RATIS-2245. Ratis should wait for all apply transaction futures before taking snapshot and group remove #1218
+119
−47
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -28,47 +28,106 @@ | |
| import org.apache.ratis.statemachine.impl.SimpleStateMachine4Testing; | ||
| import org.apache.ratis.statemachine.StateMachine; | ||
| import org.apache.ratis.statemachine.TransactionContext; | ||
| import org.junit.*; | ||
| import org.mockito.MockedStatic; | ||
| import org.mockito.Mockito; | ||
| import org.slf4j.Logger; | ||
| import org.slf4j.LoggerFactory; | ||
|
|
||
| import java.util.*; | ||
| import java.util.concurrent.*; | ||
| import java.util.concurrent.atomic.AtomicLong; | ||
|
|
||
| public abstract class StateMachineShutdownTests<CLUSTER extends MiniRaftCluster> | ||
| extends BaseTest | ||
| implements MiniRaftCluster.Factory.Get<CLUSTER> { | ||
| public static Logger LOG = LoggerFactory.getLogger(StateMachineUpdater.class); | ||
| private static MockedStatic<CompletableFuture> mocked; | ||
|
Contributor
There was a problem hiding this comment. This is actually unused. Now, it is causing problems in JDK17; see RATIS-2344. My bad that I did not see this earlier. |
||
| protected static class StateMachineWithConditionalWait extends | ||
| SimpleStateMachine4Testing { | ||
| boolean unblockAllTxns = false; | ||
| final Set<Long> blockTxns = ConcurrentHashMap.newKeySet(); | ||
| private final ExecutorService executor = Executors.newFixedThreadPool(10); | ||
| public static Map<Long, Set<CompletableFuture<Message>>> futures = new ConcurrentHashMap<>(); | ||
| public static Map<RaftPeerId, AtomicLong> numTxns = new ConcurrentHashMap<>(); | ||
| private final Map<Long, Long> appliedTxns = new ConcurrentHashMap<>(); | ||
|
|
||
| private synchronized void updateTxns() { | ||
| long appliedIndex = this.getLastAppliedTermIndex().getIndex() + 1; | ||
| Long appliedTerm = null; | ||
| while (appliedTxns.containsKey(appliedIndex)) { | ||
| appliedTerm = appliedTxns.remove(appliedIndex); | ||
| appliedIndex += 1; | ||
| } | ||
| if (appliedTerm != null) { | ||
| updateLastAppliedTermIndex(appliedTerm, appliedIndex - 1); | ||
| } | ||
| } | ||
|
|
||
| @Override | ||
| public void notifyTermIndexUpdated(long term, long index) { | ||
| appliedTxns.put(index, term); | ||
| updateTxns(); | ||
| } | ||
|
|
||
| @Override | ||
| public CompletableFuture<Message> applyTransaction(TransactionContext trx) { | ||
| final RaftProtos.LogEntryProto entry = trx.getLogEntryUnsafe(); | ||
|
|
||
| CompletableFuture<Message> future = new CompletableFuture<>(); | ||
| futures.computeIfAbsent(Thread.currentThread().getId(), k -> new HashSet<>()).add(future); | ||
| executor.submit(() -> { | ||
| synchronized (blockTxns) { | ||
| if (!unblockAllTxns) { | ||
| blockTxns.add(entry.getIndex()); | ||
| } | ||
| while (!unblockAllTxns && blockTxns.contains(entry.getIndex())) { | ||
| try { | ||
| blockTxns.wait(10000); | ||
| } catch (InterruptedException e) { | ||
| throw new RuntimeException(e); | ||
| } | ||
| } | ||
| } | ||
| numTxns.computeIfAbsent(getId(), (k) -> new AtomicLong()).incrementAndGet(); | ||
| appliedTxns.put(entry.getIndex(), entry.getTerm()); | ||
| updateTxns(); | ||
| future.complete(new RaftTestUtil.SimpleMessage("done")); | ||
| }); | ||
| return future; | ||
| } | ||
|
|
||
| public void unBlockApplyTxn(long txnId) { | ||
| synchronized (blockTxns) { | ||
| blockTxns.remove(txnId); | ||
| blockTxns.notifyAll(); | ||
| } | ||
| } | ||
|
|
||
| public void unblockAllTxns() { | ||
| unblockAllTxns = true; | ||
| synchronized (blockTxns) { | ||
| for (Long txnId : blockTxns) { | ||
| blockTxns.remove(txnId); | ||
| } | ||
| blockTxns.notifyAll(); | ||
| } | ||
| } | ||
| } | ||
|
|
||
| @Before | ||
| public void setup() { | ||
| mocked = Mockito.mockStatic(CompletableFuture.class, Mockito.CALLS_REAL_METHODS); | ||
| } | ||
|
|
||
| @After | ||
| public void tearDownClass() { | ||
| if (mocked != null) { | ||
| mocked.close(); | ||
| } | ||
|
|
||
| } | ||
|
|
||
| @Test | ||
| public void testStateMachineShutdownWaitsForApplyTxn() throws Exception { | ||
| final RaftProperties prop = getProperties(); | ||
|
|
@@ -82,10 +141,9 @@ public void testStateMachineShutdownWaitsForApplyTxn() throws Exception { | |
|
|
||
| //Unblock leader and one follower | ||
| ((StateMachineWithConditionalWait)leader.getStateMachine()) | ||
| .unblockAllTxns(); | ||
| ((StateMachineWithConditionalWait)cluster. | ||
| getFollowers().get(0).getStateMachine()).unblockAllTxns(); | ||
| cluster.getLeaderAndSendFirstMessage(true); | ||
|
|
||
| try (final RaftClient client = cluster.createClient(leaderId)) { | ||
|
|
@@ -107,16 +165,30 @@ public void testStateMachineShutdownWaitsForApplyTxn() throws Exception { | |
| final Thread t = new Thread(secondFollower::close); | ||
| t.start(); | ||
|
|
||
|
|
||
|
|
||
| // Now unblock the second follower | ||
| long minIndex = ((StateMachineWithConditionalWait) secondFollower.getStateMachine()).blockTxns.stream() | ||
| .min(Comparator.naturalOrder()).get(); | ||
| Assert.assertEquals(2, StateMachineWithConditionalWait.numTxns.values().stream() | ||
| .filter(val -> val.get() == 3).count()); | ||
| // The second follower should still be blocked in apply transaction | ||
| Assert.assertTrue(secondFollower.getInfo().getLastAppliedIndex() < minIndex); | ||
| for (long index : ((StateMachineWithConditionalWait) secondFollower.getStateMachine()).blockTxns) { | ||
| if (minIndex != index) { | ||
| ((StateMachineWithConditionalWait) secondFollower.getStateMachine()).unBlockApplyTxn(index); | ||
| } | ||
| } | ||
| Assert.assertEquals(2, StateMachineWithConditionalWait.numTxns.values().stream() | ||
| .filter(val -> val.get() == 3).count()); | ||
| Assert.assertTrue(secondFollower.getInfo().getLastAppliedIndex() < minIndex); | ||
| ((StateMachineWithConditionalWait) secondFollower.getStateMachine()).unBlockApplyTxn(minIndex); | ||
|
|
||
| // Now wait for the thread | ||
| t.join(5000); | ||
| Assert.assertEquals(logIndex, secondFollower.getInfo().getLastAppliedIndex()); | ||
| Assert.assertEquals(3, StateMachineWithConditionalWait.numTxns.values().stream() | ||
| .filter(val -> val.get() == 3).count()); | ||
|
|
||
| cluster.shutdown(); | ||
| } | ||
|
|
||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Trying to understand changes and impact,
Q1: Step 3 changes checkAndTakeSnapshot() moving out of stop() and force checking to take snapshot do have performance impact?
Q2: Let No snapshot to be taken, in this case, future.get() is never called, do this is intended ?
Q3: We can refactor code as above behavior, that,
This have similar impact instead of passing future deep inside takeSnapshot().
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
ratis/ratis-server/src/main/java/org/apache/ratis/server/impl/StateMachineUpdater.java
Line 196 in e25ba08
Future.get() will be called inconsequential of whether the snapshot is taken or not.
Uh oh!
There was an error while loading. Please reload this page.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We need to do a future.get() just before taking a snapshot. We cannot be having transactions still being applied in the background which could cause some inconsistency. It makes more sense to wait just before taking a snapshot that is why I moved the future.get() inside takeSnapshot method
ratis/ratis-server/src/main/java/org/apache/ratis/server/impl/StateMachineUpdater.java
Line 286 in e25ba08