Hi Avinash, getChildren returns a list in lexicographic order, so if you are
updating the children of the election node concurrently, then you may get a
different first node with different clients. If you are using the sequence flag
to create nodes, then you may consider stripping the prefix of
Mark, Please use a port for electionPort different from the one you're using
in the server configuration.
Thanks,
-Flavio
> -Original Message-
> From: mark harwood [mailto:[EMAIL PROTECTED]
> Sent: Wednesday, August 27, 2008 1:12 PM
> To: zookeeper-user@hadoop.apache.org
> Subject: Migrat
With the new leader election, we require a third port. So, there is the
clientPort, there is the port servers use for communication upon regular
ZooKeeper operation, and a third port for leader election among the
ZooKeeper servers. The previous leader election algorithm used UDP, and the
regular co
Austin, Please check:
https://issues.apache.org/jira/browse/ZOOKEEPER-140
Thanks,
-Flavio
> -Original Message-
> From: Austin Shoemaker [mailto:[EMAIL PROTECTED]
> Sent: Tuesday, September 16, 2008 12:22 PM
> To: zookeeper-user@hadoop.apache.org
> Subject: Re: Leader election stalled
>
Hi Thomas, We currently don't have such a feature of adding and removing
servers dynamically, although we would like to, so we'll have it eventually.
Without a dynamic mechanism for adding and removing servers, your example is
problematic. Suppose that you configure your ensemble to have 3 servers
HiPath Applications
>
> SEN LIP DA 11
> Schertlinstr. 8
> 81379 Munich, Germany
>
> -Ursprüngliche Nachricht-
> Von: Flavio Junqueira [mailto:[EMAIL PROTECTED]
> Gesendet: Montag, 17. November 2008 13:49
> An: zookeeper-user@hadoop.apache.org
> Betreff: RE: Dynamic server m
Hi Kevin, The admin doesn't need to create "myid". This is created
automaticaly after parsing the configuration file (check
QuorumPeerConfig.parse(String[]) if you are interested in the
internals).
-Flavio
On Jan 5, 2009, at 1:10 AM, Kevin Burton wrote:
This wasn't clear in the documentat
uorumPeerConfig code tomorrow..
Kevin
On Sun, Jan 4, 2009 at 11:58 PM, Flavio Junqueira inc.com>wrote:
Hi Kevin, The admin doesn't need to create "myid". This is created
automaticaly after parsing the configuration file (check
QuorumPeerConfig.parse(String[]) if you are in
new ClientCnxn(host, sessionTimeout, this, watchManager,
sessionId, sessionPasswd);
}
On Mon, Jan 5, 2009 at 1:51 AM, Flavio Junqueira
wrote:
Are you guys passing one server to the ZooKeeper constructor or a
list of
servers? If possible, could you provide your part of the
Are you guys passing one server to the ZooKeeper constructor or a list
of servers? If possible, could you provide your part of the code in
which you create a ZooKeeper object?
Thanks,
-Flavio
On Jan 5, 2009, at 10:46 AM, David Yee wrote:
I'm seeing this behavior as well, and I'm dealing wit
If I understand it correctly, you propose two mechanisms:
1- Have one single node, and modify the data of that znode;
2- Have a znode, say "/broadcast", and have clients creating a new
child znode under "/broadcast" for every new message they want to
broadcast.
In case 1), if you are propos
ages ahead. To avoid synchronization, clients have to start at
different points of the message sequence.
This might not be useful to you, but I wonder if this is useful to
others in this list.
-Flavio
On Jan 6, 2009, at 6:48 PM, Flavio Junqueira wrote:
If I understand it correctly, you propose two
On Jan 6, 2009, at 6:55 PM, Kevin Burton wrote:
In case 1), if you are proposing to overwrite the content of the
znode,
then you would need first to make sure that all receivers have
already
received the previous message. This doesn't seem a good solution to
me
because a client that want
You can't simply leave an element in the queue until a consumer
finishes processing it, otherwise multiple consumers may end up
processing it. What about the following:
- Use a failure detector to detect which consumers are up;
- Before removing an element from the queue, a consumer creates a
Andrew, It sounds right to me that this is a bug. In any case, I'd
like to suggest that you open a jira issue and propose your patch.
-Flavio
On Jan 21, 2009, at 6:52 AM, Andrew Carman wrote:
We were looking through the ZK server code and came across a
possible bug
that occurs when shutti
Hi Bill, I'm sorry, I missed this message initially. I'm sending below
a table that gives you throughput figures for BookKeeper. The rows
correspond to distinct BookKeeper configuration (ensemble size, quorum
size, entry type), and the columns to different values for the length
of an entry
Also, you may consider checking a graph that we posted comparing the
performance of BookKeeper with the one of HDFS using a local file
system and local+NFS in the jira issue 5189 (https://issues.apache.org/jira/browse/HADOOP-5189
).
-Flavio
On Feb 20, 2009, at 10:05 AM, Flavio Junqueira
I was wondering if anyone in this list would be willing to share
ZooKeeper logs for research purposes, even if anonymized. As there are
a few research groups in different universities conducting research
using ZooKeeper, it would be useful for them to have workloads they
can work with. If y
Todd, I don't think a feature right now that allows you to do exactly
what you're requesting. However, we have been working on a couple of
features that might give you what you want:
1- Hierarchical quorums: this feature allows you to split servers into
groups (perhaps mapping groups to com
requirements.
Do bear in mind that the patch on the jira is only for discussion
purposes;
I would not consider it currently fit for production use. I hope to
put
up a
much better patch this week.
Henry
On Sat, Jul 18, 2009 at 7:38 PM, Ted Dunning
wrote:
Can you submit updates via an observ
better patch this week.
Henry
On Sat, Jul 18, 2009 at 7:38 PM, Ted Dunning
wrote:
Can you submit updates via an observer?
On Sat, Jul 18, 2009 at 6:38 AM, Flavio Junqueira >
wrote:
2- Observers: you could have one computing center containing an
ensemble
and observers around the edge just learning committed values.
--
Ted Dunning, CTO
DeepDyve
Servers in a quorum need to be able to talk to each other to elect a
leader. Once a leader is elected, followers only talk to the leader.
Of course, if the leader fails, servers in some quorum will need to
talk to each other again. If no quorum can be formed, the system is
stalled.
-Flavi
Just a few quick observations:
On Jul 24, 2009, at 4:40 PM, Ted Dunning wrote:
On Fri, Jul 24, 2009 at 4:23 PM, Todd Greenwood
wrote:
Could you explain the idea behind the Observers feature, what this
concept is supposed to address, and how it applies to the WAN
configuration problem in parti
hat Flavio mentions below.
-Original Message-----
From: Flavio Junqueira [mailto:f...@yahoo-inc.com]
Sent: Friday, July 24, 2009 4:50 PM
To: zookeeper-user@hadoop.apache.org
Subject: Re: Zookeeper WAN Configuration
Just a few quick observations:
On Jul 24, 2009, at 4:40 PM, Ted Dunning wrote
Todd, Answers inline:
On Jul 26, 2009, at 11:05 AM, Todd Greenwood wrote:
Flavio, thank you for the suggestion.
I have looked at the documention (relevant snippets pasted in
below), and looked at the presentations (http://wiki.apache.org/hadoop/ZooKeeper/ZooKeeperPresentations
),
but I sti
Todd, Some more answers. Please check out carefully the information at
the bottom of this message.
On Jul 27, 2009, at 4:02 PM, Todd Greenwood wrote:
I'm assuming that you're setting the weight of ZooKeeper servers in
PODs to zero, which means that their votes when ordering updates do
not co
Todd,
On Jul 30, 2009, at 5:08 PM, Todd Greenwood wrote:
The build succeeds, but not the all of the tests. In previous test
runs,
I noticed an error in org.apache.zookeeper.test.FLETest. It was not
able
to bind to a port or something. Now, after a machine reboot, I'm
getting
different fai
You're missing 491 from your set of patches.
-Flavio
On Jul 31, 2009, at 7:15 PM, Todd Greenwood wrote:
This repro's in both branch-3.2, and branch-3.2+patches(473, 479,
481).
Basically, it seems like the nodes are electing pd4-zook02 to be the
leader. However, pd4-zook02 seems to realize i
er Elections in WAN deploy
Ok, I'll apply that patch and report back.
-Todd
-Original Message-
From: Flavio Junqueira [mailto:f...@yahoo-inc.com]
Sent: Friday, July 31, 2009 7:18 PM
To: zookeeper-user@hadoop.apache.org
Subject: Re: Unending Leader Elections in WAN deploy
You're
: Flavio Junqueira [mailto:f...@yahoo-inc.com]
Sent: Friday, July 31, 2009 7:48 PM
To: zookeeper-user@hadoop.apache.org
Subject: Re: Unending Leader Elections in WAN deploy
It should be in 479. Perhaps you have a stale version of the patch.
-Flavio
On Jul 31, 2009, at 7:46 PM, Todd Greenwood
Hi Qian, It would be useful to have access to the log and to know
which version of ZooKeeper you're using. You may want to open a jira
and attach the log there.
Thanks,
-Flavio
On Aug 13, 2009, at 7:51 AM, Qian Ye wrote:
Hi all:
My friend encountered a problem when using ZooKeeper. He bui
I agree with Mahadev that it sounds like a stretch. I just wanted to
point out that we have been working on a new feature that would
certainly help in this case: Observers (ZOOKEEPER-368). I don't think
we have decided yet in which release we will include it, but my
current guess is 4.0.0 a
I forgot to mention this. You may also consider adding more zookeeper
servers and setting the weight of such servers to zero. We will be
introducing this possibility in 3.2.1 (the upcoming release). Zero-
weight servers simulate observers, but they do not behave exactly as a
observers, since
We just need a unique identifier for every server. If such an
identifier "magically" appears somehow, then I believe our protocols
will be equally happy. Now, a mechanism to assign ids would also have
to take into consideration the group scheme we have for hierarchical
quorums. To assign se
Hi Nick, Your assessment sounds correct, the issue seems to be caused
by the bug described in ZOOKEEPER-427. Can't you upgrade to a newer
release? Killing the leader should do it, but the bug will still be
there, so I recommend upgrading.
Thanks,
-Flavio
On Jan 12, 2010, at 10:52 PM, Nick
Hi Vijay, I'm just curious: why exactly you want all voting nodes in a
single data center? Are you concerned about latency?
It might not be possible in your case, but if you have a third
location available, you would be able to tolerate one location going
down.
-Flavio
On Jan 14, 2010, a
2010, at 11:00 PM, Vijay wrote:
Hi Falvio,
Yes i am concerned about the latency between the DC's (Across
continents),
We actually have 6 locations but how exactly are we going to do it
if we
have the 3rd DC?
Regards,
On Thu, Jan 14, 2010 at 1:46 PM, Flavio Junqueira inc.com> wrot
Hi, Mahadev said it all, we have been thinking about it for a while,
but haven't had time to work on it. I also don't think we have a jira
open for it; at least I couldn't find one. But, we did put together
some comments:
http://wiki.apache.org/hadoop/ZooKeeper/PartitionedZookeeper
On top of Ben's description, you probably need to set initLimit to
several minutes to transfer 700MB (worst case). The value of
syncLimit, however, does not need to be that large.
-Flavio
On Mar 15, 2010, at 7:24 PM, Benjamin Reed wrote:
it is a bit confusing but initLimit is the timer tha
Hi Chen, Let's say that the value of a znode "/test" is initially v
and client A writes value v' to znode "/test". If the server that
client B is connected to has not persisted the update operation of A,
it will read v. If it submits sync before the read, client B will read
v'.
-Flavio
O
Hi Sudipto, I'm sorry but I don't have slides I can share that include
a description of leader election. I'll work on it when I have a
chance. In any case, you may consider inspecting the code if you need
it urgently.
-Flavio
On May 13, 2010, at 10:54 PM, Sudipto Das wrote:
Thanks Mahade
Just in case anyone is interested, I've posted some BookKeeper
performance figures here:
http://wiki.apache.org/hadoop/BookKeeperPerfPage
I'll be adding more numbers soon.
-Flavio
Hi Andre, To guarantee that two clients that read from a ledger will
read the same sequence of entries, we need to make sure that there is
agreement on the end of the sequence. A client is still able to read
from an open ledger, though. We have an open jira about informing
clients of the pr
t I'd certainly be happy to hear a different perspective.
Having said that, we have interesting projects to get folks involved
with BK, but I don't have it clear that this is one of them.
-Flavio
On May 20, 2010, at 1:36 AM, Patrick Hunt wrote:
On 05/19/2010 01:23 PM, Flavio Junqu
Hi Charity, This is certainly not expected. It would be very useful if
you could provide us with as much information about your issue as
possible. I would suggest that either you create a new jira and link
it to ZOOKEEPER-335, or that you add to 335 directly.
We'll be looking further into w
I would recommend opening a separate jira issue. I'm not convinced the
issues are the same, so I'd rather keep them separate and link the
issues if it is the case.
-Flavio
On Jun 17, 2010, at 12:16 AM, Patrick Hunt wrote:
We are unable to reproduce this issue. If you can provide the server
Hi Travis, Do you think it would be possible for you to open a jira and upload your logs?Thanks,-FlavioOn Jul 1, 2010, at 8:13 AM, Travis Crawford wrote:Hey zookeepers -We just experienced a total zookeeper outage, and here's a quickpost-mortem of the issue, and some questions about preventing it g
Hi Sergei, I'm not sure what the implementation of QuorumVerifier you have in mind would look like to make your setting work. Even if you don't have partitions, variation in message delays can cause inconsistencies in your ZooKeeper cluster. Keep in mind that we make the assumption that quorums int
? Would it result in considerable performance impact due to network latency? I hope that at least in theory since quorum can be reached without ack from EC2 node performance impact might be manageable.Regards,SergeiOn 07/14/2010 04:52 PM, Flavio Junqueira wrote:Hi Sergei, I'm not sure wha
Avinash, The sync() call flushes the pending updates between the leader and a follower. Check the "consistency guarantees" section here:http://hadoop.apache.org/zookeeper/docs/r3.3.1/zookeeperProgrammers.html-FlavioOn Sep 17, 2010, at 10:29 PM, Avinash Lakshman wrote:What is a Sync OpCode stand for
Thanks for your questions, Amit.
On Sep 28, 2010, at 6:37 PM, amit jaiswal wrote:
Hi,
I am experimenting with BookKeeper and have a question on
LedgerHandler class.
The readEntries(firstEntry, lastEntry) method takes the indexes of
first and
last entries. Also, the LedgerSequence object r
We don't have dynamic configuration yet, but it is on our todo list: http://wiki.apache.org/hadoop/ZooKeeper/ClusterMembershipso for now I believe you would have to reconfigure manually and restart the cluster. For Zab, you should be looking at org.apache.zookeeper.server.quorum.Cheers,-FlavioOn Oc
I thought we had agreed at some point that the application should do it in the case it needs this feature. That is, every so often the app writer either writes to ZooKeeper its last confirmed write or it sends directly to the reader. Knowing a confirmed write x enables the reader to read up to x.-F
Hi Ruifang, It is not clear to me if you verified that leader1 restarted correctly. Was it able to join the ensemble by following leader2?-FlavioOn Nov 1, 2010, at 8:09 PM, Ruifang Ge wrote:Hi,I started a 5-node zookeeper cluster, then killed the leader (leader1). One of the other server became th
54 matches
Mail list logo
