public UserMailDto getUserByEmail(String email) throws UserExceptionMessage {
try {
return userRepository.searchByMail(email);
} catch (DataAccessException | JDBCConnectionException accessException) {
com.example.user_service.config.log.Logger.errorLog("UserService", accessException.getMessage());
throw new DataAccessExceptionMessage(Messages.ERROR_TRY_AGAIN + accessException.getMessage());
}
}
In my view you should handle it, but not with a try..catch block in what seems to be your service. Why are you catching the Exception and then rethrowing you own custom Exception with a message? You will need to handle that Exception at some point in your code to. So you are not really handling it here.
It looks like you are building a web app, so I would recommend that you handle your Exceptions in one central place in a #ControllerAdvice class. You can read about it here. This way you can really handle the Exception, by returning a corresponding status code to the user. 503 in your case.
Related
I am using Avro and Schema registry with my Spring Kafka setup.
I would like to somehow handle the SerializationException, which might be thrown during deserialization.
I found the following two resource:
https://github.com/spring-projects/spring-kafka/issues/164
How do I configure spring-kafka to ignore messages in the wrong format?
These resources suggest that I return null instead of throwing an SerializationException when deserializing and listen for KafkaNull. This solution works just fine.
I would however like to be able to throw an exception instead of returning null.
KIP-161 and KIP-210 provide better features to handling exceptions. I did find some resources mentioning KIP-161 in Spring Cloud, but nothing specific about Spring-Kafka.
Does anyone know how to catch SerializationException in Spring Boot?
I am using Spring Boot 2.0.2
Edit: I found a solution.
I would rather throw an exception and catch it than having to return null or KafkaNull. I am using my custom Avro serializer and deserializer in multiple different project, some of which are not Spring. If I changed my Avro serializer and deserializer then some of the other projects would need to be changed to expect the deserializer to return null.
I would like to shutdown the container, such that I do not lose any messages. The SerializationException should never be expected in production. The SerializationException should only be able to happen if Schema Registry is down or if an unformatted message somehow is sent to the production kafka. Either way, SerializationException should only happen very rarely, and if it happens then I want to shutdown the container such that no messages are lost and I can investigate the issue.
Just take into consideration that will catch all exceptions from your consumer container. In my specific case I just want to only shutdown if it is a SerializationException
public class SerializationExceptionHandler extends ContainerStoppingErrorHandler {
#Override
public void handle(Exception thrownException, List<ConsumerRecord<?, ?>> records, Consumer<?, ?> consumer,
MessageListenerContainer container) {
//Only call super if the exception is SerializationException
if (thrownException instanceof SerializationException) {
//This will shutdown the container.
super.handle(thrownException, records, consumer, container);
} else {
//Wrap and re-throw the exception
throw new KafkaException("Kafka Consumer Container Error", thrownException);
}
}
}
This handler is passed to the consumer container. Below is an example of a
KafkaListenerContainerFactory bean.
#Bean
public KafkaListenerContainerFactory<ConcurrentMessageListenerContainer<Integer, String>>
kafkaListenerContainerFactory(JpaTransactionManager jpa, KafkaTransactionManager<?, ?> kafka) {
ConcurrentKafkaListenerContainerFactory<Integer, String> factory =
new ConcurrentKafkaListenerContainerFactory<>();
factory.setConsumerFactory(consumerFactory());
factory.setConcurrency(1);
factory.getContainerProperties().setPollTimeout(3000);
factory.getContainerProperties().setErrorHandler(new SerializationExceptionHandler());
factory.getContainerProperties().setTransactionManager(chainedTxM(jpa, kafka));
return factory;
}
There is nothing Spring can do; the deserialization occurs before the consumer gets any data. You need to enhance the deserializer.
I would however like to be able to throw an exception instead of returning null.
That won't help anything since Kafka won't know how to deal with the exception. Again; this all happens before the data is available so returning null (or some other special value) is the best technique.
EDIT
In 2.2, we added an error handling deserializer which delegates to the actual deserializer and returns null, with the exception in a header; the listener container then passes this directly to the error handler instead of the listener.
As per Customizing Error Handling "Throwing the exception in the catch block will forward the message to the error queue. If that's not desired, remove the throw from the catch block to indicate that the message has been successfully processed." That's not true for me even if I simply swallow any kind of exception in a behavior:
public override async Task Invoke(IInvokeHandlerContext context, Func<Task> next)
{
try
{
await next().ConfigureAwait(false);
}
catch (Exception ex)
{
}
}
I put a breakpoint there and made sure execution hit the catch block. Nevertheless after intimidate and delayed retries messages inevitably ends up in error queue. And I have no more Behaviours in the pipeline besides this one.
Only if I run context.DoNotContinueDispatchingCurrentMessageToHandlers(); inside the catch block it prevents sending error to the error queue, but it also prevents any further immediate and delayed retries.
Any idea on why it works in contravention of Particular NserviceBus documentation is very appreciated
NserviceBus ver. used: 6.4.3
UPDATE:
I want only certain type of exceptions not being sent to an error queue in NServiceBus 6, however to make test case more clear and narrow down the root cause of an issue I use just type Exception. After throwing exception, execution certainly hits the empty catch block. Here is more code to that:
public class EndpointConfig : IConfigureThisEndpoint
{
public void Customize(EndpointConfiguration endpointConfiguration)
{
endpointConfiguration.DefineEndpointName("testEndpoint");
endpointConfiguration.UseSerialization<XmlSerializer>();
endpointConfiguration.DisableFeature<AutoSubscribe>();
configure
.Conventions()
.DefiningCommandsAs(t => t.IsMatched("Command"))
.DefiningEventsAs(t => t.IsMatched("Event"))
.DefiningMessagesAs(t => t.IsMatched("Message"));
var transport = endpointConfiguration.UseTransport<MsmqTransport>();
var routing = transport.Routing();
var rountingConfigurator = container.GetInstance<IRountingConfiguration>();
rountingConfigurator.ApplyRountingConfig(routing);
var instanceMappingFile = routing.InstanceMappingFile();
instanceMappingFile.FilePath("routing.xml");
transport.Transactions(TransportTransactionMode.TransactionScope);
endpointConfiguration.Pipeline.Register(
new CustomFaultMechanismBehavior(),
"Behavior to add custom handling logic for certain type of exceptions");
endpointConfiguration.UseContainer<StructureMapBuilder>(c => c.ExistingContainer(container));
var recoverability = endpointConfiguration.Recoverability();
recoverability.Immediate(immediate =>
{
immediate.NumberOfRetries(2);
});
endpointConfiguration.LimitMessageProcessingConcurrencyTo(16);
recoverability.Delayed(delayed =>
{
delayed.NumberOfRetries(2);
});
endpointConfiguration.SendFailedMessagesTo("errorQueue");
...
}
}
public class CustomFaultMechanismBehavior : Behavior<IInvokeHandlerContext>
{
public override async Task Invoke(IInvokeHandlerContext context, Func<Task> next)
{
try
{
await next().ConfigureAwait(false);
}
catch (Exception ex)
{
}
}
}
UPDATE 2
I think I know what's going on: message is handled by first handler that throws an exception which is caught by the Behavior catch block, but then NServiceBus runtime tries to instantiate second handler class which is also supposed to handle the message (it handles class the message is derived from). That's where another exception is thrown in a constructor of one of dependent class. StructureMap tries to instantiate the handler and all its dependent services declared in the constructor and in the process runs into the exception. And this exception is not caught by CustomFaultMechanismBehavior.
So my I rephrase my question now: Is there any way to suppress errors (ignore error queue) occurring inside constructor or simply during StructureMap classes initialization? Seems like the described way does not cover this kind of situations
Your behavior is activated on Handler invocation. This means you are catching exceptions happening inside the Handle method so any other exception, e.g. in the Constructor of the handler would not be caught.
To change the way you 'capture' the exceptions, you can change the way the behavior is activated, e.g. change it from Behavior<IInvokeHandlerContext> to Behavior<ITransportReceiveContext> which is activated when the transport receives a message. You can investigate on different stages and behaviors to see which one suits your purpose best.
Scenario is a Silverlight client using Wcf service & custom authentication. To mitigate the 500/200 status code problem (avoid EndPointNotFound exception) I've applied the SilverLightFaultBehaviour. However, this does not work with UserNamePasswordValidator - When a FaultException is thrown from Validate(), it is not caught by the SilverLightFaultMessageInspector's implementation of BeforeSendReply.
So far, the only workaround I've found is using the alternative client stack instead ( WebRequest.RegisterPrefix("http://", WebRequestCreator.ClientHttp);), but there are complications with using it which can no longer be ignored as a lot of our clients are on virtual machines, the silverlight client keeps crashing ( Silverlight 5 - Debugging npctrl.dll crash , http://communities.vmware.com/thread/394306?tstart=0 ).
My primary motivation is that I want to be able to distinguish a failed login from a connection error (the following code is from a client-side async callback method, and only works with the Client stack):
if (e.Error is MessageSecurityException)
{
this.HasLoginFailed.Value = Captions.Login_FailedLogin;
}
else
{
this.HasLoginFailed.Value = Captions.Login_FailedConnection;
}
Is there any other way of modifying the message sent when throwing a FaultException from UserNamePasswordValidator? Or any conceptually different way of doing custom authentication rather than what I am using which enables me to modify the message status or to keep it 200, or just to be able to distinguish a connection failure from bad credentials?
my server-side code for usernamepassword reg:
var serviceCredential = host.Description.Behaviors.Find<ServiceCredentials>();
serviceCredential.UserNameAuthentication.UserNamePasswordValidationMode =
UserNamePasswordValidationMode.Custom;
serviceCredential.UserNameAuthentication.CustomUserNamePasswordValidator =
new MyValidator();
When you throw a FaultException from MyValidator, it is wrapped as the InnerException of a MessageSecurityException, that's probably why you weren't able to catch it directly as a FaultException.
To add some information to the fault you are throwing, what you can do is adding a FaultCode:
throw new FaultException(
"Invalid user name or bad password.",
new FaultCode("BadUserNameOrPassword")
);
Then, catch the exception client-side and retrieve your FaultCode:
try { ... }
catch (MessageSecurityException e)
{
FaultException fault = (FaultException) e.InnerException;
String faultCode = fault.Code.Name;
// you can now display a meaningful error with the faultCode
}
I hope it will help!
I have a web service that runs perfectly when i reference it from within the project solution. As soon as i upload it to the remote server, it starts blowing up. Unfortunately, the only error message I get is on the client side "faultexception was unhandled by user code". Inside of the web service, I have exceptions handled in all of the methods, so I'm pretty sure it's getting caught somewhere, but I don't know how to see it. I suspect that the problem is permissions related, but I can't see where it's happening.
I tried placing an error message into object returns, but it's still not making it out; something like this:
public bool SetDirectReports(ADUser user)
{
try
{
var adEntry = new DirectoryEntry(string.Format("LDAP://<GUID={0}>", user.Guid), "administrator", "S3cur1ty");
if (adEntry.Properties["directReports"].Count > 0)
{
user.DirectReports = new List<ADUser>();
foreach (string directReport in adEntry.Properties["directReports"]) //is being returned as full distinguished name
{
var dr = new DirectoryEntry(string.Format("LDAP://{0}", directReport), "administrator", "S3cur1ty");
user.DirectReports.Add(GetUserByGuid(dr.NativeGuid));
}
return true;
}
else
{
user.DirectReports = new List<ADUser>();
return false;
}
}
catch (Exception ex)
{
user.HasError = true;
user.ErrorMessage = "Error setting direct reports: " + ex.Message;
return false;
}
}
but its' still not catching. I was hoping for a better approach. I'm not sure if I could add something that would output the exception to the console or what. Any help would be appreciated. TIA
P.S. this isn't necessarily the method thats crashing, there's a web of them in the service.
You should dump all of your exceptions to a log file on the server side; exposing error information to the client is a potential security risk, which is why it's turned off by default.
If you really want to send exception information to the client, you can turn it on. If you are using a WCF service you should set the "includeExceptionDetailsInFaults" property on for the service behavior, as described in this MSDN article on dealing with unhandled exceptions in WCF. Once you do so, you will have a property on the FaultException called Detail that should itself be a type of Exception.
For better error handling you should also take a look at typed faults using the FaultContract and FaultException<> class; these have the benefit that they don't throw the channel into a faulted state and can be handled correctly:
try
{
// do stuff here
}
catch (Exception ex)
{
var detail = new CustomFaultDetail
{
Message = "Error setting direct reports: " + ex.Message
};
throw new FaultException<CustomFaultDetail>(detail);
}
If you are using an ASP.NET Web Service, you should set the customErrors mode to "Off" in your web.config. This will send back the entire exception detail as HTML, which the client should receive as part of the SOAP exception that it receives.
The error your are seeing ("faultexception was unhandled by user code") is happening because this is a remote exception and it is standard behavior to only display exceptions on the local computer by default. In order to make it work how you intend, you need to change the customErrors section of the web.config and set it to Off
UPDATE: I found a related question: c# exception not captured correctly by jquery ajax
(Three years later..)
Here's the solution I came up with, along with some sample WCF code, and Angular code to catch, and display the exception message:
Catching exceptions from WPF web services
Basically, you just need to wrap your WCF service in a try..catch, and when something goes wrong, set a OutgoingWebResponseContext value.
For example, in this web service, I've slipped in an Exception, which will make my catch code set the OutgoingWebResponseContext value.
It looks odd... as I then return null, but this works fine.
public List<string> GetAllCustomerNames()
{
// Get a list of unique Customer names.
//
try
{
throw new Exception("Oh heck, something went wrong !");
NorthwindDataContext dc = new NorthwindDataContext();
var results = (from cust in dc.Customers select cust.CompanyName).Distinct().OrderBy(s => s).ToList();
return results;
}
catch (Exception ex)
{
OutgoingWebResponseContext response = WebOperationContext.Current.OutgoingResponse;
response.StatusCode = System.Net.HttpStatusCode.Forbidden;
response.StatusDescription = ex.Message.Replace("\r\n", "");
return null;
}
}
What is brilliant about this try..catch is that, with minimal changes to your code, it'll add the error text to the HTTP Status, as you can see here in Google Chrome:
If you didn't have this try..catch code, you'd just get an HTTP Status Error of 400, which means "Bad Request".
So now, with our try..catch in place, I can call my web service from my Angular controller, and look out for such error messages coming back.
$http.get('http://localhost:15021/Service1.svc/getAllCustomerNames')
.then(function (data) {
// We successfully loaded the list of Customer names.
$scope.ListOfCustomerNames = data.GetAllCustomerNamesResult;
}, function (errorResponse) {
// The WCF Web Service returned an error
var HTTPErrorNumber = errorResponse.status;
var HTTPErrorStatusText = errorResponse.statusText;
alert("An error occurred whilst fetching Customer Names\r\nHTTP status code: " + HTTPErrorNumber + "\r\nError: " + HTTPErrorStatusText);
});
Cool, hey ?
Incredibly simple, generic, and easy to add to your services.
Shame some readers thought it was worth voting down. Sorry about that.
You have several options:
1) If you are using WCF, throw a FaultException on the server and catch it on the client. You could, for instance, implement a FaultContract on your service, and wrap the exception in a FaultException. Some guidance to this here.
2) You could use the Windows Server AppFabric which would give you more details to the exception within IIS. (requires some fiddling to get it working, though)
3) Why not implement some sort of server-side logging for the exceptions? Even if to a file, it would be invaluable to you to decipher what is really happening. It is not a good practice (especially for security reasons) to rely on the client to convey the inner workings of the server.
I have a WCF service configured to use custom UserName validation via the overriden Validate() method of the System.IdentityModel.Selectors.UserNamePasswordValidator class.
All methods of the contract have been decorated with the FaultContractAttribute to specify a custom SOAP fault as being returnable.
When throwing FaultException<T>, where T is the type specified in the FaultContractAttribute, everything behaves as expected and I get the custom fault in the response XML.
However, if I try and throw FaultException<T> in the overriden Validate() method of the username authentication class, I get a generic SOAP fault with the following reason:
"The creator of this fault did not specify a Reason."
However, if I change the code to throw the general SOAP fault as in:
throw new FaultException("Authentication failed.");
I will at least get "Authentication failed." in the reason element.
My questions are:
Why aren't the FaultException<T> exceptions treated the same if they're thrown in the Validate() as they are within the service implementation?
Is it possible to have exceptions thrown in the Validate() method conform to the FaultContractAttribute specified on the contract methods?
Any help greatly appreciated. My own guess is that the authentication comes before the message is associated with any method of the contract, and therefore, is not associated with the FaultContractAttribute, but any article confirming this and giving a workaround would be very useful.
Tali
It's a bit annoying but I got round it by doing this:
SecurityTokenValidationException stve
= new SecurityTokenValidationException("Invalid username or password");
throw new FaultException<SecurityTokenValidationException>(stve, stve.Message);
Including the message additionally means that you don't get the silly "did not specify a reason" message.
The problem is that the custom validation code is running outside of the context of any specific OperationContract, so there is no FaultContract is place for WCF to handle. So the short answer is no, you cannot get the exceptions thrown from your custom validator to honor the FaultContract.
You have a few options here. The one I prefer is to throw the non-generic FaultException and provide a pre-determined FaultCode; this way my catch blocks can differentiate contract faults from "plumbing" faults. Note that any exception you throw from a custom validator should come back as a MessageSecurityException, as shown below:
// Custom Validator:
public override void Validate(string userName, string password)
{
throw new FaultException(
"Invalid username or password.",
new FaultCode("AUTHENTICATION_FAILURE"));
}
// Client Code:
try
{
client.DoSomething();
}
catch ( MessageSecurityException ex )
{
var inner = ex.InnerException as FaultException;
if (inner != null && inner.Code.Name.Equals("AUTHENTICATION_FAILURE"))
{
// Security failure.
}
}
catch ( FaultException<SomethingFault> ex )
{
// Exception from the method itself.
}