I've never worked on tremendously huge projects and the workflow we use at work is check-out/code/compile locally to test/commit. I was wondering how a build server would change this process. How do developer test their code when the application is too huge to compile locally? They just code, commit and pray?
Absolutely not.
The developer usually has a build file which can build the project for him or her, which has some "targets" defined which do the testing. If you have a really big project, you may have certain portions of it precompiled for you, so you don't have to build the whole thing in one big chunk. You usually do your testing locally before you commit to your repository. Breaking the build in big projects can mark you as an object of ridicule and scorn. Breaking the build in really important, really big projects can be career limiting... ;-)
The build SERVER itself doesn't change this. The build server only runs your build file and the targets you tell it to.
There are also build components (I've just started using TeamCity - no affiliation) that allow "personal builds".
I haven't used it yet as we haven't got it set up properly but my understanding is that TeamCity allows running a build (and tests if they are any run on the server) with your changes before committing (and optionally the server will commit your changes if the build is succesful). in TeamCity this is called a Pre-Tested Commit.
Related
I haven't done any cshtml front-end development for a few years.
What's the current, generally accepted way for ASP.NET Core front-end developers to work across a range of tools on Windows?
By that, I mean a way to have the front-end JS build and the .NET project(s) also build and to work rapidly in the browser and code.
My thinking is.
We have much better command line story around dotnet today.
Some folk like VS Code.
Some folk prefer VS 2019, and some like either, depending.
We need to work on UI aspects sometimes.
But we also need to attach a debugger and debug the server logic sometimes.
The build server should have no problem, be simple, and rely mostly on build logic held in the repo.
Tooling, and kicking off the whole build and serve process should be understandable and familiar.
It should be pretty simple to get going after a team noob clones the repo.
My initial thought would be to setup NPM then use something like Gulp to kick off everything, including running dotnet run.
Then when running under the Visual Studio 2019 debugger, use the Task Runner Explorer to kick off the Gulp stuff but skip the dotnet run part.
(shame there doesn't seem to be a command line for start VS(Code or 2019) and attach debugger)
Now I'm expecting to get a "primarily opinion based" SO beating, but there are general trends and ideas that go into designing all these tools for how they can all play ball together and what the dev story looks like.
You've pretty much already described the process. However, I'll add a few things:
You don't need the dotnet run bit. Visual Studio and VS Code are both capable of debugging directly.
You can assign the gulp tasks to build tasks in Task Runner Explorer, so you really don't even then to think about running those directly. I'm not as sure on this aspect of VS Code, but I'm sure there's probably some extension to handle it, if it's not already built-in.
If you want true ease of development, the best thing you can do is use Docker. Just add a Dockerfile to each project that actually runs (i.e. not a class library) and set up the steps to build and run it there. In Visual Studio, you can right-click the project and choose Add > Docker Support, and it will actually generate a ready-made Dockerfile, though you may need to add a step or two to handle the client-side build steps. In any case, this then becomes truly click and run, with nothing to worry about. The story is even better when you use docker-compose, as then Visual Studio and VS Code can spin up your entire application stack all at once, including external dependencies such as a database, Redis instance, etc. If you haven't used Docker before, start now. It's absolutely revolutionary for development.
One note for CI/CD, as much as possible, you should add a YAML file to describe your CI/CD pipeline. Depending on the the actual provider you're using for build/release, there might be some differences, so consult the relevant documentation. (Azure DevOps, for example, doesn't currently support describing release pipelines in yaml, though you can still do your build that way.) In any case, this allows you to configure all this in code, and have it committed to source control.
You may consider the same for your infrastructure. Azure has ARM templates, AWS has CloudFormation, GCP has Deployment Manager. There's also third-party tools like Terraform or Ansible. All of these, in some form or fashion (usually JSON or YAML) allow you to define all the characteristics of the infrastructure you're going to deploy to and commit that to source control. This makes deployment and things like creating new environments as breeze.
While I only have a github repository that I'm pushing to (alone), I often forget to run tests, or forget to commit all relevant files, or rely on objects residing on my local machine. These result in build breaks, but they are only detected by Travis-CI after the erroneous commit. I know TeamCity has a pre-commit testing facility (which relies on the IDE in use), but my question is with regards to the current use of continuous integration as opposed to any one implementation. My question is
Why aren't changes tested on a clean build machine - such as those which Travis-CI uses for post-commit tesing - before those changes are committed?
Such a process would mean that there would never be build breaks, meaning that a fresh environment could pull any commit from the repository and be sure of its success; as such, I don't understand why CI isn't implemented using post-commit testing.
I preface my answer with the details that I am running on GitHub and Jenkins.
Why should a developer have to run all tests locally before committing. Especially in the Git paradigm that is not a requirement. What if, for instance, it takes 15-30 minutes to run all of the tests. Do you really want your developers or you personally sitting around waiting for the tests to run locally before your commit and push your changes?
Our process usually goes like this:
Make changes in local branch.
Run any new tests that you have created.
Commit changes to local branch.
Push local changes remotely to GitHub and create pull request.
Have build process pick up changes and run unit tests.
If tests fail, then fix them in local branch and push them locally.
Get changes code reviewed in pull request.
After approval and all checks have passed, push to master.
Rerun all unit tests.
Push artifact to repository.
Push changes to an environment (ie DEV, QA) and run any integration/functional tests that rely on a full environment.
If you have a cloud then you can push your changes to a new node and only after all environment tests pass reroute the VIP to the new node(s)
Repeat 11 until you have pushed through all pre-prod environments.
If you are practicing continuous deployment then push your changes all the way to PROD if all testing, checks, etc pass.
My point is that it is not a good use of a developers time to run tests locally impeding their progress when you can off-load that work onto a Continuous Integration server and be notified of issues that you need to fix later. Also, some tests simply can't be run until you commit them and deploy the artifact to an environment. If an environment is broken because you don't have a cloud and maybe you only have one server, then fix it locally and push the changes quickly to stabilize the environment.
You can run tests locally if you have to, but this should not be the norm.
As to the multiple developer issue, open source projects have been dealing with that for a long time now. They use forks in GitHub to allow contributors the chance to suggest new fixes and functionality, but this is not really that different from a developer on the team creating a local branch, pushing it remotely, and getting team buy-in via code review before pushing. If someone pushes changes that break your changes then you try to fix them yourself first and then ask for their help. You should be following the principle of "merging early and often" as well as merging in updates from master to your branch periodically.
The assumption that if you write code and it compiles and tests are passed locally, no builds could be broken is wrong. It is only so, if you are the only developer working on that code.
But let's say I change the interface you are using, my code will compile and pass tests
as long as I don't get your updated code That uses my interface.
Your code will compile and pass tests as long as you don't get my update in the interface.
And when we both check in our code, the build machine explodes...
So CI is a process which basically say: put your changes in as soon as possible
and test them in the CI server (it should be of course compiled and tested locally first).
If all developers follow those rules,
the build will still break, but we will know about it sooner rather than later.
The CI server is not the same as the version control system. The CI server, too, checks the code out of the repository. And therefore the code has already been committed when it gets tested on the CI server.
More extensive tests may be run periodically, rather than at time of checking in, on whatever is the current version of the code at the time of testing. Think of multi-platform tests or load tests.
Generally, of course, you'll unit test your code on your development machine before checking it in.
Our msbuild process creates a variety of zip packages for deployment (mostly web sites, but other things as well). We have a variety of recurring problems that keep sneaking back - files included that shouldn't be, missing resources. This screams for automated validation. The criteria to test for are simple
Validation of foosite package:
Resource files are present.
No test result files, obj files, or other build artifacts
And so forth.
Ideally, I could use nunit or mstest, which everone is familiar with. Msbuild knows where the packages are. We have a lot of packages, possible concurrent builds on different branches. Ergo, the location of the packages and names of the packages are not deterministic - so the tests don't know where the packages are.
What is the simplest way to feed msbuild information to mstest or nunit? The answer to this question would one possible answer, however, that question got architectural advice instead of an answer. I know this isn't a unit test, but the test framework is handy, anyway. I could create an exe to validate the build - but why add a couple hours to the project?
Or, do you have a better suggestion for automatically validating build packages? (MSIs, zips, whatever)?
What I've ended up doing is having a bunch of custom MS build tasks which spin up a virtual machine on Virtual Server, copy the MSI onto the machine, silently deploy it and then validate against it. I used PSExec to start the MSI. It could then use the MSTest command line runner to use MSTest and run your test bits.
This is probably overkill for you, but using a VM allows you to start clean and not be affected by any previous installs on your dev box.
If you want a fast fail, like a unit test, then I suggest you create unit tests against your packages. Such a test would unzip the .ZIP packages, and run some asserts against the contents.
You could even use some TDD techniques against the packages. For instance, if you have a deployment fail because a particular file is missing, then write a unit test that fails because the file is missing; change the build so that the file is present; then make sure the unit test succeeds.
But in general, deployment issues are broader than that, and I echo the suggestion from blowdart. Deploy into one or more virtual machines, then run automates tests over the deployed environments. These tests would not only test for simple things like was there an error returned during the installation itself; they would also check things like were the IIS virtual directories set up correctly, with the correct properties and contents, and does the web site basically run.
I'd use several different virtual machines to test different deployment scenarios: one for a clean deploy; one for an upgrade from version .-1, etc. It's possible that the same, or similar IVT tests could be run for each environment.
Even if you can't do this all at once, the thought process involved in this exercise should lead to a more formal definition of what your deployment environment really is. You this will be helpful when you get a chance to embody this formal definition in actual tests.
Currently I'm tasked with doing the daily build. We have an ASP.NET 2005 website with a SQL Server 2005 backend. Our current source control is Visual Source Safe 2005.
At this point, I use the brute-force method of daily builds.
Get Latest version of source code
Get Latest version of Database release script
Backup old website files to a directory
Publish new code to my local machine
Run on my server to keep the test/stage site working
Push newly created files to the website
Run SQL Script on test database (assuming updates, otherwise I don't bother)
Test website on the Test Server.
Looking at the idea of automated builds intrigues me since it means that I do less each morning. How would you recommend I proceed? I want to have a fully fleshed out idea before I present it to my boss.
Ditch VSS, move to Subversion, and check out CruiseControl.NET. Alternatively, if you have a MSDN developer license, you can run TFS workgroup edition and set up a build server on any old XP box. Its what we do at our shop.
As Assaf noted, you can use CC.NET with VSS directly. Nice.
TeamCity has worked well for me. It has a very simple setup. Combine it with an MsBuild script for your operations and you're auto-matic.
For build management I wholeheartedly recommend TeamCity. It doesn't require IIS6 (like CC.net does) since it runs on it's own copy of Tomcat and the setup is all done thru various forms. This is a big deal to me since the build server is just an XPPro box. It integrates well with SVN and there is no crazy XML file manipulation like I had to do with CruiseControl.Net. Big win for me.
For a build runner we use NAnt to send emails to various people, copy the packaged builds where they're supposed to go, run NUnit and NCover, and deploy the software to our web farm.
For automated testing we use Watin.
http://www.nunit.org/index.php
http://www.jetbrains.com/teamcity
http://ncover.sourceforge.net/
http://subversion.tigris.org/
http://nant.sourceforge.net/
http://watin.sourceforge.net/
Try CruiseControl.Net. It's free, and whatever customized daily/continuous routine you want it to perform you can always add with scripts.
Remember, it's not just about daily (nightly) builds, but also about letting you catch build errors in time (since it continuously builds after every source commit/check-in). You don't necessarily test every code chance on every possible platform and build configuration, but CC can do exactly that for you (in the background).
http://confluence.public.thoughtworks.org/display/CCNET/Visual+Source+Safe+Source+Control+Block
All of what you are doing can be performed by a set of batch files, depending on how automated your test environment is. The main batch file can be started as a 'scheduled task' at midnight or whatever. That's how we 'do it cheap' here and at other places I've worked. If you need help with a particular batch, I can provide a sample.
I second (or third) the reccomendation for Subversion/CruiseControl.net. Also, if it is appropriate, check out hosted services for SVN like CVSDude. You'll probably become well versed with MSBuild in the process too. Once you get it setup it is great.
The cost doesn't come from licensing of the tools or even hardware necessarily, but from your time building and maintaining the system - and depending on what you are doing, that could become significant.
Start with the basics and incrementally improve it over time. Like anything else, if you try to come out of the gate with lots of automation and functionality you could find yourself mired in it fulltime for weeks.
Whatever tools you use, house them in a virtual machine (ie., vmware).
When the equipment inevitably goes south, you can copy the image onto any machine and not miss a beat because your build server decided to take the day off, assuming of course, you back up.
Working in a team environment, we have a Team Foundation Server that also contains a Team Build component. It is configured to automatically build all projects and solutions at specific times or on request.
We develop a product that is built with several solultions that depend on eachother. When things have been changed in one solution, it has to be rebuilt locally manually in both debug and release mode so that changes take effect in another solution that depends on it.
Also when a developer retrieves all sources the first time, he has to build all solutions manually in the correct order to get a working environment.
What is the best way to automate things like this? Create .cmd files that trigger the correct msbuild files? Using a program such as CruiseControl.NET?
What do you people do to maintain a clean local development environment?
What I did for our Team was to provide a Visual Studio Solution which contains all projects. Then I created a simple .cmd file which uses the commmandline tools of Visual Studio to build this solution with their respective debug/release/profile configurations. This is a one step build solution that can be used from every engineering machine.
The next level is the continuous integration system that is setup to check for changes every 15 min and start a build if there are changes in the VCS. I'm using hudson as our CI system. The CI system is used to build the native projects, the java projects as well as the flex stuff. Since everything can be build from the commandline it's pretty easy to use it with hudson or CruiseControl.NET.