[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Flavio Paiva Junqueira updated ZOOKEEPER-569: - Resolution: Fixed Status: Resolved (was: Patch Available) +1, thanks Henry! I have just committed it (Committed revision 912119). > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: zookeeper-569.patch, ZOOKEEPER-569.patch, > zookeeper-569.patch, zookeeper-569.patch, zookeeper-569.patch, > zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Patrick Hunt updated ZOOKEEPER-569: --- Status: Patch Available (was: Open) > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: zookeeper-569.patch, ZOOKEEPER-569.patch, > zookeeper-569.patch, zookeeper-569.patch, zookeeper-569.patch, > zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Patrick Hunt updated ZOOKEEPER-569: --- Status: Open (was: Patch Available) > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: zookeeper-569.patch, ZOOKEEPER-569.patch, > zookeeper-569.patch, zookeeper-569.patch, zookeeper-569.patch, > zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Henry Robinson updated ZOOKEEPER-569: - Status: Patch Available (was: Open) > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: zookeeper-569.patch, ZOOKEEPER-569.patch, > zookeeper-569.patch, zookeeper-569.patch, zookeeper-569.patch, > zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Henry Robinson updated ZOOKEEPER-569: - Attachment: zookeeper-569.patch I'm making really heavy weather of this :) Flavio, thanks, you're completely right. You also exposed a weird issue that the listener was actually causing the test to succeed no matter what messages were sent - but if the fix for this issue was not in place the test would fail. This is fixed by using the right port for the DatagramSocket. New patch hopefully has the DatagramSocket listening on the right port, with everything else in order. > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: zookeeper-569.patch, ZOOKEEPER-569.patch, > zookeeper-569.patch, zookeeper-569.patch, zookeeper-569.patch, > zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Flavio Paiva Junqueira updated ZOOKEEPER-569: - Attachment: ZOOKEEPER-569.patch Henry, you do need an instance of QCM to get the listener, but you don't need the listener because you're using LE. I'm uploading a patch in which I have commented out some lines of the new test you wrote. This is how I think it should look like. If you agree, please remove those lines and submit a new patch. Note that I have also changed the leader election parameter in the new instance of QuorumPeer in mockServer to 0. > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: ZOOKEEPER-569.patch, zookeeper-569.patch, > zookeeper-569.patch, zookeeper-569.patch, zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Flavio Paiva Junqueira updated ZOOKEEPER-569: - Status: Open (was: Patch Available) > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: ZOOKEEPER-569.patch, zookeeper-569.patch, > zookeeper-569.patch, zookeeper-569.patch, zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Henry Robinson updated ZOOKEEPER-569: - Attachment: zookeeper-569.patch Slightly updated patch - Flavio, I believe you do need to instantiate QuorumCnxManager as QCM.Listener is not static and therefore can't be instantiated without an enclosing instance. I have removed the check for listener == null and set the thread timeout to 15s to tally with LETest.java. > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: zookeeper-569.patch, zookeeper-569.patch, > zookeeper-569.patch, zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Henry Robinson updated ZOOKEEPER-569: - Status: Open (was: Patch Available) > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: zookeeper-569.patch, zookeeper-569.patch, > zookeeper-569.patch, zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Henry Robinson updated ZOOKEEPER-569: - Status: Patch Available (was: Open) > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: zookeeper-569.patch, zookeeper-569.patch, > zookeeper-569.patch, zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Henry Robinson updated ZOOKEEPER-569: - Status: Open (was: Patch Available) > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: zookeeper-569.patch, zookeeper-569.patch, > zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Henry Robinson updated ZOOKEEPER-569: - Status: Patch Available (was: Open) > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: zookeeper-569.patch, zookeeper-569.patch, > zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Henry Robinson updated ZOOKEEPER-569: - Attachment: zookeeper-569.patch Here's a patch with tests that appears to fix the issue (test fails without fix, test succeeds with). All tests pass for me with this patch on my laptop. I have replaced one kludge with another here. QuorumPeer.electionAlg is set to null when electionType==0 until the election is actually run. This causes problems if you want to retrieve the electionAlg object via getElectionAlg() beforehand for tests. I've set it up so that makeLEStrategy always creates a new LeaderElection if electionType == 0, but also that createElectionAlgorithm sets electionAlg=new LeaderElection(this) instead of null, so that as long as startLeaderElection has been called, getElectionAlg() won't return null. I've checked to see if this will cause any obvious problems for the call sites of getElectionAlg and couldn't find anything that expected null. It seems more consistent to me this way. The question I have is over why LeaderElection needs re-instantiating each time when FLE does not. If this sounds confusing, it's because the code really is! The interaction of createElectionAlgorithm, startLeaderElection and makeLEStrategy is hard to discern. > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: zookeeper-569.patch, zookeeper-569.patch, > zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Patrick Hunt updated ZOOKEEPER-569: --- Fix Version/s: 3.3.0 Status: Patch Available (was: Open) > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Fix For: 3.3.0 > > Attachments: zookeeper-569.patch, zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Patrick Hunt updated ZOOKEEPER-569: --- Status: Open (was: Patch Available) > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Attachments: zookeeper-569.patch, zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Henry Robinson updated ZOOKEEPER-569: - Attachment: zookeeper-569.patch Forgot --no-prefix from git. Hopefully this will apply. > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Attachments: zookeeper-569.patch, zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Henry Robinson updated ZOOKEEPER-569: - Status: Patch Available (was: Open) Running this by Hudson to verify the build, despite not including new tests. (It will probably +1 if tests pass because I have slightly edited LETest) > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Attachments: zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Updated: (ZOOKEEPER-569) Failure of elected leader can lead to never-ending leader election
[ https://issues.apache.org/jira/browse/ZOOKEEPER-569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Henry Robinson updated ZOOKEEPER-569: - Attachment: zookeeper-569.patch Here's a functional patch for this issue. Looking into making a mock test now - would like to use Mockito or similar, but will probably have to use hand-rolled mocks for now. > Failure of elected leader can lead to never-ending leader election > -- > > Key: ZOOKEEPER-569 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-569 > Project: Zookeeper > Issue Type: Bug >Reporter: Henry Robinson >Assignee: Henry Robinson > Attachments: zookeeper-569.patch > > > It is possible for basic LeaderElection to enter a situation where it never > terminates. > As an example, consider a three node cluster A, B and C. > 1. In the first round, A votes for A, B votes for B and C votes for C > 2. Since C > B > A, all nodes resolve to vote for C in the second round as > there is no first round winner > 3. A, B vote for C, but C fails. > 4. C is not elected because neither A nor B hear from it, and so votes for it > are discarded > 5. A and B never reset their votes, despite not hearing from C, so continue > to vote for it ad infinitum. > Step 5 is the bug. If A and B reset their votes to themselves in the case > where the heard-from vote set is empty, leader election will continue. > I do not know if this affects running ZK clusters, as it is possible that the > out-of-band failure detection protocols may cause leader election to be > restarted anyhow, but I've certainly seen this in tests. > I have a trivial patch which fixes it, but it needs a test (and tests for > race conditions are hard to write!) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.