I have a ASP.Net web site that get all business logic from a WCF hosting site (IIS). Once in a while, it looks like the backend WCF frozen and that makes web front-end stop responding. I have to recycle both application pools to make it work again.
It happens more often lately, could be because we have more & more customers using the site. Before, it happens once every month, now it happens once every week. Maybe more.
We DO Close connection after each SVC call.
Error message received in Event Log after I recycle the pool
Message: System.ServiceModel.CommunicationException: An error occurred while receiving the HTTP response to .........../.....BusinessServices.svc. This could be due to the service endpoint binding not using the HTTP protocol. This could also be due to an HTTP request context being aborted by the server (possibly due to the service shutting down). See server logs for more details. ---> System.Net.WebException: The underlying connection was closed: An unexpected error occurred on a receive. ---> System.IO.IOException: Unable to read data from the transport connection: An existing connection was forcibly closed by the remote host. ---> System.Net.Sockets.SocketException: An existing connection was forcibly closed by the remote host
Web.config of the web app:
<bindings>
<wsHttpBinding>
<binding name="WSHttpBinding_IBusinessServices" closeTimeout="00:31:00" openTimeout="00:31:00"
receiveTimeout="00:10:00" sendTimeout="00:31:00"
bypassProxyOnLocal="false" transactionFlow="false"
hostNameComparisonMode="StrongWildcard" maxBufferPoolSize="524288"
maxReceivedMessageSize="2147483647" messageEncoding="Text"
textEncoding="utf-8" useDefaultWebProxy="true" allowCookies="false">
<readerQuotas maxDepth="999" maxStringContentLength="2147483647" maxArrayLength="1638400"
maxBytesPerRead="4096" maxNameTableCharCount="16384" />
<reliableSession ordered="true" inactivityTimeout="00:10:00" enabled="false" />
<security mode="Message">
<transport clientCredentialType="Windows" proxyCredentialType="None" realm="" />
<message clientCredentialType="Windows" negotiateServiceCredential="true" algorithmSuite="Default" />
</security>
</binding>
</wsHttpBinding>
</bindings>
<client>
<endpoint address="http://localhost:8890/MyBusinessServices.svc" binding="wsHttpBinding"
bindingConfiguration="WSHttpBinding_IBusinessServices"
contract="MyBusinessServices.IBusinessServices"
name="WSHttpBinding_IBusinessServices" />
</client>
Web.config of the WCF service, with throttling set to 1500
.....
<behavior name="AppServiceBehaviors">
<serviceMetadata httpGetEnabled="true" />
<serviceDebug includeExceptionDetailInFaults="true" />
<dataContractSerializer maxItemsInObjectGraph="2147483647" />
<serviceTimeouts transactionTimeout="00:15:00" />
<serviceThrottling maxConcurrentCalls="1500" maxConcurrentSessions="1500"
maxConcurrentInstances="2147483647" />
</behavior>
<behavior name="">
<serviceMetadata httpGetEnabled="true" />
<serviceDebug includeExceptionDetailInFaults="true" />
<dataContractSerializer maxItemsInObjectGraph="2147483647" />
</behavior>
</serviceBehaviors>
</behaviors>
This problem is intermittent but it's driving me crazy. Any ideas/suggestions is appreciated.
Eric
The issue you are facing right now can be due to many other factors that needs a lot more description than what we currently have.
The thrown exception is a kind of generic exception that may point to many scenarios possible. For example, if you have a load balancer in front of your service servers and one of them has an error in runtime but the other does not. Another example is the sql server is creating an exception at server side, but your wrapping exceptions may not send the related information to your client. So you cannot know exactly what is going on.
I have a few suggestions for you to check:
I can see that you have expressed that you are closing connection after each call. But you need to make sure that the calls that receive an exception is also closed. So your svc calls should look like this:
try
{
if (this.client != null)
{
IClientChannel channel = this.client as IClientChannel;
if (channel.State == CommunicationState.Faulted)
{
channel.Abort();
}
else
{
channel.Close();
}
}
}
finally
{
this.client = null;
}
Other than that, you may also want to check your Application Pool settings on IIS. You can make sure from event logs, that you are not forced to recycle/shutdown with Maximum Failures setting. (In advanced app pool settings - rapidFailProtectionMaxCrashes)
My 2 cents for you,
Service abort - error due to fatal error occurred inside,
Make sure while debugging WCF in VS IDE in Dev, keep the Exceptions(Ctrl D,E), check all "unhandled exception" columns. This enables IDE to break the application code line exactly, inspite of the CATCHes..this will unearth any bugs.
Please do let us know, if any findings here...
Good Luck,
HydTechie
Related
Using VS2019, ASP.NET project running .net 4.0.
I created the soap client adding the Service Reference to the wsdl file. Now I'm configuring the certificate and calls method.
This way worked in the old soap server, but now the soap server changed. I tested with SoapUI using the same Basic Auth configuration and works perfectly, but not with my .Net 4.0 client...
Web.config
<system.serviceModel>
<bindings>
<customBinding>
<binding name="PLATAFORMA">
<textMessageEncoding messageVersion="Soap11WSAddressing10" />
<security
authenticationMode="MutualCertificateDuplex"
messageProtectionOrder="SignBeforeEncrypt"
messageSecurityVersion="WSSecurity10WSTrustFebruary2005WSSecureConversationFebruary2005WSSecurityPolicy11BasicSecurityProfile10">
<secureConversationBootstrap />
</security>
<httpsTransport />
</binding>
</customBinding>
</bindings>
<behaviors>
<endpointBehaviors>
<behavior name="CERT">
<clientCredentials>
<clientCertificate findValue="ClientCert" storeLocation="CurrentUser" storeName="My" x509FindType="FindBySubjectName" />
<serviceCertificate>
<defaultCertificate findValue="*.ServerCert.com" storeLocation="CurrentUser" storeName="My" x509FindType="FindBySubjectName" />
</serviceCertificate>
</clientCredentials>
</behavior>
</endpointBehaviors>
</behaviors>
<client>
<endpoint
address="endpointURI"
binding="customBinding"
bindingConfiguration="PLATAFORMA"
behaviorConfiguration="CERT"
contract="ServiceReference1.RequestPort1"
name="Request.Request1">
<identity>
<dns value="*.ServerCert.com" />
</identity>
</endpoint>
</client>
</system.serviceModel>
XML Outoing header (catched with intercerptor):
<s:Envelope xmlns:a="http://www.w3.org/2005/08/addressing" xmlns:s="http://schemas.xmlsoap.org/soap/envelope/">
<s:Header>
<a:Action s:mustUnderstand="1">peticionSincrona</a:Action>
<a:MessageID>urn:uuid:e3a5c4bd-f159-48c1-8f3f-cf22da6b7e3b</a:MessageID>
<ActivityId CorrelationId="035f2491-0772-4b1e-a286-9be30720d5ea" xmlns="http://schemas.microsoft.com/2004/09/ServiceModel/Diagnostics">4d7f76c2-0486-4a81-93ad-32aecf02b035</ActivityId>
<a:ReplyTo>
<a:Address>http://www.w3.org/2005/08/addressing/anonymous</a:Address>
</a:ReplyTo>
<VsDebuggerCausalityData xmlns="http://schemas.microsoft.com/vstudio/diagnostics/servicemodelsink">uIDPo9ph/BcvvnBFuLQYdch+LyYAAAAAHsOKxYeqzk+Do5pQmamNIPUdiXOiYjpBl1dsV5pp+SMACQAA</VsDebuggerCausalityData>
</s:Header>
I understand that some information is needed, how can I specify to sign as XMLDsig? why is not soap/envelope/encoding? I need help to configure the soap client.
The server return error 500 with a default tomcat error, is not even soapenv:fault or similar. I think the request envelope is not generating properly.
EDIT: Must be over soap11
Is this the client configuration generated after changing the SOAP server? If it requires the Basic authentication, why do we not need to provide username/password, but just need to provide a client certificate(according to the binding type)?
I suggest you re-generate a client proxy class by adding service reference. this also generates a proper configuration in the webconfig file.
Besides, since the server changes, the server’s certificate used to implement HTTPS security may also change, so the default certificate we provide on the client-side should also be changed.
<serviceCertificate>
<defaultCertificate findValue="*.ServerCert.com" storeLocation="CurrentUser" storeName="My" x509FindType="FindBySubjectName" />
</serviceCertificate>
Mutual certificate authentication requires a trust relationship between the client and the server.
<security
authenticationMode="MutualCertificateDuplex"
Not only do we need to install a client certificate on the new server in order to trust the client, we also need to install the server’s certificate on the client-side.
For details,
https://learn.microsoft.com/en-us/dotnet/framework/wcf/feature-details/transport-security-with-certificate-authentication
https://learn.microsoft.com/en-us/dotnet/framework/wcf/feature-details/message-security-with-a-certificate-client
Feel free to let me know if there is anything I can help with.
I have a library that contains a Service Reference generated proxy to a remote ASMX web service. When I try to call methods on that proxy from an ASP.NET application, I get the following error:
System.ServiceModel.EndpointNotFoundException: There was no endpoint
listening at https://domain.com/path/to/the.asmx that could accept the
message. This is often caused by an incorrect address or SOAP action.
See InnerException, if present, for more details. --->
System.Net.WebException: The remote server returned an error: (404)
Not Found.
However, if I call the same methods from a simple console app dropped into the bin folder (and therefore referencing the exact same libraries), it works. I have the same, simple configuration in web.config and the console app's app.config:
<system.serviceModel>
<bindings>
<customBinding>
<binding name="ServicesSoap12">
<textMessageEncoding messageVersion="Soap12" />
<httpsTransport />
</binding>
</customBinding>
</bindings>
<client>
<endpoint address="https://domain.com/path/to/the.asmx"
binding="customBinding" bindingConfiguration="ServicesSoap12"
contract="TheContract" name="ServicesSoap12" />
</client>
</system.serviceModel>
Tracing revealed no further useful information. The ASMX endpoint is definitely available. What could cause this behavior, and what other troubleshooting steps can I take?
This turned out to be a parameter issue. My production code was passing null for an optional string parameter, but my test code was passing a guid. The parameter is optional in the contract, so I don't understand why it triggered a 404, but I'm able to consistently reproduce the problem by passing null.
The contract parameter is defined as follows:
<s:element minOccurs="0" maxOccurs="1" name="userdef1" type="s:string"/>
The 404 error threw me off from the beginning. It's a reminder to always write your test code exactly like the code that doesn't work.
I have created a WCF service hosted in an ASP.NET web application, with netTcpBinding. I want to make it run, but I'm always getting the "object not set to a reference"-error.
I first created my service and tested it using the default bindings. It worked, I got my expected results returned.
As I don't want to run it through http, but through tcp, I changed my bindings as follow:
<system.serviceModel>
<services>
<service name="be.xxx.xxx.WCF.Leasing">
<endpoint address="" binding="netTcpBinding" bindingConfiguration="" contract="be.xxx.xxx.WCF.ILeasing" />
<endpoint address="mex" binding="mexTcpBinding" bindingConfiguration="" contract="IMetadataExchange" />
<host>
<baseAddresses>
<add baseAddress="net.tcp://localhost/be.xxx.xxx.WCF/Leasing.svc"/>
</baseAddresses>
</host>
</service>
</services>
<behaviors>
<serviceBehaviors>
<behavior>
<serviceMetadata httpGetEnabled="false" httpsGetEnabled="false" />
<serviceDebug includeExceptionDetailInFaults="true" />
</behavior>
</serviceBehaviors>
</behaviors>
<serviceHostingEnvironment aspNetCompatibilityEnabled="true" multipleSiteBindingsEnabled="true" />
In the project properties, I've changed under the web tab that it should run under "Local IIS". Set the project URL to http://localhost/be.xxx.xxx.WCF and clicked the "Create Virtual Directory"-button.
I have also changed my IIS settings:
Application pool to .Net Framework v4.0.30319, Integrated
Under "Advanced Settings", I've added "net.tcp" for Enabled Protocols.
My Windows services "Net.Tcp Listener Adapter" and "Net.TCP Port Sharing Service" are both running.
I can browse to http://localhost/be.xxx.xxx.WCF/Leasing.svc, telling me that I have created a service.
So when I start debugging by hitting F5 (VS is running under administrator rights), I get the following screen:
OK, when I start the stand-alone WCF Test Client, and I reference to http://localhost/be.xxx.xxx.WCF/Leasing.svc/mex, it loads the functions:
First, it was totally giving me the nothing saying "Object reference not set to an instance of an object" exception.
Just while setting up this question, it ran, giving me back the expected results.
I couldn't believe why it ran this time, so I closed the WCF Test Client again, and retried my effort. Now it is giving me again the "Object reference not set to an instance of an object" exception.
I also tried adding the service reference to a test console application, but this is giving me the same problems.
So, why can't I debug in Visual Studio (~metdata not found), and why doesn't seem it to be stable (running once)?
Have you tried to define the http mex binding
I have the following WCF :
[OperationContract]
IList<OperationRoomDto> GetOperationRooms();
and I have a website that uses this WCF, when I use Visual studio (Asp.Net development server) it works fine, but when I deploy the WCF to IIS I get this exception:
The remote server returned an error: (405) Method Not Allowed.
I made a simple website to test this WCF (on IIS) and It worked.
this is a part of the website web.config file:
<endpoint address="http://server/WcfService"
binding="basicHttpBinding" bindingConfiguration="BasicHttpBinding_IMessageService"
contract="MyProject.Web.Wcf.IMessageService" name="BasicHttpBinding_IMessageService">
</endpoint>
and this part of the web service config file:
<system.serviceModel>
<behaviors>
<serviceBehaviors>
<behavior>
<!-- To avoid disclosing metadata information, set the value below to false and remove the metadata endpoint above before deployment -->
<serviceMetadata httpGetEnabled="true" />
<!-- To receive exception details in faults for debugging purposes, set the value below to true. Set to false before deployment to avoid disclosing exception information -->
<serviceDebug includeExceptionDetailInFaults="true" />
</behavior>
</serviceBehaviors>
</behaviors>
<serviceHostingEnvironment multipleSiteBindingsEnabled="true" />
I tried to debug the w3wp.exe service to get any hidden exception,I put a break point at the web service methods, but the debugger didn't even reach there, and the exception happened before reaching to the web service class.
I figured out the problem, it was because the Url of the wcf is not correct, it misses the service file name
<endpoint address="http://server/WcfService"
it should become :
<endpoint address="http://server/WcfService/Service.svc"
I have a wcf service that randomly begins to fail, when requesting the autogenerated javascript that wcf supports in making. But I have no luck tracking down why,the js thing is part of the wcf featureset.So I dont know how it can suddenly begin to fail and be unable to work until IIS is recycled.
The http log gives me:
2010-06-10 09:11:49 W3SVC2095255988 myip GET /path/myservice.svc/js _=1276161113900 80 - ip browser 500 0 0
So its an error 500, and that is about the only thing I can figure out. The event log contains no information.
Requests to /path/myservice.svc works just fine. After recycling IIS it works again, and some days later it begins to fail until I recycle IIS.
<service
name="path.myservice"
behaviorConfiguration="b">
<endpoint
address=""
behaviorConfiguration="eb"
binding="webHttpBinding"
contract="path.Imyservice" />
</service>
...
<endpointBehaviors>
<behavior name="eb">
<enableWebScript />
</behavior>
</endpointBehaviors>
<serviceBehaviors>
<behavior name="b">
<dataContractSerializer maxItemsInObjectGraph="2147483647" />
<serviceMetadata httpGetEnabled="true" />
<serviceDebug includeExceptionDetailInFaults="true" />
</behavior>
</serviceBehaviors>
I dont see any problems in the web.config settings either.
Any clues how I can track down what the problem is?
Edit:
Just to make it clear - It is the generation of javascript that fails, my code is never invoked. Calls to the service works just fine.
I would recommend you to enable tracing in your service and look at the generated logs with Service Trace Viewer Tool. Trace only errors to keep your logs small.
It seems your WCF Service throws an Exception. You can catch this unhandled Exception in the Application_Error method in Global.asax, than write it to a log file.
protected void Application_Error(object sender, EventArgs e)
{
Exception ex = Server.GetLastError().GetBaseException();
//TODO: write to log
}