Azure Interview Questions

False. Azure does not support this.

• Azure Cloud Service is a classic example of a platform as a service (PaaS). This was designed to support those applications which demand high scalability, reliability, and availability all within the constraints of reduced cost of operations. These are hosted on virtual VMs and Azure provides more control over them by letting the developers install the necessary software and enabling them to control remotely.
• Azure cloud services are used for deploying multi-tier web-based applications in Azure by means of creating an instance of cloud service. It is also possible to define multiple roles such as web roles, worker roles, etc for the purpose of distributed processing. Azure cloud services help in the easier and flexible scalability of the application.
• Each role of the cloud service has its own purpose and thereby its own configuration and application files.

There are 3 models available for cloud deployment:

• Public Cloud: In this model, the cloud infrastructure is owned publicly by the cloud provider and there are chances that the server resources could be shared between multiple applications.
• Private Cloud: Here, the cloud infrastructure is owned exclusively by us or exclusive service is provided by the cloud provider to us.
  • This includes hosting our applications on our own on-premise servers or hosting the application on a dedicated server provided by the cloud provider.
• Hybrid Cloud: As the name itself says, this model is the hybrid combination of private cloud and the public cloud.
  • This might include the scenario of using on-premise servers for processing confidential, sensitive data and using public cloud features for hosting public-facing applications.

Here, we use the best of both worlds to our requirements and advantage.

A role instance is nothing but a virtual machine where the application code runs with the help of running role configurations. There can also be multiple instances of a role as per the definition in the cloud service configuration files.

Cloud service roles comprise a set of application and configuration files. There are 2 kinds of roles provided by Azure:

• Web role: This provides a dedicated web server belonging to IIS (Internet Information Services) that is used for automatic deployment and hosting of front-end websites.
• Worker role: These roles help the applications hosted within them to run asynchronously for longer durations and are independent of the user interactions and generally do not use IIS. They are also ideal for performing background processes. The applications are run in a standalone manner.

• Azure Diagnostics API helps us collect diagnostic data such as performance monitoring, system event logs, etc from the applications that are running on Azure.
• For the verbose monitoring of the data, Azure Diagnostics has to be enabled for the cloud service roles.
• The diagnostics data can be used for building visual chart representations for better monitoring and also for creating performance metric alerts.

• The Azure SLA is a contract that ensures or guarantees that when two or more role instances of a role are deployed on Azure, access to that cloud service is guaranteed for at least 99.95% of the time.
• It also states that if the role instance process is not in the running state, then the detection of such processes and corrective action for the same will be taken 99.9% percent of the time.
• If the mentioned guarantees are not satisfied at any point in time, then Azure credits a percentage of monthly fees to us depending on the pricing model of the respective Azure services.

Azure Resource Manager is a service provided by Azure to provide management and application deployment in Azure.

The resource manager provides the management layer that helps the developer to create, modify or delete the resources in the Azure subscription account. This feature comes in handy when we have requirements like managing access controls, locks, ensuring the security of the resources post-deployment, and organization of those resources.

NSG stands for Network Security Group that has a list of ACL (Access Control List) rules which either allows/denies network traffic to subnets or NICs (Network Interface Card) connected to a subnet or both. When NSG is linked with a subnet, then the ACL rules are applied to all the Virtual Machines in that subnet. 

Restrictions of traffic to individual NIC can be done by associating NSG directly to that NIC.

• Cloud computing refers to the usage of computing resources (servers) on the internet (refers to the term cloud) for the purpose of storing, managing, analyzing, and processing the data. Here, instead of maintaining our own servers, we use the infrastructure provided and maintained by third-party vendors such as Microsoft, AWS, etc, and pay them based on the server usage time duration.
• Cloud computing enhances the speed of execution, ensures flexibility of resources, and easier scalability.
• Cloud computing can be used to attain high fault tolerance and high system availability and this can also be done dynamically as per the infrastructural requirements of the application.

• Azure Functions service can be used for executing the code without a server.
• Serverless Azure Functions are used for simplifying complex orchestration and challenging resolutions. They are meant for being stateless and short-lived.
• They help to connect with other services without the need for hard coding of the integrations thereby making the development process faster.
• It helps the developer to write and concentrate on the business logic code thereby saving time and effort.
• They also provide the features of monitoring and analyzing code performance by means of Azure Application Insights that help in identifying bottlenecks and failure points across the components of the application.

• Azure Blob storage is the object storage solution provided by Microsoft for the cloud. Blob stands for “Binary Large Object”. Blob-based storage is used to store massive unstructured data in terms of text or binary format. It is ideal for serving documents/images/audio/video/text directly to browser.
• The data stored in the blob storage is accessible from anywhere in the world. The blobs are tied to user accounts by grouping them into containers. The Azure Blob Service has 3 components:
  • Storage Account: This can be a General Storage Account or Blob Storage Account registered in Microsoft Azure.
  • Container: Container is used for grouping blobs. We can store an unlimited number of blobs in a container. The name of the container should start in lowercase.
  • Blob: A blob is a Binary Large Object like a file or document of any type and size. There are 3 kinds of Blobs supported by Azure:
    • Block blobs: These are intended for text and binary files and can support up to 195GB, i.e up to 50k blocks of up to 4MB each.
    • Append blobs: These are used for appending operations like logging data in log files.
    • Page blobs: These are meant for frequent read/write operations.

No! As the name itself says, Azure Internal Load Balancer supports only Private IP addresses, and hence the assignment of a public IP address or DNS name is not possible.

In case of maximum failed attempts, the azure account would get locked and the method of locking is dependent on the protocol that analyzes the entered password and the IP addresses of the login requests.

The main difference  between Azure Scale Sets and Availability Sets are given below:

Yes, it is possible by making use of the Key Vault mapping to any Admin VM, we can log in to another VM without the need for a password.

Azure provides a service called Azure File System which is used as a common repository system for sharing the data across the Virtual Machines configured by making use of protocols like SMB, FTPS, NFS, etc.

• It is an open-source, in-memory Redis cache system provided and maintained by Azure.
• It helps the web applications to improve the performance by fetching data from the backend database and storing it into the Redis cache for the first request and then fetching data from the Redis cache for all subsequent requests.
• Azure Redis Cache provides powerful and secure caching mechanisms by making use of the Azure cloud.

• These are the Azure computation resources that can be used to deploy and manage sets of identical Virtual Machines (VMs).
• These scale sets are configured in the same manner and are designed to support the autoscaling of the applications without the need for pre-provisioning of the VMs.
• They help to build large-scale applications targeting big data and containerized workloads in an easier manner.

• Availability Set is nothing but a logical grouping of VMs (Virtual Machines) that allows Azure cloud to understand how the application was developed for providing availability and redundancy.
• Each VM in the availability set is assigned 2 kinds of domains by Azure:
  • Fault Domain: These define the grouping of VMs that would share a common power source and common network switch. The VMs within availability sets are separated across up to 3 fault domains by default. This separation of VMs in fault domains helps our applications to be available by reducing impacts of network outages, power interruptions, and certain hardware failures.
  • Update Domain: These indicate the grouping of VMs and underlying hardware which are eligible to be rebooted at the same time. Only one update domain can be rebooted at a time, however, the order of reboot does not proceed in a sequential manner. Before the maintenance of another update domain, the previously rebooted domain is given a recovery time of 30 minutes to ensure that the domain is up.
• Azure provides flexibility to configure up to 3 fault domains and 20 update domains for an availability set.

Azure provides two deployment environments, they are:

• Staging Environment: This environment is used for validating the changes of our application before making them live into the main environment.
  • Here, the application is identified by means of GUID (Globally Unique Identifier) of Azure which has the URL as: GUID.cloudapp.net
• Production Environment: This is the main environment where our application goes live and can be accessed by the target audience which can be accessed by means of DNS friendly URL: appName.cloudapp.net

The following steps need to be performed when the drive failure occurs:

• To make sure that the Azure Storage functions without fail, we need to ensure that the drive is not mounted.
• Replace the drive so that the drive gets remounted and formatted.

Yes, it is possible and is done by means of the Transient Fault Handling Block. There can be multiple causes of transient failures while using the cloud environment:

• Due to the presence of more load balancers, we can see that the application to database connections fail periodically.
• While using multi-tenant services, the calls get slower and eventually time out because other applications are using resources to hit the same resource heavily.
• The last cause can be we ourselves as the user trying to hit the resource very frequently which causes the service to deliberately deny the connection to us to support other tenants in the architecture.

Instead of showing errors to the user periodically, the application can recognize the errors that are transient and automatically try to perform the same operation again typically after some seconds with the hope of establishing the connection. By making use of the Transient Fault Handling Application Block mechanism, we can generate the retry intervals and make the application perform retries. In the majority of the cases, the error would be resolved on the second try and hence the user need not be made aware of these errors unnecessarily.

Following is the sample code that can be used for the retry policy. Here, if the connection is not successful, then the action is retried based on the retry policy defined. There are 3 retry strategies - Fixed Interval, Incremental Interval, Exponential Backoff Strategy.

/***
* Class to detect Transient Blocks - Here
* OperationCancelledException is 
* detected and then the retry strategy is employed.
*/
internal class AppTransientDetection : ITransientErrorDetectionStrategy
{
    bool IsTransient(Exception exception) => 
        exception is OperationCanceledException;
}

/***
* Retry Strategy - Here Fixed Interval Strategy is employed and is retried for 5 times.
*/
RetryStrategy retryStrategy = new FixedInterval(retryCount: 5, retryInterval: TimeSpan.FromSeconds(2));

RetryPolicy retryPolicy = new RetryPolicy(new AppTransientDetection(), retryStrategy);
retryPolicy.ExecuteAction(() => { 
    try { 
        string commandText = @”USE FEDERATION User_Federation(ShardId =” + shardId + “) WITH RESET, FILTERING=ON”; 
        userEntity.Connection.Open();
        userEntity.ExecuteStoreCommand(commandText); 
    } catch (Exception e) {
        userEntity.Connection.Close();
        SqlConnection.ClearAllPools(); 
    } 
});

• Azure storage key is used for authentication for validating access for the azure storage service to control access of data based on the project requirements.
• 2 types of storage keys are given for the authentication purpose -
  • Primary Access Key
  • Secondary Access Key
• The main purpose of the secondary access key is for avoiding downtime of the website or application.

It is a command-line tool that is used for generating service package files. The tool also helps in preparing the application for deployment in Microsoft Azure or compute emulator.

Every project of cloud service type has the .cscfg file which is basically the cloud service configuration file that is generated by means of cspack tool and is primarily used to store:

• The number of role instances for the deployment of each role in the project.
• The thumbprint of the certificates.
• User-defined configuration and settings.

Azure Scheduler helps us to invoke certain background trigger events or activities like calling HTTP/S endpoints or to present a message on the queue on any schedule.

By using this Azure Schedule, the jobs present in the cloud call services present within and outside of the Azure to execute those jobs on-demand that are routinely on a repeated regular schedule or start those jobs at a future specified date.

Yes, it can be done by defining a separate Load Balancer Role in Azure.

Yes, it is totally possible by making use of the Site Recovery Service provided by Azure. It is capable of handling fail-over and fail-back scenarios between On-Prem Servers and Azure environments.

This issue can be stopped by making use of VM Scale sets by defining proper configuration and conditions to provision a new VM whenever the load to the application increases.

• Azure VM Scale Sets lets the developer create and manage a group of VMs that are load balanced. The scale sets can be configured in such a way that the count of VMs can automatically be increased or decreased based on the application demand or based on a pre-defined schedule.
• Usage of Scale Sets ensures high availability of the applications and allows the developers to manage, update and configure large VMs centrally and also help them support the development of large-scale applications supporting big data, big workloads, and compute loads.
• Azure scale sets can support up to 1,000 VMs. If the custom VM images are created and uploaded, then the limit is 600 VMs.

Azure provides overall 4 types of storage services - Blob Service, Table Storage, Queue Storage, and File Storage Services as shown in the figure below:

• Azure Table Storage: This type of storage lets user deploy their applications with semi-structured data and a NoSQL-based key-value store.
  • This is used when there is a need for applications that follow a flexible schema of data.
  • Table Storage focuses on enterprise-level data and follows strongly consistent models.
  • The data is represented in terms of Entities grouped under tables.
• Azure Queue Storage: This storage provides a message queue system for handling large workloads by letting users develop and build flexible and modular applications.
  • This storage ensures that the application becomes less prone to failure of individual components and is scalable.
  • With the help of message queues, it provides the queue monitoring feature for helping the application to ensure the user demands are met.
• Azure File Storage: This storage type provides features of file sharing that are accessible using SMB (Server Message Block) Protocol. The data in this storage is protected by HTTPS and SMB 3.0 Protocol.
  • They are used for improving the performance and capabilities of on-premise applications.
  • The OS deployments and hardware management is taken by Azure itself.

IaaS: This stands for “Infrastructure as a Service” which provides a set of capabilities like OS, network connectivities, etc which are at the infrastructural level and are delivered as pay per use policy. The infrastructure is used for hosting applications. Examples include Azure VM, VNET, etc.

PaaS: PaaS stands for “Platform as a Service” which is mostly about underlying infrastructure abstraction to the developers for enabling quicker development of the applications without the need for worry about hosting management. Examples include Azure web apps, Storage services, cloud services, etc.

SaaS: SaaS stands for “Software as a Service” and are those applications which are delivered using the service delivery model where the applications are simply consumed and used by an organization. These applications are generally mobilized by making the organization pay for their usage or through ads. Examples include applications like Office 365, Gmail, SharePoint Online, and so on.

The following table shows the difference between the On-Prem Service, IaaS, PaaS, and SaaS services. We can observe that as we go right, the level of control the developer or the user has over the application reduces.

The main difference between Azure Table Storage and Azure SQL Service is given below:

• Possibility 1: Azure VNET based “Point to Site” service can be a correct choice for this scenario of connecting one on-premise DB to an Azure-hosted app. “Point to Site” is valid for cases where the count of resources to be connected via VPN is very limited.
• Possibility 2: In case there is a large number of resources for connection, then “Site to Site” or “Express routes” are the other options that could be considered.
  • There might be chances that using “Site to Site” might lead to network latency as VPN due to these work only via Internet (public infrastructure). In such cases, “Express Routes” are used as it provides dedicated leased line for overcoming latency issues.
• Possibility 3: In case the customer is not willing to work via VNET, then Windows Communication Foundation (WCF) service can be developed and hosted on-premise which would have CRUD operations meant only for the database hosted on-premise. This works by means of using the “Service bus relay” option for developing communication between the Azure-hosted app to the WCF service for database access.

The main difference between Azure Storage Queue and the Azure Service Bus Queue is given below:

There can be 2 possible causes:

• Client-side causes:
  • The application might have been redeployed.
  • The application might have just performed a scaling operation.
  • The client-side networking layer has been changed.
  • There might be transient errors in the client or the network between the client and the server.
  • Another possible reason could be the bandwidth threshold limits have been crossed.
• Server-side causes:
  • It might occur if the Azure Redis Cache service itself might undergo a failover from the primary to the secondary node.
  • The server instance where the cache was deployed might have undergone patching or maintenance.

az vm create `   
--resource-group myResourceGroupName `   
--name myVM --image win2016datacenter `   
--admin-username AzureuserNAME `   
--admin-password AzurePASSWORD

Conclusion

Microsoft Azure has proven itself to be the fastest-growing cloud platform due to its more than 200 service offerings and benefits with pay per use pricing strategy. The revenue generated by Microsoft Azure has been growing constantly from $880 million in 2015 to a whopping $14.6 billion in 2020.

This tremendous growth in Azure has paved the path to many businesses by creating lots of opportunities in both tech and non-tech domains thereby making it a very lucrative domain for building one’s career.

Additional Resources

• Learn Azure
• Practice Coding
• Terraform Interview