[AppVeyor CI][1] provides hosted continuous integration for .NET developers.

Disclaimer: I&#39;m the developer of this service.


  [1]: http://www.appveyor.com

If your open source project you can get a free account at Code Better http://codebetter.com/jameskovacs/2009/02/24/announcing-teamcity-codebetter-com/  I don&#39;t know of any for non open source.

Just came across [BuildBox][1] (now [Buildkite][4]) from a [tweet][2] from [someone I follow][3]. Seems like a suitable solution to this.

  [1]: https://buildbox.io/
  [2]: https://twitter.com/toolmantim/status/504062359900413952
  [3]: https://twitter.com/toolmantim
  [4]: https://buildkite.com

[OnCheckin.com][1] offers Cloud Powered Continuous Deployment for.Net websites and services.

Disclosure: I am the developer and founder of this service.


  [1]: https://oncheckin.com

Example of scatterplot matrix

![enter image description here][1]

Is there such a function in matplotlib.pyplot?


  [1]: http://i.stack.imgur.com/Sz6SK.jpg

Is there a function to make scatterplot matrices in matplotlib?

I&#39;m attempting to install ImagicK as its required for a Yii Extension I want to use for my project. 

I&#39;ve checked around and followed a few different suggestions but nothing that has worked so far, ultimately the framework extension is still throwing me an error (Must have ImagicK installed). 

I&#39;ve tried things like:
        
    sudo apt-get update
    sudo apt-get install imagick-php5 .. etc

as well as the PECL install which ends up throwing me the following error:

    Checking ImageMagick MagickWand API configuration program... configure: error: 
    not found. Please provide a path to MagickWand-config or Wand-config program. 

Any help as to how to get ImagicK installed and integrated with PHP on Ubuntu 11.10 would be great.

Thanks in advance!

Howto: Install Imagick (for php) on Ubuntu 11.10

I&#39;m looking to set up a TeamCity server for continuously building a .NET web application. I already have hosting, so I don&#39;t want to get a whole new hosting account such as AppHarbor.

I don&#39;t maintain my own physical server, nor do I want to.

I also don&#39;t want to have to pay $50 or more per month for an entire dedicated Windows machine, just to host TeamCity.

I really don&#39;t care if it&#39;s slow and on a shared machine, as it&#39;s just continuous build which will be running in the background.

I&#39;ll want to have the outputs automatically deployed to a server of my choice through FTP.

Is there anyone on the market providing hosted TeamCity environments? 

Anyone know of a hosted TeamCity build provider?

I'm looking to set up a TeamCity server for continuously building a .NET web application. I already have hosting, so I don't want to get a whole new hosting account such as AppHarbor.
I don't maintain my own physical server, nor do I want to.
I also don't want to have to pay $50 or more per month for an entire dedicated Windows machine, just to host TeamCity.
I really don't care if it's slow and on a shared machine, as it's just continuous build which will be running in the background.
I'll want to have the outputs automatically deployed to a server of my choice through FTP.
Is there anyone on the market providing hosted TeamCity environments?

I have a **dll-type project** that contains MSTest integration tests. On my machine the tests pass, and I want the same to happen on a CI server (I use TeamCity). But the tests fail, because I need to tweak some settings in app.config. This is why I was thinking to have a separate second app.config file that will hold the settings for CI server.

So I would like to have
&lt;pre&gt;
/Sln
 /Proj
  app.config (I think this is required by VS)
  app.Release.config (This is a standalone independent config file)
&lt;/pre&gt;

Thus if I select Release configuration in build config on CI, I would like to use app.Release.config file instead of app.config

**Problem**&lt;br/&gt;
This doesn&#39;t seem to be straightforward for simple .dll type projects. For web projects, I can do web config transformations. I found a hack how to do these transformations for a dll type project, but I am not a big fan of hacks.

**Question** &lt;br/&gt;
What is a standard approach to tweak app.config files depending on build config for .NET projects (such as Debug, Release, ...)?

How to select different app.config for several build configurations

We have a need to be able to skip a submodule in certain environments.

The module in question contains integration tests and takes half an hour to run.  So we want to include it when building on the CI server, but when developers build locally (and tests get run), we want to skip that module.

Is there a way to do this with a profile setting?  I&#39;ve done some googling and looked at the other questions/answers here and haven&#39;t found a good solution.

I suppose one option is to remove that submodule from the parent `pom.xml` entirely, and just add another project on our CI server to just build that module.

Suggestions?

Skip a submodule during a Maven build

In order to get the fastest feedback possible, we occasionally want Jenkins jobs to run in Parallel. Jenkins has the ability to start multiple downstream jobs (or &#39;fork&#39; the pipeline) when a job finishes. However, Jenkins doesn&#39;t seem to have any way of making a downstream job only start of all branches of that fork succeed (or &#39;joining&#39; the fork back together).

Jenkins has a &quot;Build after other projects are built&quot; button, but I interpret that as &quot;start this job when **any** upstream job finishes&quot; (not &quot;start this job when **all** upstream jobs succeed&quot;).

Here is a visualization of what I&#39;m talking about. Does anyone know if a plugin exists to do what I&#39;m after?
![Build Pipeline][1]

---

# Edit:

When I originally posted this question in 2012, Jason&#39;s answer (the Join and Promoted Build plugins) was the best, and the solution I went with.

However, dnozay&#39;s answer (The Build Flow plugin) was made popular a year or so after this question, which is a much better answer. For what it&#39;s worth, if people ask me this question today, I now recommend that instead.



  [1]: http://i.stack.imgur.com/IywrJ.png

How do I make a Jenkins job start after multiple simultaneous upstream jobs succeed?

I&#39;m looking to run automated NUnit tests for a C# application, nightly and on each commit to svn.

Is this something that Jenkins-CI can do?&lt;br/&gt;
Is there an online tutorial or how-to document which documents a similar setup that I can look at?

How do you run NUnit tests from Jenkins?

I&#39;m running Ubuntu 11.10 and have run `sudo apt-get install jenkins` to install Jenkins on this system.

I&#39;ve seen some tutorials on how to setup a reverse proxy (Apache, Nginx, etc), however this is a VM dedicated for just jenkins and I&#39;d like keep it as lean as possible while having jenkins running on port 80.

I&#39;ve found the upstart config in `/etc/init/jenkins.conf` and modified the port to 80 `env HTTP_PORT=80`

When I start jenkins via `service jenkins start`, `ps` reveals that it runs for a few seconds then terminates.

Is this because jenkins is running as the `jenkins` user on a privileged port? If so, how do I fix this? Any other ideas a welcome.

Here is the upstart config:

    description &quot;jenkins: Jenkins Continuous Integration Server&quot;
    author &quot;James Page &lt;james.page@ubuntu.com&gt;&quot;
    
    start on (local-filesystems and net-device-up IFACE!=lo)
    stop on runlevel [!2345]
    
    env USER=&quot;jenkins&quot;
    env GROUP=&quot;jenkins&quot;
    env JENKINS_LOG=&quot;/var/log/jenkins&quot;
    env JENKINS_ROOT=&quot;/usr/share/jenkins&quot;
    env JENKINS_HOME=&quot;/var/lib/jenkins&quot;
    env JENKINS_RUN=&quot;/var/run/jenkins&quot;
    env HTTP_PORT=80
    env AJP_PORT=-1
    env JAVA_OPTS=&quot;&quot;
    env JAVA_HOME=&quot;/usr/lib/jvm/default-java&quot;
    
    limit nofile 8192 8192
    
    pre-start script
        test -f $JENKINS_ROOT/jenkins.war || { stop ; exit 0; }
        $JENKINS_ROOT/bin/maintain-plugins.sh 	
        mkdir $JENKINS_RUN &gt; /dev/null 2&gt;&amp;1  || true
        chown -R $USER:$GROUP $JENKINS_RUN || true
    end script
    
    script
        JENKINS_ARGS=&quot;--webroot=$JENKINS_RUN/war --httpPort=$HTTP_PORT --ajp13Port=$AJP_PORT&quot;
        exec daemon --name=jenkins --inherit --output=$JENKINS_LOG/jenkins.log --user=$USER \
            -- $JAVA_HOME/bin/java $JAVA_OPTS -jar $JENKINS_ROOT/jenkins.war $JENKINS_ARGS \
            --preferredClassLoader=java.net.URLClassLoader
    end script



How to configure Jenkins to run on port 80

I&#39;m working on a little webapp (all client-side) I want to host it on Amazon S3. I&#39;ve found several guides on this and have managed to create myself a bucket (with the same name as my domain), set it as a website and upload some content.

Where I&#39;m struggling, and where all the documentation starts to get a bit vague, is how to properly configure my DNS.

All my registrar ([123-reg][1]) could suggest was web forwarding which gives me mydomain.com.s3.amazonaws.com

What do I have to configure, and where (ie. 123-reg / Amazon) can I get a clean URL?

  [1]: http://www.123reg.com/

Static hosting on Amazon S3 - DNS Configuration

Here&#39;s a Github repository of mine: https://github.com/n1k0/casperjs

There&#39;s a `gh-pages` branch to hold the project documentation, which is basically the project website: https://github.com/n1k0/casperjs/tree/gh-pages

This branch setups the documentation site at http://n1k0.github.com/casperjs/ — hurray.

In the meanwhile, I&#39;ve bough the `casperjs.org` domain to get this website available through it, so I put a `CNAME` file as [recommended in the docs](http://pages.github.com/#custom_domains): https://github.com/n1k0/casperjs/blob/gh-pages/CNAME — in their example, the operation is supposed to *create redirects from www.example.com and charlie.github.com to example.com*…

While the website now points to http://casperjs.org/, there&#39;s no 301 redirect from http://n1k0.github.com/casperjs/ (the old site url) to the new domain name.

Any idea how to setup such a redirect, if it&#39;s even possible? Is it a bug? If it is, where should I open an issue?


301 redirect for site hosted at github?

I am looking for a git hosting environment for several users. Therefore i&#39;ve searched for comparisons between **Gitolite**, **Gitlab** and **Gitorius**. But i get nothing what could be useful.

Is there anybody, who has experiences with different hosting tools and could provide an advice? 

Which is the best git hosting sw? - Gitolite vs. Gitlab vs. Gitorius

How to deploy a meteor application to my own server?

flavour 1: the development and deployment server are the same;

flavour 2: the development server is one (maybe my localhost) and the deployment server is another (maybe a VPS in the cloud);

flavour 3: I want to make a &quot;meteor hosting&quot; domain, just like &quot;meteor.com&quot;. Is it possible? How?

**Update**:

I&#39;m running Ubuntu and I don&#39;t want to &quot;demeteorize&quot; the application. Thank you.

How to deploy a meteor application to my own server?

What is the difference between EC2 and Beanstalk? I want to know regarding SaaS, PaaS and IaaS.

To deploy a web application in Wordpress I need a scalable hosting service. If there anything better than my purpose, please let me know as well.

FYI - I want to host and deploy multiple Wordpress and Drupal sites. 

I do not want to give more time for the server and focus on development. But the cloud hosting needs to be auto scalable. 

Difference between Amazon EC2 and AWS Elastic Beanstalk

New to TeamCity.  I have multiple build steps.  Step 3 generates an id that is needed in step 4.  What is the best way to pass the id (a string) between step 3 and step 4?  The build steps are written in Ruby.  Can I set an environment variable?

TeamCity, passing an id generated in one build step to a later build step

I am playing around with a TeamCity install and developing a plugin that extends a BuildServerAdapter.  When I package it up and install it to the server, teamcity-server.log contains entries for my plugin:

* found when scanning the plugins (.BuildServer\plugins) folder
* registering agent plugin from server plugin
* loads in the shared classloader
* is loaded

The plugin is also listed on the plugins page in the server administration.

Beyond that...nothing.  I&#39;ve put in various log statements, via both the logger and System.out, and I don&#39;t see them.  I&#39;ve even added an exception into the constructor, and I see no evidence of that either in the system logs.  When a build happens, there&#39;s again no evidence that my code is called.

&lt;!-- language: java --&gt;

    public class CustomBuildServerAdapter extends BuildServerAdapter {
    	
    	private SBuildServer myBuildServer;
    	private static final Logger LOG = Logger.getLogger(CustomBuildServerAdapter.class);
    	
    	private void debug(String msg) { LOG.debug(msg); System.out.println(msg); }
    	
    	public CustomBuildServerAdapter(SBuildServer aBuildServer) throws Exception {
    		throw new Exception(&quot;constructor is being called, at least we know that...&quot;);
    
    		//myBuildServer = aBuildServer;
    		//debug(&quot;constructor&quot;);
    	}
    	
    	public void register() {
    		debug(&quot;registering&quot;);
    		myBuildServer.addListener(this);
    		debug(&quot;registered&quot;);
    	}
    
    	public void buildFinished(SRunningBuild build) {
    		debug(&quot;build finished&quot;);
    		postMessage(build.getFullName() + &quot; - &quot; + build.getStatusDescriptor().getText());
    		debug(&quot;message posted&quot;);
    	}
    
    ...

The zip that I copy to `.BuildServer\plugins` has the following structure:

* MyTeamCityPlugin.zip
  * teamcity.plugin.xml
  * server
  * MyTeamCityPlugin.jar
     * package folders containing class files
     * META-INF
         * build-server-plugin.xml
         * MANIFEST.MF

Looking at other plugins, they use the following structure, so I&#39;ve tried that as well.

* MyTeamCityPlugin.zip
  * MyTeamCityPlugin
     * teamcity.plugin.xml
     * MyTeamCityPlugin.jar
         * package folders containing class files
         * META-INF
             * build-server-plugin.xml
             * MANIFEST.MF



My build-server-plugin.xml contains the following:

&lt;!-- language: xml --&gt;

    &lt;?xml version=&quot;1.0&quot; encoding=&quot;UTF-8&quot;?&gt;
    &lt;!DOCTYPE beans PUBLIC &quot;-//SPRING//DTD BEAN//EN&quot; &quot;http://www.springframework.org/dtd/spring-beans.dtd&quot;&gt;
    
    &lt;beans default-autowire=&quot;constructor&quot;&gt;
    	&lt;bean id=&quot;myplugin&quot; class=&quot;com.blah.blah.blah.CustomBuildServerAdapter&quot; init-method=&quot;register&quot;/&gt;
    &lt;/beans&gt;


I&#39;m part of the way there since the teamcity-server.log does indicate it knows about the plugin, and no longer throws an exception trying to load it.  Unfortunately, not erring isn&#39;t the same working.  

Using the ant build scripts from the sample plugin, I get the following error, so I&#39;ve been packing things manually.  This results in the seemingly successful load per above.

&gt; Failed to initialize spring context for plugin MyTeamCityPlugin.  Error creating bean with name &#39;simpleRunnerRunType&#39;:  instantiation of bean failed.

Can anyone give me the kick I need to get this running correctly?


TeamCity loads custom plugin, but there&#39;s no evidence the plugin code runs

Does anyone know to remove a specific version of a package on a TeamCity Nuget Feed?

How to remove a specific version of a package on a TeamCity Nuget Feed?

I have a VS 2012 web project /sln that I am trying to build in TeamCity. it uses .NET 4.5 which is installed on TeamCity.

The TeamCity server has **VS  2010 installed only**.

I get this error when the build runs:

    C:\BuildAgent\work\d5bc4e1b8005d077\CUSAAdmin.Web\CUSAAdmin.Web.csproj(799, 3): 
    error MSB4019: 
    The imported project 
        &quot;C:\Program Files (x86)\MSBuild\Microsoft\VisualStudio\v11.0\WebApplications\Microsoft.WebApplication.targets&quot; was not found. 
     Confirm that the path in the &lt;Import&gt; declaration is correct, and that the file exists on disk. Project CUSAAdmin.Web\CUSAAdmin.Web.csproj failed. 
     Project CUSAAdmin.sln failed. 


It is trying to use Visual Studio 2012 (v11.0) to build. 

I have set the **VisualStudioVersion** to be **10** in the **build.xml** though??

     &lt;Target Name=&quot;BuildPackage&quot;&gt;
       &lt;MSBuild Projects=&quot;CUSAAdmin.sln&quot; ContinueOnError=&quot;false&quot; 
         Targets=&quot;Rebuild&quot; 
          Properties=&quot;Configuration=$(Configuration); VisualStudioVersion=10.0&quot;  /&gt;


As well inside the project it is defaulting to VS2010

      &lt;PropertyGroup&gt;
    &lt;VisualStudioVersion Condition=&quot;&#39;$(VisualStudioVersion)&#39; == &#39;&#39;&quot;&gt;10.0&lt;/VisualStudioVersion&gt;
    &lt;VSToolsPath 
        Condition=&quot;&#39;$(VSToolsPath)&#39; == &#39;&#39;&quot;&gt;$(MSBuildExtensionsPath32)\Microsoft\VisualStudio\v$(VisualStudioVersion)&lt;/VSToolsPath&gt;

MSBuild in TeamCity of Visual Studio 2012 solution

I&#39;ve got a PowerShell script as follows

    ##teamcity[progressMessage &#39;Beginning build&#39;]
    # If the build computer is not running the appropriate version of .NET, then the build will not run. Throw an error immediately.
    if( (ls &quot;$env:windir\Microsoft.NET\Framework\v4.0*&quot;) -eq $null ) {
        throw &quot;This project requires .NET 4.0 to compile. Unfortunately .NET 4.0 doesn&#39;t appear to be installed on this machine.&quot;
        ##teamcity[buildStatus status=&#39;FAILURE&#39; ]
    }

    ##teamcity[progressMessage &#39;Setting up variables&#39;]
    # Set up variables for the build script
    $invocation = (Get-Variable MyInvocation).Value
    $directorypath = Split-Path $invocation.MyCommand.Path
    $v4_net_version = (ls &quot;$env:windir\Microsoft.NET\Framework\v4.0*&quot;).Name
    $nl = [Environment]::NewLine

    Copy-Item -LiteralPath &quot;$directorypath\packages\NUnit.2.6.2\lib\nunit.framework.dll&quot; &quot;$directorypath\Pandell.Tests\bin\debug&quot; -Force

    ##teamcity[progressMessage &#39;Using msbuild.exe to build the project&#39;]
    # Build the project using msbuild.exe.
    # Note we&#39;ve already determined that .NET is already installed on this computer.
    cmd /c C:\Windows\Microsoft.NET\Framework\$v4_net_version\msbuild.exe &quot;$directorypath\Pandell.sln&quot; /p:Configuration=Release
    cmd /c C:\Windows\Microsoft.NET\Framework\$v4_net_version\msbuild.exe &quot;$directorypath\Pandell.sln&quot; /p:Configuration=Debug

    # Break if the build throws an error.
    if(! $?) {
        throw &quot;Fatal error, project build failed&quot;
        ##teamcity[buildStatus status=&#39;FAILURE&#39; ]
    }

    ##teamcity[progressMessage &#39;Build Passed&#39;]
    # Good, the build passed
    Write-Host &quot;$nl project build passed.&quot;  -ForegroundColor Green


    ##teamcity[progressMessage &#39;running tests&#39;]
    # Run the tests.
    cmd /c $directorypath\build_tools\nunit\nunit-console.exe $directorypath\Pandell.Tests\bin\debug\Pandell.Tests.dll

    # Break if the tests throw an error.
    if(! $?) {
        throw &quot;Test run failed.&quot;
        ##teamcity[buildStatus status=&#39;FAILURE&#39; ]
    }

    ##teamcity[progressMessage &#39;Tests passed&#39;]

From what I&#39;m lead to believe, [an uncaught `Throw`][1] will result in an exit code of `1`, but unfortunately TeamCity is saying otherwise.

    [19:32:20]Test run failed.
    [19:32:20]At C:\BuildAgent\work\e903de7564e599c8\build.ps1:44 char:2
    [19:32:20]+     throw &quot;Test run failed.&quot;
    [19:32:20]+     ~~~~~~~~~~~~~~~~~~~~~~~~
    [19:32:20]    + CategoryInfo          : OperationStopped: (Test run failed.:String) [],
    [19:32:20]   RuntimeException
    [19:32:20]    + FullyQualifiedErrorId : Test run failed.
    [19:32:20]
    [19:32:20]Process exited with code 0
    [19:32:20]Publishing internal artifacts
    [19:32:20][Publishing internal artifacts] Sending build.finish.properties.gz file
    [19:32:20]Build finished

It might also be important to note that my `Execution Mode` is set to `Execute .ps1 script with &quot;-File&quot; argument`.

I tried changing it to `Put script into PowerShell stdin with &quot;-Command -&quot; arguments`, but then it failed with an exit code of `1` even with passing tests. I&#39;m sure that running it as `-File` is going to be the right way.

If I open up the script located at `C:\BuildAgent\work\e903de7564e599c8\build.ps1` and run it manually in CMD, it does the same thing... I.e., the failing tests fail, and the `%errorlevel%` is still `0`.

Yet, if I run it in PowerShell and call `$LASTEXITCODE`, it returns the right code every time.

 [1]: https://devblogs.microsoft.com/powershell/windows-powershell-exit-codes?Redirected=true



Content Type	Original Author	Original Content on Stackoverflow
Question	Jonathan	View Question on Stackoverflow
Solution 1 - Continuous Integration	Feodor Fitsner	View Answer on Stackoverflow
Solution 2 - Continuous Integration	Orn Kristjansson	View Answer on Stackoverflow
Solution 3 - Continuous Integration	Jonathan	View Answer on Stackoverflow
Solution 4 - Continuous Integration	Doug	View Answer on Stackoverflow

Anyone know of a hosted TeamCity build provider?

Continuous Integration Problem Overview

Continuous Integration Solutions

Solution 1 - Continuous Integration

Solution 2 - Continuous Integration

Solution 3 - Continuous Integration

Solution 4 - Continuous Integration

Howto: Install Imagick (for php) on Ubuntu 11.10

Is there a function to make scatterplot matrices in matplotlib?

Attributions