[jira] Created: (ZOOKEEPER-815) fill in "TBD"s in overview doc

2010-07-15 Thread Patrick Hunt (JIRA) 
fill in "TBD"s in overview doc
--

 Key: ZOOKEEPER-815
 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-815
 Project: Zookeeper
  Issue Type: Bug
  Components: documentation
Affects Versions: 3.3.1
Reporter: Patrick Hunt
Priority: Minor
 Fix For: 3.3.2, 3.4.0


Funny: "Ephemeral nodes are useful when you want to implement [tbd]." there are 
a few others in that doc that are should really be fixed.


-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

[jira] Commented: (ZOOKEEPER-790) Last processed zxid set prematurely while establishing leadership

2010-07-15 Thread Vishal K (JIRA) 

[ 
https://issues.apache.org/jira/browse/ZOOKEEPER-790?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=1243#action_1243
 ] 

Vishal K commented on ZOOKEEPER-790:


Likewise.

> Last processed zxid set prematurely while establishing leadership
> -
>
> Key: ZOOKEEPER-790
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-790
> Project: Zookeeper
>  Issue Type: Bug
>  Components: quorum
>Affects Versions: 3.3.1
>Reporter: Flavio Paiva Junqueira
>Assignee: Flavio Paiva Junqueira
>Priority: Blocker
> Fix For: 3.3.2, 3.4.0
>
> Attachments: ZOOKEEPER-790.patch, ZOOKEEPER-790.travis.log.bz2
>
>
> The leader code is setting the last processed zxid to the first of the new 
> epoch even before connecting to a quorum of followers. Because the leader 
> code sets this value before connecting to a quorum of followers 
> (Leader.java:281) and the follower code throws an IOException 
> (Follower.java:73) if the leader epoch is smaller, we have that when the 
> false leader drops leadership and becomes a follower, it finds a smaller 
> epoch and kills itself.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

[jira] Commented: (ZOOKEEPER-790) Last processed zxid set prematurely while establishing leadership

2010-07-15 Thread Travis Crawford (JIRA) 

[ 
https://issues.apache.org/jira/browse/ZOOKEEPER-790?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=1242#action_1242
 ] 

Travis Crawford commented on ZOOKEEPER-790:
---

Good find! I can certainly help test a patch, or collect more info if needed.

> Last processed zxid set prematurely while establishing leadership
> -
>
> Key: ZOOKEEPER-790
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-790
> Project: Zookeeper
>  Issue Type: Bug
>  Components: quorum
>Affects Versions: 3.3.1
>Reporter: Flavio Paiva Junqueira
>Assignee: Flavio Paiva Junqueira
>Priority: Blocker
> Fix For: 3.3.2, 3.4.0
>
> Attachments: ZOOKEEPER-790.patch, ZOOKEEPER-790.travis.log.bz2
>
>
> The leader code is setting the last processed zxid to the first of the new 
> epoch even before connecting to a quorum of followers. Because the leader 
> code sets this value before connecting to a quorum of followers 
> (Leader.java:281) and the follower code throws an IOException 
> (Follower.java:73) if the leader epoch is smaller, we have that when the 
> false leader drops leadership and becomes a follower, it finds a smaller 
> epoch and kills itself.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

[jira] Commented: (ZOOKEEPER-790) Last processed zxid set prematurely while establishing leadership

2010-07-15 Thread Flavio Paiva Junqueira (JIRA) 

[ 
https://issues.apache.org/jira/browse/ZOOKEEPER-790?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=1234#action_1234
 ] 

Flavio Paiva Junqueira commented on ZOOKEEPER-790:
--

Thank you both for all the information. I have been able to reproduce and find 
the source of the bug, but I don't have a patch yet. The problem is deeper than 
I thought originally. Let me show you what is going on:

I'm including an excerpt of logs from two runs: one good and one bad. The first 
run is good (see the excerpt below). I have killed the follower and restarted 
it as Vishal suggested. When it comes back, it declares itself as leader, also 
as Vishal and Travis observed.  However, different from what Vishal and Travis 
observed, it drops leadership and follows happily the leader right after.

{noformat}
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:quorump...@654] - LEADING
INFO  - [WorkerReceiver Thread:fastleaderelect...@496] - Notification: 3 
(n.leader), 0 (n.zxid), 1 (n.round), LEADING (n.state), 3 (n.sid), LOOKING (my 
state)
INFO  - [WorkerReceiver Thread:fastleaderelect...@496] - Notification: 3 
(n.leader), 0 (n.zxid), 1 (n.round), LEADING (n.state), 3 (n.sid), LEADING (my 
state)
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:lea...@54] - TCP NoDelay set to: true
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:zookeeper.version=3.4.0--1, built on 07/15/2010 10:36 GMT
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:host.name=XX.com
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:java.version=1.6.0_04
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:java.vendor=Sun Microsystems Inc.
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:java.home=/usr/java/jdk1.6.0_04/jre
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:java.class.path=.X
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:java.library.path= X
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:java.io.tmpdir=/tmp
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:java.compiler=
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:os.name=Linux
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:os.arch=amd64
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:os.version=2.6.18-53.1.21.el5
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:user.name=X
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:user.home=X
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:environm...@97] - Server 
environment:user.dir=X
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:zookeeperser...@151] - Created 
server with tickTime 3000 minSessionTimeout 6000 maxSessionTimeout 6 
datadir /X/zookeeper/version-2 snapdir //zookeeper/version-2
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:files...@82] - Reading snapshot 
//zookeeper/version-2/snapshot.100113340
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:files...@82] - Reading snapshot 
//zookeeper/version-2/snapshot.100113340
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:filetxnsnap...@208] - Snapshotting: 
10011f748
INFO  - [SessionTracker:zookeeperser...@315] - Expiring session 
0x229d6a9e0ca, timeout of 1ms exceeded
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:lea...@394] - Shutdown called
java.lang.Exception: shutdown Leader! reason: Waiting for a quorum of 
followers, only synced with: 2: 
at org.apache.zookeeper.server.quorum.Leader.shutdown(Leader.java:394)
at org.apache.zookeeper.server.quorum.Leader.lead(Leader.java:317)
at 
org.apache.zookeeper.server.quorum.QuorumPeer.run(QuorumPeer.java:657)
INFO  - [Thread-10:leader$learnercnxaccep...@243] - exception while shutting 
down acceptor: java.net.SocketException: Socket closed
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:finalrequestproces...@378] - 
shutdown of request processor complete
INFO  - [SyncThread:2:syncrequestproces...@151] - SyncRequestProcessor exited!
INFO  - [CommitProcessor:2:commitproces...@148] - CommitProcessor exited loop!
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:quorump...@620] - LOOKING
INFO  - [QuorumPeer:/0:0:0:0:0:0:0:0:10218:files...@82] - Reading snapshot 
/X/zookeeper/version-2/snapshot.10011f748
INFO  - [SessionTracker:sessiontrackeri...@158] - SessionTrackerImpl exited 
loop!
INFO  - [ProcessThread:-1:preprequestproces...@385] - Processed session 
termination for sessionid: 0x229d6a9e0ca
ERROR - [ProcessThread:-1:nioservercnxn$factor...@87] - Thread 
Thread[ProcessThread:-1,5,main] died
java.lang.NullPointerException