Skip to content

CASSANDRA-21633 Node with empty data directory is allowed to rejoin the ring on the s… #5094

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

Open
krummas wants to merge 1 commit into
base: trunk
Choose a base branch
from
Open

CASSANDRA-21633 Node with empty data directory is allowed to rejoin the ring on the s… #5094

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
10 changes: 8 additions & 2 deletions src/java/org/apache/cassandra/tcm/Startup.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -232,10 +232,15 @@ public static void initializeAsNonCmsNode(Function<Processor, Processor> wrapPro
UUID currentHostId = SystemKeyspace.getLocalHostId();
if (nodeId != NodeId.UNREGISTERED && !Objects.equals(nodeId.toUUID(), currentHostId))
{
//Returning here if no host id was found in the system.local table means startup will fall
//through to the path taken by a new node attempting to join the cluster with a broadcast
//address that is already associated with an existing member. We could choose to report an
//IP collision here unless StorageService.isReplacingSameAddress() is true, but that would
//lead to inconsistent messages/logging across multiple restarts as it would be dependent
//on whether the node had performed catch up from a peer or not.
if (currentHostId == null)
{
logger.info("Taking over the host ID: {}, replacing address {}", nodeId.toUUID(), FBUtilities.getBroadcastAddressAndPort());
SystemKeyspace.setLocalHostId(nodeId.toUUID());
logger.debug("No host id set for node {} in system.local", nodeId);
return;
}

Expand Down Expand Up @@ -689,6 +694,7 @@ public static void startup(Supplier<Transformation> initialTransformation, boole
throw new IllegalStateException("Cannot replace same address when accord transactions are enabled.");
}

SystemKeyspace.setLocalHostId(self.toUUID());
ReplaceSameAddress.streamData(self, metadata, shouldBootstrap, finishJoiningRing);
}

Expand Down
46 changes: 43 additions & 3 deletions test/distributed/org/apache/cassandra/distributed/test/log/BounceResetHostIdTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,25 +18,30 @@

package org.apache.cassandra.distributed.test.log;

import java.io.IOException;
import java.net.InetAddress;
import java.nio.file.Files;
import java.nio.file.Path;
import java.util.Arrays;
import java.util.Comparator;
import java.util.concurrent.TimeUnit;

import org.junit.Assert;
import org.junit.Test;

import org.apache.cassandra.distributed.Cluster;
import org.apache.cassandra.distributed.api.ConsistencyLevel;
import org.apache.cassandra.distributed.api.Feature;
import org.apache.cassandra.distributed.shared.AssertUtils;
import org.apache.cassandra.distributed.shared.ClusterUtils;
import org.apache.cassandra.distributed.shared.Uninterruptibles;
import org.apache.cassandra.distributed.test.TestBaseImpl;
import org.apache.cassandra.locator.NoOpProximity;
import org.apache.cassandra.locator.SimpleLocationProvider;
import org.apache.cassandra.tcm.membership.NodeId;

import static org.apache.cassandra.distributed.Constants.KEY_DTEST_API_STARTUP_FAILURE_AS_SHUTDOWN;
import static org.apache.cassandra.distributed.shared.AssertUtils.row;
import static org.junit.Assert.assertTrue;
import static org.junit.Assert.fail;

public class BounceResetHostIdTest extends TestBaseImpl
Expand Down Expand Up @@ -120,7 +125,7 @@ public void swapIpsDirectlyTest() throws Exception
}
catch (Throwable t)
{
Assert.assertTrue(t.getMessage().contains("NodeId does not match locally set one"));
assertTrue(t.getMessage().contains("NodeId does not match locally set one"));
}
try
{
Expand All @@ -129,7 +134,7 @@ public void swapIpsDirectlyTest() throws Exception
}
catch (Throwable t)
{
Assert.assertTrue(t.getMessage().contains("NodeId does not match locally set one"));
assertTrue(t.getMessage().contains("NodeId does not match locally set one"));
}
}
}
Expand All @@ -138,4 +143,39 @@ public static Object[][] sortHelper(Object[][] rows)
Arrays.sort(rows, Comparator.comparing(r -> ((InetAddress)r[0]).getHostAddress()));
return rows;
}

@Test
public void bounceWipedDatadirectoryTest() throws Exception
{
try (Cluster cluster = init(builder().withNodes(3)
.withDataDirCount(1)
.withConfig(c -> c.with(Feature.GOSSIP, Feature.NATIVE_PROTOCOL)
.set(KEY_DTEST_API_STARTUP_FAILURE_AS_SHUTDOWN, false))
.start()))
{
cluster.get(3).shutdown().get();
wipeDir(((String[]) cluster.get(3).config().get("data_file_directories"))[0]);
wipeDir((String) cluster.get(3).config().get("commitlog_directory"));
for (int i = 0; i < 5; i++)
{
try
{
cluster.get(3).startup();
fail();
}
catch (Exception e)
{
assertTrue(e.getMessage(), e.getMessage().contains("Use cassandra.replace_address if you want to replace this node"));
cluster.get(3).shutdown().get();
}
Uninterruptibles.sleepUninterruptibly(5, TimeUnit.SECONDS);
}
}
}
private static void wipeDir(String dir) throws IOException
{
Path source = Path.of(dir);
Files.move(source, Path.of(source+"backup")); // remove potential trailing / from dir
Files.createDirectories(source);
}
}