I have a website running 2 WCF services. Working awesome until I needed to move them from the "Default Web Site" in IIS7 to a dedicated website with a new IP address. Now I get the error message:
"...could not be loaded because more than one endpoint configuration for that contract was found. Please indicate the preferred endpoint configuration section by name"
ALL I did was change:
net.tcp://localhost/...
TO:
net.tcp://10.1.2.204/...
I have changed every spot in my configs which once used localhost to specified IP address. The new website is setup exactly the same with the proper net.tcp bindings and it is bound to the single IP address of 10.1.2.204.
I cannot seem to figure out what is going on. I am looking at the svc trace logs and all I get is a bunch of the same.
My new website is called "core" whereas previously I was hosting on "Default Web Site". I added a new FQDN/public IP to route to this new site (backwards to the mentioned private IP address). Now my services are busted.
I know baseAddress is ignored when using IIS/WAS hosting, so I am not sure what else to check. I changed absolutely nothing else in the configs.
What else can I provide to help troubleshoot this?
--
I am also seeing this in the service trace viewer:
Activity Name Receive bytes on connection 'Listen at 'net.tcp://hole.myserver.com/...'.'.
...but the WCF services are actually hosted on that new "core" website which translates to my new private IP address. I have change FQDN names for the example. To make it clear:
I used to have 1 FQDN on this server using Default Web Site.
I now added a 2nd FQDN to this server, new website, second IP. It seems to me that the listening should be directed towards: net.tcp://core.myserver.com instead of net.tcp://hole.myserver.com. Let's pretend the server name is hole.myserver.com and my new website is core.myserver.com on the same server.
Hopefully this is not confusing and you can make sense of it.
I'm using the default net.tcp port of 808 and binding of 808:*
--
This is the first exception thrown that I can tell:
The socket connection was aborted. This could be caused by an error processing your message or a receive timeout being exceeded by the remote host, or an underlying network resource issue. Local socket timeout was '00:00:10'.
--
None of these combos will activate the service and all produce the same exception:
net.tcp://localhost
net.tcp://localhost:808
net.tcp://10.1.2.204
net.tcp://10.1.2.204:808
even using a different port which I have done before without issue:
net.tcp://localhost:75
net.tcp://10.1.2.204:75
none of these work. :(
Thanks so much!
Its rights issue. Right click on your new website --> then Edit Permissions. On properties window click security tab and Edit button. Then Add IIS_IUSRS and give Read & Execute, List folder contents, Read permissions. Now iisreset and start to host and connect net tcp on new website.
Related
I hosted silverlight application iis and that can be accessed in multiple ways..
When I access using Qualified name of server IP address IT WORKS fine:
http://[QUALIFIEDNAME]/WebUAT/App.aspx
When I access using IP address like below:
http://[IPADDRESS]/WebUAT/App.aspx
It gives error below:
[Async_ExceptionOccurred]
Arguments:
Debugging resource strings are unavailable. Often the key and arguments provide sufficient information to diagnose the problem. See http://go.microsoft.com/fwlink/?linkid=106663&Version=5.1.20913.00&File=System.dll&Key=Async_ExceptionOccurred StackTrace : at System.ComponentModel.AsyncCompletedEventArgs.RaiseExceptionIfNecessary()
Follow these,
(i) Try to use fiddler and find the request which is making crossdomainpolicy xml request.
(ii)Check whether it is failing, if so add the crossdomain and clientaccesspolicy to the root folder of your Application
(iii)Check whether the service which is invoked by your application is blocked by firewall.
I have added a WCF service to an existing ASP.Net web project on our test website (only accessible on company network). I can access the service by typing the URL in a browser, but if I add a service reference to a web project, I get a runtime error that 'the remote name could not be resolved'.
In my web.config, I have the bypassProxyOnLocal binding attribute set to 'true' (we use a proxy on the company network), but it seems as though our test domain is not recognized as a local domain (or recognized at all, really). Again, this is only when I add it as a service reference. Obviously the domain is valid, because I can add it as a service reference, and can event paste the URL into a browser, and it works. I just get a runtime error after I add it as a service reference. Anyone have any advice?
According to MSDN WSHttpBindingBase.BypassProxyOnLocal Property, "An Internet resource is local if it has a local address. A local address is one that is on same computer, the local LAN or intranet and is identified, syntactically, by the lack of a period (.) as in the URIs "http://webserver/" and "http://localhost/".
If your test domain is "http://test.mycompany.com/", then it won't be viewed as a local address. Perhaps using "http://test/" in this case would work?
Finally got this resolved. The request was somehow being blocked by the company network.
I'm doing my dev work on a Window 7 x64 machine, deploying to a Windows 2008 x32 server. At the moment I'm adding WCF services to some internal apps so that we can build smaller clients using net.tcp bindings that report to the user what the server is doing without running multiple instances of the server. To cut back on how much administration the apps will require, I tried enabling port sharing on my first server app. I'm using the app to self-host the WCF service so they can be easily moved from one server to another if necessary. Here's the code starting the server:
Dim _service_host As ServiceHost
Dim active_server_address As Uri = New UriBuilder("net.tcp", "localhost", CInt(My.Settings.ServerPort)).Uri
_service_host = New ServiceHost(GetType(UpdateServiceOps), active_server_address)
_service_host.AddServiceEndpoint(GetType(IUpdateService), New NetTcpBinding With {.Name = "endpoint_tcp"}, "MiddlewareEndpoint")
_service_host.Description.Behaviors.Add(New ServiceMetadataBehavior)
_service_host.AddServiceEndpoint(GetType(IMetadataExchange), MetadataExchangeBindings.CreateMexTcpBinding, "mex")
_service_host.Open()
That works great. When I change it to what's below, however, I get an error. Here's the code:
Dim _service_host As ServiceHost
Dim active_server_address As Uri = New UriBuilder("net.tcp", "localhost", CInt(My.Settings.ServerPort)).Uri
_service_host = New ServiceHost(GetType(UpdateServiceOps), active_server_address)
_service_host.AddServiceEndpoint(GetType(IUpdateService), New NetTcpBinding With {.Name = "endpoint_tcp", .PortSharingEnabled = True}, "MiddlewareEndpoint")
_service_host.Description.Behaviors.Add(New ServiceMetadataBehavior)
_service_host.AddServiceEndpoint(GetType(IMetadataExchange), MetadataExchangeBindings.CreateMexTcpBinding, "mex")
_service_host.Open()
I get the following error when I put a breakpoint at the last line: Unable to automatically step into the server. Connecting to the server machine 'nettcpportsharing' failed. The requested name is valid, but no data of the requested type was found. The Try...Catch block housing that code says the exception type is AddressAlreadyInUseException...but when I run netstat I don't see anything else listening on that address. There's no occurrence of 'nettcpportsharing' anywhere in my solution. I've checked to make sure that the Net.Tcp Port Sharing service is started. Any ideas?
I think this may be a permissions issue. Please see the associated article which explains how to configure the port sharing service to support self-hosted services.
(For production purposes I would strongly recommend using IIS Hosting with WAS anyway - it makes management of the services much cleaner and you get dynamic startup/shutdown for free.)
I'm trying to get WCF to use SSL with ANYTHING for FIVE DAYS now. I've gone through countless walkthroughs, generated more certificates than a mail order diploma company, even tried hot fixes. After working with MS dev tools since VB1, I am now considering flipping burgers as a career option. WCF, as far as I can see, is a complete lemon.
Anyway, to get to my actual question: If I run through this walkthrough:
http://msdn.microsoft.com/en-us/library/ff648840.aspx
I get to step 11 (adding the service reference) and get "There was an error downloading metadata from the address. Please verify that you have entered a valid address".
Details of the error gives: There was an error downloading 'https://localhost/SSL6/Service.svc'.
Unable to connect to the remote server
No connection could be made because the target machine actively refused it 127.0.0.1:443
I'm using VS2008 on Windows 7 with IIS7. I followed the walkthrough exactly (apart from step 5 which was different on IIS7- I went into "SSL Settings" for the VD), so it shows my config (yes I've used httpsGetEnabled and mexHttpsBinding).
Anyone care to save my sanity and job?
EDIT: If I go into IIS, select the VD in content view, right-click on the svc file and browse, I get "Internet Explorer cannot display the webpage". Chrome gives "Google Chrome could not connect to localhost".
IE troubleshooting gives "the remote device or resource won't accept the connection".
If I browse using the IP address rather than using localhost via http, it says that it's secured with https ok. If I browse using the IP and https, I get HTTP error 503. The service is unavailable.
So it looks to me like a DNS issue combined with... something.
When I try to just run the service site project, I get "Unable to start debugging on the web server. Unable to connect to the web server. Verify that the web server is running and that incoming HTTP requests are not blocked by a firewall". I've checked the firewall and it's ok.
Finally cracked it. There were at least three issues at play.
1) A DNS issue of some kind with localhost. It's still unresolved on my machine, but I can work around it by using the IP addy.
2) Another issue may have been that apparently, WCF doesn't work with IIS 7 OOTB. So you need to run command prompt as administrator, and run the following command -
"%windir%\Microsoft.NET\Framework\v3.0\Windows Communication Foundation\ServiceModelReg.exe" -r -y
3) After I got through the certificate stuff I was still getting HTTP error 503, "Service unavailable". That ended up being leftovers from my previous attempts, still listening to ports:
http://blogs.msdn.com/webtopics/archive/2010/02/17/a-not-so-common-root-cause-for-503-service-unavailable.aspx
So to get a hello world level service happening with WCF and SSL took me a whole week, and in my travels I discovered many pilgrims who had taken about the same amount of time. Microsoft: You have failed.
Are you using IIS or self hosting? If you're using IIS, it sounds like it's incorrectly configured, because it seems it's not accepting connections on port 443; I guess you're probably missing a protocol binding (https to port 443). There's a detailed discussion of setting up SSL on IIS7 here that might be useful.
Of course, you could easily verify this using the browser, you should be able to connect to the site using SSL from it.
When I have had this error occur, I have found it very useful to run the service from Visual Studio to get additional information. Right-click on the service, and then select Debug -> Start New Instance from the pop-up menu. VS will launch the service using the WcfTestClient.exe.
WcfTestClient will display all the services and endpoints in your service project. A healthy launch will list your services in the a panel on the left, and provide a Start Page tab in a panel on the right. What will happen in your case, however, is that you'll get a list of services in a panel on the top and an "Additional Information" box along the bottom. Your problem service(s) will have a Status of Error.
Click on a problem service in the upper panel, and the Additional Information box will fill with an error message and stack trace. The message will tell you exactly what that problem mex (metadata exchange) address is. That may be enough of a hint for you to solve the problem. If not, post the Additional Information here and I'll be happy to take a look at it.
I've got a rather confusing problem.
Web Service A - Called directly by Win32 App, and various other web applications
Web Service B - Called directly by same Win32 App and a different set of other Web Applications
I'm working on adding a reference to B into A, so that we can use the functionality B provides without rewriting/reintegrating that code manually into A. I have a develop copy of A on my PC that I'm working with, and am calling a test copy of B on a remote server (the test server is accessed over the internet at our hosting provider)
I believe to have my code ready for testing, but every time I go to test it on my local copy, I get the following error.
No connection could be made because the target machine actively refused it 10.1.X.XX:8080
at System.Net.Sockets.Socket.DoConnect(EndPoint endPointSnapshot, SocketAddress socketAddress)
at System.Net.Sockets.Socket.InternalConnect(EndPoint remoteEP)
at System.Net.ServicePoint.ConnectSocketInternal(Boolean connectFailure, Socket s4, Socket s6, Socket& socket, IPAddress& address, ConnectSocketState state, IAsyncResult asyncResult, Int32 timeout, Exception& exception)
The internal IP listed in the error message was our old BorderManager gateway. It is no longer in use, I've checked all Machine.Configs, Web.Configs, registry entries, even gone so far as doing a full text search of all files on my PC under 2MB for the IP listed there. I double checked the Internet Options Connections settings, but that is managed by a group policy company wide. I've called in a networking guy to verify that the IP is not listed anywhere in the currently applied policy.
I can't find it anywhere except when trying to step over the code that uses it, and getting the exception shown above.
The code I have for using Web Service B is:
Dim oZip As New ZipTerr.ZIPTerr
oZip.Url = "https://test.test.com/ZIPTerr/ZIPTerr.asmx" ' This really shouldn't be hardcoded...
oZip.PreAuthenticate = True
oZip.Credentials = New System.Net.NetworkCredential("username", "password")
aCity = oZip.GetCities(pstrZip, Date.Now())
It fails on the call to oZip.GetCities() with the above error. The IP in the error message does not in any way indicate it's made it out of our local network, it's a 10.* private IP, whereas the test server has a valid live publicly accessible IP.
Any ideas??
Update:
Both web services are in .NET 2.0
The underlying System.Net libraries will not be using the Internet Settings values in this case since the code is running as a service.
MS are a little sketchy on the details of how its proxy settings are derived in the absence of specific settings in the .config files.
Here are some things to try:-
One command line use:-
ProxyCfg -d
This would remove any proxy settings used by the WinHTTP component (which was designed to be used in just such a server setting by code prior to the advent of .NET).
ProxyCfg -u
Copies the current users Internet proxy settings to the system-wide setting used by WinHTTP.
Better yet configure the proxy settings yourself in the web.config of the calling application:-
<system.net>
<defaultProxy enabled="true">
<!-- if above set to true configure below accordingly if false delete below-->
<proxy proxyaddress="http://127.0.0.1:8888" bypassonlocal="False"/>
</defaultProxy>
</system.net>
Best Guess:
Either the error is coming from code that is not on your machine. Is Web service A and B running on your machine?
Or it is hard coded in a dll where the source for the dll is not on your machine.
Hope this helps
Shiraz