
Prevent new vulnerabilities from making it to production.
Schedule scans for specific days and times, or trigger scans when a specific event occurs in the repository, such as a push.

The CodeQL action part of the code scanning features of GitHub is good even though comes at a price (of course free for open source). And I prefer the security-and-quality query suite for the CodeQL.

Code Scanning Alerts

Marketplace and integrations

Both Azure DevOps and GitHub have many integrations available. But IMHO, GitHub has a better collection, growing too.

JIRA

We manage our work items through Jira. One of the main things we are missing in Azure DevOps is the Jira integration whereas there’s an official integration available between JIRA and GitHub.

Linking work items to code have many benefits.

GitHub issues, boards all are awesome but JIRA is a bit more advanced as well as customizable for software project management, so can’t ask to switch to GitHub at this time (still I prefer GitHub).

GitHub Checks

There are several GitHub apps which is capable of running powerful checks against the code changes in the repository. Apps that perform continuous integration, code linting, or code scanning services and provide detailed feedback on commits.

GitHub Checks

Release Notes

The GitHub releases are a great way to share the latest version of your code. The release notes associated with the release are a great way to share the changes in the code.

Do you know that GitHub can help in making release notes?

Having the release notes part of each release, and helping with making the release notes is another plus side of Github.

Azure Pipelines vs GitHub Actions

Azure Pipelines are way matured compared to GitHub Actions (👈 this may change in the future).

But yes, a complete switch to GitHub from Azure DevOps might be possible by replacing Azure DevOps pipelines with GitHub actions. But people mostly hate change when things are working fine with the current setup. So one step at a time in terms of change.

i.e First code migration, then slowly pipelines. The first candidate for GitHub Actions can be PR builds.

While comparing both, I love the following facts about the GitHub action,

Easy to create by just placing the workflow file in the .github/workflows directory. Whereas Azure DevOps requires you to create a new pipeline in the UI.
Suggestions for the workflow file are provided in the GitHub action UI based on your project’s type.

Mobile Application

GitHub’s mobile application is a great way to manage your code.

You can do PR review,
Merge code,
And so on,

All through the mobile app.

GitHub Mobile App

Azure DevOps doesn’t have such a mobile application but you can access the DevOps portal through the mobile browser, the UI is somewhat responsive, but the experience is still way better in GitHub’s mobile app, and the app is getting new features on every release.

Powerful web-based editor (VS Code on Web)

I never liked "." this much.

We can open any #github repo on #vscode, directly on your browser, by simply pressing "." (no quotes, just that dot).

The repo will open in https://t.co/P2oEcfcGFA, a kind of Visual Studio Code on the browser. pic.twitter.com/ZUFNJe443s
— Abhith Rajan (@AbhithRajan) March 18, 2022

The github.dev and the shortcut ”.”, 👌. I am using it more often, even this blog am maintaining most of the time via the web editor of GitHub.

And

PR review is much better on GitHub. The sticky checkbox on each file is way better than the sidebar static checkboxes in the Azure DevOps.
The GitHub organization landing page with repositories sorted by last updated gives a better insight to the team.
The total number of repositories indicator in GitHub is a good metric, not sure how to find the total number of repositories in the Azure DevOps organization since repositories are kept under projects in the Azure DevOps organization.
The GitHub community. You can follow external projects, and people and can learn from them.

Pricing

Looks complicated on GitHub. In most places, it is like Contact Sales. The change would be an easier task if the billing could be done through an Azure Subscription but it looks like it is not possible at the moment.

Anyway, the free plan offers enough feature sets to start with GitHub.

If you plan to migrate to GitHub, the following sections will help you.

Code Migration From Azure DevOps to GitHub

The code repository migration from Azure DevOps to GitHub is pretty easy.

Import repository
Give your Azure DevOps Repo URL, and fill in the rest.
And on the next screen, it will ask for credentials to access your DevOps repo. Git credentials for the Azure DevOps repo will work fine. You can generate that one on the clone window on Azure DevOps as shown below. Git Credentials" loading="lazy" decoding="async" width="1920" height="929">

That’s it. You will not miss any code history after migration.

Change Azure Pipeline Repository From Azure DevOps to GitHub

Since in our case, we have all pipelines already configured in the DevOps. And don’t want to change any of them other than the repository location. This can be done by following the below steps.

Edit the existing pipeline
In the upper right corner open the … menu
Pick the ⚡ Triggers option
Click the YAML tab
Click the Get Sources option
Reconfigure the source repo to GitHub.

You might need to choose a GitHub connection to pick the repo. For that, better to install the Azure Pipelines GitHub app and give it access to the repository. Which will create a connection back to your Azure DevOps account.

Final Notes

After reading this,

Are you convinced that GitHub is better compared to Azure DevOps for code?
If you are a GitHub pro, did I miss any goodness of GitHub over Azure DevOps in terms of code management?

Let me know in the comments below 👇

Additional Resources

Changelog

2022-07-20

Added the Powerful web-based editor section

2022-07-13

Added the And section

2022-06-07

Added Release Notes and Mobile Application sections.
Updated Azure Pipelines vs GitHub Actions section.

#github vs #AzureDevOps - for code.

Comparing GitHub and Azure DevOps for code management and related features. Hopefully, this post will encourage you to switch to GitHub.https://t.co/ZNnEPhnKgQ
— Abhith Rajan (@AbhithRajan) February 1, 2022

2021 Year In Review

Abhith Rajan — Tue, 25 Jan 2022 09:26:00 GMT

A bit late to review 2021 work-life but “better late than never”, right?

In 2021, my focus was on mainly Cloud Native solutions. And one of my favorites of last year is, Dapr.

Other interesting findings of last year are,

And the list goes on.

Let’s talk about work for now.

More Azure

I’m an Azure guy. Interacting with it almost every working day. My 2021 facts related to Azure are,

Started using more Azure Logic Apps
Used Azure Cosmos DB the first time - it is our event store now.
Integrated Azure maps
Used Azure Cognitive Search
Baby steps with Azure Synapse + Power BI
Azure app configuration + Feature flags - Advocating colleagues to use feature flags.
Release annotation in Azure Application Insights - Want to make it part of CD, TODO for 2022.
Started switching to Azure Managed Identity
Played with ML Studio

Rest of the work

Some of the other notable things I was doing last year are,

Introduced geofence-based solutions - Hoo, the left-hand rule.
Migrated to JIRA cloud and set up some easy JIRA automation.
Now everywhere CloudEvents - I like that spec.
Done CI/CD setup for Java, Angular projects other than .NET.
More into Event-driven architectures.

And yeah, we switched to Postman cloud. The Postman is a good tool but for a microservice guy like me, the tabs are not enough.

Tab groups in #Chrome are so good for devs.
I wish #postman had something similar. pic.twitter.com/QIk22kkjGZ
— Abhith Rajan (@AbhithRajan) June 25, 2021

Pure Work

During this period, I have introduced around 12 RFCs in my workplace out of that 9 got landed in the production already. Rest are on the way. If you wonder what is an RFC, it is where we propose a new feature (major) to be developed, and the way it should be developed, some more deep dives into technical plus user stories on the same.

And I closed almost 263 JIRA work items during the same period.

A major role that I handled last year is to team up with a Java resource where I managed the work. Even though I am not a Java guy, I was reviewing the PRs (even corrected some, 😁). And it went really well, together we delivered 7 java microservice components.

The components term came to me from Backstage

Regarding PRs, previously I was following Gitflow, now switched to GitHub flow. And for release, Release Flow as recommended by Microsoft VSTS Team.

When I switch any of my workflow or tools, I always recommend my colleagues to follow the same by pointing out the benefits. My interest in Backstage is because I care about developer experience too. Always looking for great ways to organize work, improve the workflows.

Plans for 2022

Even though I’m working mostly on the Azure, never went for a certification. This year, I will.
After delivering several components (stable, reliable, and feature-rich) successfully by just managing the work, I realized that way I can contribute more to the work. Feels like time to go up a level, let’s see.
Code migration from Azure DevOps to GitHub: Advocating team about the benefits of GitHub over Azure DevOps for code.

Wish me a good year 😊

Wish me a good year 😊https://t.co/kSgkIfIMyK
— Abhith Rajan (@AbhithRajan) January 25, 2022

Create New Database Level User

Abhith Rajan — Sun, 23 Jan 2022 09:26:00 GMT

Connect to your SQL-Server using privileged account credentials (eg. sa) in your favorite IDE. I still use SSMS together with Azure Data Studio.

Set the query target database, In most IDE, right click on the DB, then new query.

CREATE USER [username]
WITH PASSWORD = 'password'
-- add user to role(s) in db
ALTER ROLE [db_datareader] ADD MEMBER [username];
ALTER ROLE [db_datawriter] ADD MEMBER [username];
ALTER ROLE [db_ddladmin] ADD MEMBER [username];
GRANT EXECUTE TO [username]

Execute the above statement after replacing the username and password according to your choice. You can also manage the roles as per your needs.

Some of the Fixed-database roles are,

Fixed-Database role name	Description
db_owner	Members of the db_owner fixed database role can perform all configuration and maintenance activities on the database, and can also drop the database in SQL Server. (In SQL Database and Azure Synapse, some maintenance activities require server-level permissions and cannot be performed by db_owners.)
db_ddladmin	Members of the db_ddladmin fixed database role can run any Data Definition Language (DDL) command in a database.
db_datawriter	Members of the db_datawriter fixed database role can add, delete, or change data in all user tables.
db_datareader	Members of the db_datareader fixed database role can read all data from all user tables and views. User objects can exist in any schema except sys and INFORMATION_SCHEMA.

Note: In order to connect to the DB using the new credentials, you need to specify the database name in the advanced options from your IDE since the above credentials don’t have the privilege to list the DB’s in the server.

Additional Resources

ASP.NET Core - OAuth 2.0 Client Access Token Management

Abhith Rajan — Sun, 09 Jan 2022 14:11:27 GMT

In a microservice world, the machine (microservice A) to machine (microservice B) communications can be secured using an OAuth 2.0 compatible token service, IdentityServer in our case. i.e To successfully call microservice B, microservice A needs to get an access token first via client credentials grant type from the IdentityServer with the right scope for the operation.

microservice A calls IdentityServer to get an access token.
microservice A then calls microservice B with the access token in the authorization header (bearer token).
microservice B validates the access token, responds to microservice A with a valid response.

The above steps are fine for a single call, but in case of multiple calls from A to B, A doesn’t need to get an IdentityServer token every time since each token has its validity, so basically, A can reuse the token until it expires.

For the token reusing, you don’t need to implement the caching and refreshing token logic anywhere since some good people who are maintaining IdentityModel.AspNetCore already done that. All we need to do is set up it, and consume it.

The library will take care of the following,

caching abstraction for access tokens
automatic renewal of expired access tokens
token lifetime automation for HttpClient

let’s jump into the setup. Following are our assumptions,

microservice B has endpoints,
- POST /orders and it requires IdentityServer token with orders.write scope to create new orders,
- and to fetch the orders GET /orders endpoint, the scope for that one is orders.read.
microservice A has valid client credentials configured on the IdentityServer with the above scopes added to the allowed scopes. i.e microservice A is allowed to call the microservice B.

So let’s configure microservice A,

Install IdentityModel.AspNetCore package

In the Startup.cs > ConfigureServices

// Setup token management service
services.AddAccessTokenManagement(options =>
{
  options.Client.Clients.Add(AppConsts.StsClientName, new ClientCredentialsTokenRequest
  {
    Address = $"{Configuration.GetValue<string>("Authentication:Sts:BaseUrl")}/connect/token", // Token service (IdentityServer) base URL
    ClientId = Configuration.GetValue<string>("Authentication:Sts:ClientId"),
    ClientSecret = Configuration.GetValue<string>("Authentication:Sts:ClientSecret"),
    Scope = $"{AppConsts.OrdersWriteScope} {AppConsts.OrdersReadScope}" // Any number of scopes, space separated
  });
});

// Adds a named HTTP client for the factory that automatically sends the client access token
services.AddClientAccessTokenClient(AppConsts.StsClientName, AppConsts.StsClientName, client => client.BaseAddress = new Uri(Configuration.GetValue<string>("ApiBaseUrl"))); // microservice B base URL

Here we setup the token management service and a named HttpClient that uses the same. Also you can observe, we moved the static string params to a AppConsts.cs file,

public class AppConsts
{
    public const string OrdersWriteScope = "orders.write";
    public static string OrdersReadScope = "orders.read";
    public static string StsClientName = "microservice_a_api_client";
}

And the configurations like client credentials, token service endpoint, and rest are kept in the appsettings.json or somewhere more private like Azure App Configuration, keyvault, and loaded to the aspnetcore configurations.

  "ApiBaseUrl": "https://microservice-b.com",
  "Authentication": {
    "Sts": {
      "BaseUrl": "https://my-identityserver.com",
      "ClientId": "microservice_a_api",
      "ClientSecret": ""
    }
  },

Configuration is done, now let’s consume the HttpClient. Here I’m using Flurl to make the HTTP request, if you prefer direct HttpClient usage, you can do that way. You don’t need to do anything special to get the access token since it is already taken care of.

using Flurl.Http;

public class GetOrdersQueryHandler : IRequestHandler<SomeInput, SomeOutput>
{
    private readonly IFlurlClient _flurlClient;

    public GetOrdersQueryHandler(IHttpClientFactory httpClientFactory)
    {
        var httpClient = httpClientFactory.CreateClient(AppConsts.StsClientName);
        _flurlClient = new FlurlClient(httpClient);

    }

    public async Task<SomeOutput> Handle(SomeInput request, CancellationToken cancellationToken)
    {
        return await _flurlClient
                .Request($"/orders")
                .GetJsonAsync<SomeOutput>(cancellationToken);
    }
}

As you can see, we use the named httpClient, and that client is already configured to have the access token with the scopes. That’s it. You can use the same way to call the create order endpoint since the client has an access token with both scopes.

Additional Resources

Identitymodel For Workers And Web Apps

UTC to UAE Time (Arabian Standard Time)

Abhith Rajan — Tue, 07 Sep 2021 09:26:00 GMT

SELECT CONVERT(datetime, SWITCHOFFSET(GETUTCDATE(), DATEPART(TZOFFSET,
GETUTCDATE() AT TIME ZONE 'Arabian Standard Time'))) AS CreationTime

Replace GETUTCDATE() with your table column that has the UTC time.

Azure Cognitive Search - GeographyPoint to GeoJSON in .NET 5

Abhith Rajan — Sat, 13 Mar 2021 17:11:27 GMT

Azure Cognitive Search Service supported data types includes Edm.GeographyPoint.

While creating index definition using attributes in C#, to specify the field type as Edm.GeographyPoint we can use the GeographyPoint as the type.

using Azure.Search.Documents.Indexes;
using Microsoft.Spatial;

public class RestaurantIndexModel
{
    ...

    [SimpleField(IsFilterable = true, IsSortable = true)]
    public GeographyPoint Location { get; set; }

    ...
}

We need to have the NuGet package Microsoft.Spatial to use the GeographyPoint type.

  <PackageReference Include="Microsoft.Spatial" Version="7.8.3" />

And with that, you can create the index definition,

Index definition

But when you try to index a document with the same type,

restaurant.Location = GeographyPoint.Create(restaurantResult.Value.Latitude, restaurantResult.Value.Longitude);

You may encounter the following issue,

{
  "error": {
    "code": "",
    "message": "The request is invalid. Details: parameters : Invalid GeoJSON. The 'type' member is required, but was not found.\r\n"
  }
}

The way you can fix this is by,

Add the following NuGet package,

  <PackageReference Include="Microsoft.Azure.Core.Spatial" Version="1.0.0" />

And then annotate your index model to use the MicrosoftSpatialGeoJsonConverter as the JsonConverter by,

using System.Text.Json.Serialization;
using Azure.Core.Serialization;
using Azure.Search.Documents.Indexes;
using Microsoft.Spatial;

public class RestaurantIndexModel
{
    ...

    [JsonConverter(typeof(MicrosoftSpatialGeoJsonConverter))]
    [SimpleField(IsFilterable = true, IsSortable = true)]
    public GeographyPoint Location { get; set; }

    ...
}

That will do.

Indexd document

Additional Resources

NDepend - The Best-in-class Static Code Analyzer for Dotnet

Abhith Rajan — Wed, 09 Dec 2020 15:11:27 GMT

If you check my about page, I have mentioned that,

My role involves prototyping, developing backend as well as front end apps with modern workflows, frameworks. Keeping an eye on the patterns and organized code.

To keep the code organized and to make sure the quality is met, I depend on NDepend.

NDepend is the best-in-class static code analyzer for dotnet. And the best come with a price too, i.e it’s a PAID tool. Someone was kind enough to send me a licensed version.

When I tried NDepend the first time, it was overwhelming, the amount of information it throws out after analysis was too much 😲. Still am exploring many sides of the information it shares. In this post, I will share how I use it and the areas where it helps.

Initial Setup

Once you download the package, getting started is easy. Run the app and its Start page got everything you need to get started,

Start Page

In my case, I installed the visual studio extension. With that, you can easily attach a new or existing NDepend project with the opened solution.

Eh, what is an NDepend project?

This is mentioned in their getting started docs.

👉 Getting Started with NDepend

A NDepend project can be attached to several Visual Studio solutions.
This is useful if you wish to analyze several .NET assemblies compiled with several Visual Studio solutions.
A benefit will be the possibility to navigate across several Visual Studio solutions opened in several Visual Studio instances.

Once the initial setup and first analysis of the project are done, you will be presented with the analysis report. And it will be part of its dashboard.

Dashboard

I love metrics and staring at them is one of my hobbies. So do I love the NDepend dashboard since it shows all the metrics nicely.

Dashboard Metrics

Want to check how the code evolved compared to a previous date? Just choose the baseline to any previous analysis date.

Different Baseline Metrics

Want some charts?

Here we go. Following are the default ones ready for you to see the trend.

Dashboard Trend Charts

Noticed the Create Trend Chart?

We can create more trend charts as per our needs and metrics can be selected from a categorized list. In which one category I like is the Third-Party Usage,

Third-Party Usage Trend

So far, I just scratched the surface of the Ndepend tool only. More about it is on the way.

In the meantime check out these great articles about NDepend, which share a lot (really, a lot) about its features.

Azure Logic Apps - Get Application Insights/Azure Monitor Logs Query Result on Microsoft Teams

Abhith Rajan — Tue, 17 Nov 2020 22:11:27 GMT

I love to monitor things. Application Insights is one of my favorite tools in this category. Checking Failures and Performance sections on the Application Insights were my first thing in the morning for some time in the past. Since some apps especially built by third parties lack Application Insights integration, we check Application Gateway access logs too.

Instead of manually checking every day, to ease the process what I did is configured a couple of logic apps to send a message on the Teams channel with a list of results from Application Insights and Azure Monitor Logs.

The configured daily reports include,

Application Insights - Performance
Application Gateway - Failures

The steps involved in configuring these are almost similar.

First, you need to define the query, in the logs section of each service try out your query and see if the expected results coming. Then we need to design an Adaptive card for teams. This can be easily done on adaptivecards.io.

Adaptive Cards

Adaptive Cards are platform-agnostic snippets of UI, authored in JSON, that apps and services can openly exchange. When delivered to a specific app, the JSON is transformed into native UI that automatically adapts to its surroundings. It helps design and integrate light-weight UI for all major platforms and frameworks.

Adaptive Cards

The easiest way to design the adaptive card is to pick the matching sample from the adaptivecards.io website and edit it in the live designer there itself to adjust it to your needs.

Couple of cards which I designed that suited my needs shared below,

Simple card

A simple adaptive card

The schema of the same given below,

{
  "$schema": "http://adaptivecards.io/schemas/adaptive-card.json",
  "type": "AdaptiveCard",
  "version": "1.2",
  "speak": "<s>Order update.</s>",
  "body": [
    {
      "type": "ColumnSet",
      "columns": [
        {
          "type": "Column",
          "width": "stretch",
          "items": [
            {
              "type": "TextBlock",
              "text": "Order Update",
              "horizontalAlignment": "right",
              "isSubtle": true
            },
            {
              "type": "TextBlock",
              "text": "@{triggerBody()?['Label']}",
              "horizontalAlignment": "Right",
              "spacing": "None",
              "size": "Large",
              "color": "Attention"
            }
          ]
        }
      ]
    },
    {
      "type": "ColumnSet",
      "spacing": "Medium",
      "separator": true,
      "columns": [
        {
          "type": "Column",
          "width": 1,
          "items": [
            {
              "type": "TextBlock",
              "text": "Order",
              "isSubtle": true,
              "weight": "Bolder"
            },
            {
              "type": "TextBlock",
              "text": "@{json(base64ToString(triggerBody()?['ContentData'])).OrderId}",
              "spacing": "Small"
            }
          ]
        }
      ]
    }
  ],
  "actions": [
    {
      "type": "Action.OpenUrl",
      "title": "View",
      "url": "https://my-website.com/path/to/resource"
    }
  ]
}

The above card is used to notify when there is a new message in an Azure Service Bus Topic and the @{triggerBody()?['Label']} is used to get the Message Label.

Similarly, @{json(base64ToString(triggerBody()?['ContentData'])).OrderId} is used to get the OrderId from the message body.

Our focus in this article is on a card with a list of items.

Card with a list of items

Initial design with a single row of information will look like,

Card with single row

And it’s schema,

{
  "type": "AdaptiveCard",
  "body": [
    {
      "type": "Container",
      "style": "emphasis",
      "items": [
        {
          "type": "ColumnSet",
          "columns": [
            {
              "type": "Column",
              "items": [
                {
                  "type": "TextBlock",
                  "size": "Large",
                  "weight": "Bolder",
                  "text": "**Performance (App Insights)**"
                }
              ],
              "width": "stretch"
            }
          ]
        }
      ],
      "bleed": true
    },
    {
      "type": "Container",
      "spacing": "Large",
      "style": "emphasis",
      "items": [
        {
          "type": "ColumnSet",
          "columns": [
            {
              "type": "Column",
              "spacing": "Large",
              "items": [
                {
                  "type": "TextBlock",
                  "weight": "Bolder",
                  "text": "Operation Name"
                }
              ],
              "width": "stretch"
            },
            {
              "type": "Column",
              "items": [
                {
                  "type": "TextBlock",
                  "weight": "Bolder",
                  "text": "Count"
                }
              ],
              "width": "auto"
            },
            {
              "type": "Column",
              "items": [
                {
                  "type": "TextBlock",
                  "weight": "Bolder",
                  "text": "Avg",
                  "horizontalAlignment": "Right"
                },
                {
                  "type": "TextBlock",
                  "text": "Percentiles 50 / 99",
                  "horizontalAlignment": "Right"
                }
              ],
              "width": "stretch",
              "horizontalAlignment": "Right"
            }
          ]
        }
      ],
      "bleed": true
    },
    {
      "type": "Container",
      "items": [
        {
          "columns": [
            {
              "items": [
                {
                  "text": "some operation",
                  "type": "TextBlock",
                  "wrap": true
                },
                {
                  "size": "Small",
                  "text": "my app cloud role name",
                  "type": "TextBlock",
                  "wrap": true
                }
              ],
              "spacing": "Large",
              "type": "Column",
              "width": "stretch"
            },
            {
              "items": [
                {
                  "horizontalAlignment": "Right",
                  "separator": true,
                  "text": "321 ",
                  "type": "TextBlock"
                }
              ],
              "type": "Column",
              "width": "auto"
            },
            {
              "items": [
                {
                  "color": "Attention",
                  "horizontalAlignment": "Right",
                  "text": "123ms",
                  "type": "TextBlock"
                },
                {
                  "horizontalAlignment": "Right",
                  "size": "Small",
                  "text": "456 / 789",
                  "type": "TextBlock"
                }
              ],
              "type": "Column",
              "width": "stretch"
            }
          ],
          "type": "ColumnSet"
        }
      ],
      "separator": true
    }
  ],
  "$schema": "http://adaptivecards.io/schemas/adaptive-card.json",
  "version": "1.2",
  "fallbackText": "This card requires Adaptive Cards v1.2 support to be rendered properly."
}

In the schema, there are three Container,

First one contains the main heading
Second column headings
The last one with the data.

We will be dynamically populating the third container items. And the result looks like,

Card with rows

And for a reference, A slightly altered design also given below,

Card with a row of data

And the schema for this one is,

{
  "type": "AdaptiveCard",
  "body": [
    {
      "type": "Container",
      "style": "emphasis",
      "items": [
        {
          "type": "ColumnSet",
          "columns": [
            {
              "type": "Column",
              "items": [
                {
                  "type": "TextBlock",
                  "size": "large",
                  "weight": "bolder",
                  "text": "**Application Gateway - Failed Requests**"
                }
              ],
              "width": "stretch"
            }
          ]
        }
      ],
      "bleed": true
    },
    {
      "type": "Container",
      "spacing": "Large",
      "style": "emphasis",
      "items": [
        {
          "type": "ColumnSet",
          "columns": [
            {
              "type": "Column",
              "spacing": "Large",
              "items": [
                {
                  "type": "TextBlock",
                  "weight": "Bolder",
                  "text": "URL"
                }
              ],
              "width": "stretch"
            },
            {
              "type": "Column",
              "items": [
                {
                  "type": "TextBlock",
                  "weight": "Bolder",
                  "text": "Status",
                  "horizontalAlignment": "Right"
                }
              ],
              "width": "auto"
            },
            {
              "type": "Column",
              "items": [
                {
                  "type": "TextBlock",
                  "weight": "Bolder",
                  "text": "Count",
                  "horizontalAlignment": "Right"
                }
              ],
              "width": "auto"
            }
          ]
        }
      ],
      "bleed": true
    },
    {
      "type": "Container",
      "items": [
        {
          "columns": [
            {
              "items": [
                {
                  "text": "/path/to/resource",
                  "type": "TextBlock",
                  "wrap": true
                },
                {
                  "color": "Attention",
                  "size": "Small",
                  "text": "demo.myapp.com",
                  "type": "TextBlock",
                  "wrap": true
                }
              ],
              "spacing": "Large",
              "type": "Column",
              "width": "stretch"
            },
            {
              "items": [
                {
                  "horizontalAlignment": "Right",
                  "separator": true,
                  "text": "404 ",
                  "type": "TextBlock"
                }
              ],
              "type": "Column",
              "width": "auto"
            },
            {
              "items": [
                {
                  "horizontalAlignment": "Right",
                  "separator": true,
                  "text": "123 ",
                  "type": "TextBlock"
                }
              ],
              "type": "Column",
              "width": "auto"
            }
          ],
          "type": "ColumnSet"
        }
      ],
      "separator": true
    }
  ],
  "$schema": "http://adaptivecards.io/schemas/adaptive-card.json",
  "version": "1.2",
  "fallbackText": "This card requires Adaptive Cards v1.2 support to be rendered properly."
}

When fed with data, the result is,

Application Gateway - Failed Requests

Once we are done with adaptive card design, it’s time to move on to Logic App designer.

Logic App

The only tricky part of the Logic app was to generate the rows of items. The other steps are pretty much straight forward. And the overall steps are,

All steps

Let’s go through each step.

1. Recurrence trigger

Configured it to trigger daily.

Recurrence trigger

2. Getting a query result

For Application Insights query results, use the Application Insights connector and action Run Analytics Query

Run Analytics Query

Query to get performance metrics from Application insights

let dataset=requests
| where timestamp > ago(1d);
dataset
| summarize count_=sum(itemCount), avg(duration), percentiles(duration, 50, 99) by operation_Name, cloud_RoleName
| union(dataset
| summarize count_=sum(itemCount), avg_duration=avg(duration), percentiles(duration, 50, 99)
| extend operation_Name="Overall")
| top 30 by avg_duration desc

And in case of Application Gateway access logs, we need to use the Azure Monitor Logs connector and it’s Run query and list results action, details here.

Run query and list results

Note: Diagnostic setting for Application Gateway should be configured to feed access log data to the log analytics workspace before.

Query to get failed requests summary from Application Gateway

AzureDiagnostics
| where ResourceProvider == "MICROSOFT.NETWORK" and Category == "ApplicationGatewayAccessLog"
| where httpStatus_d >= 400
| where TimeGenerated > ago(1d)
| summarize count_= count(httpStatus_d) by httpStatus_d,requestUri_s, host_s
| order by count_ desc
| top 30 by count_

3. Initialize an array variable

Since we are expecting a list of rows from the above query, to map it to rows in the Teams card message, we use this.

Initialize an array variable

4. For each row in the query result, we map the result item to an Adaptive card row.

For each

Here the value is an array of query results from step 2. And for each item in the array, we are appending the JSON object that represents the row of data in the adaptive card (under the third Container items).

5. Append the rows to the rest of the Adaptive card schema in the Teams connector channel.

Initialize an array variable

The full Message is,

{
    "type": "AdaptiveCard",
    "body": [
        {
            "type": "Container",
            "style": "emphasis",
            "items": [
                {
                    "type": "ColumnSet",
                    "columns": [
                        {
                            "type": "Column",
                            "items": [
                                {
                                    "type": "TextBlock",
                                    "size": "Large",
                                    "weight": "Bolder",
                                    "text": "**Performance (App Insights)**"
                                }
                            ],
                            "width": "stretch"
                        }
                    ]
                }
            ],
            "bleed": true
        },
        {
            "type": "Container",
            "spacing": "Large",
            "style": "emphasis",
            "items": [
                {
                    "type": "ColumnSet",
                    "columns": [
                        {
                            "type": "Column",
                            "spacing": "Large",
                            "items": [
                                {
                                    "type": "TextBlock",
                                    "weight": "Bolder",
                                    "text": "Operation Name"
                                }
                            ],
                            "width": "stretch"
                        },
                        {
                            "type": "Column",
                            "items": [
                                {
                                    "type": "TextBlock",
                                    "weight": "Bolder",
                                    "text": "Count"
                                }
                            ],
                            "width": "auto"
                        },
                        {
                            "type": "Column",
                            "items": [
                                {
                                    "type": "TextBlock",
                                    "weight": "Bolder",
                                    "text": "Avg",
                                    "horizontalAlignment": "Right"
                                },
                                {
                                    "type": "TextBlock",
                                    "text": "Percentiles 50 / 99",
                                    "horizontalAlignment": "Right"
                                }
                            ],
                            "width": "stretch",
                            "horizontalAlignment": "Right"
                        }
                    ]
                }
            ],
            "bleed": true
        },
        {
            "type": "Container",
            "items": @{variables('rows')},
            "separator": true
        }
    ],
    "$schema": "http://adaptivecards.io/schemas/adaptive-card.json",
    "version": "1.2",
    "fallbackText": "This card requires Adaptive Cards v1.2 support to be rendered properly."
}

Notice the @{variables('rows')}, that’s where the magic happens. So the array we prepared in the previous step gets appended here, and hence we get to see the multiple rows in the adaptive card. To understand better, cross-check the above one with the original schema mentioned in the very beginning here.

That’s all. Run the logic app and see a message popping up on Teams.

Simplifying work by automating routine tasks 🧑‍💻#azurelogicapps
+
( #azuremonitor || #applicationinsights)
👇
#microsoftteams https://t.co/ZyZc7Kh85i
— Abhith Rajan (@AbhithRajan) November 23, 2020

Microsoft Clarity - Setting Up with GatsbyJS & First Impressions

Abhith Rajan — Wed, 04 Nov 2020 00:00:00 GMT

I’m impressed. From a long-term Google Analytics user point, Clarity provides more info about the actual user interactions on the website.

Getting started was pretty easy.

Sign-up here, it’s FREE.
Create a Clarity project for your website.
Integrate with your website.

There are already some direct integrations available. In my case, I have opted for the Install tracking code manually.

Integrating Microsoft Clarity with GatsbyJS website

Official plugin

Step 1 Navigate to the root of your Gatsby project.

Step 2 Open your preferred command line and run one of the following commands.

npm install gatsby-plugin-clarity

yarn add gatsby-plugin-clarity

Step 3 Navigate to gatsby-config.js.

Step 4 Add the following to your plugins array

{
    resolve: `gatsby-plugin-clarity`,
    options: {
      // String value for your clarity project ID
      clarity_project_id: YOUR-CLARITY-PROJECT-ID,
      // Boolean value for enabling clarity while developing
      // true will enable clarity tracking code on both development and production environments
      // false will enable clarity tracking code on production environment only
      enable_on_dev_env: true
    },
}

The Clarity tracking code is now installed.

Legacy way of installing [DEPRECATED]

It is very similar to adding Google AdSense as described in my older post here,

👉 GatsbyJS - Add Google AdSense

All you have to do is, update your gatsby-ssr.js with the below code (if there is no gatsby-ssr.js file on your project directory then create one at the root).

import React from "react";

export const onRenderBody = ({ setHeadComponents, setPostBodyComponents }) => {
  const pluginOptions = {
    head: true,
  };
  if (process.env.NODE_ENV !== `production`) {
    return null;
  }

  const setComponents = pluginOptions.head
    ? setHeadComponents
    : setPostBodyComponents;
  return setComponents([
    <script
      async
      type="text/javascript"
      dangerouslySetInnerHTML={{
        __html: `
      (function(c,l,a,r,i,t,y){
        c[a]=c[a]||function(){(c[a].q=c[a].q||[]).push(arguments)};
        t=l.createElement(r);t.async=1;t.src="https://www.clarity.ms/tag/"+i;
        y=l.getElementsByTagName(r)[0];y.parentNode.insertBefore(t,y);
    })(window, document, "clarity", "script", "<YOUR-CLARITY-PROJECT-ID>");
      `,
      }}
    />,
  ]);
};

Replace <YOUR-CLARITY-PROJECT-ID> with your Clarity project ID. The embedded script you can view at Clarity Project > Settings > Setup > Install tracking code manually.

onRenderBody is a Gatsby SSR API, called after every page Gatsby server renders while building HTML so you can set head and body components to be rendered in your html.js.

👉 SSR APIs | GatsbyJS

Build the project (gatsby build) and see whether the AdSense snippet is present in the build output HTML (gatsby serve). You can also refer to the same used in my repo itself.

👉 abhith/abhith.net

Once you have done the integrations and if the site is live, after a few minutes you will start seeing the metrics getting populated in the Clarity Dashboard.

Dashboard

It is pretty clean and minimal. And it looks like,

Dashboard

Recordings

It’s a dope feature. Exactly what each user did, like actual screencast of user interactions on that page with clicks and all. There is a sound indication for mouse clicks as well, Nice.

Recordings

It will help website managers to understand how the user is interacting with each element/section on the web page.

Also, sensitive data masking and other privacy concern are taken care of by Clarity automatically.

Heatmaps

As the name suggests, most interacted areas will be visualized here.

Heatmaps

If you are concerned about the data it collects, privacy, etc, one fact is that it is an opensource project. So you can have a look at it,

👉 GitHub-microsoft/clarity

I’m impressed with Microsoft Clarity. It’s not(yet) a replacement for Google Analytics, but I would recommend you to try it out and see how you can make use of it.

Additional Resources

Clarity Documentation

Microsoft Clarity - I'm impressed
https://t.co/JqYbcIZlwz #Analytics #MicrosoftClarity
— Abhith Rajan (@AbhithRajan) November 3, 2020

DNS Lookup

Abhith Rajan — Wed, 23 Sep 2020 09:26:00 GMT

for i in {1..10}; do curl -k -o /dev/null -s -w "DNS-Lookup [%{time_namelookup}] Time-Connect [%{time_connect}] Time-PreTransfer [%{time_pretransfer}] Time-StartTransfer [%{time_starttransfer}] Total-Time [%{time_total}] Response-Code [%{http_code}]\n" https://www.abhith.net; done

Sample Output

root@pod-name-69d96c5494-bd97v:/app# for i in {1..10}; do curl -k -o /dev/null -s -w "DNS-Lookup [%{time_namelookup}] Time-Connect [%{time_connect}] Time-PreTransfer [%{time_pretransfer}] Time-StartTransfer [%{time_s
tarttransfer}] Total-Time [%{time_total}] Response-Code [%{http_code}]\n" https://www.abhith.net; done
DNS-Lookup [5.465393] Time-Connect [5.542967] Time-PreTransfer [5.626041] Time-StartTransfer [6.024135] Total-Time [6.261565] Response-Code [200]
DNS-Lookup [0.458571] Time-Connect [0.536812] Time-PreTransfer [0.619370] Time-StartTransfer [0.960386] Total-Time [1.199219] Response-Code [200]
DNS-Lookup [0.005218] Time-Connect [0.083305] Time-PreTransfer [0.167227] Time-StartTransfer [0.505865] Total-Time [0.745717] Response-Code [200]
DNS-Lookup [5.066491] Time-Connect [5.144106] Time-PreTransfer [5.227078] Time-StartTransfer [5.585845] Total-Time [5.825170] Response-Code [200]
DNS-Lookup [0.004125] Time-Connect [0.081587] Time-PreTransfer [0.165687] Time-StartTransfer [0.246201] Total-Time [0.481887] Response-Code [200]
DNS-Lookup [0.005214] Time-Connect [0.083619] Time-PreTransfer [0.168375] Time-StartTransfer [0.510846] Total-Time [0.823167] Response-Code [200]
DNS-Lookup [0.016983] Time-Connect [0.094932] Time-PreTransfer [0.177621] Time-StartTransfer [0.522677] Total-Time [0.758300] Response-Code [200]
DNS-Lookup [0.005210] Time-Connect [0.083526] Time-PreTransfer [0.166203] Time-StartTransfer [0.246943] Total-Time [0.481547] Response-Code [200]
DNS-Lookup [0.010488] Time-Connect [0.088643] Time-PreTransfer [0.172099] Time-StartTransfer [0.517206] Total-Time [0.829188] Response-Code [200]
DNS-Lookup [0.005282] Time-Connect [0.083156] Time-PreTransfer [0.167259] Time-StartTransfer [0.499211] Total-Time [0.734148] Response-Code [200]
root@pod-name-69d96c5494-bd97v:/app#

Noticed the 5sec+ responses above? Explore more about the intermittent delay in Kubernetes in these articles,

Azure Application Gateway - List of Failed Requests

Abhith Rajan — Sat, 19 Sep 2020 15:11:27 GMT

Azure application gateway serves an important role in our infrastructure. All the public inbound traffic happening through Application Gateway.

Part of monitoring, a year ago, I asked the following question in SO,

“We can see the failed request count in Azure application gateway but how can we see the list of failed requests in the Logs section? All the logs enabled in Diagnostics settings of Application Gateway. We need to see at least the requested path (URL) and the response code for the failed requests in the Logs section in the Application Gateway. Because of the way access log stored in the storage account (folder structure), it is a bit difficult to access the required data through the storage account.”

At that time there was no good solution for that. But that changed.

List of failed requests

AzureDiagnostics
| where ResourceProvider == "MICROSOFT.NETWORK" and Category == "ApplicationGatewayAccessLog"
| where httpStatus_d >= 400
| project httpStatus_d, requestUri_s, TimeGenerated

List of failed requests

Also, you can include other fields like host_s in the query which will be useful if you have multiple domains configured in the Application Gateway.

You can do much more with the data. In our case, we added a widget to our Azure dashboard where it displays the summary of failed requests per hour. And the query for the same given below,

AzureDiagnostics
| where ResourceProvider == "MICROSOFT.NETWORK" and Category == "ApplicationGatewayAccessLog"
| where httpStatus_d >= 400
| summarize count(httpStatus_d) by httpStatus_d,requestUri_s, bin(TimeGenerated, 1h)
| order by count_httpStatus_d desc
| project httpStatus_d, requestUri_s, TimeGenerated, count_httpStatus_d

And the result looks like (highlighted one),

Application Gateway Azure Dashboard

In fact, Azure made all this easy. When you navigate to the logs section of application gateway, the example queries by default include a similar query,

Application Gateway Logs

// Errors by URI
// Number of errors by URI.
AzureDiagnostics
| where ResourceType == "APPLICATIONGATEWAYS" and OperationName == "ApplicationGatewayAccess" and httpStatus_d > 399
| summarize AggregatedValue = count() by requestUri_s
| sort by AggregatedValue desc

Truncate String Without Word Break

Abhith Rajan — Mon, 14 Sep 2020 09:26:00 GMT

/// <summary>
/// Truncate string at a word near to the limit specified. Avoid word split.
/// </summary>
/// <param name="input">string</param>
/// <param name="length"></param>
/// <param name="appendDots"></param>
/// <returns></returns>
public static string TruncateAtWord(this string input, int length, bool appendDots = false)
{
    if (input == null || input.Length < length)
        return input;
    var iNextSpace = input.LastIndexOf(" ", length, StringComparison.Ordinal);
    var trimmedInput = string.Format("{0}", input.Substring(0, (iNextSpace > 0) ? iNextSpace : length).Trim());

    if (appendDots)
    {
        return trimmedInput + "...";
    }
    return trimmedInput;
}

Tests

[Fact]
public void TruncateAtWordTests()
{
    "This is a long sentence".TruncateAtWord(6).Should().Be("This");
    "This is a long sentence".TruncateAtWord(7).Should().Be("This is");
}

Reference

Code.Library

Domain Name From URL

Abhith Rajan — Sun, 13 Sep 2020 09:26:00 GMT

export const domainFromURL = (url) => {
  return url
    .replace("http://", "")
    .replace("https://", "")
    .replace("www.", "")
    .split(/[/?#]/)[0];
};

Mask String

Abhith Rajan — Sat, 12 Sep 2020 09:26:00 GMT

/// <summary>
/// Mask the string.
/// </summary>
/// <param name="value">String that need to be masked</param>
/// <param name="startIndex">zero index indicating mask start position</param>
/// <param name="mask">mask that need to be applied, eg. ***</param>
/// <returns>Usage: "123456789".Mask(3, "****") => "123****89"</returns>
public static string Mask(this string value, int startIndex, string mask)
{
    if (string.IsNullOrEmpty(value))
        return string.Empty;

    var result = value;
    var starLength = mask.Length;

    if (value.Length < startIndex) return result;

    result = value.Insert(startIndex, mask);

    result = result.Length >= (startIndex + (starLength * 2)) ? result.Remove(startIndex + starLength, starLength) : result.Remove(startIndex + starLength, result.Length - (startIndex + starLength));

    return result;
}

Tests

"123456789".Mask(3, "****").Should().Be("123****89");
"123456789".Mask(3, "****---").Should().Be("123****---");

Reference

Mask Mobile Number Using C# - JIN Weijie

Azure DevOps - CI/CD Pipeline involving Helm 3, ACR & AKS

Abhith Rajan — Tue, 18 Aug 2020 15:11:27 GMT

Azure is my favorite cloud provider. We use Azure for most of the infra & services. Our code goes to Azure DevOps, we use Azure Container Registry (ACR) to host our docker container images and our Kubernetes clusters running in Azure Kubernetes Service (AKS).

We configured the CI/CD pipelines in Azure DevOps. In my case, we have a monorepo that contains several ASP.NET Core microservices and the folder structure looks like below, which we inherited from eShopOnContainers.

- build
  - azure-devops
    - common
      - ci-steps-template.yml
      - ci-vars-template.yml
    - project-one
      - ci-pipeline.yml
    - project-two
- deploy

  - azure-devops
    - common
      - cd-steps-template.yml
      - cd-steps-template-prod.yml
      - cd-vars-template.yml
    - project-one
      - cd-pipeline.yml
    - project-two
  - k8s
    - helm
      - project-one
      - project-two

- src
  - Services
    - Project-One
    - Project-Two

One of the great articles that helped me to initial setup the CI/CD is given below. Its kind of outdated now since it is using az acr helm commands which were deprecated later. But it is still worth reading. So definitely check it out.

👉 Tutorial: Using Azure DevOps to setup a CI/CD pipeline and deploy to Kubernetes

CI Pipeline

The CI pipeline does the following,

Build a Docker image and push to ACR
Build Helm chart and push to ACR

Prerequisites

Helm chart for your project. Here my chart directory is located at deploy > k8s > helm. To create a new chart for your project, refer Helm Create.
acr-connection-name: ACR service connection in Azure DevOps. You can add it under Azure DevOps > Project > Project Settings > Service Connections.

The ACR credentials I stored in the Azure DevOps Variable Groups (acr-variable-group).

Name	Value
registryName	Your ACR name
registryLogin	ACR login
registryPassword	ACR password

Variable group definition

Common

ci-vars-template.yml

parameters:
  projectName: ""
  dockerRegistryServiceConnectionName: ""
  dockerfile: ""
  buildContext: ""

variables:
  helmVersion: 3.2.3
  HELM_EXPERIMENTAL_OCI: 1
  registryServerName: "$(registryName).azurecr.io"
  dockerRegistryServiceConnectionName: ${{ parameters.dockerRegistryServiceConnectionName }}
  dockerfile: ${{ parameters.dockerfile }}
  buildContext: ${{ parameters.buildContext }}
  projectName: ${{ parameters.projectName }}
  imageName: ${{ parameters.projectName }}
  imageTag: $(build.sourceBranchName)
  helmChartVersion: $(build.sourceBranchName)
  helmfrom: $(Build.SourcesDirectory)/deploy/k8s/helm
  helmto: $(Build.ArtifactStagingDirectory)/deploy/k8s/helm

Few things to note here,

HELM_EXPERIMENTAL_OCI is to enable OCI support in the Helm 3 client. Currently, this support is experimental.
build.sourceBranchName as the image tag and chart version is handy if you are following Gitflow (which we follow) or similar git branching convention, so each release (eg, refs/tags/project-one/2.2.6) will generate Docker image and Helm chart with the same version.

ci-steps-template.yml

steps:
  - task: Docker@2
    displayName: Build and push an image to container registry
    inputs:
      command: buildAndPush
      repository: $(imageName)
      dockerfile: $(dockerfile)
      containerRegistry: $(dockerRegistryServiceConnectionName)
      buildContext: $(buildContext)
      tags: |
        $(imageTag)

  - task: HelmInstaller@1
    displayName: "install helm"
    inputs:
      helmVersionToInstall: $(helmVersion)
  - bash: |
      echo $(registryPassword) | helm registry login $(registryName).azurecr.io --username $(registryLogin) --password-stdin
      cd deploy/k8s/helm/
      helm chart save $(helm package --app-version $(imageTag) --version $(helmChartVersion) ./$(projectName) | grep -o '/.*.tgz') $(registryName).azurecr.io/charts/$(projectName):$(imageTag)
      helm chart push $(registryName).azurecr.io/charts/$(projectName):$(helmChartVersion)
      echo $(jq -n --arg version "$(helmChartVersion)" '{helmChartVersion: $version}') > $(build.artifactStagingDirectory)/variables.json
    failOnStderr: true
    displayName: "helm package"
  - task: CopyFiles@2
    inputs:
      sourceFolder: $(helmfrom)
      targetFolder: $(helmto)
  - publish: $(build.artifactStagingDirectory)
    artifact: build-artifact

The steps in the CI pipeline we moved to a common template file ci-steps-template.yml so that we can reuse it on other pipelines as well, and the steps include,

Build and push the docker image
Installs Helm client
A series of script which does
- Authenticate to ACR
- Creates and push Helm chart to ACR.
- Creates variables.json which contain the newly created Helm chart version. Which we will use to fetch the right chart version during CD.
Copy some additional files to the artifact. Which we can use to override Helm chart values.

ci-pipeline.yml

trigger:
  branches:
    include:
      - refs/tags/project-one/*
  paths:
    include:
      - src/Services/ProjectOne/*

pr: none

pool:
  vmImage: "ubuntu-latest"

variables:
  - group: acr-variable-group
  - template: ../common/ci-vars-template.yml
    parameters:
      projectName: "project-one"
      dockerRegistryServiceConnectionName: "acr-connection-name"
      dockerfile: "src/Services/Project-One/Dockerfile"
      buildContext: "$(System.DefaultWorkingDirectory)"

steps:
  - template: ../common/ci-steps-template.yml

If everything went well, you will have two repositories under your ACR.

project-one which contains the Docker image
chart/project-one for the Helm chart

CD Pipeline

The CD pipeline will install the Helm chart on AKS. The CD pipeline stage requires following details,

Name	Value
aks	AKS name
rg	AKS resource group
aksSpTenantId	Subscription tenant id
aksSpId	Service principal Id
aksSpSecret	Service principal password

These credentials I stored in another varible group named aks-variable-group.

Helpful commands

Service principal credentials

Create new service principal aks-name-deploy by

az ad sp create-for-rbac -n aks-name-deploy --scopes aks-resource-id --role "Azure Kubernetes Service Cluster User Role" --query password -o tsv

Where aks-resource-id is,

az aks show -n $aks -g $rg --query id -o tsv

The above command will output service principal password aksSpSecret.

To get service principal id aksSpId,

az ad sp show --id http://aks-name-deploy --query appId -o tsv

Also we need to attach ACR with AKS so that AKS can pull our private docker images from our ACR.

Attach ACR with AKS

az aks update -g $rg -n $aks --attach-acr acr-resource-id

Where acr-resource-id is the output of,

az acr show -n $registryName -g acr-resource-group-name --query id -o tsv

Get Azure Tenant Id

To get tenantId aksSpTenantId,

az account show --query tenantId -o tsv

Now lets explore the pipeline YAML files.

Common

cd-vars-template.yml

parameters:
  projectName: ""

variables:
  helmVersion: 3.2.3
  HELM_EXPERIMENTAL_OCI: 1
  registryServerName: "$(registryName).azurecr.io"
  projectName: ${{ parameters.projectName }}

cd-steps-template.yml

steps:
  - checkout: none
  - task: HelmInstaller@1
    displayName: "install helm"
    inputs:
      helmVersionToInstall: $(helmVersion)
  - download: ci-pipeline
    artifact: build-artifact
  - bash: |
      az login \
           --service-principal \
           -u $(aksSpId) \
           -p '$(aksSpSecret)' \
           --tenant $(aksSpTenantId)
       az aks get-credentials \
           -n $(aks) \
           -g $(rg)
       echo $(registryPassword) | helm registry login $(registryServerName) --username $(registryLogin) --password-stdin
       helmChartVersion=$(jq .helmChartVersion $(pipeline.workspace)/ci-pipeline/build-artifact/variables.json -r)
       helm chart pull $(registryServerName)/charts/$(projectName):$helmChartVersion
       helm chart export $(registryServerName)/charts/$(projectName):$helmChartVersion --destination $(pipeline.workspace)/install
       helm upgrade \
           --namespace $(k8sNamespace) \
           --create-namespace \
           --install \
           --wait \
           --version $helmChartVersion \
           --set image.repository=$(registryServerName)/$(projectName) \
           -f $(pipeline.workspace)/ci-pipeline/build-artifact/deploy/k8s/helm/app.yaml \
           -f $(pipeline.workspace)/ci-pipeline/build-artifact/deploy/k8s/helm/inf.yaml \
           $(projectName) \
           $(pipeline.workspace)/install/$(projectName)
    failOnStderr: true
    displayName: "deploy helm chart"

The common CD steps include a series of script which does,

Authenticate to Azure using the service principal credentials
Set the specified AKS cluster as the context.
Authenticate ACR with the ACR credentials (The same credentials we used in CI pipeline defined in the acr-variable-group)
Extract the Helm chart version that need to install
Pulls the Helm chart and installs (or upgrade) it. Here we are overriding the chart image repository to our ACR repository and some additional common values (app.yaml & inf.yaml).

cd-pipeline.yml

trigger: none
pr: none

# define variables: registryName, registryLogin and registryPassword in the Azure pipeline UI definition
variables:
  - group: acr-variable-group
  - template: ../common/cd-vars-template.yml
    parameters:
      projectName: "project-one"
  - name: k8sNamespace
    value: myteam

resources:
  pipelines:
    - pipeline: ci-pipeline
      source: "project-one-ci"
      trigger:
        enabled: true
        branches:
          include:
            - refs/tags/project-one/*

# define 5 variables: aks, rg, aksSpId, aksSpSecret and aksSpTenantId in the Azure pipeline UI definition
stages:
  - stage: test
    displayName: test
    jobs:
      - deployment: test
        variables:
          - group: aks-variable-group
        displayName: deploy helm chart into AKS
        pool:
          vmImage: ubuntu-latest
        environment: test-$(projectName)
        strategy:
          runOnce:
            deploy:
              steps:
                - template: ../common/cd-steps-template.yml
  - stage: production
    displayName: production
    jobs:
      - deployment: production
        variables:
          - group: aks-prod-variable-group
        displayName: deploy helm chart into AKS
        pool:
          vmImage: ubuntu-latest
        environment: production-$(projectName)
        strategy:
          runOnce:
            deploy:
              steps:
                - template: ../common/cd-steps-template-prod.yml

In the CD pipeline above, I have defined two stages, one for TEST and one for PROD. The main difference between them is in the variable group used. aks-variable-group has the TEST cluster values and you guessed right, aks-prod-variable-group has the PROD cluster values. And the difference between cd-steps-template.yml and cd-steps-template-prod.yml is that prod file has some additional chart value overrides with respect to our PRODUCTION environment.

cd-steps-template-prod.yml

steps:
  - checkout: none
  - task: HelmInstaller@1
    displayName: "install helm"
    inputs:
      helmVersionToInstall: $(helmVersion)
  - download: ci-pipeline
    artifact: build-artifact
  - bash: |
      az login \
           --service-principal \
           -u $(aksSpId) \
           -p '$(aksSpSecret)' \
           --tenant $(aksSpTenantId)
       az aks get-credentials \
           -n $(aks) \
           -g $(rg)
       echo $(registryPassword) | helm registry login $(registryServerName) --username $(registryLogin) --password-stdin
       helmChartVersion=$(jq .helmChartVersion $(pipeline.workspace)/ci-pipeline/build-artifact/variables.json -r)
       helm chart pull $(registryServerName)/charts/$(projectName):$helmChartVersion
       helm chart export $(registryServerName)/charts/$(projectName):$helmChartVersion --destination $(pipeline.workspace)/install
       helm upgrade \
           --namespace $(k8sNamespace) \
           --create-namespace \
           --install \
           --wait \
           --version $helmChartVersion \
           --set image.repository=$(registryServerName)/$(projectName) \
           -f $(pipeline.workspace)/ci-pipeline/build-artifact/deploy/k8s/helm/app.yaml \
           -f $(pipeline.workspace)/ci-pipeline/build-artifact/deploy/k8s/helm/inf.yaml \
           -f $(pipeline.workspace)/ci-pipeline/build-artifact/deploy/k8s/helm/inf-prod.yaml \
           -f $(pipeline.workspace)/ci-pipeline/build-artifact/deploy/k8s/helm/$(projectName)/values-prod.yaml \
           $(projectName) \
           $(pipeline.workspace)/install/$(projectName)
    failOnStderr: true
    displayName: "deploy helm chart"

Few more Notes

CD pipeline is also YAML based (You gonna like it), hence create it like a regular pipeline (Not as RELEASE) in the Azure DevOps, and choose the cd-pipeline.yml after choosing to create pipeline based on Existing Azure Pipelines YAML file.
Once you create the CD pipeline, check the Environments under Azure DevOps Pipelines. There will be two environments as per the above example, test-project-one and production-project-one. Inside each, you can configure the approvals and more for the respective CD stages.

A sample reference source code is also pushed to here.

If you have any grey area in this article, feel free to shoot it in the comments below 👇, I will try to shed some light on that part.

Setup Continuous Integration in Azure DevOps which push Docker image, Helm 3 chart to Azure Container Registry(ACR), and Continuous Deployment pipeline that deploys the chart to Azure Kubernetes Service (AKS).https://t.co/S96mzQXn4M #AzureDevOps #helm #Kubernetes
— Abhith Rajan (@AbhithRajan) August 21, 2020

Cloudflare and AspNet Boilerplate

Abhith Rajan — Sat, 18 Jul 2020 11:26:00 GMT

I am an early adopter of aspnetboilerplate. I learned AngularJS as part of developing aspnetboilerplate project and it all started by reading an awesome blog post Introduction to ASP.NET Boilerplate by Halil ibrahim Kalkan. Since then I always recommended using the paid aspnetzero template for all the major projects in my workplace. And I used the free aspnetboilerplate template for some of the projects including my side project, piggyvault, a family finance management app, and its APIs is built on aspnetboilerplate.

Similarly, I use Cloudflare for most of the hosted websites as a firewall-cum-CDN. Recently after hosting the aspnetboilerplate project on SmarterASP.net, I started to face a login issue where it was working fine on local (as always 😆). The traffic to the PROD app was routed via Cloudflare.

Later the troubleshooting journey ended up here.

Cloudflare strips any header that contains dots (.) from origin web server responses. For example, the origin web server header test.header:data is removed by Cloudflare’s proxy.

By default, ABP uses Abp.TenantId header as the TenantIdResolveKey, and Cloudflare strips it resulting in null value always. Thankfully, aspnetboilerplate provides an option to change the TenantIdResolveKey. To change,

In the CoreModule,

// Enable this line to create a multi-tenant application.
Configuration.MultiTenancy.IsEnabled = PiggyvaultConsts.MultiTenancyEnabled;
Configuration.MultiTenancy.TenantIdResolveKey = "Piggy-TenantId"; // your key without "."

It resolved the login issue.

Additional Resources

ASP.NET Core - Using Mutliple Authentication Schemes

Abhith Rajan — Sat, 11 Jul 2020 19:26:00 GMT

In some use cases, you might want your endpoints to be authorized using multiple schemes. In my case, I had to allow some of the endpoints for authorized clients (using Identity Server) as well as for requests with a custom token which is generated by a custom service for authorized users.

To achieve this, we need to create the custom AuthenticationScheme and configure a policy to use our custom scheme as well as JwtBearer.

Custom AuthenticationScheme

Creating a custom authentication scheme will validate the custom token using the [Authorize] attribute.

To create a custom authentication scheme, we need to define the following,

CustomAuthenticationDefaults
CustomAuthenticationHandler
CustomAuthenticationOptions

Let’s start with the defaults, where we describe the name of the scheme.

public static class CustomAuthenticationDefaults
{
    public const string AuthenticationScheme = "Custom";
}

Next AuthenticationSchemeOptions,

using Microsoft.AspNetCore.Authentication;

public class CustomAuthOptions : AuthenticationSchemeOptions
{
    public string UserInfoEndpoint { get; set; }
}

To validate the custom token, I need to send an HTTP request to an endpoint and the URL for that endpoint needs to be configurable. By defining AuthenticationSchemeOptions, we can pass these values while setting up the scheme in the Startup.

Let’s move on to AuthenticationHandler, which validates the token.

using Flurl.Http;
using Microsoft.AspNetCore.Authentication;
using Microsoft.Extensions.Logging;
using Microsoft.Extensions.Options;
using System;
using System.Collections.Generic;
using System.Security.Claims;
using System.Text.Encodings.Web;
using System.Threading.Tasks;
...


public class CustomAuthenticationHandler : AuthenticationHandler<CustomAuthOptions>
{
    public CustomAuthenticationHandler(
        IOptionsMonitor<CustomAuthOptions> options,
        ILoggerFactory logger,
        UrlEncoder encoder,
        ISystemClock clock
        )
        : base(options, logger, encoder, clock)
    {
    }

    protected override async Task<AuthenticateResult> HandleAuthenticateAsync()
    {
        if (!Request.Headers.ContainsKey("Authorization"))
            return AuthenticateResult.Fail("Unauthorized");

        string authorizationHeader = Request.Headers["Authorization"];
        if (string.IsNullOrEmpty(authorizationHeader))
        {
            return AuthenticateResult.NoResult();
        }

        if (!authorizationHeader.StartsWith(CustomAuthenticationDefaults.AuthenticationScheme, StringComparison.OrdinalIgnoreCase))
        {
            return AuthenticateResult.Fail("Unauthorized");
        }

        string token = authorizationHeader.Substring(CustomAuthenticationDefaults.AuthenticationScheme.Length).Trim();

        if (string.IsNullOrEmpty(token))
        {
            return AuthenticateResult.Fail("Unauthorized");
        }

        try
        {
            return await ValidateTokenAsync(token);
        }
        catch (Exception ex)
        {
            return AuthenticateResult.Fail(ex.Message);
        }
    }

    private async Task<AuthenticateResult> ValidateTokenAsync(string session)
    {
        // getting user info using HTTP request made using Flurl
        var user = await Options.UserInfoEndpoint
            .WithHeader("some-id", session)
            .GetJsonAsync<User>();

        if (user == null)
        {
            return AuthenticateResult.Fail("Unauthorized");
        }

        var claims = new List<Claim>
        {
            new Claim(ClaimTypes.Name, $"{user.Name} {user.Surname}"),
            new Claim(ClaimTypes.GivenName, $"{user.Name}"),
            new Claim(ClaimTypes.Surname, surname),

            new Claim("scope", "orders:write"),
            new Claim(ClaimTypes.NameIdentifier, user.id)
            new Claim(ClaimTypes.Role, "User")
        };

        var identity = new ClaimsIdentity(claims, Scheme.Name);
        var principal = new ClaimsPrincipal(identity);
        var ticket = new AuthenticationTicket(principal, Scheme.Name);
        return AuthenticateResult.Success(ticket);
    }
}

In the AuthenticationHandler, you can use your way to validate your tokens.

Now we need to configure our project to use the custom authentication, for that, in the ConfigureServices of startup.cs,

services
    .AddCustomAuthentication(Configuration);

public static IServiceCollection AddCustomAuthentication(this IServiceCollection services, IConfiguration configuration)
{

    // Identity Server Configuration
    var identityUrl = configuration.GetValue<string>("Authentication:IdentityServerBaseUrl");
    services.AddAuthentication(options =>
    {
        options.DefaultAuthenticateScheme = JwtBearerDefaults.AuthenticationScheme;
        options.DefaultChallengeScheme = JwtBearerDefaults.AuthenticationScheme;
    }).AddJwtBearer(options =>
    {
        options.Authority = identityUrl;
        options.RequireHttpsMetadata = false;
        options.Audience = "your_api";
    });

    // Custom Authentication configuration
    services.AddAuthentication(CustomAuthenticationDefaults.AuthenticationScheme)
        .AddScheme<CustomAuthOptions, CustomAuthenticationHandler>(CustomAuthenticationDefaults.AuthenticationScheme,
        o => o.UserInfoEndpoint = configuration.GetValue<string>("Authentication:Custom:UserInfoEndpoint"));

    // we define policies here where we configure which scheme or combinations we need for each of our policies.
    services.AddAuthorization(options =>
    {
        // authorize using custom auth scheme only
        options.AddPolicy("UserRole", policy =>
        {
            policy.AuthenticationSchemes.Add(CustomAuthenticationDefaults.AuthenticationScheme);
            policy.RequireRole("User");
        });

        // authorize using custom auth scheme as well as identity server
        options.AddPolicy("OrdersWrite", policy =>
        {
            policy.AuthenticationSchemes.Add(JwtBearerDefaults.AuthenticationScheme);
            policy.AuthenticationSchemes.Add(CustomAuthenticationDefaults.AuthenticationScheme);
            policy.RequireClaim("scope", "orders:write");
        });
    });

    return services;
}

You can have different combinations in the policy defined above like based on scheme, claim, etc.

The configuration in the appsettings.json look like,

"Authentication": {
    "Custom": {
      "UserInfoEndpoint": "https://yourcustomauthwebsite.com/user-info-path"
    },
    "IdentityServerBaseUrl": "https://url-of-idserver"
}

Done, let’s enable the multi authorization to our endpoint. In the controller action,

[Authorize(Policy = "OrdersWrite")]
public async Task<ActionResult<OrderResult>> CreateOrder(OrderRequest orderRequest)
{
    var clientIdClaim = HttpContext.User.FindFirst("client_id"); // identity server client
    var userIdClaim = HttpContext.User.FindFirst(ClaimTypes.NameIdentifier); // user authenticated using custom auth handler.
    ...
}

Now we can invoke our create order endpoint with valid bearer token as well as with our custom token. The general format for authorization header is,

Authorization: <type> <credentials>

So bearer token request header looks like,

Authorization: bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiIxMjM0NTY3ODkwIiwibmFtZSI6IkpvaG4gRG9lIiwiaWF0IjoxNTE2MjM5MDIyfQ.SflKxwRJSMeKKF2QT4fwpMeJf36POk6yJV_adQssw5c

And our custom token request header looks like,

Authorization: custom abcasdjasdjlaksdjlasjdlasjd

Additional Resources

Serilog - Console Sink Themes

Abhith Rajan — Sat, 20 Jun 2020 19:17:00 GMT

Serilog is my favorite logger. The sinks I was using are ApplicationInsights, File, and Seq (for local). I was not using Console sink but when we switched to docker, I thought it is nice to have Console too.

I was checking the colored output for Console and I realized that the serilog-sinks-coloredconsole is retired and the Serilog.Sinks.Console by default has the colored output. There were few color themes listed on the readme so I want to see how each looks like, and see it for yourself below.

Built-in Themes

SystemConsoleTheme.Literate (Default)

.WriteTo.Console()

.WriteTo.Console(theme: SystemConsoleTheme.Literate)

SystemConsoleTheme.Literate

ConsoleTheme.None

.WriteTo.Console(theme: ConsoleTheme.None)

ConsoleTheme.None

SystemConsoleTheme.Grayscale

.WriteTo.Console(theme: SystemConsoleTheme.Grayscale)

SystemConsoleTheme.Grayscale

AnsiConsoleTheme.Literate

.WriteTo.Console(theme: AnsiConsoleTheme.Literate)

AnsiConsoleTheme.Literate

AnsiConsoleTheme.Grayscale

.WriteTo.Console(theme: AnsiConsoleTheme.Grayscale)

AnsiConsoleTheme.Grayscale

AnsiConsoleTheme.Code

.WriteTo.Console(theme: AnsiConsoleTheme.Code)

AnsiConsoleTheme.Code

Final Notes

In the above list, my favorite so far is the default one. I didn’t try any custom themes yet. if you know any good looking custom themes, let me know in the comments 👇

Additional Resources

serilog/serilog-sinks-console

Best FREE Form Provider for Static Websites

Abhith Rajan — Wed, 10 Jun 2020 09:17:00 GMT

If you have a static website like a portfolio without a contact form then you are missing so many opportunities. I realized it after adding a contact form on my website here.

Now you might be thinking about how difficult it is to add a form to your website, and I can tell you that it is easier than you think. Without any backend codes, we can consume some of the awesome FORM BACKEND as a Service providers. Some of them I explored yet are listed here.

Services

Getform

Form backend platform for designers and developers

My current contact form is powered by Getform. Their free plan details given below,

Getform pricing

~~It indicates 100 Submissions in the FREE plan which I thought renews every month but the FAQ says it’s not. That triggered me to search for another provider who offers a better FREE plan.~~

The pricing model of Getform changed and I am happy with the change.

If you want to check how I created the contact form in this Gatsby website using Getform, check my commit here

👉 Getform.io | Easy form endpoints for your forms

Kwes forms

It looks like Kwes offers more features and a generous 50 submissions per month for FREE.

Kwes pricing

👉 Kwes forms | Form Endpoint Service

SmartForms

SmartForms also offers 50 submissions per month for free.

SmartForms pricing

👉 SmartForms - Powerful and easy to setup form backend that alerts you on E-email, Slack and Telegram!

formcarry.

If you want more submissions per month for FREE, look for formcarry. It offers 100 submissions per month for FREE.

formcarry pricing

👉 Formcarry. | Hassle-Free Form Backend

formsubmit

If you want unlimited submissions for FREE, Formsubmit is the one.

Formsubmit sends all your form submissions to your email without having a need for any coding or a backbend. SignUp (optional) to get additional advantages.

And they don’t even have a pricing page to show 💙

👉 Formsubmit

Conclusion

I didn’t switch from Getform yet and am still reviewing other services. I will update this article whenever I find better solutions. If you know any good services in this category, let me know in the comments 👇

Additional Resources

2020 Pluralsight Free April Month Review

Abhith Rajan — Sat, 02 May 2020 09:26:00 GMT

I completed the #FreeApril of Pluralsight and it was very productive. In this one month, I completed around 16 courses, and they are,

Completed Courses

And I tested my Skills IQ,

Skills IQ

And there are many other courses which I started watching but yet not finished,

History

Also, you can check my public profile on Pluralsight,

👉 Abhith Rajan | Pluralsight Profile

Extending Pluralsight Subscription For Another One Free Month

I extended my Pluralsight subscription for one more month by activating the Visual Studio Dev Essentials benefit, its FREE.

Visual Studio Dev Essentials Benefits

If you don’t have an account on Visual Studio Dev Essentials, join on the link below, its also free and have many other benefits like shown above,

👉 Visual Studio Dev Essentials

Discounts

If you have enough budget, you can signup for their paid subscription, and get some DISCOUNTS by using my referral,

👉 Pluralsight Referral - Abhith

What is the plan, After Subscription Expiry

After this extra free month of Pluralsight, I plan to start the free month of LinkedIn Learning and will compare the course quality with Pluralsight, then I might switch to a paid plan on either of these two platforms.

I know many companies provide access to these e-learning platforms to their employees, if you are getting such a benefit from your firm, let me know in the comments 👇

.NET Core - NuGet Packages CI & CD Using GitHub Actions

Abhith Rajan — Thu, 02 Apr 2020 00:11:27 GMT

I always likes to automate the repeatative part of my work. At work, Azure DevOps taking care of the most of the CI & CD. I do maintain a small Class Library project in GitHub which has some helpers and stuffs you know like we reuse on most projects. And is been published to NuGet so that it can be consumed easily.

Recently I added one more project to the same repo which has some common stuff for ASP.NET Core projects. Previously I used to manually upload the .nuget pack file to NuGet since the releases were not that frequent. But now, the number of projects doubled, and sometime we need to deploy both packages since the ASP.NET Core project depends on the other library. So I did automated the whole process using GitHub actions. Here is how I did that,

You can check my repo where you can find the GitHub Actions configuration yml files under the .github/workflows/ directory.

👉 Abhith/Code.Library

I have two actions,

CI
Release to NuGet

CI

The CI (Continuous Integration) action basically runs on every commit to the master branch as well as for every PR to master.

And the jobs are pretty self-explanatory,

Setting up dotnet on the Linux
Builds the solution file
Run tests

# This is a basic workflow to help you get started with Actions

name: CI

# Controls when the action will run. Triggers the workflow on push or pull request
# events but only for the master branch
on:
  push:
    branches: [master]
  pull_request:
    branches: [master]

# A workflow run is made up of one or more jobs that can run sequentially or in parallel
jobs:
  # This workflow contains a single job called "build"
  build:
    # The type of runner that the job will run on
    runs-on: ubuntu-latest

    # Steps represent a sequence of tasks that will be executed as part of the job
    steps:
      - uses: actions/checkout@v2
      - name: Setup .NET Core
        uses: actions/setup-dotnet@v1
        with:
          dotnet-version: 3.1.101
      - name: Build with dotnet
        run: dotnet build --configuration Release
        working-directory: ./src
      - name: Test
        run: dotnet test
        working-directory: ./src

Continuous Integration

Release to NuGet

This action runs on every new release. And the jobs include,

Setting up dotnet on the Linux
Create the .nupkg files
Uploads artifacts
Upload the Nuget packages

# This is a basic workflow to help you get started with Actions

name: Release to NuGet

# Controls when the action will run. Triggers the workflow on push or pull request
# events but only for the master branch
on:
  release:
    types: [published]

# A workflow run is made up of one or more jobs that can run sequentially or in parallel
jobs:
  # This workflow contains a single job called "build"
  build:
    # The type of runner that the job will run on
    runs-on: ubuntu-latest

    # Steps represent a sequence of tasks that will be executed as part of the job
    steps:
      - uses: actions/checkout@v2
      - name: Setup .NET Core
        uses: actions/setup-dotnet@v1
        with:
          dotnet-version: 3.1.101
      - name: Create NuGet Package
        run: dotnet pack -c Release /p:Version=${{ github.event.release.tag_name }} /p:PackageReleaseNotes="See https://github.com/Abhith/Code.Library/releases/tag/${{ github.event.release.tag_name }}"
        working-directory: ./src
      - name: Archive NuGet Package
        uses: actions/upload-artifact@v1
        with:
          name: Code.Library
          path: ./src/Code.Library/bin/Release/Code.Library.${{ github.event.release.tag_name }}.nupkg
      - name: Archive NuGet Package
        uses: actions/upload-artifact@v1
        with:
          name: Code.Library.AspNetCore
          path: ./src/Code.Library.AspNetCore/bin/Release/Code.Library.AspNetCore.${{ github.event.release.tag_name }}.nupkg
      - name: Publish Nuget Package
        run: dotnet nuget push **/*.nupkg --api-key ${{ secrets.nuget_api_key }} --source https://api.nuget.org/v3/index.json --no-symbols true

Release to NuGet

You can see that I am pushing 2 NuGet packages and both have the same version number, which is same as my release tag on GitHub.

During Create NuGet Package stage, initially it was failing with following error

.dll' to be packed was not found on disk

This was because, in one of the project, it had following Settings in the .csproj file

<GeneratePackageOnBuild>true</GeneratePackageOnBuild>

Once I removed the above, everything went well. You can see a similar issue posted here.

And the “Publish NuGet Package” job requires an API key from NuGet. You can get the API key from NuGet by logging into your NuGet account, create a new API key.

NuGet API Key

And I stored the key under my repo secrets.

Manage Secrets in GitHub Repository

To manage secrets, on your repo go to the “Settings” tab, under that “Secrets”. There I added a secret with name “nuget_api_key” and we can access those secrets inside actions like ${{ secrets.nuget_api_key }}.

Best Visual Studio Extensions

Abhith Rajan — Tue, 17 Mar 2020 17:50:10 GMT

During my work, I spend the most time with Visual Studio. Now because of VSCode, Visual Studio is getting little more free time. There was a time when I just thought that I couldn’t be productive without ReSharper for Visual Studio. But those times changed since Visual Studio is getting better with every new release. The number of extensions in the marketplace also helps. The following are my favorite tools and extensions for Visual Studio 2022.

Free Tools

CodeMaid VS2022

CodeMaid VS2022 is an open source Visual Studio extension to cleanup and simplify our C#, C++, F#, VB, PHP, PowerShell, R, JSON, XAML, XML, ASP, HTML, CSS, LESS, SCSS, JavaScript and TypeScript coding.

EF Core Power Tools

EF Core Power Tools adds useful design-time DbContext features to the Visual Studio Solution Explorer context menu.

Productivity Power Tools 2022

Productivity Power Tools 2022 installs the individual extensions of Productivity Power Tools 2022

This is an extension pack installer will install each of the individual components of Productivity Power Tools for Visual Studio 2022.

The following components will be installed:

Align Assignments 2022
Double-Click Maximize 2022
Copy As Html 2022
Fix Mixed Tabs 2022
Match Margin 2022
Middle-Click Scroll 2022
Peek Help 2022
Shrink Empty Lines 2022
Solution Error Visualizer 2022
Time Stamp Margin 2022

Tweaks 2022

Tweaks 2022 contains collection of minor fixes and tweaks for Visual Studio to reduce the paper cuts and make you a happier developer.

WakaTime (For VS2022 Preview)

WakaTime does metrics, insights, and time tracking automatically generated from your programming activity. This extension will be replaced by WakaTime when VS 2022 is officially released.

xUnit.net.TestGenerator2022

xUnit.net.TestGenerator2022 adds xUnit.net Test Generator extensions for Visual Studio 2022. Creates Unit tests with xUnit.net 2.0 framework. Works on Visual Studio 2022.

Visual Studio 2019

Following are the ones which I previously used with Visual Studio 2019. Those are not brought to the upper list due to any of the following reasons,

Incompatibility with VS 2022
Not tried myself in VS 2022 (Especially the paid ones)

Free Tools

Better Comments

Better Comments is a Visual Studio extension that gives you the ability to customize the font and opacity of your comments independently of the editor’s font settings. It also adds four additional comment classifications, each classification with its own customizable foreground.

BuildVision

BuildVision visualize the building process in a table format.

Open in Visual Studio Code

Open in Visual Studio Code adds a menu command that lets you open any solution, project, folder and file in Visual Studio Code.

Roslynator 2019

Roslynator 2019 is a collection of 500+ analyzers, refactorings and fixes for C#, powered by Roslyn.

The Essentials

The Essentials is an extension pack for Visual Studio 2019 containing extensions no developer should be without. These extensions will add important functionality that will make you more productive and a happier developer.

Extensions

Trailing Whitespace Visualizer

Trailing Whitespace Visualizer keeps your code files clean by making it easier than ever to identify and remove any trailing whitespace.

Visual Studio IntelliCode

Visual Studio IntelliCode(IntelliCode) is a set of AI-assisted capabilities that improve developer productivity with features like contextual IntelliSense, argument completion, code formatting, and style rule inference.

IntelliCode augments existing developer workflows with machine-learning services that provide an understanding of code and its context. It’s applicable for C#, C++ (in preview), JavaScript/TypeScript (in preview), and XAML code today, and will be updated in the future to support more languages.

Visual Studio Spell Checker (VS2017 and Later)

Visual Studio Spell Checker (VS2017 and Later) is an editor extension that checks the spelling of comments, strings, and plain text as you type or interactively with a tool window. It can also spell check an entire solution, project, or selected items. Options are available to define multiple languages to spell check agains…

Web Essentials 2019

Web Essentials 2019 is the easiest way to set up Visual Studio for the ultimate web development experience.

Extensions

After installing the Web Essentials and restarting Visual Studio, the following extensions will be installed:

Paid

NDepend

NDepend is a Visual Studio and Azure DevOps/TFS extension that makes it easy to manage Code Quality on large and complex .NET code base

NCrunch for Visual Studio

NCrunch for Visual Studio is Distributed continuous build and test runner. Persistent Inline Code Coverage, Performance Metrics. Navigation + more!

OzCode

OzCode is a revolutionary debugging extension which complements Visual Studio and supercharges the C# debugging experience!.

ReSharper

ReSharper is a legendary .NET productivity tool: find and fix errors and code smells; navigate and refactor; run unit tests and write quality code faster.

Older Visual Studio Extensions List

Visual Studio 2010

Did I miss something? If you think so, drop a line with the extension name in the comments below 👇. I will be updating this post whenever I find a good extension for Visual Studio.

Changelog

2022-01-16

VS 2022 extensions added

2020-11-14

Added EF Core Power Tools

2020-08-20

Added The Essentials

2020-08-09

Added Open in Visual Studio Code
Moved NDepend to Paid

2020-04-08

Added xUnit.net.TestGenerator

2020-03-20

Added Roslynator 2019.

2020-03-18

Added Better Comments.

Style Guide - SQL Server Auditing Column Names

Abhith Rajan — Wed, 05 Feb 2020 08:26:00 GMT

For most of the tables in our DB will have some kind of auditing fields. Following are the column names I used to use for those auditing fields.

Creation Audited

Column Name	Type	Nullable	Description
CreationTime	datetime2(7)	NOT NULL	eg. 2019-09-15 13:56:13.8955178
CreatorUserId		NULL

Modification Audited

Column Name	Type	Nullable
LastModificationTime	datetime2(7)	NULL
LastModifierUserId		NULL

Soft Delete

Column Name	Type	Nullable
IsDeleted	bit	NOT NULL
DeleterUserId		NULL
DeletionTime	datetime2(7)	NULL

If you have worked with ABP before, you know from where I adopted the naming convention.

When I work on a non-ABP project, I used to follow the same column names.

Let me know what you used to name for auditing columns in the comments. 👇

Additional Resources

datetime2 (Transact-SQL) - SQL Server

Azure DevOps - GatsbyJs Build Pipeline

Abhith Rajan — Tue, 04 Feb 2020 15:11:27 GMT

For long I was using Netlify to do CI/CD for this website and was using their free plan. But when they introduced the build minutes limit, I had to switch to another option.

Bad news 📰 in the morning
New limit on #netlify build minutes.
Looking for alternative options! pic.twitter.com/W12MwlyOcn
— Abhith Rajan (@AbhithRajan) December 17, 2019

I already have a paid account with SmarterASP.NET where I host Piggy and is very much suitable for personal side projects since it is very cheap. So I thought of moving back to SmarterASP.NET and for CI/CD Azure DevOps Build & Release Pipelines since am using the same at work.

Build Pipeline

We used the YAML based pipeline and steps are self-explanatory.

trigger:
  - master

pool:
  vmImage: "ubuntu-latest"

steps:
  - task: NodeTool@0
    inputs:
      versionSpec: "10.x"
    displayName: "Install Node.js"

  - script: |
      npm install -g gatsby-cli
    displayName: "Install gatsby-cli"

  - script: |
      yarn install
    displayName: "Install dependencies"

  - script: |
      yarn build
    displayName: "Gatsby build"

  - task: CopyFiles@2
    inputs:
      sourceFolder: "public"
      contents: "**"
      targetFolder: "$(Build.ArtifactStagingDirectory)"
      cleanTargetFolder: true
    displayName: "Copy built site"
    condition: and(succeeded(), ne(variables['Build.Reason'], 'PullRequest'))

  - task: PublishBuildArtifacts@1
    inputs:
      PathtoPublish: "$(Build.ArtifactStagingDirectory)"
      ArtifactName: "drop"
      publishLocation: "Container"
    condition: and(succeeded(), ne(variables['Build.Reason'], 'PullRequest'))
    displayName: "Publish Build Artifacts"

This pipeline will run for pull request as well. In those cases we don’t need to build the artifacts. That’s why we put the below condition for CopyFiles@2 and PublishBuildArtifacts@1 tasks.

condition: and(succeeded(), ne(variables['Build.Reason'], 'PullRequest'))

And a successfull build looks like,

Azure DevOps Build Pipeline for Gatsby

Release Pipeline

Once we configured our build pipeline, created a release pipeline.

Build pipeline vs Release pipeline

A build is a pipeline that has no side-effects, it only takes in code and puts out compiled “artifacts”. A release is a pipeline that has side-effects, it takes in artifacts and publishes or deploys them to environments.

For the release pipeline, in case of SmarterASP.NET, the stage contain one job and it has only one task, the FTP Upload.

Release Pipeline for Gatsby

And the yml for the same look like below,

steps:
- task: FtpUpload@2
  displayName: 'FTP Upload to SmarterASP.NET'
  inputs:
    credentialsOption: inputs
    serverUrl: '$(FTP.ServerUrl)'
    username: '$(FTP.Username)'
    password: '$(FTP.Password)'
    rootDirectory: '$(System.DefaultWorkingDirectory)/_Abhith.abhith.net/drop'
    remoteDirectory: /sitefolder/
    preservePaths: true
    trustSSL: true
  condition: and(succeeded(), ne(variables['Build.Reason'], 'PullRequest'))
``

Notes

All the $(---) are pipeline variables
rootDirectory can be browsed through the … button near to the Root folder text input (check the comments below)

Even though I have successfully configured the CI/CD. But the time it took to complete the deployment was too long since the artifact size was high. The gatsby-images generate so many versions of the image files resulting increase in the artifact size. So the FTP upload takes so much time to finish.

Tried #AzureDevOps to do the build and FTP upload task to push the artifact to @smarterasp, it worked fine but the time took to complete the release is almost 2 hrs. (artifact is around 500 MB in size & number of files 📂 around 3k). pic.twitter.com/wcltxbObRj
— Abhith Rajan (@AbhithRajan) December 27, 2019

Because of this, I searched for another hosting provider and found out Render. They have a free plan for static sites and offers almost similar experience when comparing to Netlify.

If you know any better hosting solution for GatsbyJs website, let me know in the comments. 👇

Azure DevOps Build - Could not locate the assembly

Abhith Rajan — Thu, 23 Jan 2020 06:59:27 GMT

We were configuring CI/CD for one of our .NET Core 3.1 API project in the Azure DevOps. When trying dotnet build from our local workstation, it was successfull but when CI triggered the build on the Azure DevOps, it was failing.

The Azure Pipeline yml looks like,

pool:
  vmImage: "windows-latest"

variables:
  buildConfiguration: "Release"

steps:
  - task: DotNetCoreCLI@2
    displayName: "dotnet build $(buildConfiguration)"
    inputs:
      command: "build"
      projects: |
        path/to/project/*.csproj
      arguments: "--configuration $(buildConfiguration)"

The build step successfully restored the NuGet packages which we can see in the logs,

Restore completed in 14.41 sec for d:\a\1\s\path\to\project\Project.Name.csproj.

But the step was failing with the following reference missing error.

C:\Program Files\dotnet\sdk\3.1.100\Microsoft.Common.CurrentVersion.targets(2106,5): warning MSB3245: Could not resolve this reference. Could not locate the assembly “System.ComponentModel.Annotations”. Check to make sure the assembly exists on disk. If this reference is required by your code, you may get compilation errors. [d:\a\1\s\path\to\project\Project.Name.csproj]

Azure DevOps Build Error

When checked the project reference, found out that the reference to System.ComponentModel.Annotations was added from assembly instead of Nuget. Fixed it by removing the assembly reference and adding the NuGet package of System.ComponentModel.Annotations.

DotNet Core Reference Fix

Commited the changes and initiated a hotfix release. Tada, the build succeeded.

GatsbyJS - Add Google AdSense

Abhith Rajan — Sun, 05 Jan 2020 17:00:00 GMT

One of the major strengths of Gatsby is its evolving plugins catalog. And there are a couple of plugins found for Google AdSense integration but don’t seem to be maintained for so long. But peeking into the source code of the one pave the path to do the same without any plugins.

All you have to do is, update your gatsby-ssr.js with the below code (if there is no gatsby-ssr.js file on your project directory then create one at the root).

import React from "react";

export const onRenderBody = ({ setHeadComponents, setPostBodyComponents }) => {
  const pluginOptions = {
    googleAdClientId: `ca-pub-XXXXXXXXXXXXXX`, //TODO: Replace with your client-Id
    head: true, // Set to false if you prefer to have your adsense script loaded at the end of body instead of head.
  };

  if (process.env.NODE_ENV !== `production`) {
    return null;
  }
  if (pluginOptions.googleAdClientId === undefined) {
    return null;
  }
  const setComponents = pluginOptions.head
    ? setHeadComponents
    : setPostBodyComponents;
  return setComponents([
    <script
      async
      type="text/javascript"
      src="https://pagead2.googlesyndication.com/pagead/js/adsbygoogle.js"
    />,
    <script
      key={`gatsby-plugin-google-adsense`}
      dangerouslySetInnerHTML={{
        __html: `
        (adsbygoogle = window.adsbygoogle || []).push({
            google_ad_client: "${pluginOptions.googleAdClientId}",
            enable_page_level_ads: true
        });
        `,
      }}
    />,
  ]);
};

onRenderBody is a Gatsby SSR API, called after every page Gatsby server renders while building HTML so you can set head and body components to be rendered in your html.js.

👉 SSR APIs | GatsbyJS

Build the project (gatsby build) and see whether the AdSense snippet is present in the build output HTML.

You can also refer the same used in my repo itself.

👉 abhith/abhith.net: Personal website running on Gatsby

.NET Core API Gateway Ocelot - Logging HTTP Requests & Response Including Headers & Body

Abhith Rajan — Thu, 26 Dec 2019 17:11:27 GMT

We have many microservices running under one umbrella on production which is Azure Application Gateway and has only one endpoint for all. We are using path-based rules to route to each service.

During development, we may need to simulate the same experience, i.e one endpoint instead of we referring separate URL endpoint for each service.

For the microservices, I used to refer dotnet-architecture/eShopOnContainers. And I read about the Ocelot API gateway from there. So I set up a local API gateway using Ocelot, for that

Created a new ASP.NET Core 3.1 web application and choose the Empty template.
Added the following NuGet packages

<PackageReference Include="Ocelot" Version="13.8.0" />
<PackageReference Include="Serilog" Version="2.9.0" />
<PackageReference Include="Serilog.AspNetCore" Version="3.2.0" />
<PackageReference Include="Serilog.Sinks.File" Version="4.1.0" />

Updated the Program.cs

using System.IO;
using Microsoft.AspNetCore;
using Microsoft.AspNetCore.Hosting;
using Microsoft.Extensions.Configuration;
using Microsoft.Extensions.Hosting;
using Microsoft.Extensions.DependencyInjection;
using Serilog;
using Serilog.Events;
using Ocelot.DependencyInjection;
using Ocelot.Middleware;
using Microsoft.AspNetCore.Builder;
using OcelotApiGw.Middlewares;

namespace OcelotApiGw
{
    public class Program
    {
        public static IWebHost BuildWebHost(string[] args)
        {
            var builder = WebHost.CreateDefaultBuilder(args);

            // Ocelot configuration file
            builder.ConfigureAppConfiguration(
                          ic => ic.AddJsonFile(Path.Combine("configuration",
                                                            "configuration.json")))
                    .ConfigureServices(s =>
                    {
                        s.AddSingleton(builder);
                        s.AddOcelot();
                    })
                    .UseStartup<Startup>()
                    .UseSerilog((_, config) =>
                    {
                        config
                            .MinimumLevel.Information()
                            .MinimumLevel.Override("Microsoft", LogEventLevel.Warning)
                            .Enrich.FromLogContext()
                            .WriteTo.File(@"Logs\log.txt", rollingInterval: RollingInterval.Day);
                    })
                    .Configure(app =>
                    {
                        app.UseMiddleware<RequestResponseLoggingMiddleware>();
                        app.UseOcelot().Wait();
                    });

            var host = builder.Build();
            return host;
        }

        public static void Main(string[] args)
        {
            BuildWebHost(args).Run();
        }
    }
}

Here we are configuring the Ocelot middleware as well as Serilog for logging. The Ocelot configuration file is added under “configuration” folder in the root directory.

A sample Ocelot configuration file looks like this.

configuration.json

{
  "ReRoutes": [
    {
      "DownstreamPathTemplate": "/{version}/{everything}",
      "DownstreamScheme": "http",
      "DownstreamHostAndPorts": [
        {
          "Host": "localhost",
          "Port": 5010
        }
      ],
      "UpstreamPathTemplate": "/d/{version}/{everything}",
      "UpstreamHttpMethod": ["POST", "PUT", "GET"]
    },
    {
      "DownstreamPathTemplate": "/b/{everything}",
      "DownstreamScheme": "http",
      "DownstreamHostAndPorts": [
        {
          "Host": "localhost",
          "Port": 5011
        }
      ],
      "UpstreamPathTemplate": "/a/b/{everything}",
      "UpstreamHttpMethod": ["POST", "PUT", "GET"]
    },
    {
      "DownstreamPathTemplate": "/{everything}",
      "DownstreamScheme": "http",
      "DownstreamHostAndPorts": [
        {
          "Host": "localhost",
          "Port": 5013
        }
      ],
      "UpstreamPathTemplate": "/c/{everything}",
      "UpstreamHttpMethod": ["POST", "PUT", "GET", "OPTIONS"]
    }
  ],
  "GlobalConfiguration": {
    "BaseUrl": "https://localhost:44390"
  }
}

Apart from that, we are also adding one custom middleware RequestResponseLoggingMiddleware and its purpose is clear from its name, log request and response.

The Logging Middleware

This is a slightly modified version of middleware mentioned in the article, Using Middleware in ASP.NET Core to Log Requests and Responses.

using Microsoft.AspNetCore.Http;
using Microsoft.Extensions.Logging;
using System.IO;
using System.Text;
using System.Threading.Tasks;

namespace OcelotApiGw.Middlewares
{
    public class RequestResponseLoggingMiddleware
    {
        private readonly ILogger<RequestResponseLoggingMiddleware> _logger;
        private readonly RequestDelegate _next;

        public RequestResponseLoggingMiddleware(RequestDelegate next, ILogger<RequestResponseLoggingMiddleware> logger)
        {
            _next = next;
            _logger = logger;
        }

        public async Task InvokeAsync(HttpContext context)
        {
            context.Request.EnableBuffering();

            var builder = new StringBuilder();

            var request = await FormatRequest(context.Request);

            builder.Append("Request: ").AppendLine(request);
            builder.AppendLine("Request headers:");
            foreach (var header in context.Request.Headers)
            {
                builder.Append(header.Key).Append(':').AppendLine(header.Value);
            }

            //Copy a pointer to the original response body stream
            var originalBodyStream = context.Response.Body;

            //Create a new memory stream...
            using var responseBody = new MemoryStream();
            //...and use that for the temporary response body
            context.Response.Body = responseBody;

            //Continue down the Middleware pipeline, eventually returning to this class
            await _next(context);

            //Format the response from the server
            var response = await FormatResponse(context.Response);
            builder.Append("Response: ").AppendLine(response);
            builder.AppendLine("Response headers: ");
            foreach (var header in context.Response.Headers)
            {
                builder.Append(header.Key).Append(':').AppendLine(header.Value);
            }

            //Save log to chosen datastore
            _logger.LogInformation(builder.ToString());

            //Copy the contents of the new memory stream (which contains the response) to the original stream, which is then returned to the client.
            await responseBody.CopyToAsync(originalBodyStream);
        }

        private async Task<string> FormatRequest(HttpRequest request)
        {
            // Leave the body open so the next middleware can read it.
            using var reader = new StreamReader(
                request.Body,
                encoding: Encoding.UTF8,
                detectEncodingFromByteOrderMarks: false,
                leaveOpen: true);
            var body = await reader.ReadToEndAsync();
            // Do some processing with body…

            var formattedRequest = $"{request.Scheme} {request.Host}{request.Path} {request.QueryString} {body}";

            // Reset the request body stream position so the next middleware can read it
            request.Body.Position = 0;

            return formattedRequest;
        }

        private async Task<string> FormatResponse(HttpResponse response)
        {
            //We need to read the response stream from the beginning...
            response.Body.Seek(0, SeekOrigin.Begin);

            //...and copy it into a string
            string text = await new StreamReader(response.Body).ReadToEndAsync();

            //We need to reset the reader for the response so that the client can read it.
            response.Body.Seek(0, SeekOrigin.Begin);

            //Return the string for the response, including the status code (e.g. 200, 404, 401, etc.)
            return $"{response.StatusCode}: {text}";
        }
    }
}

Usually, the request body can be read only once. Here we are making use of the new EnableBuffering extension method to read the request body multiple times. You can read more about the feature here, Re-reading ASP.Net Core request bodies with EnableBuffering().

That’s it, run your projects, make some requests and then check the logs directory.

This logging middleware can be used in any other ASP.NET Core project.

Here we used in the Ocelot API Gateway so that we don’t need to check the logs of each microservices.

Bonus

I am using a free tool Tailviewer to go through my logs. Check it out, it is small in size also an opensource project.

SQL Server - Delete Duplicate Rows

Abhith Rajan — Wed, 04 Dec 2019 20:00:00 GMT

Deleting rows from a SQL Server table is no big deal. But it becomes a little complicated when we are in a situation like a table with no primary keys or unique column and it has duplicate records which we need to delete.

When tried to delete rows using the context menu in the Microsoft SQL Server Management Studio, it failed with an error message indicating the action affects multiple records. But if you try Azure Data Studio to do the same, it will delete all the similar rows, not just the one you choose to delete.

So, in this case, we can use a query like below,

 WITH CTE AS(
   SELECT [Col1]
      ,[Col2]
      ,...
      ,ROW_NUMBER() OVER(PARTITION BY [Col1] ORDER BY [Col1]) AS RN
   FROM [dbo].[Table]
)
SELECT * FROM CTE

In the above query, we are using a CTE to list the data along with an additional column RN which will have a value higher than 1 for the duplicate entries.

It is always a good idea to “SELECT” and double-check the data before doing a hard “DELETE”

So first run the above query and inspect the rows with RN>1, Once you are okay with the results, execute the delete.

 WITH CTE AS(
   SELECT [Col1]
      ,[Col2]
      ,...
      ,ROW_NUMBER() OVER(PARTITION BY [Col1] ORDER BY [Col1]) AS RN
   FROM [dbo].[Table]
)
DELETE FROM CTE WHERE RN > 1

Angular - Add Custom Font

Abhith Rajan — Tue, 12 Nov 2019 12:59:27 GMT

To add custom fonts to your Angular project, put the font files under src.

Here I am following the folder pattern which I found on ABP Zero boilerplate.

In my case src > assets > fonts > fontname > font-files (.woff).

Then create a CSS file, src > assets > fonts > fonts-fontname.css.

And on the CSS, add the font-face referring to our new font-flies, an example is given below.

@font-face {
  font-family: "fontname";
  src: url("./fontname/fontname.woff");
}

@font-face {
  font-family: "fontname-Medium";
  src: url("./fontname/fontname-medium.woff");
}

@font-face {
  font-family: "fontname-Light";
  src: url("./fontname/fontname-light.woff");
}

Once the CSS file is ready, now we have to include that too in the angular build. To do that, update the angular.json to have references to our new CSS file like in below example,

{
 ...
  "projects": {
    "project-name": {
     ...
      "architect": {
        "build": {
          ...
          "options": {
            ...
            "styles": [
              ...
              "src/assets/fonts/fonts-fontname.css",
              ...
            ],
            ...
          },
          ...
        },
        ...,
        "test": {
          ...
          "options": {
            ...
            "styles": [
              ...
              "src/assets/fonts/fonts-fontname.css",
              ...
            ],
            ...
          }
        },
        ...
      }
    },
}

Note: the above code is tried and tested on Angular 8

Once we updated the angular.json, start a new build and see the new font applied in the places where it is used.

C# - Convert DateTime to Long and Vice Versa

Abhith Rajan — Tue, 29 Oct 2019 18:26:00 GMT

Story behind this post is when I started working with an existing MSSQL DB, I had to convert the DateTime to long since action auditing fields such as CreationTime, LastUpdated etc was stored as numeric.

eg, 20191028091320.

First of all, I don’t like storing DateTime in numeric format like above, which brings so much overhead during dealing with the same field, like querying based on date, formatting for displaying etc.

The way I handled these conversions given below,

DateTime to Long

long.Parse(DateTime.UtcNow.ToString("yyyyMMddHHmmss"))

Long to DateTime

The below code can handle 14 digits (eg. 20191028091320), 16 digits (eg. 2019102809132000 ) and 8 digits (eg. 20191028).

private static DateTime? GetDateTimeFromInt(long? dateAsLong, bool hasTime = true)
{
    if (dateAsLong.HasValue && dateAsLong > 0)
    {
        if (hasTime)
        {
            // sometimes input is 14 digit and sometimes 16
            var numberOfDigits = (int)Math.Floor(Math.Log10(dateAsLong.Value) + 1);

            if (numberOfDigits > 14)
            {
                dateAsLong /= (int)Math.Pow(10, (numberOfDigits - 14));
            }
        }

        if (DateTime.TryParseExact(dateAsLong.ToString(), hasTime ? "yyyyMMddHHmmss" : "yyyyMMdd",
                                  CultureInfo.InvariantCulture,
                                  DateTimeStyles.None, out DateTime dt))
        {
            return dt;
        }
    }

    return null;
}

Gatsby - See All Generated Pages

Abhith Rajan — Sat, 05 Oct 2019 20:00:00 GMT

If you want to see the list of pages generated by Gatsby, you can use the GraphiQL playground part of the Gatsby Dev (usually served on http://localhost:8000/___graphql).

Run the below query and you will be presented with the list of pages,

query MyQuery {
  allSitePage {
    edges {
      node {
        path
      }
    }
  }
}

Azure - Configuring End-to-End SSL for IIS Website by using Application Gateway

Abhith Rajan — Tue, 24 Sep 2019 09:26:00 GMT

For most cases we do SSL offloading at the Application Gateway end. i.e, HTTPS traffic between public and application gateway, HTTP traffic between application gateway and websites hosted under it. But in some cases we need end to end SSL. In our case, it was for the Identity Server.

So lets start, We pointed the domain to the Application Gateway Public IP and We assume you purchased the SSL for your domain and we need the SSL certificate in two formats.

PFX format for application gateway HTTPS listener.
CER format for application gateway HTTP Settings.

Now we need to do the following,

Add an HTTPS listener for that domain with the Certificate.
Create a backend pool with the VM where we will publish the website in the IIS.
Add an HTTP Settings with the Certificate as shown in the picture.

Application Gateway HTTP Settings

Create a rule which connects the listener (1) and backend pool(2) via the newly created HTTP settings (3).

Once all this done, user will be able to hit Application Gateway with HTTPS and probably will see 404 since the VM not responding to the requests. So lets configure the VM to accept the traffic through 443 port. In the VM,

Install the SSL certificate.
Configure the HTTPS website in the IIS with that certificate and the hostname (domain).

After this done, check whether you are able to reach the website successfully. If you are getting any 502 errors, check the following

In the Application Gateway, check the unhealthy backend pools.
If our backend pool with the HTTPS HTTP settings name is in that unhealthy list, that indicates the VM in the backend pool not responding 200 back to the health probe check.
By default health probe check for HTTPS will sent health check requests to the VM’s IP on 443 port, and in the IIS, no website will be listening for 443 port (Without any hostname specified).
To overcome, all we have to do is, on the default website (if there is no default site, add another binding to our site) as shown below,

IIS default 443 port binding

Once this done, watch the unhealthy backend pools and you may notice that our backend pool is healthy now (It will take at least 30 seconds to update). Now application gateway will route traffic to our VM over HTTPS and the VM will serve the website and we are done.

Additional Resources

Quickstart - Configure end-to-end SSL encryption with Azure Application Gateway - Azure portal | Microsoft Docs

C# - Get Last N characters from a string

Abhith Rajan — Sun, 22 Sep 2019 19:26:00 GMT

In C#, Substring method is used to retrieve (as the name suggests) substring from a string. The same we can use to get the last ‘N’ characters of a string.

String.Substring Method

As described in String.Substring Method (System) | Microsoft Docs, Substring method has two overloads,

Overload	Description
Substring(Int32)	Retrieves a substring from this instance. The substring starts at a specified character position and continues to the end of the string.
Substring(Int32, Int32)	Retrieves a substring from this instance. The substring starts at a specified character position and has a specified length.

With the starting position only overload, we can do.

mystring.Substring(mystring.Length - N);

But the above code will fail in case mystring length is lower than the ‘N’. So considering that case, lets have an extension method,

public static string GetLast(this string source, int numberOfChars)
{
    if(numberOfChars >= source.Length)
      return source;
    return source.Substring(source.Length - numberOfChars);
}

And you can use the above like,

mystring.GetLast(5)

This extension method is now part of the Code.Library Nuget package.

ASP.NET Core - Return 500 (Internal Server Error) or any other Status Code from API

Abhith Rajan — Mon, 16 Sep 2019 19:26:00 GMT

HTTP response status codes have so much importance in REST API’s. In any case if you want to return a status code from your ASP.NET Core API, all you have to do is in your controller method,

return  StatusCode(StatusCodes.Status500InternalServerError);

StatusCode is from Microsoft.AspNetCore.Mvc.ControllerBase.StatusCode and StatusCodes from Microsoft.AspNetCore.Http.StatusCodes. The above code will return a 500 status code. Similar way, you can return any other status code. For the complete list,

👉 StatusCodes Class (Microsoft.AspNetCore.Http) | Microsoft Docs

HTTP status code 5xx indicates server error. 5xx series includes,

Status Code	Description	ASP.NET Core 2.2
500	Internal Server Error	Status500InternalServerError
501	Not Implemented	Status501NotImplemented
502	Bad Gateway	Status502BadGateway
503	Service Unavailable	Status503ServiceUnavailable
504	Gateway Timeout	Status504GatewayTimeout
505	HTTP Version Not Supported	Status505HttpVersionNotsupported
506	Variant Also Negotiates	Status506VariantAlsoNegotiates
507	Insufficient Storage	Status507InsufficientStorage
508	Loop Detected	Status508LoopDetected
510	Not Extended	Status510NotExtended
511	Network Authentication Required	Status511NetworkAuthenticationRequired
599	Network Connect Timeout Error

You can also return some additional data along with the status code. Here is an example,

[Route("{code}")]
[HttpGet]
[ProducesResponseType(typeof(Merchant), 200)]
public async Task<ActionResult> GetMerchant(string code)
{
    var input = new EntityRequestDto<string>
    {
        Id = code
    };

    var result = await _merchantService.GetMerchant(input);

    if (result.IsSuccess)
    {
        return Ok(result.Value);
    }

    _logger.LogError("FAILED: GetMerchant - ${result.Error}");
    return StatusCode(StatusCodes.Status500InternalServerError, result.Error);

}

If another service which is a client of the above API, they can get the additional info by,

public async Task<MerchantPreviewDto> GetMerchant(string merchantCode)
{
    try
    {
        var merchant = await $"{_apiBaseUri}/api/v1/merchants/{merchantCode}"
            .WithOAuthBearerToken(...)
            .GetJsonAsync<Merchant>().ConfigureAwait(false);

        return ObjectMapper.Map<MerchantPreviewDto>(merchant);
    }
    catch (FlurlHttpException ex)
    {
        var error = await ex.GetResponseStringAsync();
        throw new UserFriendlyException("Oops! There is a problem!", error);
    }
}

Bonus

On the client service, I am using Flurl, which is my favorite HTTP client.

Flurl is a modern, fluent, asynchronous, testable, portable, buzzword-laden URL builder and HTTP client library for .NET.

Which one is your favorite HTTP client library for .NET?

C# - Get Web.Config Connection String in a Class Library

Abhith Rajan — Sun, 15 Sep 2019 19:26:00 GMT

This is relatively old topic but looks like still relevant. You need to know why?

Story

Recently one of our vendor deployed an app in one of our environment and when tested, found that the some of the entries going to wrong DB. Checked the connection strings in Web.Config, everything was fine. Due to the lack of proper logging in the application, we planned to hook up Microsoft Azure Application Insights, and we did discovered something, thanks to AI.

Application Map in the #Azure #applicationinsights saved us.

We were able to catch a wrong connection string hardcoded in a vendor dll by checking the Application Map in the AI.@AppInsights 👏 👏 👏
— Abhith Rajan (@AbhithRajan) September 14, 2019

Hard coding something like Connection String is a BAD PRACTICE, things like that should come as a parameter to the class library methods. Since we are not in a position to rewrite everything, what we did was, we replaced the hard coded connection string with the actual connection string value from the web.config.

using  System.Configuration;
...

public static string ConnectionString
{
  get { return ConfigurationManager.ConnectionStrings["Default"].ConnectionString; }
}

In the above code, Default is the connection string name in the web.config.

Using AutoMapper in a .NET Core Class Library

Abhith Rajan — Wed, 11 Sep 2019 12:59:27 GMT

AutoMapper is a handy helper library to do mapping between objects. In most scenarios, you might have an Entity Class and you want to map it to some DTO class or vice versa. With minimal configuration, AutoMapper can do this mapping without you creating a method where you map each field one by one.

Suppose I have to map Entity class to EntityDto and they look like,

public class Entity
{
  public string Name {get;set;}
  public Guid Id {get;set;}
  public string CurrencyCode {get;set;}
  public decimal CurrencyValue {get;set;}
}

public class EntityDto
{
  public string Name {get;set;}
  public Guid Id {get;set;}
  public Currency Currency {get;set;}
}

public class Currency
{
  public string Code {get;set;}
  public string Value {get;set;}
}

using AutoMapper;
...

public class SomeService : ISomeService
{
  private IMapper _iMapper;

  public SomeService()
  {
    // One time configuration
      var config = new MapperConfiguration(cfg =>
      {
          cfg.CreateMap<Entity, EntityDto>().ForMember(dto => dto.Currency, map => map.MapFrom(source => new Currency
          {
              Code = source.CurrencyCode,
              Value = source.CurrencyValue.ToString("0.00")
          }));
      });

      _iMapper = config.CreateMapper();
  }
...
}


public async Task<DtoClass> GetAsync(Guid id)
{
    Entity entity = await GetEntityByIdAsync(id);

    // using auto mapper to do the mapping
    var dto = _iMapper.Map<DtoClass>(entity);

    return dto;
}

In the mapping configuration, we didn’t specify anything related Id, Name properties. AutoMapper will map them automatically since the source and destination property name matches for those.

In the above, we defined the mappings in the same class, a better solution would be a common place where we define all the mappings. Let’s do that,

public static class ObjectMapper
{
    private static readonly Lazy<IMapper> Lazy = new Lazy<IMapper>(() =>
    {
        var config = new MapperConfiguration(cfg =>
        {
            // This line ensures that internal properties are also mapped over.
            cfg.ShouldMapProperty = p => p.GetMethod.IsPublic || p.GetMethod.IsAssembly;
            cfg.AddProfile<CustomDtoMapper>();
        });
        var mapper = config.CreateMapper();
        return mapper;
    });

    public static IMapper Mapper => Lazy.Value;
}

All the mapping configuration goes to,

public class CustomDtoMapper : Profile
{
    public CustomDtoMapper()
    {
        CreateMap<Entity, EntityDto>().ForMember(dto => dto.Currency, map => map.MapFrom(source => new Currency
          {
              Code = source.CurrencyCode,
              Value = source.CurrencyValue.ToString("0.00")
          }));

        // All other mappings goes here
    }
}

And use the ObjectMapper where ever you wanted,

var dto = ObjectMapper.Mapper.Map<DtoClass>(entity);

Read more about the AutoMapper configuration on the official website of AutoMapper.

Additional Resources

Lazy Class (System) | Microsoft Docs

Changelog

2020-03-15

Added common configuration model (ObjectMapper).

Dapper - Execute Multiple Stored Procedures

Abhith Rajan — Thu, 05 Sep 2019 12:59:27 GMT

Dapper is my preferred ORM for performance constraint applications. It is one of the best solution in case of a database first approach as well as when there is already a database in place. Dapper works with SQL Server as well as so many other DB’s and I have tried Oracle DB apart from SQL Server.

Dapper have so many features/extension methods, the documentation of these features can be accessed here.

👉 Dapper Dapper Tutorial | Dapper Tutorial and Documentation .

To execute multiple stored procedures or multiple SQL statements, we can use the QueryMultipleAsync extension method. See the sample code below,

using Dapper;
...

public async Task<SomeClass> GetAsync(Guid someId, Guid anotherId)
{
    using (var connection = new SqlConnection(_settings.ConnectionString()))
    {
        connection.Open();

        const string sql = @"exec uspTableOneAction @someId;
                             exec uspTableTwoAnotherAction @anotherId;";

        using (var multi = await connection.QueryMultipleAsync(sql, new { someId, anotherId }))
        {
            var firstTableItem = multi.Read<TableOneItem>().First();
            var secondTableItem = multi.Read<TableTwoItem>().FirstOrDefault();
           ...
        }

        ...
    }
}

In the above code, my stored procedure are,

uspTableOneAction
uspTableTwoAnotherAction

The naming convention which we follows for stored procedures are,

usp{TableName}{Action}

eg: uspUserInsert, uspUserUpdate

And in the above example, first stored procedure expects a parameter @someId. And the second stored procedure expects @anotherId. We passed these args by

new { someId, anotherId }

The same can be written as,

new { someId = someId, anotherId = anotherId }

Why we omitted the right hand side part in the example is because of the arguments name matching the parameter.

Just in case if both stored procedures expecting same parameter, say @someId. Then our args will look like,

new { someId }

That’s it. You have the result of the two queries.

Additional Resources

IIS - Disable CORS

Abhith Rajan — Wed, 14 Aug 2019 16:54:00 GMT

For any reason you wish to disable CORS for any website hosted on IIS, one way you can do this by allowing all origins. To do that,

Make sure you installed IIS CORS Module on the server.
Update the Web.Config of the website to have the cors section as given below,

Note: code tested on IIS 10

<?xml version="1.0" encoding="UTF-8"?>
<configuration>
    <system.webServer>
        <cors enabled="true" failUnlistedOrigins="true">
          <add origin="*">
            <allowHeaders allowAllRequestedHeaders="true" />
          </add>
        </cors>
    </system.webServer>
</configuration>

As you can see, we are allowing all origin’s by specifying * as the origin.

After just allowing all origins alone, if you encounter error like,

Request header field content-type is not allowed by Access-Control-Allow-Headers in preflight response.

To solve that, we are setting allowAllRequestedHeaders="true" in the allowHeaders for all the origins.

Remember: CORS is a security feature. Disable only if the resource is totally public.

Additional Resources

CORS Module Configuration Reference | Microsoft Docs

Azure Web App - Web Deploy to a Sub-folder

Abhith Rajan — Tue, 30 Jul 2019 15:11:27 GMT

I always recommends Microsoft Azure. Recently, We needed to deploy an ASP.NET Core app to a sub-folder inside Azure Web App (App Service).

By default Azure provides publish profiles for the Web App (To put files to the root of the domain). In our case, We wanted to deploy inside a sub-folder.

Suppose our web app name was demo-site and we want to web deploy the app under demo-site\folder\sub-folder. To do that,

Download the publish profile
Import in Visual Studio
Edit the web-deploy profile(Normally the publish profile will have Web Deploy as well as FTP profile)
- Change Site Name from demo-site to demo-site\folder\sub-folder
- Change the Destination URL from http://demo-site.azurewebsites.net to http://demo-site.azurewebsites.net/folder/sub-folder
Publish

But the publish failed with error,

System.TypeLoadException: Method ‘get_Settings’ in type ‘Microsoft.Web.LibraryManager.Build.HostInteraction’ from assembly ‘Microsoft.Web.LibraryManager.Build

This is specific to our application and may not have anything to do with your app, so can skip this part.

After referring a similar issue posted in GitHub aspnet/LibraryManager, We resolved the issue by updating Microsoft.Web.LibraryManager.Build Nuget package in my project.

Now next error,

Error - Web Deployment failed ERROR_NOT_SUPPORTED

More details about web deployment error can be found from here. In order to solve that,

Go to portal > demo-site App Service > Configuration > Path Mappings > Virtual applications and directories. And add the following,

Virtual path	Physical Path	Type
/folder	site\wwwroot\folder	Folder
/folder/sub-folder	site\wwwroot\folder\sub-folder	Application

Now publish from Visual Studio. If you only need to publish to a first level folder, i.e to demo-site\folder, then all you have to do is, change the Type to Application in the Path mappings for /folder, and skip the sub-folder entry since you don’t need it. And correct the Site Name and Destination URL in the publish profile accordingly.

Azure Web App - Missing MIME types

Abhith Rajan — Thu, 11 Jul 2019 12:59:27 GMT

When you deploy an app to Azure app service (Web App), you may notice that the custom font isn’t showing up or some specific files is getting 404 even if they exist on the right place.

This is because of the missing Mime type definitions.

Mime Types

Its definition is,

A media type (also known as a Multipurpose Internet Mail Extensions or MIME type) is a standard that indicates the nature and format of a document, file, or assortment of bytes.

And if you want to know why it is causing the 404, read this article.

Missing Mime Types

Font Files

  <mimeMap fileExtension="woff" mimeType="application/font-woff" />
  <mimeMap fileExtension="woff2" mimeType="application/font-woff" />

Json

  <mimeMap fileExtension=".json" mimeType="application/json" />

SVG

  <remove fileExtension=".svg"/>
  <mimeMap fileExtension=".svg" mimeType="image/svg+xml" />

  <mimeMap fileExtension=".webmanifest" mimeType="application/manifest+json" />
  <mimeMap fileExtension=".webp" mimeType="image/webp" /> <!-- Not all browser supports webp />

Sample web.config file

<?xml version="1.0" encoding="UTF-8"?>
<configuration>
    <system.webServer>
        <staticContent>
            <remove fileExtension="woff"/>
            <mimeMap fileExtension="woff" mimeType="application/font-woff" />
            <remove fileExtension="woff2"/>
            <mimeMap fileExtension="woff2" mimeType="application/font-woff" />
            <remove fileExtension=".json"/>
            <mimeMap fileExtension=".json" mimeType="application/json" />
            <remove fileExtension=".svg"/>
            <mimeMap fileExtension=".svg" mimeType="image/svg+xml" />
         </staticContent>
    </system.webServer>
</configuration>

Curious about remove fileExtension ? MimeMap collection inherits but it won’t allow duplicate entries. If a duplicate mimeMap definition is added, it will result,

IIS7 Error: cannot add duplicate collection entry of type ‘mimeMap’ with unique key attribute ‘fileExtension’

So for the safe side, we are removing the global definition for that extension before changing it.

Best Laptop for Programming Under $2500 Dollars 2019

Abhith Rajan — Sat, 08 Jun 2019 22:50:00 GMT

For the past month, I have been in the search for the best laptop, that will help me to do things faster. Best in the sense which worth every penny and yet strong in specs, not a high-end gaming machine.

History

Laptops I used earlier are,

Dell Inspiron (Exact series don’t remember)
HP Pro Book
Lenovo Y520 (256 SSD)
Lenovo G40

In the above list, none of them owned by myself, first 3 out of the 4 was given by my ex-company during the employment period and the other is rented out from my brother-in-law. Out of all Lenovo Y520 was my favorite just because of its specs and speed.

Minimum Requirements

Right now its time to own a machine, and the minimum features am looking for are,

Thunderbolt 3 support.
SSD
Intel i7-9750H processor
RTX 2060 Graphics card
Fingerprint sensor

Shortlisted Laptops

The main YouTube channels which I rely for laptop reviews are,

And After watching tons of reviews and all, my shortlisted items are,

AERO 15

Model Name: WA-7US5130SP Official Website

On the pro side, it ticks all my requirement 5/5.

Cons

Web Camera position
Thermal throttling

youtube: https://www.youtube.com/watch?v=W64J7_Md1I4

Purchase AERO 15

Amazon

ThinkPad X1 Extreme (2nd Gen) (15.6”) Laptop

Official Website

Acer Predator Triton 500

Model Name: PT515-51-73EG Official Website

Watch the reviews,

youtube: https://www.youtube.com/watch?v=pXkE0BlTBmI youtube: https://www.youtube.com/watch?v=rOCiNSZIhmM youtube: https://www.youtube.com/watch?v=qwGTHLeG2dY

On the pro side, it matches my requirement 4/5. But on the con side, it is not easily upgradable because of its inverted motherboard construction. See the below video,

youtube: https://www.youtube.com/watch?v=6-PmEaytJOA

Other than that, Only zoned RGB keyboard, NO per key RGB.

Couldn’t find the exact one in Amazon but similar spec with intel 8750H found here.

HP Omen 15 (Latest)

Check out the below review.

youtube: https://www.youtube.com/watch?v=EloUckKqAKc

Its the cheapest in this list. And it got a bit more gaming look.

Any Refurbished WORKSTATION

To save some bucks.

Skipped Laptops

Acer Helios 300 (2019) - Lack of thunderbolt support.
Lenovo Y740 - Big Charger and small battery.
Asus ROG Zephyrus S GX701 - Lack of thunderbolt support.
ASUS ROG Zephyrus S GX502GW - Lack of thunderbolt support.

2019 Laptop Awards

youtube: https://www.youtube.com/watch?v=O8zcB_EcAyY youtube: https://www.youtube.com/watch?v=icRz80kKZBA

Conclusion

And my search ended when I found a good deal on Dell Precision 7720. Even though it only ticked 2 out of my 5 requirements, am pretty satisfied with it.

Create SQL Server Database From a Script in Docker-Compose

Abhith Rajan — Tue, 21 May 2019 14:55:00 GMT

In one of our project which using Microsoft Orleans, We used ADO.NET clustering with SQL Server. When we tried to containerize the application suite, we need a SQL Server instance in a container (for development environment) with a database initialized with some tables and all.

Note: The below setup is not recommended for a PRODUCTION environment.

For that solution, our docker-compose.yml looks like below,

version: '3.4'

services:
  other.service:
    ...
    environment:
      - ConnectionString=Data Source=CONTAINER_NAME_OF_db;Initial Catalog=DB_NAME;User Id=sa;Password=Your_password123;Pooling=False;Max Pool Size=200;MultipleActiveResultSets=True
    depends_on:
      - db
  db:
    image: my/db/image/name
    build:
      context: .
      dockerfile: data/Dockerfile
    # ports:
    #   - "1444:1433"
    volumes:
     - mssqldata:/var/opt/mssql

volumes:
  mssqldata:

In the above, the other.service can be anything (WEB App/Console App etc) that requires the sql server db instance. We marked the dependency between these services properly using depends_on in the docker-compose.yml.

The volumes is specified in the docker-compose.yml in-order to persist the SQL Server data.

In the same directory where docker-compose.yml is, there is a data folder which contains following,

Dockerfile
entrypoint.sh
setup.sql

Dockerfile Which basically get the latest image of mcr.microsoft.com/mssql/server (linux) and configures the SQL Server container.

FROM mcr.microsoft.com/mssql/server

ENV ACCEPT_EULA=Y
ENV SA_PASSWORD=Your_password123

COPY ./data /

ENTRYPOINT [ "/bin/bash", "entrypoint.sh" ]
CMD [ "/opt/mssql/bin/sqlservr" ]

Entry point in the Dockerfile is pointed to a custom shell entrypoint.sh which is,

#!/bin/bash
set -e

if [ "$1" = '/opt/mssql/bin/sqlservr' ]; then
  # If this is the container's first run, initialize the application database
  if [ ! -f /tmp/app-initialized ]; then
    # Initialize the application database asynchronously in a background process. This allows a) the SQL Server process to be the main process in the container, which allows graceful shutdown and other goodies, and b) us to only start the SQL Server process once, as opposed to starting, stopping, then starting it again.
    function initialize_app_database() {
      # Wait a bit for SQL Server to start. SQL Server's process doesn't provide a clever way to check if it's up or not, and it needs to be up before we can import the application database
      sleep 15s

      #run the setup script to create the DB and the schema in the DB
      /opt/mssql-tools/bin/sqlcmd -S localhost -U sa -P Your_password123 -d master -i setup.sql

      # Note that the container has been initialized so future starts won't wipe changes to the data
      touch /tmp/app-initialized
    }
    initialize_app_database &
  fi
fi

exec "$@"

And the setup.sql contains all the SQL statements to generate the DB and all. An example given below,

CREATE DATABASE DB_NAME;
GO
USE DB_NAME;

...

Microsoft SQL Server Guy Trying Oracle Database

Abhith Rajan — Wed, 08 May 2019 15:11:27 GMT

I have worked with Microsoft SQL Server for the past 7 years and still enjoying working with SQL Server. Now I got a chance to explore Oracle Database and here is my experience.

Noticed Differences

There is NO DATABASE NAME, instead there is this user or schema wise separation.
Preferred IDE is Oracle SQL Developer
After executing query, you need to commit in order to retain the changes.
SELECT TOP N is not that easy compared to SQL Server.

Docs

Tablespaces

An Oracle database consists of one or more logical storage units called tablespaces, which collectively store all of the database’s data.

VARCHAR vs VARCHAR2

VARCHAR is reserved by Oracle to support distinction between NULL and empty string in future, as ANSI standard prescribes.

VARCHAR2 does not distinguish between a NULL and empty string, and never will.

If you rely on empty string and NULL being the same thing, you should use VARCHAR2.

Problems

Show Line Numbers in Oracle SQL Developer

This can be done easily by right clicking the line number section in the IDE, then Toggle Line Numbers.

Create User

CREATE USER <UserName> IDENTIFIED BY <Password>;
GRANT CONNECT , RESOURCE, DBA TO <UserName>;

VARCHAR to BLOB

We can use the inbuilt function utl_raw.cast_to_raw which transforms a varchar2 into a BLOB value,

utl_raw.cast_to_raw('SOME_STRING')

Update BLOB via Query

Update TABLE_NAME SET COL_NAME = utl_raw.cast_to_raw('SOME_STRING');

Oracle localhost connection string for an ASP.NET CORE & Dapper project

private readonly string ConnectionString = @"Data Source = (DESCRIPTION = (ADDRESS_LIST = (ADDRESS = (PROTOCOL = TCP)(HOST = localhost)(PORT = 1521)))(CONNECT_DATA = (SERVER = DEDICATED)(SERVICE_NAME = XE))); user id = YOUR_USER_ID; Password= YOUR_USER_PASSWORD";

using (var dbConn = new OracleConnection(ConnectionString))
{
  ...
}

Replace YOUR_USER_ID and YOUR_USER_PASSWORD with appropriate values.

ORA-00933: SQL command not properly ended

This is happened when my query contained semicolon ”;” at the end. Removing the semicolon ”;” from the end of the query solved the problem.

Conclusion

I just started exploring Oracle Database, so this post can be considered as a draft for now, more coming.

Docker - SQL Error on ASP.NET Core Alpine

Abhith Rajan — Tue, 07 May 2019 09:26:00 GMT

My ASP.NET Core app was running fine on mcr.microsoft.com/dotnet/core/runtime:2.2-stretch-slim on my linux container. And the build image size was 195MB. In order to reduce the file size, I switched to mcr.microsoft.com/dotnet/core/sdk:2.2.203-alpine3.9 and the new image size was 98.6MB. WOW !!!

But then there came this problem,

System.InvalidOperationException: Internal connection fatal error

The SQL client was having trouble in this build. After the regular research, find out that in the alpine image

The microsoft/dotnet Alpine images are purposefully configured to use the .NET Core 2.0 Globalization Invariant Mode. This is done to reduce the size of the image substantially. It seems as though the SqlClient does not support running with the Globalization Invariant Mode.

If you are thinking what is .NET Core 2.0 Globalization Invariant Mode,

.NET Core 2.0 Globalization Invariant Mode enables you to remove application dependencies on globalization data and globalization behavior. This mode is an opt-in feature that provides more flexibility if you care more about reducing dependencies and the size of distribution than globalization functionality or globalization-correctness.

But there is a small catch,

The drawback of running in the invariant mode is applications will get poor globalization support.

So solution for this was to add the following in Dockerfile just before the ENTRYPOINT,

RUN apk add --no-cache icu-libs
ENV DOTNET_SYSTEM_GLOBALIZATION_INVARIANT=false

In the first step, we are adding the icu-libs Alpine linux package, which is the international components for unicode library (libraries) and in the next step, we are setting DOTNET_SYSTEM_GLOBALIZATION_INVARIANT to false to enable the globalization support.

Additional Resources

.NET Core 2.0 Globalization Invariant Mode

Gatsby - Cookbook

Abhith Rajan — Fri, 26 Apr 2019 20:00:00 GMT

I 💟 Gatsby. This article contains some recipes which will be useful if you are into Gatsby for your next static site. Recipes are,

Using Google fonts in Gatsby
Using icons like Font Awesome in Gatsby
Disqus integration in a Gatsby website

Using Google fonts in Gatsby

Import the font css on your site main css file.

Using gatsby-plugin-prefetch-google-fonts.

Install the plugin gatsby-plugin-prefetch-google-fonts

yarn add gatsby-plugin-prefetch-google-fonts

Modify gatsby-config.js. Inside the plugins array, add

{
  resolve: `gatsby-plugin-prefetch-google-fonts`,
  options: {
    fonts: [
      {
        family: `Oswald`,
        subsets: [`latin`],
      },
      {
        family: `Open Sans`,
        variants: [`400`, `700`]
      },
    ],
  },
}

You can change the font family and variants or number of fonts in the above configuration according to your website needs.

Using icons like Font Awesome in Gatsby

Using react-icons.

Add it to your packages.

npm install react-icons --save

And in your react component

import { FaBeer } from "react-icons/fa";

class Question extends React.Component {
  render() {
    return (
      <h3>
        {" "}
        Lets go for a <FaBeer />?{" "}
      </h3>
    );
  }
}

Disqus integration in a Gatsby website

Since Gatsby is built on top of React, we can use the React package for Disqus.

yarn add disqus-react

And in your post component

import { DiscussionEmbed } from "disqus-react";

Define the Disqus configuration like

const disqusConfig = {
  shortname: `yourDisqusShortName`,
  config: { identifier: slug, title },
};

Here identifier is the unique id for your post, it can be anything.

And now you can render the Disqus section by adding

<DiscussionEmbed {...disqusConfig} />

More recipes will be added as I explore Gatsby.

Abhith Rajan — everything

Cost Optimization: Auto Shutdown and Startup of Tagged VMs with Azure Automation

The Significance of Cost Optimization#

Azure Automation: Streamlining Cost Optimization#

Implementing VM Shutdown with Azure Automation#

Implementing VM Startup with Azure Automation#

Additional Considerations#

Decode JWT Token

Method#

Usage#

Output Environment Variable

Delete All Local/Remote Git Tags

Recommended Steps#

1. Delete all local tags#

2. Fetch all remote tags#

3. Delete All remote tags#

4. Delete All local tags#

Unnecessary Details#

How We Use Jira Automation

Workflow#

Automation rules#

1. Auto-assign to whoever created the issue#

2. Close duplicate issues#

3. Link issues that are mentioned in the comments#

4. When a commit is created and the status is ‘To do’ → then transition this issue to ‘In Dev’#

5. On Pull Request created → comment the details#

6. On Pull Request merged → comment the details#

7. Comment on Ready for QA items if not assigned to a QA#

8. Assign issue back to Developer once Ready To Deploy#

9. Assign issue back to Developer once Done#

10. When issue is resolved → comment on related issues#

11. Label issues as stale that have not had recent interaction#

Is closing stale issues really a good idea?#

Planned Improvements#

Final Thoughts#

Additional Resources#

GitHub vs Azure DevOps - For Code

Dependabot#

Code Scanning#

Marketplace and integrations#

JIRA#

GitHub Checks#

Release Notes#

Azure Pipelines vs GitHub Actions#

Mobile Application#

Powerful web-based editor (VS Code on Web)#

And#

Pricing#

Code Migration From Azure DevOps to GitHub#

Change Azure Pipeline Repository From Azure DevOps to GitHub#

Final Notes#

Additional Resources#

Changelog#

2022-07-20#

2022-07-13#

2022-06-07#

2021 Year In Review

More Azure#

Rest of the work#

Pure Work#

Plans for 2022#

Create New Database Level User

Additional Resources#

ASP.NET Core - OAuth 2.0 Client Access Token Management

Additional Resources#

UTC to UAE Time (Arabian Standard Time)

Azure Cognitive Search - GeographyPoint to GeoJSON in .NET 5

Additional Resources#

NDepend - The Best-in-class Static Code Analyzer for Dotnet

Initial Setup#

Dashboard#

Azure Logic Apps - Get Application Insights/Azure Monitor Logs Query Result on Microsoft Teams

Adaptive Cards#

Simple card#

Card with a list of items#

Logic App#

1. Recurrence trigger#

2. Getting a query result#

Query to get performance metrics from Application insights#

Query to get failed requests summary from Application Gateway#

The Significance of Cost Optimization

Azure Automation: Streamlining Cost Optimization

Implementing VM Shutdown with Azure Automation

Implementing VM Startup with Azure Automation

Additional Considerations

Method

Usage

Recommended Steps

1. Delete all local tags

2. Fetch all remote tags

3. Delete All remote tags

4. Delete All local tags

Unnecessary Details

Workflow

Automation rules

1. Auto-assign to whoever created the issue

2. Close duplicate issues

3. Link issues that are mentioned in the comments

4. When a commit is created and the status is ‘To do’ → then transition this issue to ‘In Dev’

5. On Pull Request created → comment the details

6. On Pull Request merged → comment the details

7. Comment on Ready for QA items if not assigned to a QA

8. Assign issue back to Developer once Ready To Deploy

9. Assign issue back to Developer once Done

10. When issue is resolved → comment on related issues

11. Label issues as stale that have not had recent interaction

Is closing stale issues really a good idea?

Planned Improvements

Final Thoughts

Additional Resources

Dependabot

Code Scanning

Marketplace and integrations

JIRA

GitHub Checks

Release Notes

Azure Pipelines vs GitHub Actions

Mobile Application

Powerful web-based editor (VS Code on Web)

And

Pricing

Code Migration From Azure DevOps to GitHub

Change Azure Pipeline Repository From Azure DevOps to GitHub

Final Notes

Additional Resources

Changelog

2022-07-20

2022-07-13

2022-06-07

More Azure

Rest of the work

Pure Work

Plans for 2022

Additional Resources

Additional Resources

Additional Resources

Initial Setup

Dashboard

Adaptive Cards

Simple card

Card with a list of items

Logic App

1. Recurrence trigger

2. Getting a query result

Query to get performance metrics from Application insights

Query to get failed requests summary from Application Gateway

3. Initialize an array variable

4. For each row in the query result, we map the result item to an Adaptive card row.

5. Append the rows to the rest of the Adaptive card schema in the Teams connector channel.

Integrating Microsoft Clarity with GatsbyJS website

Official plugin

Legacy way of installing [DEPRECATED]

Dashboard

Recordings

Heatmaps

Additional Resources

Sample Output

List of failed requests

Tests

Reference