Digital Edition

SYS-CON.TV
What Should Be in Your .NET Toolkit? [#DotNet]
The easiest way to cover the list of deployment steps and the best tools to use

These days, implementing automatic deployment for .NET web projects is easier than ever. Drastic improvements started in Visual Studio 2010 when basic deployment strategies and tools were incorporated into VS itself. Yet, documentation was quite poor at that time, so you had to scour the Internet to find good tutorials in blogs or conference videos. Things have been constantly improving since then; now, we have even more functionality available out-of-the-box and documentation provided in a way that allows you to get started from zero understanding of the process to working deployment in less than two days. Some good places to start are this tutorial for an end-to-end guide on configuration and MSDN's overview.

In this article, I will show you the easiest way to cover the list of deployment steps and the best tools to use.

Trigger your deployment from build server
Whatever build server you're using for doing continuous integration (CI), it is an ideal tool to trigger your deployments, as it provides core benefits of audit (you always know who/when triggered deployment at which version). Just set up deployment build plans for each platform to which you want to deploy. The amount of deployment planning largely depends on branching and application lifecycle management (ALM) strategies you employ. For example, you could have something like the following:

  • Integration - Builds from master branch to deploy to integration platform where UI tests are executed as of the latest developed source code. Triggers whenever a new set of commits appears in the repository.
  • QA - Builds from master branch to deploy new features to QA environment whenever QA and Dev teams agree they need a new version validated. Triggered manually.
  • Release - Builds from a particular commit of master branch to deliver a validated version of the application to production environment.

You will want to add similar build plans for release or feature branches if you are using them in your workflow. You could also add a nightly build plan to deliver a nightly version of your application for preview (this is what we do when stakeholders need better visibility of the progress to get faster feedback without distracting the team).

A build server guarantees that a build plan is triggered on the source code exactly as of specific commit, so you're safe from "wrong source code version deployed" kind of errors.

Deployment out-of-the-box (configuring deployment in several clicks)
Visual Studio and MSBuild teams have been working hard to make the deployment experience a smooth one in the past few years, and we have some pretty neat features these days as a result (for example, you can configure basic auto-deployment in a few steps).

Visual Studio has a concept of publish profiles. In a nutshell, a publish profile is a versioned (you should store it in the repository) XML file that holds information about the destination environment and deployment settings. In a simple scenario, you'll have a publish profile for each of your platforms.

Right-click your project in solution explorer and go to the publish dialog:

You can add a publish profile using manage profiles dialog (this is what you do for your on-premises environments). Just fill in WebDeploy endpoint address and password - and you're done.

For those using Azure: ready-to-use publish profiles can be downloaded directly from Azure, which is even easier.

Now check-in your freshly generated profile and modify your deployment build plan to execute the following:

Msbuild.exe Web.csproj /p:DeployOnBuild=true /p:Configuration=Release /p:PublishProfile=staging /p:Password=p0ssw0rd

And you're done. MSBuild will take care of building your application in the desired configuration (release) and publishing it to the desired environment (staging in this case).

You can extend/customize packaging to address your specific needs (e.g., do not Remove /Uploads folder). Check out Sayed Hashimi's blog for some interesting examples.

Version your configuration
One of the most frequent deployment errors when done manually is incorrect configuration. Connection strings, timeouts, custom errors, external services endpoints, and much more can be different in production compared to development; moreover, configuration evolves along with the application, so every time a new version arrives to production, intelligent merge of dev and production config files is required. I don't know a single dev/ops person who likes merging these files by hand. If you're still doing that, stop it now. You can spend your time doing something much more useful. Config transformations to the rescue.

Config transformations have been introduced in VS 2010. Essentially, config transformation is a simplified version of xslt to transform your .config files in an efficient way. A transformation is just another .config file sitting next to your original .config file and applied upon publish. Transformation files use a very simple language.

You can have configuration-specific (release/debug) and publish-profile-specific (QA/production) transformations. MSBuild chains transformation files so that first configuration-specific transformation is applied and then profile-specific transformations finish the job:

For example:

  • Configuration-specific transformation turns on custom errors for release (needed in release regardless of the environment used);
  • Publish-profile-specific transformation changes connect strings (which are different across different environments).

Now your process of configuration maintenance is simple and defined:

  • Production configuration is always built automatically from dev configuration;
  • Transformations are versioned in the version control system and evolve naturally with source code and dev configuration changes.

No more manual merging of dev and production configuration. No more storing of configuration changes in emails/wiki pages. No more headache.

Database updates
Your database changes as your source code changes, so it makes perfect sense to version both together (otherwise they will get out of sync at some point). There are a couple of ways to do so, but these days, most of our teams tend to use migration frameworks (as they provide fine-grained control over what's happening). We have plenty of migration frameworks for .NET (e.g., FluentMigrator, EntityFramework migrations); simply use the one you like most.

Migrations can be configured to update the database on the first use (when application starts), or you can execute a console migrations runner in your migration script. While the former option is easier, I strongly recommend the latter (running database migrations as part of you deployment script), as this allows you to detect errors before users hit your application.

To do this, just add this to your deployment build:

Migrate.exe MyMvcApplication.dll /startupConfigurationFile="..\web.config"

Deployment scripting with MSBuild
MSBuild is your best friend in executing all those deployment steps. It has some built-in capabilities to copy files, run executables and manage configuration properties. For example, use WebDeploy client to control IIS (recycle ApplicationPool, take the application offline before deployment, bring it online after deployment).

Use extension libraries to do the rest (set ACL, backup database, etc):

For example, use the SqlExecute task to back up the database before deployment. Or use a folder task to set ACL permissions to the uploads folder. Check out the list of tasks in these libraries, and you can do pretty much everything your deployment process requires without much effort so that no manual interaction is needed.

Summary
Every new practice takes time to adopt, but with modern tools, implementing automatic deployment can be quite easy. If you're sold on the idea but are unsure on where to start (or can't sell it to your manager), I suggest you to start with automating QA builds, as they require less custom steps (e.g. you don't' necessarily need to take the application down during QA deployment). Once this is in place, you'll have the proof you need to confidently expand deployment to production.

About Ivan Antsipau
Ivan Antsipau is a senior .NET developer at Itransition specializing in architecting and implementation of business-specific web applications. With a specialist degree in Radiophysics and Computer Science, a knack for team leading, and years of hands-on programming experience under his belt, he sees the key to sustainable and accelerated delivery of software projects in elimination of stressful manual efforts with the help of continuous integration and automated testing.

In order to post a comment you need to be registered and logged in.

Register | Sign-in

Reader Feedback: Page 1 of 1



ADS BY GOOGLE
Subscribe to the World's Most Powerful Newsletters

ADS BY GOOGLE

The need for greater agility and scalability necessitated the digital transformation in the form of ...
ChatOps is an emerging topic that has led to the wide availability of integrations between group cha...
In his keynote at 18th Cloud Expo, Andrew Keys, Co-Founder of ConsenSys Enterprise, provided an over...
The cloud era has reached the stage where it is no longer a question of whether a company should mig...
You know you need the cloud, but you’re hesitant to simply dump everything at Amazon since you know ...
Coca-Cola’s Google powered digital signage system lays the groundwork for a more valuable connection...
In his session at 21st Cloud Expo, Raju Shreewastava, founder of Big Data Trunk, provided a fun and ...
As DevOps methodologies expand their reach across the enterprise, organizations face the daunting ch...
While some developers care passionately about how data centers and clouds are architected, for most,...
"Since we launched LinuxONE we learned a lot from our customers. More than anything what they respon...
Is advanced scheduling in Kubernetes achievable?Yes, however, how do you properly accommodate every ...
DevOps is under attack because developers don’t want to mess with infrastructure. They will happily ...
"As we've gone out into the public cloud we've seen that over time we may have lost a few things - w...
In his session at 21st Cloud Expo, Michael Burley, a Senior Business Development Executive in IT Ser...
Sanjeev Sharma Joins June 5-7, 2018 @DevOpsSummit at @Cloud Expo New York Faculty. Sanjeev Sharma is...
We are given a desktop platform with Java 8 or Java 9 installed and seek to find a way to deploy hig...
"I focus on what we are calling CAST Highlight, which is our SaaS application portfolio analysis too...
"Cloud4U builds software services that help people build DevOps platforms for cloud-based software a...
The question before companies today is not whether to become intelligent, it’s a question of how and...
Kubernetes is an open source system for automating deployment, scaling, and management of containeri...