public interface DOMDataTreeWriteTransaction extends DOMDataTreeTransaction, DOMDataTreeWriteOperations
Initial state of write transaction is a stable snapshot of the current data tree. The state is captured when the transaction is created and its state and underlying data tree are not affected by other concurrently running transactions.
Write transactions are isolated from other concurrent write transactions. All writes are local to the transaction and represent only a proposal of state change for the data tree and it is not visible to any other concurrently running transaction.
Applications make changes to the local data tree in the transaction by via the put, merge, and delete operations.
Performing the following put operations:
1) container { list [ a ] }
2) container { list [ b ] }
will result in the following data being present:
container { list [ b ] }
Performing the following merge operations:
1) container { list [ a ] }
2) container { list [ b ] }
will result in the following data being present:
container { list [ a, b ] }
This also means that storing the container will preserve any
augmentations which have been attached to it.
After applying changes to the local data tree, applications publish the changes proposed in the
transaction by calling commit() on the transaction. This seals the transaction
(preventing any further writes using this transaction) and commits it to be
processed and applied to global conceptual data tree.
The transaction commit may fail due to a concurrent transaction modifying and committing data in
an incompatible way. See commit() for more concrete commit failure examples.
Implementation Note: This interface is not intended to be implemented by users of MD-SAL, but only to be consumed by them.
| Modifier and Type | Method and Description |
|---|---|
boolean |
cancel()
Cancels the transaction.
|
@NonNull com.google.common.util.concurrent.FluentFuture<? extends CommitInfo> |
commit()
Commits this transaction to be asynchronously applied to update the logical data tree.
|
delete, merge, put@CheckReturnValue @NonNull com.google.common.util.concurrent.FluentFuture<? extends CommitInfo> commit()
FluentFuture conveys the result of applying the data changes.
This call logically seals the transaction, which prevents the client from further changing the data tree using
this transaction. Any subsequent calls to put(LogicalDatastoreType, Path, Object),
merge(LogicalDatastoreType, Path, Object), delete(LogicalDatastoreType, Path) will fail
with IllegalStateException. The transaction is marked as committed and enqueued into the data store
back-end for processing.
Whether or not the commit is successful is determined by versioning of the data tree and validation of registered commit participants if the transaction changes the data tree.
The effects of a successful commit of data depends on listeners and commit participants that are registered with the data broker.
private void doWrite(final int tries) {
WriteTransaction writeTx = dataBroker.newWriteOnlyTransaction();
MyDataObject data = ...;
InstanceIdentifier<MyDataObject> path = ...;
writeTx.put(LogicalDatastoreType.OPERATIONAL, path, data);
Futures.addCallback(writeTx.commit(), new FutureCallback<CommitInfo>() {
public void onSuccess(CommitInfo result) {
// succeeded
}
public void onFailure(Throwable t) {
if (t instanceof OptimisticLockFailedException) {
if(( tries - 1) > 0 ) {
// do retry
doWrite(tries - 1);
} else {
// out of retries
}
} else {
// failed due to another type of TransactionCommitFailedException.
}
});
}
...
doWrite(2);
Transaction may fail because of multiple reasons, such as
OptimisticLockFailedException. It is the responsibility
of the caller to create a new transaction and commit the same modification again in order to update data
tree.
Warning: In most cases, retrying after an OptimisticLockFailedException will result in a high
probability of success. However, there are scenarios, albeit unusual, where any number of retries will
not succeed. Therefore it is strongly recommended to limit the number of retries (2 or 3) to avoid
an endless loop.
DataValidationFailedException. User
should not retry to create new transaction with same data, since it probably will fail again.
| Initial state | Tx 1 | Tx 2 | Result |
|---|---|---|---|
| Empty | put(A,1) | put(A,2) | Tx 2 will fail, state is A=1 |
| Empty | put(A,1) | merge(A,2) | A=2 |
| Empty | merge(A,1) | put(A,2) | Tx 2 will fail, state is A=1 |
| Empty | merge(A,1) | merge(A,2) | A=2 |
| A=0 | put(A,1) | put(A,2) | Tx 2 will fail, A=1 |
| A=0 | put(A,1) | merge(A,2) | A=2 |
| A=0 | merge(A,1) | put(A,2) | Tx 2 will fail, A=1 |
| A=0 | merge(A,1) | merge(A,2) | A=2 |
| A=0 | delete(A) | put(A,2) | Tx 2 will fail, A does not exists |
| A=0 | delete(A) | merge(A,2) | A=2 |
| Initial state | Tx 1 | Tx 2 | Result |
|---|---|---|---|
| Empty | put(TOP,[]) | put(TOP,[]) | Tx 2 will fail, state is TOP=[] |
| Empty | put(TOP,[]) | merge(TOP,[]) | TOP=[] |
| Empty | put(TOP,[FOO=1]) | put(TOP,[BAR=1]) | Tx 2 will fail, state is TOP=[FOO=1] |
| Empty | put(TOP,[FOO=1]) | merge(TOP,[BAR=1]) | TOP=[FOO=1,BAR=1] |
| Empty | merge(TOP,[FOO=1]) | put(TOP,[BAR=1]) | Tx 2 will fail, state is TOP=[FOO=1] |
| Empty | merge(TOP,[FOO=1]) | merge(TOP,[BAR=1]) | TOP=[FOO=1,BAR=1] |
| TOP=[] | put(TOP,[FOO=1]) | put(TOP,[BAR=1]) | Tx 2 will fail, state is TOP=[FOO=1] |
| TOP=[] | put(TOP,[FOO=1]) | merge(TOP,[BAR=1]) | state is TOP=[FOO=1,BAR=1] |
| TOP=[] | merge(TOP,[FOO=1]) | put(TOP,[BAR=1]) | Tx 2 will fail, state is TOP=[FOO=1] |
| TOP=[] | merge(TOP,[FOO=1]) | merge(TOP,[BAR=1]) | state is TOP=[FOO=1,BAR=1] |
| TOP=[] | delete(TOP) | put(TOP,[BAR=1]) | Tx 2 will fail, state is empty store |
| TOP=[] | delete(TOP) | merge(TOP,[BAR=1]) | state is TOP=[BAR=1] |
| TOP=[] | put(TOP/FOO,1) | put(TOP/BAR,1]) | state is TOP=[FOO=1,BAR=1] |
| TOP=[] | put(TOP/FOO,1) | merge(TOP/BAR,1) | state is TOP=[FOO=1,BAR=1] |
| TOP=[] | merge(TOP/FOO,1) | put(TOP/BAR,1) | state is TOP=[FOO=1,BAR=1] |
| TOP=[] | merge(TOP/FOO,1) | merge(TOP/BAR,1) | state is TOP=[FOO=1,BAR=1] |
| TOP=[] | delete(TOP) | put(TOP/BAR,1) | Tx 2 will fail, state is empty store |
| TOP=[] | delete(TOP) | merge(TOP/BAR,1] | Tx 2 will fail, state is empty store |
| TOP=[FOO=1] | put(TOP/FOO,2) | put(TOP/BAR,1) | state is TOP=[FOO=2,BAR=1] |
| TOP=[FOO=1] | put(TOP/FOO,2) | merge(TOP/BAR,1) | state is TOP=[FOO=2,BAR=1] |
| TOP=[FOO=1] | merge(TOP/FOO,2) | put(TOP/BAR,1) | state is TOP=[FOO=2,BAR=1] |
| TOP=[FOO=1] | merge(TOP/FOO,2) | merge(TOP/BAR,1) | state is TOP=[FOO=2,BAR=1] |
| TOP=[FOO=1] | delete(TOP/FOO) | put(TOP/BAR,1) | state is TOP=[BAR=1] |
| TOP=[FOO=1] | delete(TOP/FOO) | merge(TOP/BAR,1] | state is TOP=[BAR=1] |
txA = broker.newWriteTransaction(); // allocates new transaction, data tree is empty txB = broker.newWriteTransaction(); // allocates new transaction, data tree is empty txA.put(CONFIGURATION, PATH, A); // writes to PATH value A txB.put(CONFIGURATION, PATH, B) // writes to PATH value B ListenableFuture futureA = txA.commit(); // transaction A is sealed and committed ListenebleFuture futureB = txB.commit(); // transaction B is sealed and committedCommit of transaction A will be processed asynchronously and data tree will be updated to contain value
A for PATH. Returned FluentFuture will
successfully complete once state is applied to data tree.
Commit of Transaction B will fail, because previous transaction also modified path in a
concurrent way. The state introduced by transaction B will not be applied. Returned
FluentFuture object will fail with OptimisticLockFailedException
exception, which indicates to client that concurrent transaction prevented the committed
transaction from being applied.
A successful commit produces implementation-specific CommitInfo structure, which is used to communicate
post-condition information to the caller. Such information can contain commit-id, timing information or any
other information the implementation wishes to share.
TransactionCommitFailedException or an exception derived from TransactionCommitFailedException.IllegalStateException - if the transaction is already committed or was canceled.boolean cancel()
commit() already successfully completed)
will always fail (return false).Copyright © 2019 OpenDaylight. All rights reserved.