We're trying to figure out how to automate our website deployment. We've picked an existing project, and started playing around.
First, I used the Publish wizard from with VS2012 to create a Web Deploy Package. We then tried deploying the package as a website, through the IIS Manager, and that succeeded without a problem. We thought we were nearly there.
What was next was to learn how to run this from the command-line, so we could script it. And that's caused us no end of headaches.
We've been playing around with msdeploy.exe, and with the .cmd file that the publish wizard created, and while both methods seem to install the package as a virtual directory just fine, neither will install the package as a root website.
Browsing around on the web, I've run across this:
Web Deploy iisApp Provider
In a sync operation, the iisApp provider copies content to a folder under the destination site that you designate and marks the destination folder as an application. The iisApp provider cannot create a site. The iisApp provider will not create applications under sites that do not exist.
And
If you want to synchronize a Web site and its related configuration, use the appHostConfig provider.
At this point, we don't know beans about iisApp, or appHostConfig, or whatever. We didn't create a package that used one or the other, VS2012 did. We haven't a clue, at this point, how to convince VS2012's web publish to create an appHostConfig package, and for that matter, I don't know if we want to.
Here's the thing - whether a given website is installed as a root application or as a virtual directory is not something the developers control - it's a decision made by the implementation team. Or testing team usually installs any given website both ways, to ensure that both work.
And since IIS seems to be able to manage to install this package as a root site, there must be a way to get msdeploy to do it.
But how?
Related
I would be most grateful if anyone could help me solve this problem with ClickOnce Web deployment.
I have read all the threads on this subject and I have also read through all the Microsoft documentation on the subject. They seem to say a lot without actually being direct or providing helpful examples. However, perhaps I am wrong and I have not looked in the right places.
I have already used ClickOnce successfully to deploy an application on the local area network.
It works well and really isn't that complicated. However, my goal is to deploy this application to customers, who are not connected to my local network.
I have set up a web site (www.mydomain.co.za), which I can access directly or via the ftp protocol.
I have created a sub directory off the root where I intend to publish the files created by the publish function. The publish function of the application requires a Publishing Folder Location and a Installation Folder URL I don't really understand the functional difference between these two locations. If I set the Publishing Location to ftp://www.mydomain.co.za/MyProductName and the Installation Folder URL to http://www.mydomain.co.za/MyProductName, then the publish process succeeds and when I check on the web server, the files have been published successfully it would seem. A further Application Files/MyProductName subdiectory with the version number information appended was created where all the output was placed.
My next step is to then grab the URL of the setup.exe file and to run it from a browser. This downloads the setup.exe file to my downloads folder which I then try to run but I get an error
Deployment and application do not have matching security zones.>
I have seen this come up in other threads but These threads don't seem to relate directly to what I am trying to do. These threads make mention of using Internet Explorer to achieve some degree of success, but all the browser did was to download the file.
I have also noted with interest that a web page is created in the root with a button that prompts the user to install the application. This does not work either.
Does anyone know of an article that I can read on this subject which is more helpful or if anyone can offer more insights into this I would be very grateful.
Before we used TFS, we were building our packages locally using visual studio. There was a lot of projects organized into solutions. When we wanted to build the package, we simply located the ccproj project, right click on it and hit “package”.
There is couple of specific in our solution:
We use web roles with multiple web sites and virtual applications and we have them as project dependencies in VS2012 solution
We use config transformations for both web and worker roles. Worker roles transforms was achieved by adding transform target manually to the project file.
We have some additional class library projects – their outputs needs to go in a subfolder of the worker role along with correct configuration files, it is kind of plug-in architecture. We used some xcopy commands to include these non-referenced libraries in our worker roles.
Everything worked smoothly when building in VS 2012 locally.
When migrated to TFS we quickly learned that we won’t be able to replicate the same build process on build server
It turned out that TFS is not preserving solutions structure, more
details here:
http://social.msdn.microsoft.com/Forums/en-US/tfsbuild/thread/9ac815c8-5961-4670-a6d0-660a9b66da9c
The project dependency that was solving multiple web sites and
virtual applications in a single role did not work on build server,
probably because of different output directory. We had to add some
hacks in our ccproj and csproj files to get these published and
correctly included in the resulting package.
xcopy commands failed because of different directory structures on
TFS build sever.
We had to force run cspack on TFS build server by explicitly adding
/t:Publish parameter to msbuild command line.
Config transforms for worker roles did not worked, we had to force to
occur using another hacks in the ccproj and csporj files.
There was more issues but those are too detailed. I would keep it on
high level just to illustrate the whole issue. The build somehow
works now, but we have now a lot of hacks in place now.
I have two questions:
Is it possible to configure TFS build server to have exact same
behavior as the local build in VS2012?
Is there any official solution for building azure packages with multiple web sites and virtual applications in a single web role?
I haven't yet tried this on a TFS build server, but the approach outlined in my blog at http://michaelcollier.wordpress.com/2013/01/14/multiple-sites-in-a-web-role/ has been working well. The "trick" is basically to modify the .ccproj file to tap into the CoreBuildDependsOn target, adding logic that will execute MSBuild against the secondary sites. This should also allow config transforms to work.
I have a web application project in VS2012 which I'm publishing using a "Web Deploy Package". I want this package to include app-pool settings, specifically creating an IIS app-pool and assigning the newly created application to it.
I'm familiar with the option "Include application pool settings used by this Web project" available when the project is configured to use an IIS instance (not IIS Express), but IIS configuration is not part of the project file, and thus not source controlled. What happens when somebody builds a deployment package on a machine that hasn't had IIS meticulously configured? Not ideal.
How else then, can I go about getting AppPool settings into my web deploy package? I understand that the appPoolConfig provider is IIS7+ only, I'm fine with that limitation. I've banged my head against this issue in the past and never found a solution. 18 months later, we've got a new VisualStudio version, and a new web-publishing-pipeline, are there new options to address this? Or maybe something I missed when I first tackled this problem?
Edit
OK, I'm seeing the following as options:
Configure my project to sync settings from an IIS instance. As mentioned, I'm not a fan of this given that it puts settings outside of the project, meaning the environment has to be meticulously configured to build + publish. Plus it drags along other IIS settings I don't want included.
Inject something into the web-publishing-pipeline (WPP) to modify the archive.xml. I've toyed with this in the past and had limited success. One problem is the pipeline isn't exactly co-operative with working directly on the archive.xml file, another problem is some of the more cryptic attributes involved, like MSDeploy.MSDeployProviderOptions which appears to have some Base64 encoded binary? No idea what to put in there.
Find an existing "provider" that can do what I want. I might be out of luck here, the appPoolConfig provider only seems to want to read / write IIS, not, say, an XML file of settings. Does anybody know otherwise?
Write my own "provider" to produce manifest output entries. I'm not sure, is it possible to write a custom provider that writes to a manifest using the name of an existing provider? As in, MyCustomPoolProvider writes appPoolConfig sections into a manifest? This sounds like a potentially painful exercise that may or may not work. Would I still need to figure out the encoding of whatever is going into MSDeploy.MSDeployProviderOptions?
I get the feeling that the fundamental obstacle with Web Deploy for what I'm trying to accomplish, is how strictly it leans on "providers". The pre-existing providers are largely designed for IIS synchronisation, not primary development and publication. It so happens that some of these providers can be relatively easily hooked into via MSBuild, but the majority insist on pulling data from IIS, and that's that.
You are correct in your understanding of the appPoolConfig provider, in that it can only sync between App Pools and can't be provided with the configuration directly. What you could potentially do is keep a copy of the appPool in question in package form (ie. msdeploy -verb:sync -source:appPoolConfig=PoolName -dest:package=apppool.zip) and attempt to hijack the pipeline so that the MSDeploy call adds the application content into the package, leaving the existing content there.
Alternatively, you could always keep the packages separate and deploy them with different calls to MSDeploy.
FYI, MSDeploy.MSDeployProviderOptions is simply an encoded version of the parameters supplied to the provider when it was packaged. For example, -source:dirPath=c:\,ignoreErrors=0x10293847 -dest:package=package.zip would package the ignoreErrors value.
I created a wsp solution that which create 2 custom workflow actions. I want to deploy it to my 2 servers farm which have an application server and a webfront end server. SPF Web application is not activated on the app server.
My visual studio solution goes like this:
-One project which create the dll, where actions code is.
-One project which create the wsp package, feature, etc. The package references the other project as additional assembly.
My problem is, when I deploy my wsp package with Deployment Server Type WebFrontEnd, the feature is only installed on the wfe and I can't activate it. I can't see the feature in the manage feature page (the feature is farm level). When I change the Deployment Server Type of the package to ApplicationServer, I get the following message:
"This solution must be deployed to application servers, not front-end Web servers. It cannot contain a resource that is scoped to a Web application."
I did some tests. I removed the additional assembly from my package and then I can deploy my wsp solution as ApplicationServer type (but can't use my custom actions..). Then I created a dummy dll with nothing in it, added it as additional assembly to my sharepoint package and I realised I can't deploy my wsp as ApplicationServer type again.
So, can I reference an additional assembly from my wsp solution and still deploy as ApplicationServer type??
How can I deal with this? Any idea?
I did it again, I've been looking for answers for this all day, I finally post something here and one hour later, I got myself the answer. Here it is anyway.
I deployed independently both solutions using 2 wsp packages. One simply deploys the dll into the gac (and is ApplicationServer), the other one is using it without deploying it itself (and is WebFronEndServer). Now I have to deal with making sure the first one is deployed before using the other one... Feature activation dependency should do it.
Regards.
A couple of days ago, Microsoft released the engine they're using to do git deployments to Azure. I've had a task on my TODO list for a while to get that kind of functionality set up on my DEV IIS server, so I'm interested in trying out Kudu for that purpose.
The "Getting Started" document shows how to run the web front-end, but everything in there uses "http://localhost:PORTNUMBER" type URL's for git repositories, site URL's, etc.
I realize this is probably getting too far ahead of them, but I'm wondering if anyone has pointers for how to set it up using real domains on "regular" IIS instead of all of the localhost bits?
This is an old question, so I'm giving an updated answer with more current info since I just worked through setting up Kudu on an internal deployment server. The currently selected answer only deals with if you are directly running Kudu from within a development environment.
If you are deploying to a "production" type environment and don't want to install Visual Studio on the target server, there is a good guide on the project website on github.
https://github.com/projectkudu/kudu/wiki/Deploying-to-a-server
On the target server, you will need to install:
MSBuild ( http://go.microsoft.com/fwlink/?LinkId=309745 ) - comes with Visual Studio
NodeJS ( http://nodejs.org )
Git ( http://git-scm.com/downloads )
Back on your development machine, clone the git repo and build using the "build.cmd" file, following instructions in the above link.
In running build.cmd I got several test failures which blocked the build from producing artifacts. These were all related to Mercurial, which we don't use. Installing a Mercurial client didn't make them magically go away, so I disabled the tests rather than sink a bunch of time into debugging my environment.
Your build output will indicate the failures. I disabled by commenting out the [Fact] attribute.
These are the tests I disabled:
tests/Kudu.Core.Test/HgRepositoryFacts.cs (all tests)
Once you have a successful build that has created all the items in the artifacts you can move to deploying the Kudu website and web service code. The below instructions are for setting up a distinct web application instance, rather than dumping everything in c:\inetpub\wwwroot, which is how the instructions read.
Copy "artifacts\Release\KuduWeb" to the target area on the server where your website will run from. I run my kudu install with a separate host header, but you could as easily use a separate port or run as the root website. This directory will be the root for your web application.
Create an empty "App_Data" folder immediately under the KuduWeb folder.
Copy "artifacts\Release\SiteExtensions\Kudu" to the same level as the folder in step 1 and rename to "Kudu.Services.Web". This location is set as a relative path in the KuduWeb web.config file - setting serviceSitePath.
Open IIS Admin and create a website pointing to the "KuduWeb" folder from step 1.
Configure the app pool from step 4 to run as "LocalSystem". This is required to manage IIS Sites.
Create a new folder "apps" at the same level as KuduWeb. This is where deployments will be sent. Note: this location is controlled in the KuduWeb web.config file - setting "sitesPath"
Change filesystem permissions to grant "Users" full access to the "apps" folder created in the above step.
On starting my Kudu website, I got the following error.
Parser Error Message: Could not load file or assembly 'System.Web.Mvc, Version=5.1.0.0, Culture=neutral, PublicKeyToken=31bf3856ad364e35' or one of its dependencies. The system cannot find the file specified.
For some reason it didn't copy the appropriate MVC version into the deployment artifacts.
If you hit this error, the MVC 5 file can be obtained via NuGet. I found that my source code was built against 5.1.0, so this is the appropriate link:
https://www.nuget.org/packages/Microsoft.AspNet.Mvc/5.1.0
In order to extract the dll, I set up a new dummy project and used NuGet to pull down the dll via the package manager console.
Install-Package Microsoft.AspNet.Mvc -Version 5.1.0
Once you get the binary, copy it from package directory ( .\packages\Microsoft.AspNet.Mvc.5.1.0\lib\net45\System.Web.Mvc.dll ) to the website bin directory on the target machine.
At this point you are up and running. Use the web interface to create your application. It will create a subfolder under the "apps" directory with a tree that should be self explanatory. It will also have created two new websites for your application:
kudu_{your-app-name}
kudu_service_{your-app-name}
In a production situation, you should create an additional website running on appropriate port/host header that points to: .\apps\\site\wwwroot
Now you can add a git remote for your deployment. Go to your source location in a git console (ex: Git Bash) and add the remote as identified by Kudu. Note: you may need to change localhost in the url to be the appropriate server name.
git remote add deploy http://:52711/your-app-name.git
Push your code to the new "deploy" remote and see what happens. You should see all the normal push messages, plus the build output.
git push deploy master
My initial push failed to build and deploy due to "node" not being recognized. It was in the path, so a server reset convinced the path environment variable to be refreshed. You may find additional errors to work through. For instance, I had an issue with MSBuild being imported and causing a hiccup.
error MSB4019: The imported project "C:\Program Files (x86)\MSBuild\Microsoft\Visual Studio\v11.0\WebApplications\Microsoft.WebApplication.targets" was not found.
YMMV, but these are all solvable problems now. Good continuous deploying!
The project automatically sets up two websites on IIS for each application you add using the web front end. Kudu doesn't automatically map the bindings for them but it's relatively easy to open IIS and find the two sites named "kudu_appname" and "kudu_appname_service". The service website is the one that you point GIT too and the other one is the site itself. Just add public bindings to them by right-clicking and "edit bindings". You can then add public hostnames to them.
This is the easy part. The hard part that I'm still working on is getting authentication working so any random Joe isn't able to push to my kudu repository!