I won't get a delete event. I'll get a NodeChildrenChanged on the parent
path.
-JZ
On 7/14/11 5:28 PM, Ted Dunning ted.dunn...@gmail.com wrote:
Since you can see what the event is, can you test it to see if is the
delete
that was just issued?
On Thu, Jul 14, 2011 at 5:24 PM, Jordan Zimmerman
if the difference is just the item you
deleted.
If so, you don't need to reset. Likewise, if something else was deleted
that is still in your future, you can just delete it from the list of
upcoming items and continue.
On Thu, Jul 14, 2011 at 5:31 PM, Jordan Zimmerman
jzimmer...@netflix.comwrote
It seems implicit in the docs that it's not possible to add nodes (even
Observers) to an existing ensemble without restarting each node in the
ensemble. i.e. The zoo.cfg has to be changed for each node. Is this true?
Is there any work being done on a broadcast protocol that doesn't require
static
I have a three node ensemble. Each node responds to ruok with imok.
Yet, each one response to stat with This ZooKeeper instance is not
currently serving requests. What gives? Which one should I trust? How
might it get in this state and what's the correct way to fix it?
-Jordan
means that it is not currently a member of the quorum.
You'd need to check the logs to see why it's not participating.
Patrick
On Fri, Aug 26, 2011 at 11:06 AM, Jordan Zimmerman
jzimmer...@netflix.com wrote:
I have a three node ensemble. Each node responds to ruok with imok.
Yet, each one
wrote:
Not really.
Diagnosing the network is an important step. Diagnosing quorum formation
is
the next step.
On Fri, Aug 26, 2011 at 11:31 AM, Jordan Zimmerman
jzimmer...@netflix.comwrote:
Thanks - that makes ruok essentially useless. Grrr...
On 8/26/11 11:10 AM, Patrick Hunt ph
When/why does sync() need to be called? I've searched the archive and the
docs are non-existent. Any examples would be appreciated.
-JZ
/r3.3.3/zookeeperProgrammers.html#ch_zkGuar
antees
specifically the section Simultaneously Consistent Cross-Client Views
Patrick
On Tue, Aug 30, 2011 at 9:52 AM, Jordan Zimmerman
jzimmer...@netflix.com wrote:
When/why does sync() need to be called? I've searched the archive and
the
docs are non
I have a script that executes 'kill' on an instance when needed. However,
the process isn't dying. After the kill, ps still shows the process
running. Any ideas?
-JZ
Smith philip_sm...@apple.com
wrote:
kill -9
Review the man page to understand how a process responds to the
different signals.
On Sep 6, 2011, at 10:39 AM, Jordan Zimmerman wrote:
I have a script that executes 'kill' on an instance when needed.
However,
the process isn't dying. After the kill
Lol - that would explain why it isn't working
On 9/6/11 11:12 AM, Ted Dunning ted.dunn...@gmail.com wrote:
I believe (but should check) that the kill command is disabled by default.
On Tue, Sep 6, 2011 at 6:10 PM, Jordan Zimmerman
jzimmer...@netflix.comwrote:
I'm referring to the four-letter
I gave it a vote - thanks.
On 9/20/11 12:44 PM, Fournier, Camille F. camille.fourn...@gs.com
wrote:
Well, if you are locking member IDS that are generated strings that will
never be reused, the best you can do right now is clean those up after a
period of time.
If, on the other hand, your
https://github.com/Netflix/curator
What is Curator?
Curator n: a keeper or custodian of a museum or other collection - A
ZooKeeper Keeper.
Curator is a set of Java libraries that make using Apache ZooKeeper much
easier. While ZooKeeper comes bundled with a Java client, using the client
is
We'd appreciate any/all feedback, BTW.
-JZ
On 10/11/11 1:04 PM, Mahadev Konar maha...@hortonworks.com wrote:
Nice. Good to see the Apache License.
mahadev
On Mon, Oct 10, 2011 at 2:22 PM, Jordan Zimmerman
jzimmer...@netflix.comwrote:
https://github.com/Netflix/curator
What is Curator
The major raison d'être for Curator is to make adding recipes much
easier. Using the Curator Framework APIs you can easily build new
usages/recipes and not worry about connection management.
-JZ
On 10/11/11 1:46 PM, Ted Dunning ted.dunn...@gmail.com wrote:
What I prefer to see in this context
Sorry. I didn't mean to come across as prickly.
Jordan Zimmerman
On Oct 11, 2011, at 4:52 PM, Camille Fournier cami...@apache.org wrote:
You asked for feedback, I gave it.
Good 3 ed. :)
Jordan Zimmerman
On Oct 11, 2011, at 7:46 PM, Ted Dunning ted.dunn...@gmail.com wrote:
Don't worry about being prickly. Camille and I can beat you any day on that
account.
Just be good hearted and serious about making things good while you are
being
It might be possible to put a Scala wrapper around the recipe classes. I'll see
what I can do - or if you want to contribute that ;)
Jordan Zimmerman
On Oct 11, 2011, at 4:40 PM, Joe Stein charmal...@allthingshadoop.com wrote:
Its not written in Scala :( otherwise
of being unable to delete the
directories.
C
On Tue, Oct 11, 2011 at 10:53 PM, Jordan Zimmerman
jzimmer...@netflix.com wrote:
Good 3 ed. :)
Jordan Zimmerman
On Oct 11, 2011, at 7:46 PM, Ted Dunning ted.dunn...@gmail.com
wrote:
Don't worry about being
FYI - Curator checks for KeeperException.Code.NODEEXISTS in its retry loop
and just ignores it treating it as a success. I'm not sure if other
libraries do that. So, this is a case that a disconnection can be handled
generically.
-JZ
On 10/14/11 7:20 AM, Fournier, Camille F.
with sequential files since you don't know who
created any other znodes out there.
On Fri, Oct 14, 2011 at 9:39 AM, Jordan Zimmerman
jzimmer...@netflix.comwrote:
FYI - Curator checks for KeeperException.Code.NODEEXISTS in its retry
loop
and just ignores it treating it as a success. I'm not sure if other
Actually, as I think about it, it's incorrect to ignore
KeeperException.Code.NODEEXISTS. This is because create() also takes a
byte[] to set as the data. Another process may have created the node. I
need to rethink this...
On 10/14/11 9:51 AM, Jordan Zimmerman jzimmer...@netflix.com wrote
that works for every use case.
C
-Original Message-
From: Jordan Zimmerman [mailto:jzimmer...@netflix.com]
Sent: Friday, October 14, 2011 12:39 PM
To: user@zookeeper.apache.org; 'Mike Schilli'
Subject: Re: Locks based on ephemeral nodes - Handling network outage
correctly
FYI - Curator checks
This group does not support zkClient. You should bring it up on the
zkClient website.
-JZ
On 10/19/11 4:57 PM, nileader nilea...@gmail.com wrote:
Nice to hear from you. C.
So, why tricky. In zk's original api, if you add an error authInfo, then
you
will receive en exception of NoAuth. Then
I have a node that I restarted and it's not joining the ensemble. I ask it
'stat' and it says it isn't serving any requests. Here's a snippet from
the log:
2011-10-21 22:46:23,166 - INFO
[QuorumPeer:/0.0.0.0:2181:FastLeaderElection@697] - Notification time out:
800
2011-10-21 22:46:23,167 - INFO
FYI - I turned on DEBUG and here's more log info:
2011-10-21 23:33:06,732 - DEBUG
[QuorumPeer:/0.0.0.0:2181:FastLeaderElection@510] - id: 3, proposed id: 3,
zxid: 12885265585, proposed zxid: 12885265585
2011-10-21 23:33:06,732 - DEBUG
[QuorumPeer:/0.0.0.0:2181:FastLeaderElection@727] - Adding
Interesting. I restarted Server 2 in the ensemble and the problem cleared
itself.
-JZ
On 10/21/11 4:34 PM, Jordan Zimmerman jzimmer...@netflix.com wrote:
FYI - I turned on DEBUG and here's more log info:
2011-10-21 23:33:06,732 - DEBUG
[QuorumPeer:/0.0.0.0:2181:FastLeaderElection@510] - id: 3
FYI - Netflix has just open-sourced a new library that does many recipes:
https://github.com/Netflix/curator
From: zknewbie [sza...@narus.com]
Sent: Sunday, October 23, 2011 8:41 PM
To: zookeeper-u...@hadoop.apache.org
Subject: Distributed lock via
http://zookeeper.apache.org/doc/trunk/recipes.html#sc_recipes_Locks
I've just been informed that the recipe for Locking (and probably others) is
seriously incorrect. When creating a sequential ephemeral node, there is a
possibility that the ZK server can crash before the result is returned to
could you give more information about your client library? I'd like to
include
it in my list of work that people were forced to do to work around the
shortcommings of the current ZooKeeper client API. You'd be number four
after
A lot could be solved by adding a retry (injectable please)
A while back I asked about sync() and got responses that said it's only needed
for reads (getData, getChildren). I was looking through the source and it
appears that this does _not_ apply to exsists(). Am I reading that correctly?
i.e. will exists() always return an accurate (in terms of the
On 11/13/11 3:40 PM, Jérémie BORDIER jeremie.bord...@gmail.com wrote:
As noticed in ZOOKEEPER-1209, this can cause really important issues.
As Leader election is one of the most demanded feature / recipe, I
would really like to see the official recipe fixed and fully
functional.
Curator handles
On 11/13/11 4:45 PM, Jérémie BORDIER jeremie.bord...@gmail.com wrote:
Hello Jordan,
Thanks a lot for your answer. I tried to figure out where the handling
of Disconnected / Expired takes place, but so far I understood that to
have notifyClientClosing() called from the Lock, an exception needs to
/18/11 9:52 AM, Ted Dunning ted.dunn...@gmail.com wrote:
Is the background sync even necessary? The ZK client itself will
re-establish connection if it can.
I think that LOST should only be sent on session expiration.
On Fri, Nov 18, 2011 at 1:07 AM, Jordan Zimmerman
jzimmer...@netflix.comwrote
Curator has both a Queue and a Priority Queue as described on the
ZooKeeper recipes page:
https://github.com/Netflix/curator/wiki/Recipes
-JZ
On 12/6/11 11:16 AM, Mike Schilli m...@perlmeister.com wrote:
Seems like it should be possible to implement a queue service like
Amazon's SQS [1] with
FYI
https://github.com/Netflix/curator
I've decided to move Curator to 1.0 as it is feature complete and has seen
good usage since it's initial announcement. Some highlights of
developments since the initial announcement:
* An implementation for every recipe on the ZooKeeper recipe wiki (sans
at 2:12 PM, Jordan Zimmerman
jzimmer...@netflix.com wrote:
FYI
https://github.com/Netflix/curator
I've decided to move Curator to 1.0 as it is feature complete and has
seen
good usage since it's initial announcement. Some highlights of
developments since the initial announcement
Not currently. I'm working on 3.4 compatibility and should have something
soon.
-JZ
On 1/3/12 11:09 AM, Dima Gutzeit dima.gutz...@mailvision.com wrote:
I have a small question to Curator author/users, does it support batch
operations introduced in 3.4 ?
Thanks in advance.
Regards,
Dima
I'm updating Curator to 3.4.x (3.4.2 to be exact). A lot of my tests have
failed and the culprit is that it appears to be taking a lot longer to get
the initial SysConnected. Of course, this is a flaw in my tests, but I
thought I'd mention it and see if anyone else has noticed this.
-JZ
or just generally slower.
No load at all. These are just unit tests, 1 or 2 connections.
jdk versions change?
java version 1.6.0_29 (Apple)
-JZ
On 1/3/12 4:20 PM, Patrick Hunt ph...@apache.org wrote:
On Tue, Jan 3, 2012 at 4:02 PM, Jordan Zimmerman jzimmer...@netflix.com
wrote:
I'm updating Curator
://issues.apache.org/jira/browse/ZOOKEEPER-1336
Contributing docs is a great way to get started as a zk contributor
(hint hint ;-) )
Patrick
On Wed, Jan 4, 2012 at 7:30 AM, Ted Dunning ted.dunn...@gmail.com wrote:
Not really.
On Wed, Jan 4, 2012 at 12:47 AM, Jordan Zimmerman
jzimmer...@netflix.comwrote
Did the patch for https://issues.apache.org/jira/browse/ZOOKEEPER-1336
not do a good enough job on that?
Missed that - thanks.
On 1/4/12 10:20 AM, Patrick Hunt ph...@apache.org wrote:
On Wed, Jan 4, 2012 at 10:03 AM, Jordan Zimmerman
jzimmer...@netflix.com wrote:
* Can someone give more
If I read the code correctly, the server ID in zoo.cfg (i.e.
server.n=foo) does _not_ have to be sequential, 1 based, etc. i.e. it
can be anything useful to me. For example, I could do:
server.10064=foo:288:3888
server.18=bar:288:3888
server.65535=snafu:288:3888
Am I correct?
-JZ
Hi Josh,
Second use case: Distributed locking
This is one of the most common uses of ZooKeeper. There are many
implementations - one included with the ZK distro. Also, there is Curator:
https://github.com/Netflix/curator
First use case: Distributing work to a cluster of nodes
This sounds
Stone pacesysj...@gmail.com wrote:
Thanks for the response. Comments below:
On Wed, Jan 4, 2012 at 10:46 PM, Jordan Zimmerman
jzimmer...@netflix.comwrote:
Hi Josh,
Second use case: Distributed locking
This is one of the most common uses of ZooKeeper. There are many
implementations - one
Care to work on it?
On 1/5/12 12:50 AM, Ted Dunning ted.dunn...@gmail.com wrote:
This pattern would make a nice addition to Curator, actually. It comes up
repeatedly in different contexts.
: Is the distributed queue effectively located by a single
z-node? What happens when that node goes down? Will a node going down
still
clear any distributed locks?
Josh
On Thu, Jan 5, 2012 at 9:41 AM, Jordan Zimmerman
jzimmer...@netflix.comwrote:
FYI - Curator has a resilient message Queue:
https://github.com
I've just released Curator 1.1.0 that adds support for ZooKeeper transactions.
I'm now going to maintain two branches of Curator:
* 1.0.x for ZooKeeper 3.3.x
* 1.1.x+ for ZooKeeper 3.4.x+
The Curator Transaction APIs use the same oh-so-cool Fluent style as the rest
of Curator. E.g.
is highly pluggable.
This pattern would make a nice addition to Curator, actually. It comes up
repeatedly in different contexts.
On Thu, Jan 5, 2012 at 12:11 AM, Jordan Zimmerman
jzimmer...@netflix.comwrote:
OK - so this is two options for doing the same thing. You use a Leader
Election algorithm
As a backup strategy, it seems I would only want to backup snapshots from
the leader. Does that make sense?
-JZ
time you want to
take a backup? That would be the downside to this strategy I would
think.
C
From my phone
On Jan 13, 2012 5:24 PM, Jordan Zimmerman jzimmer...@netflix.com
wrote:
As a backup strategy, it seems I would only want to backup snapshots
from
the leader. Does that make sense?
-JZ
OK - I'll give you access to the repo as soon as it's in a reasonable
state.
On 1/17/12 10:10 AM, Neha Narkhede neha.narkh...@gmail.com wrote:
Jordan,
I'd be interested in previewing it. Let me know.
Thanks,
Neha
On Mon, Jan 16, 2012 at 5:42 PM, Jordan Zimmerman
jzimmer...@netflix.com wrote
durability would be acceptable. If it is not acceptable and
you still want to have a backup, then I don't see a way other than
shutting down the clients before you take a backup, which doesn't seem
to be what is being proposed here.
-Flavio
On Jan 18, 2012, at 1:38 AM, Jordan Zimmerman wrote:
Neha
you take a backup, which doesn't seem to be what is being proposed
here.
-Flavio
On Jan 18, 2012, at 1:38 AM, Jordan Zimmerman wrote:
Neha - can you send me your email address. Send it to:
jzimmer...@netflix.com
On 1/17/12 10:10 AM, Neha Narkhede neha.narkh...@gmail.com
wrote:
Jordan
Correct
On 1/19/12 11:30 AM, Flavio Junqueira f...@yahoo-inc.com wrote:
You're not talking about data corruption, are you? It is incorrect
data that has been introduced by a user or application by mistake. Am
I getting it right?
-Flavio
On Jan 19, 2012, at 8:07 PM, Jordan Zimmerman wrote
The problem with 'ruok' is that it doesn't tell you the state of the
Instance. 'ruok' might return 'imok' but the instance might not be serving
due to some other error. Only a 'stat' will tell you that.
-JZ
On 1/23/12 1:51 PM, Philip Smith philip_sm...@apple.com wrote:
There is a batch java
The Curator Test module has a class that does this:
https://github.com/Netflix/curator/blob/master/curator-test/src/main/java/c
om/netflix/curator/test/KillSession.java
On 1/26/12 2:38 PM, Benjamin Reed br...@apache.org wrote:
one easy way is to terminate the session using JMX.
ben
On Thu,
Is 'srvr' the same as 'stat' but without the clients? I'm relying on it in
a monitor app. I need to the Mode and the message not currently serving.
-JZ
On 1/25/12 3:32 PM, Patrick Hunt ph...@apache.org wrote:
Prefer srvr over stat in most cases - stat returns details on the
connections which
You can either create a new LeaderSelector or call start() again on your
existing leader instance. Whatever's easier for your use-case.
-Jordan
On 2/5/12 8:09 AM, Hartmut Lang hartmut.l...@googlemail.com wrote:
Hi,
i work on a small demo application using the Curator Leader-Election.
What i
.
But no ephemeral node in the cluster.
/Hartmut
Am 6. Februar 2012 18:46 schrieb Jordan Zimmerman jzimmer...@netflix.com:
How are you verifying that there is no ephemeral node?
-Jordan
On 2/6/12 9:28 AM, Hartmut Lang hartmut.l...@googlemail.com wrote:
Hi Jordan,
thanks for your infos.
What
, without redoing the lock in the ZK-cluster.
This seems not ok for me.
But i'm the newbie here.
Would be great if you can have a look.
/Hartmut
Am 7. Februar 2012 09:05 schrieb Jordan Zimmerman
jzimmer...@netflix.com:
I just pushed a test that simulates the situation you describe and it
works
at this point. There is no reasonable
alternative that I can think of.
I'm still thinking about this, so stay tuned...
On 2/7/12 1:26 PM, Jordan Zimmerman jzimmer...@netflix.com wrote:
I really appreciate your help Hartmut. You have, indeed, found a bug. My
test case didn't precisely replicate your
Some of the Curator recipes use async. For instance, adding a entry into
the DistributedQueue is done in the background as there's no reason for
the foreground process to wait for this.
-JZ
On 2/8/12 2:09 PM, Pierre Louis Aublin pierre-louis.aub...@inria.fr
wrote:
Hello everybody
I would like
Curator provides an in-memory ZooKeeper Server and ZooKeeper Cluster:
dependency
groupIdcom.netflix.curator/groupId
artifactIdcurator-test/artifactId
version1.1.2/version
/dependency
https://github.com/Netflix/curator/tree/master/curator-test/src/main/java/c
om/netflix/curator/test
Yes, there's an issue in Jira on this:
https://issues.apache.org/jira/browse/ZOOKEEPER-723
On 3/1/12 3:38 PM, Shelley, Ryan ryan.shel...@disney.com wrote:
I know that Ephemeral nodes can't have children, but I was curious if
there's been discussion around this lately, in particular around
I suggest starting with Curator. It would make a good foundation.
-JZ
On 3/2/12 12:39 AM, Christopher Schmidt fakod...@googlemail.com wrote:
Hi all, we plan to use Zookeeper and I wonder if there is a Java framework
out there to use ZK as a central configuration service (holding paths,
Announcing the beta release of Exhibitor.
https://github.com/Netflix/exhibitor
Exhibitor is a Java supervisor system for ZooKeeper. It provides a number
of features:
* Watches a ZK instance and makes sure it is running
* Performs periodic backups
* Perform periodic cleaning of ZK log
The ListenerContainer for PathChildrenCache allows you to pass an Executor
along with your listener. So, that will give you the desired behavior.
DistributedQueue is _supposed_ to allow this as well, but it looks like it
doesn't. I view that as a bug. If you don't mind, please post an issue on
I've been reading the archives regarding rolling upgrades. Here's the scenario,
given a stable ensemble:
ZK1 - ZK2 - ZK3
In the above, the zoo.cfg for each server looks like this (pseudo):
server.1=ZK1
server.2=ZK2
server.3=ZK3
I want to add a new server, ZK4. If I understand this correctly,
files of all servers at once and restart them,
because a majority of the new config
necessarily intersects with a majority of the old one, so a server who
has the latest state will be elected leader.
Alex
-Original Message-
From: Jordan Zimmerman [mailto:jzimmer...@netflix.com]
Sent
I'm pretty excited about this one so I want to pre-annouce it. I'm currently
working on rolling release/upgrade support in Exhibitor and Curator.
For Exhibitor:
* Ensemble list, config values, etc. are centralized (via Amazon S3, a
shared file, etc.)
* If the ensemble list changes,
It sounds like you're not setting the correct ACL in your Curator use.
Curator supports ACLs.
FYI - this issue might be better to post at
https://github.com/Netflix/curator/issues (a Curator mailing list is
coming soon).
-JZ
On 4/13/12 5:14 AM, antoniom antonio...@gmail.com wrote:
Hi there,
Introducing Exhibitor - A Supervisor System for Apache ZooKeeper
Exhibitor provides a number of features that make managing a ZooKeeper
ensemble much easier:
* Instance Monitoring
* Log Cleanup
* Backup/Restore
* Cluster-wide Configuration
* Rolling Ensemble Changes
* Visualizer
* Curator
FYI - making this process more reliable is one of the main reasons for
Exhibitor:
http://techblog.netflix.com/2012/04/introducing-exhibitor-supervisor-system
.html
On 4/18/12 8:44 AM, Ted Dunning ted.dunn...@gmail.com wrote:
As long as the old quorum constitutes a quorum in the new cluster you
Have a look at the constructor for StaticHostProvider. You'll see that it
does a Collections.shuffle(this.serverAddresses) - so each client should
connect to a random server in the ensemble.
-JZ
On 4/20/12 5:25 PM, Matthew Ward m...@pixelpipe.com wrote:
Hello,
I am running an ensemble of 3
zkCli.sh seems to use port 8080. Is there any config/argument to change
this? I couldn't find one. It's conflicting with another process on the
machine.
-JZ
Doh - wrong subject.
Is there any way to tell zkCli.sh to now register JMX or to not use port
8080 for it? It's conflicting with another process on our server.
-JZ
On 4/25/12 2:56 PM, Jordan Zimmerman jzimmer...@netflix.com wrote:
zkCli.sh seems to use port 8080. Is there any config/argument
See #6 in the FAQ: http://wiki.apache.org/hadoop/ZooKeeper/FAQ
Also, FYI, we recently open sourced a ZooKeeper supervisor app that makes
these kinds of tasks simpler:
https://github.com/Netflix/exhibitor
-Jordan
On 4/26/12 2:46 PM, Mohammad Abdul-Amir (Shamma)
mohammadsha...@gmail.com
Yes, you must start them all individually. ZooKeeper does not currently
have any cluster-wide management tools built in.
FYI - we've open sourced a cluster management tool for ZooKeeper. Please
have a look:
https://github.com/Netflix/exhibitor
On 5/8/12 2:31 PM, Something Something
:
Thanks Jordan. Will definitely look at the cluster management tool. In
the mean time, am I correct with my assumption in #4? Once all of them
are
started they would automagically start talking to each other, right?
On Tue, May 8, 2012 at 2:36 PM, Jordan Zimmerman
jzimmer...@netflix.comwrote:
Yes
Received: 4
Sent: 3
Outstanding: 0
Zxid: 0x1
Mode: follower
Node count: 11
Connection closed by foreign host.
Not sure I understand this output. I have 9 nodes in the ensemble.
On Tue, May 8, 2012 at 2:45 PM, Jordan Zimmerman
jzimmer...@netflix.comwrote:
Yes - they should. Keep in mind
understand the difference between Nodes Instance. I expected ZKNodes to
be 9. Please explain. Thanks a lot for your help.
On Tue, May 8, 2012 at 3:05 PM, Jordan Zimmerman
jzimmer...@netflix.comwrote:
You have 9 instances? That's a lot. Why so many?
I believe the Node field in the stat
Interesting - this issue has come up several times with Curator users. I
ended up writing a Tech Note on it.
https://github.com/Netflix/curator/wiki/Tech-Note-1
-JZ
On 5/9/12 1:23 PM, Patrick Hunt ph...@apache.org wrote:
I believe the issue is that there is a single thread updating
watchers.
I've decided to add an alternate version of leader selection that doesn't
use threads. It will behave somewhat like a CountDownLatch so I'm calling
it LeaderLatch. I'll report back (only on the Curator list) in the next
day or so when it's available.
-Jordan
On 5/9/12 2:51 PM, Patrick Hunt
When using SASL authentication, waiting for SysConnected isn't enough. Clients
need to wait for SaslAuthenticated as well before calling ZK methods. Is there
a way for a library such as Curator to know that SASL is enabled so that it can
know that it needs to wait for SaslAuthenticated?
I
Never mind - https://issues.apache.org/jira/browse/ZOOKEEPER-1437 fixes the
issue
From: Netflix jzimmer...@netflix.commailto:jzimmer...@netflix.com
Date: Thu, 17 May 2012 00:23:08 -0700
To: user@zookeeper.apache.orgmailto:user@zookeeper.apache.org
Subject: SaslAuthenticated
When using SASL
ZooKeeper has a telnet style interface for periodic querying.
You could also use Exhibitor and query it's REST API periodically. I
should probably add alerting to Exhibitor for this kind of thing.
-JZ
On 5/18/12 10:34 AM, Adam Rosien a...@rosien.net wrote:
We have a 5-member 3.3.3 cluster. One
I'm trying to add a new instance to the ensemble and it is throwing while
trying to sync. Any ideas?
2012-05-24 20:21:08,751 - WARN
[NIOServerCxn.Factory:0.0.0.0/0.0.0.0:2181:NIOServerCnxn@639] - Exception
causing close of session 0x0 due to java.io.IOException: ZooKeeperServer not
running
Later on, I get this exception:
2012-05-24 20:23:03,800 - WARN [QuorumPeer:/0.0.0.0:2181:QuorumPeer@497] -
Unable to load database
java.io.IOException: Transaction log:
/mnt/data/zookeeper/version-2/log.d0f357a0f has invalid magic number 0 !=
1514884167
at
We have a node that has 200K items and would like to delete them. getChildren()
keeps failing. Is there anything that can be done?
-JZ
;
if (len 0 || len maxBuffer) {
throw new IOException(Unreasonable length = + len);
}
byte[] arr = new byte[len];
in.readFully(arr);
return arr;
}
On Thu, May 24, 2012 at 11:17 PM, Jordan Zimmerman
jzimmer...@netflix.com
wrote:
We have a node
I found the problem. I needed to increase the values for initLimit and
syncLimit.
-JZ
On 5/24/12 1:23 PM, Jordan Zimmerman jzimmer...@netflix.com wrote:
Later on, I get this exception:
2012-05-24 20:23:03,800 - WARN [QuorumPeer:/0.0.0.0:2181:QuorumPeer@497]
- Unable to load database
I think we may be running into ZOOKEEPER-1367 in Production (we're still on
3.3.3). Is there a reliable way to reproduce this in a test environment?
-JZ
Our ZK data size is currently around 116GB on disk. I find that I'm needing to
set initLimit and syncLimit to very big numbers. Currently, the cluster cannot
get a quorum on restart unless I have these values:
initLimit=300
syncLimit=300
tickTime=2000
Does that seem normal?
-JZ
What's the latest snapshot size look like?
1,378,363,003
What's the size of the ensemble. How many znodes. etc...
3 nodes - AWS m1.xlarge
From: Patrick Hunt ph...@apache.org
To: user@zookeeper.apache.org; Jordan Zimmerman jor...@jordanzimmerman.com
Sent
http://zookeeper.apache.org/doc/r3.3.5/releasenotes.html
The bugs fixed in 3.3.5 seem to apply only to the server, but it isn't totally
clear. If I were using a 3.3.4 client on a 3.3.5 server might I still see any
of these issues?
-JZ
FYI - have a look at our Exhibitor system which makes upgrades/additions
easier to manage.
https://github.com/Netflix/exhibitor
On 6/18/12 11:26 AM, Ted Dunning ted.dunn...@gmail.com wrote:
Yes. Updating and restarting works fine.
One does wonder why you have *four* servers now. That
We'd appreciate knowing what companies are using Curator. So, if you don't
mind, please send me an email (jzimmer...@netflix.com). Let me know if you need
to keep it private or if we can put it on a Powered By page.
Thank you!
-Jordan
All watchers will get called with session expiration, disconnect, etc.
Jordan Zimmerman
On Jun 26, 2012, at 7:51 AM, David Nickerson
davidnickerson4mailingli...@gmail.com wrote:
In my locking implementation, if a thread wants to wait for a lock, it will
create a watcher
Even if it did (I don't actually know), I'd be nervous about having that
kind of dependency in my app. What's the reason you need this?
-JZ
On Tue, Jun 26, 2012 at 1:25 PM, David Nickerson
davidnickerson4mailingli...@gmail.com wrote:
Is there any guarantee of order? For example, does the
1 - 100 of 371 matches
Mail list logo