I am trying to configure IIS to host one set of binaries in two IIS websites. So we want to be able to access the urls:
internal.example.com and external.example.com are set up as different IIS sites to allow us to assign them different application pools. However when I added HTTPS support to our web.config, the internal HTTP support stopped working; http://internal.example.com/ADataService now returns the error:
Could not find a base address that matches scheme https for the endpoint with binding CustomBinding. Registered base address schemes are [http].
Here are the details of our web.config
<system.serviceModel>
<serviceHostingEnvironment aspNetCompatibilityEnabled="true" multipleSiteBindingsEnabled="true" />
<bindings>
<customBinding>
<binding name="jsonCustomMapper">
<webMessageEncoding webContentTypeMapperType="Service.JSONCustomMapper, Service" />
<httpTransport manualAddressing="true" />
</binding>
<binding name="httpsjsonCustomMapper">
<webMessageEncoding webContentTypeMapperType="Service.JSONCustomMapper, Service" />
<httpsTransport manualAddressing="true" />
</binding>
</customBinding>
</bindings>
<behaviors>
<endpointBehaviors>
<behavior name="jsonBehavior">
<webHttp />
</behavior>
</endpointBehaviors>
<serviceBehaviors>
<behavior name="defaultBehavior">
<serviceDebug includeExceptionDetailInFaults="true" />
</behavior>
</serviceBehaviors>
</behaviors>
<services>
<service name="Service.Service" behaviorConfiguration="defaultBehavior">
<endpoint address="json" binding="customBinding" bindingConfiguration="jsonCustomMapper" behaviorConfiguration="jsonBehavior" contract="Service.IJSONService" />
<endpoint address="json" binding="customBinding" bindingConfiguration="httpsjsonCustomMapper" behaviorConfiguration="jsonBehavior" contract="Service.IJSONService" />
</service>
</services>
</system.serviceModel>
From what I understand multipleSiteBindingsEnabled="true"
and HTTPS do not mix, but I don't understand what resources they would be sharing? If we have internal.example.com and external.example.com hosted in different app pools, I thought they would have process isolation?
seems that adding an HTTPS certificate to the site for "internal.example.com" fixed the issue. Note: without this certificate we were not able to access internal.example.com over either HTTP or HTTPS, with certificate both mechanisms work correctly.