[jira] [Commented] (ZOOKEEPER-2959) ignore accepted epoch and LEADERINFO ack from observers when a newly elected leader computes new epoch
[
https://issues.apache.org/jira/browse/ZOOKEEPER-2959?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16434923#comment-16434923
]
ASF GitHub Bot commented on ZOOKEEPER-2959:
---
Github user anmolnar commented on the issue:
https://github.com/apache/zookeeper/pull/500
Given that this change affects leader election I think it'd be very
beneficial if @fpj could take a look by any chance.
> ignore accepted epoch and LEADERINFO ack from observers when a newly elected
> leader computes new epoch
> --
>
> Key: ZOOKEEPER-2959
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-2959
> Project: ZooKeeper
> Issue Type: Bug
>Affects Versions: 3.4.10, 3.5.3
>Reporter: xiangyq000
>Assignee: Bogdan Kanivets
>Priority: Blocker
>
> Once the ZooKeeper cluster finishes the election for new leader, all learners
> report their accepted epoch to the leader for the computation of new cluster
> epoch.
> org.apache.zookeeper.server.quorum.Leader#getEpochToPropose
> {code:java}
> private final HashSet connectingFollowers = new HashSet();
> public long getEpochToPropose(long sid, long lastAcceptedEpoch) throws
> InterruptedException, IOException {
> synchronized(connectingFollowers) {
> if (!waitingForNewEpoch) {
> return epoch;
> }
> if (lastAcceptedEpoch >= epoch) {
> epoch = lastAcceptedEpoch+1;
> }
> connectingFollowers.add(sid);
> QuorumVerifier verifier = self.getQuorumVerifier();
> if (connectingFollowers.contains(self.getId()) &&
>
> verifier.containsQuorum(connectingFollowers)) {
> waitingForNewEpoch = false;
> self.setAcceptedEpoch(epoch);
> connectingFollowers.notifyAll();
> } else {
> long start = Time.currentElapsedTime();
> long cur = start;
> long end = start + self.getInitLimit()*self.getTickTime();
> while(waitingForNewEpoch && cur < end) {
> connectingFollowers.wait(end - cur);
> cur = Time.currentElapsedTime();
> }
> if (waitingForNewEpoch) {
> throw new InterruptedException("Timeout while waiting for
> epoch from quorum");
> }
> }
> return epoch;
> }
> }
> {code}
> The computation will get an outcome once :
> # The leader has call method "getEpochToPropose"
> # The number of all reporters is greater than half of participants.
> The problem is, an observer server will also send its accepted epoch to the
> leader, while this procedure treat observers as participants.
> Supposed that the cluster consists of 1 leader, 2 followers and 1 observer,
> and now the leader and the observer have reported their accepted epochs while
> neither of the followers has. Thus, the connectingFollowers set consists of
> two elements, resulting in a size of 2, which is greater than half quorum,
> namely, 2. Then QuorumVerifier#containsQuorum will return true, because it
> does not check whether the elements of the parameter are participants.
> The same flaw exists in
> org.apache.zookeeper.server.quorum.Leader#waitForEpochAck
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[GitHub] zookeeper issue #500: ZOOKEEPER-2959: ignore accepted epoch and LEADERINFO a...
Github user anmolnar commented on the issue: https://github.com/apache/zookeeper/pull/500 Given that this change affects leader election I think it'd be very beneficial if @fpj could take a look by any chance. ---
[jira] [Commented] (ZOOKEEPER-2959) ignore accepted epoch and LEADERINFO ack from observers when a newly elected leader computes new epoch
[
https://issues.apache.org/jira/browse/ZOOKEEPER-2959?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16434910#comment-16434910
]
ASF GitHub Bot commented on ZOOKEEPER-2959:
---
Github user anmolnar commented on a diff in the pull request:
https://github.com/apache/zookeeper/pull/500#discussion_r180793047
--- Diff: src/java/test/org/apache/zookeeper/server/quorum/Zab1_0Test.java
---
@@ -245,6 +245,180 @@ public void testLastAcceptedEpoch() throws Exception {
recursiveDelete(tmpDir);
}
}
+
+@Test
+public void testGetEpochToProposeWithObserver() throws Exception {
+File tmpDir = File.createTempFile("test", "dir", testData);
+tmpDir.delete();
+tmpDir.mkdir();
+Leader leader = null;
+try {
+QuorumPeer peer = createQuorumPeer(tmpDir);
+long participantId = 1;
+long observerId = peer.quorumPeers.size();
+peer.quorumPeers.put(observerId, new QuorumServer(observerId,
"0.0.0.0", 33225,
--- End diff --
I think to be consistent with `createQuorumPeer()` method this should be
something like:
```
peers.put(observerId, new QuorumServer(observerId, new
InetSocketAddress("127.0.0.1", PortAssignment.unique()),
new InetSocketAddress("127.0.0.1", PortAssignment.unique()),
new InetSocketAddress("127.0.0.1", PortAssignment.unique()),
QuorumPeer.LearnerType.OBSERVER));
```
> ignore accepted epoch and LEADERINFO ack from observers when a newly elected
> leader computes new epoch
> --
>
> Key: ZOOKEEPER-2959
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-2959
> Project: ZooKeeper
> Issue Type: Bug
>Affects Versions: 3.4.10, 3.5.3
>Reporter: xiangyq000
>Assignee: Bogdan Kanivets
>Priority: Blocker
>
> Once the ZooKeeper cluster finishes the election for new leader, all learners
> report their accepted epoch to the leader for the computation of new cluster
> epoch.
> org.apache.zookeeper.server.quorum.Leader#getEpochToPropose
> {code:java}
> private final HashSet connectingFollowers = new HashSet();
> public long getEpochToPropose(long sid, long lastAcceptedEpoch) throws
> InterruptedException, IOException {
> synchronized(connectingFollowers) {
> if (!waitingForNewEpoch) {
> return epoch;
> }
> if (lastAcceptedEpoch >= epoch) {
> epoch = lastAcceptedEpoch+1;
> }
> connectingFollowers.add(sid);
> QuorumVerifier verifier = self.getQuorumVerifier();
> if (connectingFollowers.contains(self.getId()) &&
>
> verifier.containsQuorum(connectingFollowers)) {
> waitingForNewEpoch = false;
> self.setAcceptedEpoch(epoch);
> connectingFollowers.notifyAll();
> } else {
> long start = Time.currentElapsedTime();
> long cur = start;
> long end = start + self.getInitLimit()*self.getTickTime();
> while(waitingForNewEpoch && cur < end) {
> connectingFollowers.wait(end - cur);
> cur = Time.currentElapsedTime();
> }
> if (waitingForNewEpoch) {
> throw new InterruptedException("Timeout while waiting for
> epoch from quorum");
> }
> }
> return epoch;
> }
> }
> {code}
> The computation will get an outcome once :
> # The leader has call method "getEpochToPropose"
> # The number of all reporters is greater than half of participants.
> The problem is, an observer server will also send its accepted epoch to the
> leader, while this procedure treat observers as participants.
> Supposed that the cluster consists of 1 leader, 2 followers and 1 observer,
> and now the leader and the observer have reported their accepted epochs while
> neither of the followers has. Thus, the connectingFollowers set consists of
> two elements, resulting in a size of 2, which is greater than half quorum,
> namely, 2. Then QuorumVerifier#containsQuorum will return true, because it
> does not check whether the elements of the parameter are participants.
> The same flaw exists in
> org.apache.zookeeper.server.quorum.Leader#waitForEpochAck
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (ZOOKEEPER-2959) ignore accepted epoch and LEADERINFO ack from observers when a newly elected leader computes new epoch
[
https://issues.apache.org/jira/browse/ZOOKEEPER-2959?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16434911#comment-16434911
]
ASF GitHub Bot commented on ZOOKEEPER-2959:
---
Github user anmolnar commented on a diff in the pull request:
https://github.com/apache/zookeeper/pull/500#discussion_r180789703
--- Diff: src/java/test/org/apache/zookeeper/server/quorum/Zab1_0Test.java
---
@@ -245,6 +245,180 @@ public void testLastAcceptedEpoch() throws Exception {
recursiveDelete(tmpDir);
}
}
+
+@Test
+public void testGetEpochToProposeWithObserver() throws Exception {
+File tmpDir = File.createTempFile("test", "dir", testData);
--- End diff --
Have you considered using ClientBase.createEmptyTestDir() instead?
> ignore accepted epoch and LEADERINFO ack from observers when a newly elected
> leader computes new epoch
> --
>
> Key: ZOOKEEPER-2959
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-2959
> Project: ZooKeeper
> Issue Type: Bug
>Affects Versions: 3.4.10, 3.5.3
>Reporter: xiangyq000
>Assignee: Bogdan Kanivets
>Priority: Blocker
>
> Once the ZooKeeper cluster finishes the election for new leader, all learners
> report their accepted epoch to the leader for the computation of new cluster
> epoch.
> org.apache.zookeeper.server.quorum.Leader#getEpochToPropose
> {code:java}
> private final HashSet connectingFollowers = new HashSet();
> public long getEpochToPropose(long sid, long lastAcceptedEpoch) throws
> InterruptedException, IOException {
> synchronized(connectingFollowers) {
> if (!waitingForNewEpoch) {
> return epoch;
> }
> if (lastAcceptedEpoch >= epoch) {
> epoch = lastAcceptedEpoch+1;
> }
> connectingFollowers.add(sid);
> QuorumVerifier verifier = self.getQuorumVerifier();
> if (connectingFollowers.contains(self.getId()) &&
>
> verifier.containsQuorum(connectingFollowers)) {
> waitingForNewEpoch = false;
> self.setAcceptedEpoch(epoch);
> connectingFollowers.notifyAll();
> } else {
> long start = Time.currentElapsedTime();
> long cur = start;
> long end = start + self.getInitLimit()*self.getTickTime();
> while(waitingForNewEpoch && cur < end) {
> connectingFollowers.wait(end - cur);
> cur = Time.currentElapsedTime();
> }
> if (waitingForNewEpoch) {
> throw new InterruptedException("Timeout while waiting for
> epoch from quorum");
> }
> }
> return epoch;
> }
> }
> {code}
> The computation will get an outcome once :
> # The leader has call method "getEpochToPropose"
> # The number of all reporters is greater than half of participants.
> The problem is, an observer server will also send its accepted epoch to the
> leader, while this procedure treat observers as participants.
> Supposed that the cluster consists of 1 leader, 2 followers and 1 observer,
> and now the leader and the observer have reported their accepted epochs while
> neither of the followers has. Thus, the connectingFollowers set consists of
> two elements, resulting in a size of 2, which is greater than half quorum,
> namely, 2. Then QuorumVerifier#containsQuorum will return true, because it
> does not check whether the elements of the parameter are participants.
> The same flaw exists in
> org.apache.zookeeper.server.quorum.Leader#waitForEpochAck
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[GitHub] zookeeper pull request #500: ZOOKEEPER-2959: ignore accepted epoch and LEADE...
Github user anmolnar commented on a diff in the pull request:
https://github.com/apache/zookeeper/pull/500#discussion_r180789703
--- Diff: src/java/test/org/apache/zookeeper/server/quorum/Zab1_0Test.java
---
@@ -245,6 +245,180 @@ public void testLastAcceptedEpoch() throws Exception {
recursiveDelete(tmpDir);
}
}
+
+@Test
+public void testGetEpochToProposeWithObserver() throws Exception {
+File tmpDir = File.createTempFile("test", "dir", testData);
--- End diff --
Have you considered using ClientBase.createEmptyTestDir() instead?
---
[GitHub] zookeeper pull request #500: ZOOKEEPER-2959: ignore accepted epoch and LEADE...
Github user anmolnar commented on a diff in the pull request:
https://github.com/apache/zookeeper/pull/500#discussion_r180793047
--- Diff: src/java/test/org/apache/zookeeper/server/quorum/Zab1_0Test.java
---
@@ -245,6 +245,180 @@ public void testLastAcceptedEpoch() throws Exception {
recursiveDelete(tmpDir);
}
}
+
+@Test
+public void testGetEpochToProposeWithObserver() throws Exception {
+File tmpDir = File.createTempFile("test", "dir", testData);
+tmpDir.delete();
+tmpDir.mkdir();
+Leader leader = null;
+try {
+QuorumPeer peer = createQuorumPeer(tmpDir);
+long participantId = 1;
+long observerId = peer.quorumPeers.size();
+peer.quorumPeers.put(observerId, new QuorumServer(observerId,
"0.0.0.0", 33225,
--- End diff --
I think to be consistent with `createQuorumPeer()` method this should be
something like:
```
peers.put(observerId, new QuorumServer(observerId, new
InetSocketAddress("127.0.0.1", PortAssignment.unique()),
new InetSocketAddress("127.0.0.1", PortAssignment.unique()),
new InetSocketAddress("127.0.0.1", PortAssignment.unique()),
QuorumPeer.LearnerType.OBSERVER));
```
---
ZooKeeper-trunk - Build # 3796 - Failure
See https://builds.apache.org/job/ZooKeeper-trunk/3796/ ### ## LAST 60 LINES OF THE CONSOLE ### [...truncated 259.92 KB...] [exec] Zookeeper_simpleSystem::testNullData : elapsed 1038 : OK [exec] Zookeeper_simpleSystem::testIPV6 : elapsed 1025 : OK [exec] Zookeeper_simpleSystem::testCreate : elapsed 2314 : OK [exec] Zookeeper_simpleSystem::testPath : elapsed 4130 : OK [exec] Zookeeper_simpleSystem::testPathValidation : elapsed 1185 : OK [exec] Zookeeper_simpleSystem::testPing : elapsed 18242 : OK [exec] Zookeeper_simpleSystem::testAcl : elapsed 1291 : OK [exec] Zookeeper_simpleSystem::testChroot : elapsed 3098 : OK [exec] Zookeeper_simpleSystem::testAuth ZooKeeper server started ZooKeeper server process failed ZooKeeper server NOT started : assertion : elapsed 42157 [exec] Zookeeper_simpleSystem::testHangingClient : elapsed 1001 : OK [exec] Zookeeper_simpleSystem::testWatcherAutoResetWithGlobal : assertion : elapsed 1001 [exec] Zookeeper_simpleSystem::testWatcherAutoResetWithLocal : assertion : elapsed 1000 [exec] Zookeeper_simpleSystem::testGetChildren2 : assertion : elapsed 1000 [exec] Zookeeper_simpleSystem::testLastZxid : assertion : elapsed 2005 [exec] Zookeeper_simpleSystem::testRemoveWatchers : assertion : elapsed 1001 [exec] /home/jenkins/jenkins-slave/workspace/ZooKeeper-trunk/src/c/tests/zkServer.sh: line 62: kill: (13001) - No such process [exec] Zookeeper_readOnly::testReadOnly : elapsed 4047 : OK [exec] /home/jenkins/jenkins-slave/workspace/ZooKeeper-trunk/src/c/tests/TestClient.cc:679: Assertion: assertion failed [Expression: ctx5.waitForConnected(zk)] [exec] /home/jenkins/jenkins-slave/workspace/ZooKeeper-trunk/src/c/tests/TestClient.cc:1243: Assertion: equality assertion failed [Expected: 0, Actual : -4] [exec] /home/jenkins/jenkins-slave/workspace/ZooKeeper-trunk/src/c/tests/TestClient.cc:1262: Assertion: equality assertion failed [Expected: 0, Actual : -4] [exec] /home/jenkins/jenkins-slave/workspace/ZooKeeper-trunk/src/c/tests/TestClient.cc:716: Assertion: equality assertion failed [Expected: 0, Actual : -4] [exec] /home/jenkins/jenkins-slave/workspace/ZooKeeper-trunk/src/c/tests/TestClient.cc:1291: Assertion: equality assertion failed [Expected: 0, Actual : -4] [exec] /home/jenkins/jenkins-slave/workspace/ZooKeeper-trunk/src/c/tests/TestClient.cc:1336: Assertion: equality assertion failed [Expected: 0, Actual : -4] [exec] Failures !!! [exec] Run: 74 Failure total: 6 Failures: 6 Errors: 0 [exec] FAIL: zktest-mt [exec] == [exec] 1 of 2 tests failed [exec] Please report to u...@zookeeper.apache.org [exec] == [exec] Makefile:1744: recipe for target 'check-TESTS' failed [exec] make[1]: Leaving directory '/home/jenkins/jenkins-slave/workspace/ZooKeeper-trunk/build/test/test-cppunit' [exec] Makefile:2000: recipe for target 'check-am' failed [exec] make[1]: *** [check-TESTS] Error 1 [exec] make: *** [check-am] Error 2 BUILD FAILED /home/jenkins/jenkins-slave/workspace/ZooKeeper-trunk/build.xml:1395: The following error occurred while executing this line: /home/jenkins/jenkins-slave/workspace/ZooKeeper-trunk/build.xml:1355: The following error occurred while executing this line: /home/jenkins/jenkins-slave/workspace/ZooKeeper-trunk/build.xml:1365: exec returned: 2 Total time: 19 minutes 53 seconds Build step 'Execute shell' marked build as failure [FINDBUGS] Skipping publisher since build result is FAILURE [WARNINGS] Skipping publisher since build result is FAILURE Archiving artifacts Setting JDK_1_7_LATEST__HOME=/home/jenkins/tools/java/latest1.7 Recording fingerprints Setting JDK_1_7_LATEST__HOME=/home/jenkins/tools/java/latest1.7 Recording test results Setting JDK_1_7_LATEST__HOME=/home/jenkins/tools/java/latest1.7 Publishing Javadoc Setting JDK_1_7_LATEST__HOME=/home/jenkins/tools/java/latest1.7 Setting JDK_1_7_LATEST__HOME=/home/jenkins/tools/java/latest1.7 Setting JDK_1_7_LATEST__HOME=/home/jenkins/tools/java/latest1.7 Email was triggered for: Failure - Any Sending email for trigger: Failure - Any Setting JDK_1_7_LATEST__HOME=/home/jenkins/tools/java/latest1.7 Setting JDK_1_7_LATEST__HOME=/home/jenkins/tools/java/latest1.7 ### ## FAILED TESTS (if any) ## All tests passed
Re: [VOTE] Apache ZooKeeper release 3.4.12 candidate 1
Hi, The PR for ZK-2959 already has +1. Can the PR be merged ? Thanks
ZooKeeper_branch34_openjdk7 - Build # 1877 - Failure
See https://builds.apache.org/job/ZooKeeper_branch34_openjdk7/1877/ ### ## LAST 60 LINES OF THE CONSOLE ### [...truncated 39.58 KB...] [junit] Running org.apache.zookeeper.test.SaslAuthDesignatedClientTest [junit] Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 1.544 sec [junit] Running org.apache.zookeeper.test.SaslAuthDesignatedServerTest [junit] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.505 sec [junit] Running org.apache.zookeeper.test.SaslAuthFailDesignatedClientTest [junit] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 1.533 sec [junit] Running org.apache.zookeeper.test.SaslAuthFailNotifyTest [junit] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.484 sec [junit] Running org.apache.zookeeper.test.SaslAuthFailTest [junit] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.587 sec [junit] Running org.apache.zookeeper.test.SaslAuthMissingClientConfigTest [junit] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.483 sec [junit] Running org.apache.zookeeper.test.SaslClientTest [junit] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.083 sec [junit] Running org.apache.zookeeper.test.SessionInvalidationTest [junit] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.47 sec [junit] Running org.apache.zookeeper.test.SessionTest [junit] Tests run: 6, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 32.85 sec [junit] Running org.apache.zookeeper.test.StandaloneTest [junit] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.872 sec [junit] Running org.apache.zookeeper.test.StatTest [junit] Tests run: 4, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.597 sec [junit] Running org.apache.zookeeper.test.StaticHostProviderTest [junit] Tests run: 7, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 1.316 sec [junit] Running org.apache.zookeeper.test.SyncCallTest [junit] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.539 sec [junit] Running org.apache.zookeeper.test.TruncateTest [junit] Tests run: 3, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 6.991 sec [junit] Running org.apache.zookeeper.test.UpgradeTest [junit] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 1.258 sec [junit] Running org.apache.zookeeper.test.WatchedEventTest [junit] Tests run: 4, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.092 sec [junit] Running org.apache.zookeeper.test.WatcherFuncTest [junit] Tests run: 6, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.78 sec [junit] Running org.apache.zookeeper.test.WatcherTest [junit] Tests run: 7, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 30.051 sec [junit] Running org.apache.zookeeper.test.ZkDatabaseCorruptionTest [junit] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 6.245 sec [junit] Running org.apache.zookeeper.test.ZooKeeperQuotaTest [junit] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.586 sec fail.build.on.test.failure: BUILD FAILED /home/jenkins/jenkins-slave/workspace/ZooKeeper_branch34_openjdk7/build.xml:1382: The following error occurred while executing this line: /home/jenkins/jenkins-slave/workspace/ZooKeeper_branch34_openjdk7/build.xml:1385: Tests failed! Total time: 28 minutes 13 seconds Build step 'Invoke Ant' marked build as failure Archiving artifacts Setting OPENJDK_7_ON_UBUNTU_ONLY__HOME=/usr/lib/jvm/java-7-openjdk-amd64/ Recording test results Setting OPENJDK_7_ON_UBUNTU_ONLY__HOME=/usr/lib/jvm/java-7-openjdk-amd64/ Setting OPENJDK_7_ON_UBUNTU_ONLY__HOME=/usr/lib/jvm/java-7-openjdk-amd64/ Setting OPENJDK_7_ON_UBUNTU_ONLY__HOME=/usr/lib/jvm/java-7-openjdk-amd64/ Email was triggered for: Failure - Any Sending email for trigger: Failure - Any Setting OPENJDK_7_ON_UBUNTU_ONLY__HOME=/usr/lib/jvm/java-7-openjdk-amd64/ Setting OPENJDK_7_ON_UBUNTU_ONLY__HOME=/usr/lib/jvm/java-7-openjdk-amd64/ ### ## FAILED TESTS (if any) ## 8 tests failed. FAILED: org.apache.zookeeper.server.ZxidRolloverTest.testMultipleRollover Error Message: java.net.BindException: Address already in use Stack Trace: java.lang.RuntimeException: java.net.BindException: Address already in use at org.apache.zookeeper.test.QuorumUtil.(QuorumUtil.java:116) at org.apache.zookeeper.test.QuorumUtil.(QuorumUtil.java:121) at org.apache.zookeeper.server.ZxidRolloverTest.setUp(ZxidRolloverTest.java:63) Caused by: java.net.BindException: Address already in use at sun.nio.ch.Net.bind0(Native Method)
