Skip to content

HBASE-29724: Backport "HBASE-25334: TestRSGroupsFallback.testFallback is flaky" into branch-2 and branch-2.6 #7472

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Nov 21, 2025
+13 −12
Merged

HBASE-29724: Backport "HBASE-25334: TestRSGroupsFallback.testFallback is flaky" into branch-2 and branch-2.6 #7472

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 5 additions & 8 deletions hbase-rsgroup/src/test/java/org/apache/hadoop/hbase/rsgroup/TestRSGroupsFallback.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -101,16 +101,13 @@ public void testFallback() throws Exception {
assertRegionsInGroup(tableName, FALLBACK_GROUP);

// add a new server to default group, regions move to default group
JVMClusterUtil.RegionServerThread t =
TEST_UTIL.getMiniHBaseCluster().startRegionServerAndWait(60000);
Address startRSAddress = t.getRegionServer().getServerName().getAddress();
TEST_UTIL.waitFor(3000,
() -> rsGroupAdmin.getRSGroupInfo(RSGroupInfo.DEFAULT_GROUP).containsServer(startRSAddress));
TEST_UTIL.getMiniHBaseCluster().startRegionServerAndWait(60000);
assertTrue(master.balance().isBalancerRan());
assertRegionsInGroup(tableName, RSGroupInfo.DEFAULT_GROUP);

// add a new server to test group, regions move back
t = TEST_UTIL.getMiniHBaseCluster().startRegionServerAndWait(60000);
JVMClusterUtil.RegionServerThread t =
TEST_UTIL.getMiniHBaseCluster().startRegionServerAndWait(60000);
rsGroupAdmin.moveServers(
Collections.singleton(t.getRegionServer().getServerName().getAddress()), groupName);
assertTrue(master.balance().isBalancerRan());
Expand All @@ -119,11 +116,11 @@ public void testFallback() throws Exception {
TEST_UTIL.deleteTable(tableName);
}

private void assertRegionsInGroup(TableName tableName, String group) throws IOException {
private void assertRegionsInGroup(TableName table, String group) throws IOException {
ProcedureTestingUtility
.waitAllProcedures(TEST_UTIL.getMiniHBaseCluster().getMaster().getMasterProcedureExecutor());
RSGroupInfo groupInfo = rsGroupAdmin.getRSGroupInfo(group);
master.getAssignmentManager().getRegionStates().getRegionsOfTable(tableName).forEach(region -> {
master.getAssignmentManager().getRegionStates().getRegionsOfTable(table).forEach(region -> {
Address regionOnServer = master.getAssignmentManager().getRegionStates()
.getRegionAssignments().get(region).getAddress();
assertTrue(groupInfo.getServers().contains(regionOnServer));
Expand Down
3 changes: 2 additions & 1 deletion hbase-server/src/main/java/org/apache/hadoop/hbase/master/ServerManager.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -549,7 +549,8 @@ public DeadServer getDeadServers() {
* @return true if any RS are being processed as dead, false if not
*/
public boolean areDeadServersInProgress() throws IOException {
return master.getProcedures().stream().anyMatch(p -> p instanceof ServerCrashProcedure);
return master.getProcedures().stream()
.anyMatch(p -> !p.isFinished() && p instanceof ServerCrashProcedure);
}

void letRegionServersShutdown() {
Expand Down
9 changes: 6 additions & 3 deletions hbase-server/src/test/java/org/apache/hadoop/hbase/master/TestDeadServer.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,6 @@
import static org.junit.Assert.assertFalse;
import static org.junit.Assert.assertTrue;

import java.io.IOException;
import java.util.List;
import java.util.Set;
import org.apache.hadoop.hbase.HBaseClassTestRule;
Expand Down Expand Up @@ -93,15 +92,19 @@ public void testIsDead() {
}

@Test
public void testCrashProcedureReplay() throws IOException {
public void testCrashProcedureReplay() throws Exception {
HMaster master = TEST_UTIL.getHBaseCluster().getMaster();
final ProcedureExecutor<MasterProcedureEnv> pExecutor = master.getMasterProcedureExecutor();
ServerCrashProcedure proc =
new ServerCrashProcedure(pExecutor.getEnvironment(), hostname123, false, false);

pExecutor.stop();
ProcedureTestingUtility.submitAndWait(pExecutor, proc);

assertTrue(master.getServerManager().areDeadServersInProgress());

ProcedureTestingUtility.restart(pExecutor);
ProcedureTestingUtility.waitProcedure(pExecutor, proc);
assertFalse(master.getServerManager().areDeadServersInProgress());
}

@Test
Expand Down
Loading