I have been struggling with setting up an iOS (Objective-C) app which utilizes the Service Management APIs. I've successfully was able to get Storage API calls working fine, but from what I'm reading, one big difference between the two Azure APIs seems to be a matter of authentication. For reference, what I'd ideally like to do is setup a simple API call to list the Hosted Services accounts in an app (eventually geared for public deployment).
According to the API Azure documentation, the Service Management APIs require a management certificate (.cer) to be uploaded and then for the client to utilize that cert to authenticate the request. My hunch is that this will prevent an app like the one I want to create from being feasible by any means, since public users with devices containing the downloaded app won't have that cert or the ability to attach it programatically in the objective-c code.
Is my hunch correct on this? Is this a forlorn idea that should not be pursued? I would think for this to be possible, the user would be required to upload their device's cert file to Azure somehow, and then to somehow have the app use this cert for authentication. I'm a bit lost on where to even begin on that, even if it is possible. :(
Any helpful info would be greatly appreciated. I have a lot of experience in the iOS side of things, but specifically in authentication/certificates of this type, I unfortunately have minimal experience.
Thanks in advance!!
-Vincent
Actually, I have an app that does exactly what you're talking about. CloudTools for Windows Azure is an iOS app that uses the Azure Management Service API to perform Azure management. It's been on iTunes for over a year.
Your question is somewhat broad, but I can tell you that the certificate issues were by far the biggest issues in designing/developing the app. You can't store the certificate in keychain, because keychain works in a manner such as this: a) you request a remote url that requires a certificate; b) that url tells you that it needs a certificate and c) keychain provides an appropriate certificate. It's a multi-request process. That won't work for Azure, because Azure service management APIs expect the initial call to include the certificate.
I require the users to add their certificate through iTunes File Sharing (steps here). Of course, they have to upload the public key portion to Azure. Then, I provide the certificate and private key with each request. I use the HTTP library ASIHttpRequest, although I believe that the latest (iOS 5.x) Apple libraries have similar functionality.
I'd be happy to provide any further details if you have any follow-up questions.
With Windows Azure, if you are using Service Management API then you really need to have certificate based authentication to create a SSL tunnel between your machine which is requesting the connection and Windows Azure Management Portal. I am not sure how wide your iOS application distribution is.
My first thought is that why would you want to deploy Windows Azure application from an iOS application, are you going to build application in iOS devices and deploy? Windows Azure Application deployment is mostly done on client machines so, Azure application management on iOS is great idea, however Application deployment from iOS not sure why. May be you are on something big here.. In both cases you really need Service Management Certificate on iOS device. If it is an enterprise app where you can let users to install Service Management certificate, it would be easier for iOS devices to use Service Mgmt API.
So if you want to use Service Management API from a client iOS app, I think the best solution will be to have WCF service hosted in Windows Azure which is configured to directly connect with your Windows Azure Portal. And from your iOS app, you just make call to your WCF service. This is very popular method to access service management API (through WCF Service) from any client app (WP7, iOS, Android) and the client side code is very light. On other hand you may need to pay to host a WCF service on Windows Azure.
Related
I am trying to add an ASP.NET 4.x app hosted externally (using AWS Elastic Beanstalk) into the Service-registry of an existing PCF.
Edit: Is this possible? If so, can someone give me an example about how this can be done
Assuming you have network connectivity in all directions between apps in PCF and the external app, yes this should be quite possible.
However, if you're using Spring Cloud Eureka, your externally-hosted app will need to get valid OAuth credentials so that it can authenticate prior to registering.
The comment by Daniel Mikusa is very appropriate for how I achieved this.
For Pivotal SCS, you would want to create a service instance (if you
don't have one already), then create a service key for your external
app. That will give you all of the binding info/creds you need to
connect from your remote service. A service key is the same as binding
a service to an app, except it's not tied to an app so it works well
for situations like this. Just give your service key a good name, so
you know that it's being used by an external app when you come back
and see it a year from now
I can't find an example code for publishing ASP.NET Webforms websites to Azure Functions. Months ago I tried to replicate the C# example but I ended up with only being able to use the precompiled batch function type.
I want to publish VB.NET web apps - any framework version, using Web Deploy...
Here are some important concepts you should know about Azure Web App and Azure Function:
Azure Web App:
Azure Web App is a sand box. The only way an Azure web app can be accessed via the internet is through the only two already-exposed HTTP (80) and HTTPS (443) TCP ports.
For Nodejs App deployed to Azure, Azure will create a named pipe for your server to listen, and pass the request from 443 port(as you use https) to the named pipe.
Azure Function:
Azure Functions is a solution for easily running small pieces of code, or "functions," in the cloud. You can write just the code you need for the problem at hand, without worrying about a whole application or the infrastructure to run it. Functions can make development even more productive, and you can use your development language of choice, such as C#, F#, Node.js, Python or PHP. Pay only for the time your code runs and trust Azure to scale as needed. Azure Functions lets you develop serverless applications on Microsoft Azure.
Api apps and Web apps are pretty much the same deal. Logic Apps and Functions are the same in a sense that they allow you to do something as a response to event or on a schedule, but Functions are a way to run code (or existing app) and Logic Apps are more like a workflow constructor, where you take existing actions and chain them (so no coding, or almost no)
Deploy:
Use ftp to deploy your web form to Azure Function. There will be no problems with the deployment, but the webpage will not display.
Note:
Although Azure Function and Azure Web App are very similar in many cases. But if you choose to deploy web form app, you will still find differences. Even if you can deploy your project to Azure Web App, it won’t display any webpages.
I have created UCMA trusted application using C# console application.
I want to monitor all Lync Users instant messaging calls (in one place) as well as store their conversations in the database via this console application.
Is it possible?.
If possible, please give an idea or any url.
UCMA trusted applications will only respond to traffic to the endpoints associated with that application so if you want to monitor all instant message traffic it would be the wrong API.
I would suggest using SIP Application API http://msdn.microsoft.com/en-us/library/office/hh364644(v=office.14).aspx to create a server application. This will also require the use of MSPL documented here
As Dai has asked - is the console application a requirement or will a windows service be ok?
Try our this sample application SipSnoop it basically shows all the metadata passing throught the lync server, you can tweak around it according to your requirement.
I am on a project that uses web services to communicate with hand held devices (Symbol MC70s running windows mobile 5).
We need to make sure that others on the internet can't connect to the web services and start sending info.
I have made a certificate so that the hand held will only connect to my services, but I don't know how to make sure that my services will only connect to my hand held devices.
My app is coded in the .NET Compact Framework and I am using Visual Studio 2008 SP1. My services are hosted in IIS and are coded with WCF in C# (.NET 3.5 sp1).
Any ideas?
Password-protect the services. Alternatively, authenticate the client by the client SSL certificate and issue those to your users.
No way to lock access down to a specific device. A device can be spoofed fairly easily.
When you say that you've "... made a certificate so that the hand held will only connect to my services", are you saying that the device uses https to the web service or that the device is locked down to only connect to the web service http URL?
Anyway, if you are concerned about the data pulled from the web service, you could encrypt it with a key known to your device.
I have client apps that talk to my silverlight application and its web services. So the client app is running on the client machine and making calls directly to the silverlight app running on the machine and also making web service calls.
I want the usernames/password security to be handled by the 3rd party client app.
Any idea how I can do this?
I'd try the Application Scenario's, Guidelines and How Tos sections of this CodePlex Link. You should be able to find a scenario that matches closely to yours and follow the guidelines and configuration to get yourself going.
If I've understood this correctly, the client application would pass a username/password to your silverlight app which would require a wsHttpBinding that has clientCredentials="Username". You would need to be able to authenticate this against a user store configured on your server, for example you could configure a SQL Server provider.
I'm not sure how your 3rd party client app works, but you would require a seperate security configuration for that communication. You could use a less secure binding if the apps were on the same machine and possibly use clientCredentials="Windows"/"None"/"Basic".
Difficult to advise further without knowing your exact situation. What do you have so far?