I have a service where to save a lot of data to db.
Using MYSQL
I have used like this
Domain1.withTransaction {text->
def domain1=//Create my domain object to save
if(!domain1.save()){
domain1.errors.each {
println it
}
throw new RuntimeException('unable to save domain1')
}
Domain2.withTransaction {text->
def domain2=//Create my domain object to save
if(!domain2.save()){
domain2.errors.each {
println it
}
throw new RuntimeException('unable to save domain2')
}
My problem if there occurred any problem in saving domain2 i need to roll back domain1 save also.
I need to remove domain1 from db.
Instead of using programatic transaction handling, the Service artifact allows for automatic transaction handling. This typically leads to cleaner and more maintainable code.
You can also use the failOnError:true when you save() to force a RuntimeException to be thrown.
Example below:
class SomeService {
static transactional = false
def serviceMethod() {
def domain1=//Create my domain object to save
def domain2=//Create my domain object to save
domain1.save(failOnError:true)
domain2.save(failOnError:true)
}
}
UPDATE
I'm revisiting this topic, after I read a response to another topic. Grails Integration Test Does NOT Rollback
Please verify that your dialect is setup as InnoDB as MyISAM tables aren't transactional. This is configured in your Config.groovy
dataSource {
dialect= org.hibernate.dialect.MySQLInnoDBDialect
...
}
Try removing the Domain2.withTransaction {text-> part.
You are already inside a transaction with your first call.
If you do further work inside the brackets, you should stay within the same transaction and domain1 should be rolled back if you throw and exception after checking domain2.
Put that Domain2.withTransaction closure inside Domain1.withTransaction closure, so error in Domain2 transaction will rollback both Domain1 and Domain2 transaction
Like this
Domain1.withTransaction {
//....
Domain2.withTransaction {
//....
}
}
If you just want a single transaction which is rolled back when an unhandled unchecked exception is thrown, then don't start a nested transaction. The necessaary code changes are shown below:
Domain1.withTransaction {text->
def domain1=//Create my domain object to save
if (!domain1.save()) {
domain1.errors.each {
println it
}
throw new RuntimeException('unable to save domain1')
}
def domain2=//Create my domain object to save
if (!domain2.save()) {
domain2.errors.each {
println it
}
throw new RuntimeException('unable to save domain2')
}
}
Related
Background
We use ApolloHandler to handle the exceptions in our Nest.js + GraphQL application.
Problem
Although ApolloHandler manages to create a structured GraphQL error response, every exception (plus it stack trace) also generates a console log and a logger entry [ExceptionHandler], polluting the application log with thousands of already managed input errors.
Question
How to set Nest.js to supress those ApolloHandler exceptions? Of course non ApolloHandler exceptions should remain logged.
Create your own custom logger to filter out those messages like:
export class AppLogger extends Logger {
error(message: string, trace: string, context?: string) {
if (message !== 'Validations failed!') {
super.error(message, trace, context)
}
}
}
And use it as
app.useLogger(new AppLogger())
I have a function which is running in the separate thread. The code which calls this function not waits for result of it.
def sendEmail(email: String): Future[Unit] = {
...
}
def registration: Future[User] = {
...
// I do not want to wait for result of this function, just fire email sending
// in seprate thread and continue
sendEmail(email)
...
// Do another job
}
The problem is that if something went wrong in sendEmail function, I want to see this exception in log file.
Now log file and console output are empty if some exception is thrown there.
Is there a way to log exceptions from that separate thread?
P.S.: I do not want to log exception manually in sendEmail, but force play framework to log it.
In general, you wrap exceptions in the exceptionally block.
In java, it's like :
foobar.thenComposeAsync(arg -> {
sendEmail();
}).exceptionally(throwable -> {
// Do logging
});
In short, what I am trying solve is how to recover from certain database errors in a Grails application using Hibernate and continue on with the transaction skipping over the failed row updates that are part of a batch of changes.
The application uses Grails 2.3.11 but I have also tried with version 1.3.8 with similar failed results.
Basically there is a Grails service class that iterates over a list of imported records and attempts to update associated master records appropriately. In certain situations exceptions might occur during the domain.save(flush:true) call e.g. org.hibernate.exception.DataException thrown due to issues like (Data truncation: Data too long for column ...).
At this point I have tried:
Disabling transactions
Using domainObj.withTransaction() for each individual record
Trying various #Transactional annotations
Calling domain.clearErrors() and domain.discard() after catching the exception
Tried using a nested service with Transactional annotation with noRollbackFor as shown below
A number of other approaches but nothing I've tried has worked
Example code:
#Transactional
class UpdateService {
public updateBatch(Integer batchId) {
...
list.each { record ->
record.value = 123
try {
nestedService.saveDomain()
} catch (e) {
record.clearErrors()
record.discard()
}
}
batch.status = "POSTED"
batch.save()
}
}
#Transactional
class NestedService {
#Transactional(propagation = Propagation.REQUIRED, noRollbackFor = RuntimeException.class)
public void saveDomain(domainObj) throws RuntimeException {
if (domainObj.validate() && domainObj.save(flush:true) {
log.info "domain $domain was saved"
}
}
}
Once an error occurs I can't seem to clear out the hibernate session. On each subsequent record being updated I receive the error:
org.hibernate.StaleObjectStateException: Row was updated or deleted by another transaction
where it indicates the original failed domain id.
Revision:
Vahid, Thanks for the suggestions. I have tried that. I realized one issue is that I am passing objects across transactional boundaries. So I experimented with the NestedService class do something along the lines of:
#Transactional(propagation = Propagation.REQUIRE_NEW)
public void saveDomain(domainObj) {
def newObj = new Domain.get(domainObj.id)
newObj.properties = domainObj.properties
if (newObj.validate() && newObj.save(force:true) ) { ... }
I expected that to work but the original domainObj still fails even though I'm not calling the save on it. Very strange...
A simple approach would be to loop and then use validate(). If it does fail, then just store the id of the failed entity and proceed.
if(!domainObject.validate()){
// store Id for trying it again later ?
}else{
// Save
}
I have a service method which does some operation inside a transaction.
public User method1() {
// some code...
Vehicle.withTransaction { status ->
// some collection loop
// some other delete
vehicle.delete(failOnError:true)
}
if (checkSomething outside transaction) {
return throw some user defined exception
}
return user
}
If there is a runtime exception we dont have to catch that exception and the transaction will be rolled back automatically. But how to determine that transaction rolled back due to some exception and I want to throw some user friendly error message. delete() call also wont return anything.
If I add try/catch block inside the transaction by catching the Exception (super class) it is not getting into that exception block. But i was expecting it to go into that block and throw user friendly exception.
EDIT 1: Is it a good idea to add try/catch arround withTransaction
Any idea how to solver this?? Thanks in advance.
If I understand you question correctly, you want to know how to catch an exception, determine what the exception is, and return a message to the user. There are a few ways to do this. I will show you how I do it.
Before I get to the code there are a few things I might suggest. First, you don't need to explicitly declare the transaction in a service (I'm using v2.2.5). Services are transactional by default (not a big deal).
Second, the transaction will automatically roll back if any exception occurs while executing the service method.
Third, I would recommend removing failOnError:true from save() (I don't think it works on delete()... I may be wrong?). I find it is easier to run validate() or save() in the service then return the model instance to the controller where the objects errors can be used in a flash message.
The following is a sample of how I like to handle exceptions and saves using a service method and try/catch in the controller:
class FooService {
def saveFoo(Foo fooInstance) {
return fooInstance.save()
}
def anotherSaveFoo(Foo fooInstance) {
if(fooInstance.validate()){
fooInstance.save()
}else{
do something else or
throw new CustomException()
}
return fooInstance
}
}
class FooController {
def save = {
def newFoo = new Foo(params)
try{
returnedFoo = fooService.saveFoo(newFoo)
}catch(CustomException | Exception e){
flash.warning = [message(code: 'foo.validation.error.message',
args: [org.apache.commons.lang.exception.ExceptionUtils.getRootCauseMessage(e)],
default: "The foo changes did not pass validation.<br/>{0}")]
redirect('to where ever you need to go')
return
}
if(returnedFoo.hasErrors()){
def fooErrors = returnedFoo.errors.getAllErrors()
flash.warning = [message(code: 'foo.validation.error.message',
args: [fooErrors],
default: "The foo changes did not pass validation.<br/>${fooErrors}")]
redirect('to where ever you need to go')
return
}else {
flash.success = [message(code: 'foo.saved.successfully.message',
default: "The foo was saved successfully")]
redirect('to where ever you need to go')
}
}
}
Hope this helps, or gets some other input from more experienced Grails developers.
Here are a few other ways I've found to get exception info to pass along to your user:
request.exception.cause
request.exception.cause.message
response.status
A few links to other relevant questions that may help:
Exception handling in Grails controllers
Exception handling in Grails controllers with ExceptionMapper in Grails 2.2.4 best practice
https://commons.apache.org/proper/commons-lang/javadocs/api-2.6/org/apache/commons/lang/exception/ExceptionUtils.html
I have the following code:
class ServiceA {
def save(Object object) {
if (somethingBadComesBack) {
throw new CustomRuntimeException(data)
}
}
}
class ServiceB {
def serviceA
def save(Object object) {
try {
serviceA.save(object)
// do more stuff if good to go
} catch(CustomRuntimeException e) {
// populate some objects with errors based on exception
}
}
}
class ServiceC {
def serviceB
def process(Object object) {
serviceB.save(object)
if (object.hasErrors() {
// do some stuff
}else{
// do some stuff
}
def info = someMethod(object)
return info
}
}
class SomeController {
def serviceC
def process() {
def object = .....
serviceC.save(object) // UnexpectedRollbackException is thrown here
}
}
When ServiceA.save() is called and an exception occurs, ServiceC.save() is throwing an UnexpectedRollbackException when it tries to return.
I did the following:
try {
serviceC.process(object)
}catch(UnexpectedRollbackException e) {
println e.getMostSpecificCause()
}
and I am getting:
org.springframework.transaction.UnexpectedRollbackException: Transaction rolled back because it has been marked as rollback-only
I'm not sure where to start looking for how to fix this.
You're using a runtime exception to roll back the transaction, but that's cheating - it's taking advantage of a side effect. Runtime exceptions roll back transactions automatically since you don't need to catch them so it's assumed that if one is thrown, it wasn't expected and the default behavior is to roll back. You can configure methods to not roll back for specific expected runtime exceptions, but this is somewhat rare. Checked exceptions don't roll back exceptions because in Java they must be caught or declared in the throws, so you have to have either explicitly thrown it or ducked it; either way you had a chance to try again.
The correct way to intentionally roll back a transaction is to call setRollbackOnly() on the current TransactionStatus but this isn't directly accessible in a service method (it is in a withTransaction block since it's the argument to the closure). But it's easy to get to: import org.springframework.transaction.interceptor.TransactionAspectSupport and call TransactionAspectSupport.currentTransactionStatus().setRollbackOnly(). This will required that you rework your code since there won't be an exception to catch, so you'll need to check that it was rolled back with TransactionAspectSupport.currentTransactionStatus().isRollbackOnly().
I'm not sure if it's a Grails issue or standard behavior, but when I was debugging this there were 3 commit calls with 3 different TransactionStatus instances. Only the first had the rollback flag set, but the second was aware of the first and was ok. The third one was considered a new transaction and was the one that triggered the same exception that you were seeing. So to work around this I added this to the 2nd and 3rd service methods:
def status = TransactionAspectSupport.currentTransactionStatus()
if (!status.isRollbackOnly()) status.setRollbackOnly()
to chain the rollback flag. That worked and I didn't get the UnexpectedRollbackException.
It might be easier to combine this with a checked exception. It's still overly expensive since it will fill in the stacktrace unnecessarily, but if you call setRollbackOnly() and throw a checked exception, you will be able to use the same general workflow you have now.
It seems like the default transactionality of services is biting you, and the unchecked exception thrown in Service A is dooming the transaction to rollback only, even once caught.
The above docs say the txn propagation level is PROPAGATION_REQUIRED, which should mean the same transaction is shared from your service C on up to A, if my memory serves me. Can you have Service A's save method throw a checked exception instead of RuntimeException, to avoid the auto-rollback from the latter? Or disable transactions on your services, if that's an option for you?