Transaktionsförökning och isolering på våren @ Transactional

1. Introduktion

I den här självstudien kommer vi att täcka @Transactional- anteckningen och dess isolerings- och utbredningsinställningar .

2. Vad är @Transactional?

Vi kan använda @Transactional för att sätta in en metod i en databastransaktion.

Det gör att vi kan ställa in villkor för propagering, isolering, timeout, skrivskydd och återställning för vår transaktion. Vi kan också ange transaktionshanteraren.

2.1. @Transactional Implementation Details

Spring skapar en proxy eller manipulerar klassbyte-koden för att hantera skapandet, begå och återgång av transaktionen. I fallet med en proxy ignorerar Spring @Transactional i interna metodanrop .

Enkelt uttryckt, om vi har en metod som callMethod och vi markerar den som @Transactional, skulle Spring sätta in någon transaktionskod runt anropet: @Transactional method called:

createTransactionIfNecessary(); try { callMethod(); commitTransactionAfterReturning(); } catch (exception) { completeTransactionAfterThrowing(); throw exception; }

2.2. Hur man använder @Transactional

Vi kan lägga anteckningen på definitioner av gränssnitt, klasser eller direkt på metoder. De åsidosätter varandra enligt prioritetsordningen; från lägsta till högsta har vi: Gränssnitt, superklass, klass, gränssnittsmetod, superklassmetod och klassmetod.

Spring tillämpar klassanteckningen på alla offentliga metoder i den här klassen som vi inte kommenterade med @Transactional .

Men om vi lägger anteckningen till en privat eller skyddad metod kommer Spring att ignorera den utan ett fel.

Låt oss börja med ett gränssnittsexempel:

@Transactional public interface TransferService { void transfer(String user1, String user2, double val); } 

Vanligtvis rekommenderas det inte att ställa in @Transactional på gränssnittet. Det är dock acceptabelt för fall som @Repository med Spring Data.

Vi kan lägga anteckningen på en klassdefinition för att åsidosätta gränssnittets / superklassens transaktionsinställning:

@Service @Transactional public class TransferServiceImpl implements TransferService { @Override public void transfer(String user1, String user2, double val) { // ... } }

Låt oss nu åsidosätta det genom att ställa in anteckningen direkt på metoden:

@Transactional public void transfer(String user1, String user2, double val) { // ... }

3. Transaktionsförökning

Förökning definierar vår affärslogik transaktionsgräns. Spring lyckas starta och pausa en transaktion enligt vår propagationsinställning .

Spring anropar TransactionManager :: getTransaction för att få eller skapa en transaktion enligt förökningen. Det stöder en del av förökningen för alla typer av TransactionManager , men det finns några av dem som endast stöds av specifika implementeringar av TransactionManager .

Låt oss nu gå igenom olika förökningar och hur de fungerar.

3.1. Krävs förökning

KRÄVS är standardutbredningen. Våren kontrollerar om det finns en aktiv transaktion, sedan skapar den en ny om inget fanns. Annars läggs affärslogiken till den för närvarande aktiva transaktionen:

@Transactional(propagation = Propagation.REQUIRED) public void requiredExample(String user) { // ... }

Eftersom KRÄVS är standardutbredningen kan vi förenkla koden genom att släppa den:

@Transactional public void requiredExample(String user) { // ... }

Låt oss se pseudokoden för hur transaktionsskapande fungerar för KRÄVD propagering:

if (isExistingTransaction()) { if (isValidateExistingTransaction()) { validateExisitingAndThrowExceptionIfNotValid(); } return existing; } return createNewTransaction();

3.2. STÖDAR Förökning

För SUPPORTER kontrollerar Spring först om det finns en aktiv transaktion. Om det finns en transaktion kommer den befintliga transaktionen att användas. Om det inte finns en transaktion, körs den icke-transaktionell:

@Transactional(propagation = Propagation.SUPPORTS) public void supportsExample(String user) { // ... }

Let's see the transaction creation's pseudo-code for SUPPORTS:

if (isExistingTransaction()) { if (isValidateExistingTransaction()) { validateExisitingAndThrowExceptionIfNotValid(); } return existing; } return emptyTransaction;

3.3. MANDATORY Propagation

When the propagation is MANDATORY, if there is an active transaction, then it will be used. If there isn't an active transaction, then Spring throws an exception:

@Transactional(propagation = Propagation.MANDATORY) public void mandatoryExample(String user) { // ... }

And let's again see the pseudo-code:

if (isExistingTransaction()) { if (isValidateExistingTransaction()) { validateExisitingAndThrowExceptionIfNotValid(); } return existing; } throw IllegalTransactionStateException;

3.4. NEVER Propagation

For transactional logic with NEVER propagation, Spring throws an exception if there's an active transaction:

@Transactional(propagation = Propagation.NEVER) public void neverExample(String user) { // ... }

Let's see the pseudo-code of how transaction creation works for NEVER propagation:

if (isExistingTransaction()) { throw IllegalTransactionStateException; } return emptyTransaction;

3.5. NOT_SUPPORTED Propagation

Spring at first suspends the current transaction if it exists, then the business logic is executed without a transaction.

@Transactional(propagation = Propagation.NOT_SUPPORTED) public void notSupportedExample(String user) { // ... }

The JTATransactionManager supports real transaction suspension out-of-the-box. Others simulate the suspension by holding a reference to the existing one and then clearing it from the thread context

3.6. REQUIRES_NEW Propagation

When the propagation is REQUIRES_NEW, Spring suspends the current transaction if it exists and then creates a new one:

@Transactional(propagation = Propagation.REQUIRES_NEW) public void requiresNewExample(String user) { // ... }

Similar to NOT_SUPPORTED, we need the JTATransactionManager for actual transaction suspension.

And the pseudo-code looks like so:

if (isExistingTransaction()) { suspend(existing); try { return createNewTransaction(); } catch (exception) { resumeAfterBeginException(); throw exception; } } return createNewTransaction();

3.7. NESTED Propagation

For NESTED propagation, Spring checks if a transaction exists, then if yes, it marks a savepoint. This means if our business logic execution throws an exception, then transaction rollbacks to this savepoint. If there's no active transaction, it works like REQUIRED .

DataSourceTransactionManager supports this propagation out-of-the-box. Also, some implementations of JTATransactionManager may support this.

JpaTransactionManager supports NESTED only for JDBC connections. However, if we set nestedTransactionAllowed flag to true, it also works for JDBC access code in JPA transactions if our JDBC driver supports savepoints.

Finally, let's set the propagation to NESTED:

@Transactional(propagation = Propagation.NESTED) public void nestedExample(String user) { // ... }

4. Transaction Isolation

Isolation is one of the common ACID properties: Atomicity, Consistency, Isolation, and Durability. Isolation describes how changes applied by concurrent transactions are visible to each other.

Each isolation level prevents zero or more concurrency side effects on a transaction:

  • Dirty read: read the uncommitted change of a concurrent transaction
  • Nonrepeatable read: get different value on re-read of a row if a concurrent transaction updates the same row and commits
  • Phantom read: get different rows after re-execution of a range query if another transaction adds or removes some rows in the range and commits

We can set the isolation level of a transaction by @Transactional::isolation. It has these five enumerations in Spring: DEFAULT, READ_UNCOMMITTED, READ_COMMITTED, REPEATABLE_READ, SERIALIZABLE.

4.1. Isolation Management in Spring

The default isolation level is DEFAULT. So when Spring creates a new transaction, the isolation level will be the default isolation of our RDBMS. Therefore, we should be careful if we change the database.

We should also consider cases when we call a chain of methods with different isolation. In the normal flow, the isolation only applies when a new transaction created. Thus if for any reason we don't want to allow a method to execute in different isolation, we have to set TransactionManager::setValidateExistingTransaction to true. Then the pseudo-code of transaction validation will be:

if (isolationLevel != ISOLATION_DEFAULT) { if (currentTransactionIsolationLevel() != isolationLevel) { throw IllegalTransactionStateException } }

Now let's get deep in different isolation levels and their effects.

4.2. READ_UNCOMMITTED Isolation

READ_UNCOMMITTED is the lowest isolation level and allows for most concurrent access.

As a result, it suffers from all three mentioned concurrency side effects. So a transaction with this isolation reads uncommitted data of other concurrent transactions. Also, both non-repeatable and phantom reads can happen. Thus we can get a different result on re-read of a row or re-execution of a range query.

We can set the isolation level for a method or class:

@Transactional(isolation = Isolation.READ_UNCOMMITTED) public void log(String message) { // ... }

Postgres does not support READ_UNCOMMITTED isolation and falls back to READ_COMMITED instead. Also, Oracle does not support and allow READ_UNCOMMITTED.

4.3. READ_COMMITTED Isolation

The second level of isolation, READ_COMMITTED, prevents dirty reads.

The rest of the concurrency side effects still could happen. So uncommitted changes in concurrent transactions have no impact on us, but if a transaction commits its changes, our result could change by re-querying.

Here, we set the isolation level:

@Transactional(isolation = Isolation.READ_COMMITTED) public void log(String message){ // ... }

READ_COMMITTED is the default level with Postgres, SQL Server, and Oracle.

4.4. REPEATABLE_READ Isolation

The third level of isolation, REPEATABLE_READ, prevents dirty, and non-repeatable reads. So we are not affected by uncommitted changes in concurrent transactions.

Also, when we re-query for a row, we don't get a different result. But in the re-execution of range-queries, we may get newly added or removed rows.

Moreover, it is the lowest required level to prevent the lost update. The lost update occurs when two or more concurrent transactions read and update the same row. REPEATABLE_READ does not allow simultaneous access to a row at all. Hence the lost update can't happen.

Here is how to set the isolation level for a method:

@Transactional(isolation = Isolation.REPEATABLE_READ) public void log(String message){ // ... }

REPEATABLE_READ is the default level in Mysql. Oracle does not support REPEATABLE_READ.

4.5. SERIALIZABLE Isolation

SERIALIZABLE is the highest level of isolation. It prevents all mentioned concurrency side effects but can lead to the lowest concurrent access rate because it executes concurrent calls sequentially.

In other words, concurrent execution of a group of serializable transactions has the same result as executing them in serial.

Now let's see how to set SERIALIZABLE as the isolation level:

@Transactional(isolation = Isolation.SERIALIZABLE) public void log(String message){ // ... }

5. Conclusion

In this tutorial, we explored the propagation property of @Transaction in detail. Afterward, we learned about concurrency side effects and isolation levels.

As always, you can find the complete code over on GitHub.