According to your trace I see you are using jvm 1.6.0_03-b05
One of the bugs fixed in:
http://java.sun.com/javase/6/webnotes/6u4.html
specifically:
http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6403933
seems to have a description very close to what you are seeing.
Perhaps you can try runnin
This log manifests if the client is running ahead of the server.
say you have:
1) client connects to server A and sees some changes
2) client gets disconnected from A and attempts to connect to B
3) B can be running behind A by some number of changes (it will
eventually catch up)
4) client will
Agree, created a new JIRA for this:
https://issues.apache.org/jira/browse/ZOOKEEPER-430
See the following JIRA for one example why not to do this:
https://issues.apache.org/jira/browse/ZOOKEEPER-327
In general you don't want to create large node sizes since all of the
data/nodes are stored in m
wrt bandwidth the issue there is when you do a write you end up copying
the data btw servers in the quorum:
1) client setdata("largedata") -> follower ZK server (copy data)
2) follower ZK server forwards the proposal to the ZK server leader
(copy data)
3) ZK server leader does atomic broadcast
Would love to see it - the wiki might be a better choice in terms of
visibility, I created this page fairly recently:
http://wiki.apache.org/hadoop/ZooKeeper/Troubleshooting
A section on data size impact would be great (incl general information
on the cluster configuration (ie 3 vs 5 servers i
only solution is to restart the server that experiences
this problem.
Patrick
Patrick Hunt wrote:
According to your trace I see you are using jvm 1.6.0_03-b05
One of the bugs fixed in:
http://java.sun.com/javase/6/webnotes/6u4.html
specifically:
http://bugs.sun.com/bugdatabase/view_bug.do?bu
Nitay, any luck? Feel free to create a JIRA to track this. If you point
to the test code that's experiencing the problem we'll try and take a look.
Patrick
Patrick Hunt wrote:
This log manifests if the client is running ahead of the server.
say you have:
1) client connects to se
Agree. I developed on a windows box for about 9months w/o any windows
specific issues being found. If someone were willing to do the work and
take on responsibility for long term dev/test/support I don't see why we
wouldn't add it.
Patrick
ps - notice that the dev level use of windows current
The Hadoop summit is Wednesday. If you're attending please feel free to
say hi -- Mahadev is presenting @4, Ben and I will be attending as well.
Also, regardless of whether you're attending or not we'd appreciate any
updates to the "powered by" page, if you're too busy to update it
yourself se
We probably want to allow the caller to specify which type of watch they
want to remove - a watch on the znode itself, on children of the znode,
or both.
zk.removeWatch(path, watcher, wtype)
where:
path is path to the znode
watcher may be a specific watcher or null matching all watchers
wtype i
Do we have a JIRA for this? If not we should add one for 3.3.
Patrick
Mahadev Konar wrote:
Hi Raghu,
We do have plans to enforce quota in future. Enforcing requires some more
work then just reporting. Reporting is a good enough tool for operations to
manage a zookeeper cluster but we would ce
Ted thanks for the info.
I've created a wiki page
http://wiki.apache.org/hadoop/ZooKeeper/ZooKeeperOnEC2
to capture details of running ZK on EC2. If you or anyone else would
like to update it with information please do so.
Regards,
Patrick
Ted Dunning wrote:
I disagree with the original pos
We'd be happy to take a look, create a JIRA here and attach your patch:
https://issues.apache.org/jira/browse/ZOOKEEPER
Easiest way to create the patch is to use SVN and do a "svn diff". More
details here, please feel free to join the zookeeper-dev mailing list,
that's the main list for develop
Henry Robinson wrote:
Effectively, EC2 does not introduce any new failure modes but potentially
exacerbates some existing ones. If a majority of EC2 nodes fail (in the
sense that their hard drive images cannot be recovered), there is no way to
restart the cluster, and persistence is lost. As you
The Apache ZooKeeper team is proud to announce Apache ZooKeeper version
3.2.0.
ZooKeeper is a high-performance coordination service for distributed
applications. It exposes common services - such as naming, configuration
management, synchronization, and group services - in a simple interface
Nodes are maintained un-ordered on the server. A node can store any
subnodes, not exclusively sequential nodes. If we added an ordering
guarantee then then server would have to store the children sorted for
every parent node. This is a problem for a few reasons; 1) in many cases
you don't care
Yes, this is a strong guarantee:
http://hadoop.apache.org/zookeeper/docs/current/zookeeperProgrammers.html#ch_zkGuarantees
Sync is only necessary if client A makes a change, then client B wishes
to read that change with guarantee that it will see the successfully
applied change previously made
It's hard to say, there are a number of variables. Some things to think
about: Are the tasks idempotent? do they have leases (like SQS)? Is one
process responsible for processing the tasks or will you have many vying
for the jobs? Are the tasks ordered by creation date, or weighted by
some fact
Erik, take a look at this jira:
https://issues.apache.org/jira/browse/ZOOKEEPER-423?focusedCommentId=12715558&page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#action_12715558
Patrick
Erik Holstad wrote:
Thanks Benjamin!
That helped, I guess you are using some kind of set
Erik, if you'd like enter a JIRA and take a whack at it go ahead.
Perhaps a subclass of DataNode specific for ephemerals? That way it can
handle any particulars - and should also minimize the number of
"if(children==null)" type checks that would be needed. (don't neg.
impact performance or b/w
Thanks for the report, looks like something we need to address, would
you mind going the extra step and adding a JIRA on this?
https://issues.apache.org/jira/browse/ZOOKEEPER
Thanks,
Patrick
kishore g wrote:
Hi All,
Zookeeper recipe queue code has a bug.
byte[] b = zk.getData(root + "/elem
Flavio, please enter a doc jira for this if there are no docs, it should
be in forrest, not twiki btw. It would be good if you could review the
current quorum docs (any type) and create a jira/patch that addresses
any/all shortfall.
Patrick
Flavio Junqueira wrote:
Todd, Some more answers. Ple
[Todd] What is the recommended policy regarding patching zookeeper
locally? As an external user, should I patch and compile in the trunk or
in the branch (branch-3.2)?
I've looked at :
http://wiki.apache.org/hadoop/ZooKeeper/HowToContribute
http://wiki.apache.org/hadoop/HowToRelease
And both o
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 failures.
"address in use"? That's a proble
always fail?)
I've entered a jira to address this:
https://issues.apache.org/jira/browse/ZOOKEEPER-492
Patrick
Patrick Hunt wrote:
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 wa
well try running these two tests individually and see if they always
fail or just occassionally. that will be a good start (and the env detail).
Patrick
Todd Greenwood wrote:
No edits to conf/log4j.properties.
-Original Message-
From: Patrick Hunt [mailto:ph...@apache.org]
Sent
Todd Greenwood wrote:
[Todd] Yes, I believe "address in use" was the problem w/ FLETest. I
assumed it was a timing issue w/ respect to test A not fully releasing
resources before test B started.
Might be, but actually I think it's related to this:
http://hea-www.harvard.edu/~fine/Tech/addr
n the report does not reflect the time until the
VM exit.
-Todd
-Original Message-
From: Patrick Hunt [mailto:ph...@apache.org]
Sent: Thursday, July 30, 2009 10:13 PM
To: zookeeper-user@hadoop.apache.org
Subject: Re: test failures in branch-3.2
Todd Greenwood wrote:
[Todd] Yes, I
is in our environment.
Again, sort of a moot point if you can wait a week or so...
Regards,
Patrick
Todd Greenwood wrote:
Inline.
-Original Message-
From: Patrick Hunt [mailto:ph...@apache.org]
Sent: Thursday, July 30, 2009 10:57 PM
To: zookeeper-user@hadoop.apache.org
Subject
I need to proceed with the manual patches to branch-3.2, as I am under
some time constraints to get our infrastructure deployed such that QA
can start playing with it. However, I'll switch to 3.2.1 as soon as I
can.
Understood.
Patrick
-Original Message-
From: Patrick Hunt [mail
Please do enter a JIRA. Looking at the source it seems that we log and
error, but the calling code continues. I think this is happening because
the chroot c lib code is not handling znode watches separate from state
change notifications.
The calling code just continues after logging an (invali
This is currently more of a developer tool but I thought it might be
useful for users as well -- a basic ZooKeeper ensemble configuration
generator that takes some of the drudge work out of generating configs.
I got sick of creating these by hand for the various setups I have (esp
when experime
Avinash Lakshman wrote:
Hi All
Suppose I have a znode, say /Me, and have three nodes A, B and C who have
set watches on this znode. Now suppose some process changes some value on
/Me then watches get delivered to A, B and C. Now if at that instant of time
C were down I could always read the znode
Erik, that's correct. You always see the "current" information on the
server, and as part of the request can ask for notification (one time
trigger) of any subsequent change. This guarantees that you won't miss a
change, but multiple changes may occur btw your calls to the get function.
Patric
Try doing a couple things:
1) use the java cli to do a "stat /path" on the node in question
2) try doing a "echo stat |nc localhost 2181"
(where localhost:2181 is the clientport of the zk server)
would be interesting to see this information on the jira. 1) will allow
you to verify that the node
you are using java 1.6 right? more detail on the class not found would
be useful (is that missing or just not included in your email?) Also the
command line you're using to start the app would be interesting.
Patrick
Mahadev Konar wrote:
Hi Zhang,
Are you using cygwin?
mahadev
On 8/17/09
java/lib/log4j-1.2.15.jar.
Program will exit.
$
Thank you
Jeff zhang
On Tue, Aug 18, 2009 at 12:53 PM, Patrick Hunt wrote:
you are using java 1.6 right? more detail on the class not found would be
useful (is that missing or just not included in your email?) Also the
command line you're u
b/log4j-1.2.15.jar.
Program will exit.
$
Thank you
Jeff zhang
On Tue, Aug 18, 2009 at 12:53 PM, Patrick Hunt wrote:
you are using java 1.6 right? more detail on the class not found would be
useful (is that missing or just not included in your email?) Also the
command line you're using
One more thing, please enter a jira on this so that we can track/fix it.
https://issues.apache.org/jira/browse/ZOOKEEPER
Thanks,
Patrick
Patrick Hunt wrote:
I suspect it has to do with the classpath - specifically having spaces
in the directory name. Notice that one of the lines you included
Hi Qian, it would good if you could create a jira for this:
https://issues.apache.org/jira/browse/ZOOKEEPER
include both the client logs and the server logs (for overlapping
client/server time period where you see the problem). also the server
config if you're using a quorum vs standalone. If y
Hey Pythonner, thanks for the followup!
if I understand correctly, this error message is generated by hbase and
not zk? Can you tell at all? (seems likely to me as the configuration
you re referring to is hbase specific... but would like to verify).
Can anyone confirm, is this an issue with h
Hi Jean-Daniel, not sure I get your response fully. Are you saying that
the configured ip addr was resolved to a hostname, but that hostname
didn't match the list of ip addresses used when defining the zk quorum
machines? Is there a workaround you could suggest for ppl who don't have
DNS availa
p a DNS server or to configure/start zookeeper
by hand. From what I've seen, that stuff is never easier but eh, we
still get you a quorum running in the end :P
J-D
On Mon, Aug 24, 2009 at 4:37 PM, Patrick Hunt wrote:
Hi Jean-Daniel, not sure I get your response fully. Are you saying that t
No worries. The details are actually interesting/useful, you might
consider adding to your docs in case another user runs into this.
Patrick
Jean-Daniel Cryans wrote:
Patrick,
Basically, yes. Sorry for the lengthy answer ;)
J-D
On Mon, Aug 24, 2009 at 5:09 PM, Patrick Hunt wrote:
I see
Nice!
Jean-Daniel Cryans wrote:
Added here http://wiki.apache.org/hadoop/Hbase/Troubleshooting#12
J-D
On Mon, Aug 24, 2009 at 5:20 PM, Patrick Hunt wrote:
No worries. The details are actually interesting/useful, you might consider
adding to your docs in case another user runs into this
Hi Leonard,
Between 00:43:23,035 and 00:43:23,157 I see client session
0x123730dbe6e0001 get 15 "node exists" exceptions in a row. Are you
expecting this? (ie are you attempting to create this node 15 times in a
row or is this unexpected? I can't tell from the client snippet you
included)
A
and the mini-program didn't
have the problem. That caused me to go back and look at my code even more
closely, when I discovered it was My Problem.
Thanks again for your quick reply and sorry to have wasted your time.
Leonard
On 8/31/09 9:16 PM, "Patrick Hunt" wrote:
Hi Le
What is your client timeout? It may be too low.
also see this section on handling recoverable errors:
http://wiki.apache.org/hadoop/ZooKeeper/ErrorHandling
connection loss in particular needs special care since:
"When a ZooKeeper client loses a connection to the ZooKeeper server
there may be so
ed that you are not swapping (see gc pressure), etc...)
Patrick
Satish Bhatti wrote:
Session timeout is 30 seconds.
On Tue, Sep 1, 2009 at 4:26 PM, Patrick Hunt wrote:
What is your client timeout? It may be too low.
also see this section on handling recoverable errors:
http://wiki.ap
Yes. create/set/delete/... are really the issue (non-idempotent).
Satish Bhatti wrote:
Well a bunch of the ConnectionLosses were for zookeeper.exists() calls. I'm
pretty sure dumb retry for those should suffice!
On Tue, Sep 1, 2009 at 4:31 PM, Mahadev Konar wrote:
Hi Satish,
Connectionlos
has one server. Not sure if this is exacerbating the problem?
I will check out the trouble shooting link you sent me.
On Tue, Sep 1, 2009 at 5:01 PM, Patrick Hunt wrote:
I'm not very familiar with ec2 environment, are you doing any monitoring?
In particular network connectivity btw nodes? S
How large/small are the writes?
Can you run the following, then try your test again and report the
results for the time period where your test is running?
iostat -x -d 1
also note that ZK JMX allows you to reset the latency attributes (look
under "operations" in jconsole). If you reset the l
neral advice. I strongly suspect that
something is being observed incorrectly or the machines are being massively
abused.
On Wed, Sep 2, 2009 at 12:37 PM, Patrick Hunt wrote:
I suspect that given a single disk is being used (not a dedicated disk for
the transaction log), and also given
The Apache ZooKeeper team is proud to announce Apache ZooKeeper version
3.2.1.
ZooKeeper is a high-performance coordination service for distributed
applications. It exposes common services - such as naming, configuration
management, synchronization, and group services - in a simple interface
ZooKeeper 3.2.1 was release over the weekend. If you find any issues
with it please create a JIRA. I've attached the announce email which
includes links to the release notes and detailed changelog. If you are
using 3.2.0 please consider upgrading to 3.2.1 asap.
Regards and thanks again to ever
I don't think this changed - we require cppunit to build the c libs.
Perhaps you built on a different machine that had cppunit installed?
I have these installed:
libcppunit-1.12-1 install
libcppunit-dev install
Try commenting out li
Todd, there were other responses as well. Are you seeing other traffic
from the lists? (perhaps a spam filtering issue?)
Patrick
Mahadev Konar wrote:
HI todd,
We did respond on zookeeper-user. Here is my response in case you didn't
see it...
HI todd,
From what I understand, you are sayi
7;/ACLTest\n"
r = {org.apache.zookeeper.proto.replyhea...@1389}"2,2,0\n"
xid = 2
zxid = 2
err = 0
request =
{org.apache.zookeeper.proto.createrequ...@1355}"'/ACLTest,,v{s{15,s{'wor
ld,'anyone}}},0\n"
path = {java.lang.str...@1314}"/ACLTest"
h = {org.apache.zo
wn correctly and finally
expired while running a different/new test?)
Patrick
v5:
response =
{org.apache.zookeeper.proto.createrespo...@1360}"'/ACLTest\n"
r = {org.apache.zookeeper.proto.replyhea...@1389}"2,2,0\n"
-Todd
-Original Message-
From: Patric
Erik, I think you ran into this:
https://issues.apache.org/jira/browse/ZOOKEEPER-420
Henry Robinson from Cloudera (cc'd) created the zkpython contrib, ccing
him if he has a better way, but here's how I am able to run the tests
w/o installing:
I get around it by compiling src/c and then chang
Not sure if you'll find this interesting but my zk configuration
generator is available on github:
http://github.com/phunt/zkconf
"zkconf.py will generate all of the configuration needed to run a
ZooKeeper ensemble. I mainly use this tool for localhost based testing,
but it can generate confi
When the session expires your only option is to create a new session and
re-register the watches, re-create ephemerals, etc...
Patrick
yeqian@gmail.com wrote:
Hi all:
When used Zookeeper in my project, I noticed a problem about the
session. When a session expired, the io thread of the cl
Jason Venner wrote:
I do find having to have a custom file in each zk root somewhat awkward, as
I like to rsync my configuration files around. I also would prefer not to
have to have all of my zk nodes listed in the configuration file by id.
I think I would prefer it if there was a mechanism for
Hi Stefan, two suggestions off the bat:
1) fill in something in the README, doesn't have to be final or
polished, but give some insight into the what/why/how/where/goals/etc...
to get things moving quickly for reviewers & new users.
2) you should really discuss on the dev list. It's up to you
ick look at some of the tests but that
didn't help much. Is there a test(s) in particular that I should look at
to see how zkclient is used, and the benefits incurred?
Regards,
Patrick
Patrick Hunt wrote:
Hi Stefan, two suggestions off the bat:
1) fill in something in the README, doesn&#x
Peter Voss wrote:
On 01.10.2009, at 08:57, Patrick Hunt wrote:
2) what purpose does ZkEventThread serve?
ZkClient updates it's connection state from the ZooKeeper events. Based
on these it notifies listeners, updates it's connection state or
reconnects to ZooKeeper. ZkClient h
Not to harp on this ;-) but this sounds like something that would be a
very helpful addition to the README.
Ted Dunning wrote:
I think that another way to say this is that zkClient is going a bit for the
Spring philosophy that if the caller can't (or won't) be handling the
situation, then they
Ted Dunning wrote:
You may be able to tell if the file is yours be examining the content and
ownership, but this is pretty implementation dependent. In particular, it
makes queues very difficult to implement correctly. If this happens during
the creation of an ephemeral file, the only option ma
Ted Dunning wrote:
Judging by history and that fact that only 40/127 issues are resolved, 3.3
is probably 3-6 months away. Is that a fair assessment?
Yes, that's fair.
Patrick
On Thu, Oct 1, 2009 at 11:13 AM, Patrick Hunt wrote:
One nice thing about ephemeral is that the Stat con
That detail is purposefully not exposed through the client api, however
it is output to the log on connection establishment.
Why would your client code need to know which server in the ensemble it
is connected to?
Patrick
Rob Baccus wrote:
How do I determine the server the client is connect
dress().toString();
}
Feel free to add a JIRA, I think we could make this a protected method
on ZooKeeper to make testing easier (and not expose internals).
Regards,
Patrick
Todd Greenwood wrote:
Failover testing.
-Original Message-
From: Patrick Hunt [mailto:ph...@apache.org]
testing.
-Original Message-
From: Patrick Hunt [mailto:ph...@apache.org]
Sent: Thursday, October 01, 2009 3:44 PM
To: zookeeper-user@hadoop.apache.org; Rob Baccus
Subject: Re: How do we find the Server the client is connected to?
That detail is purposefully not exposed through the client
Hi Hector, looks like a connectivity issue to me: NoRouteToHostException.
3888 is the election port
2888 is the quorum port
basically, the ensemble uses the election port for leader election. Once
a leader is elected it then uses the quorum port for subsequent
communication.
Could it be a fi
You might want to add a link to zkclient on this page:
http://wiki.apache.org/hadoop/ZooKeeper/UsefulTools
Patrick
Patrick Hunt wrote:
Ted Dunning wrote:
Judging by history and that fact that only 40/127 issues are resolved,
3.3
is probably 3-6 months away. Is that a fair assessment?
Yes
I created a short smoketest script that you might find interesting, both
as a tool to help verify new/changed ensembles, but also as an example
of how to use the zkpython bindings. A link is now on the wiki:
http://wiki.apache.org/hadoop/ZooKeeper/UsefulTools
Patrick
FYI: I've updated the smoketest project to also test operation latency.
Take a look, it will give you additional insight into the latencies seen
by zookeeper clients in your environment:
http://bit.ly/2Cs6Ee
Patrick
Patrick Hunt wrote:
I created a short smoketest script that you might
Take all of the server.# lines out, including server.1 (no other change
necessary). For standalone you don't need/want this.
Alternately you could use
org.apache.zookeeper.server.ZooKeeperServerMain
(I don't think you even need to change the config file if you do that).
for example:
java -c
I've seen this before. Either you have an old version of ant, or your
JAVA_HOME is not set, or it's set incorrectly (to 1.5 and ant is built
for 1.6, or vice versa).
Patrick
Henry Robinson wrote:
Hi Steven -
I also see that problem if I build on my Mac sometimes. I'm looking into a
proper fi
You're right, 0 should be something like "INITIALIZING_STATE" but it's
not in zookeeper.h
zookeeper_init(...) docs:
* This method creates a new handle and a zookeeper session that
corresponds
* to that handle. Session establishment is asynchronous, meaning that the
* session should not be c
tween C's ZOO_*_STATE and Java's KeeperState.
-Original Message-
From: Patrick Hunt [mailto:ph...@apache.org]
Sent: Tuesday, October 13, 2009 5:03 PM
To: zookeeper-user@hadoop.apache.org
Subject: Re: C client (via zkpython) returns unknown state
You're right, 0 should be
FWIW I noticed this on twitter last night, a third party PHP binding for
ZooKeeper is now available (I haven't tried it myself):
http://twitter.com/phunt/status/4906002271
Patrick
You might try my ZooKeeper configuration generator if you have python
handy: http://bit.ly/mBEcF
The main issue that I see with your config is that each config file
needs to contain a list of all the servers in the ensemble:
...
syncLimit=2
server.1=host1...
server.2=host2...
server.3=host3..
I've created a ZooKeeper Dashboard using Django and the zkpython
bindings, it's available on github: http://bit.ly/1kjQy0
It currently provides some basic information about the ensemble:
* Cluster summary
* Individual server detail
* Client connection detail
* Navigate & examine
immediate help!
My machines are currently down for other reasons so I'll be able to try things
again, send out the errors I'm getting and such in a day or so (hopefully!)
thanks again!
-Mark
-Original Message-
From: Patrick Hunt [mailto:ph...@apache.org]
Sent: Tuesday, Oct
bummer, donno what to say, what version of python are you using? works
for me in 2.5 and 2.6 (linux jaunty)
$ ls
LICENSE.txt README.html start.py start.pyc stop.py
stop.pyc test3 zoocfg.py zoocfg.pyc
ptestREADME.textile start.py.bak start.tmpl stop.py.bak
sto
Siddharth Raghavan wrote:
I need to restart a single zookeeper server node on the same port within
my unit tests.
Are you testing c or java client?
I tried stopping the server, having a delay and restarting it on the
same port. But the server doesn't startup. When I re-start on a
different
You have a small typo in your client command, it should be:
bin/zkCli.sh -server 10.16.50.132:2181
(a : not a . prior to the port)
Patrick
chengxiong000 wrote:
Dear zookeepers:
I am a zookeeper user and encount an problem when start zookeeper when start the server and client task .
And
That usually indicates that you have a process that's already using a
port that you've specified zookeeper to use. Try something like:
netstat -a -n -p tcp
and look at 2181/2888/3888 on the hosts you are using. Either use a
different port or stop the process that's currently using.
Patrick
of them. I've tried on a few other ports as well and it's giving me the same issue. Do I need to tamper with port permissions?
-Original Message-
From: Patrick Hunt [mailto:ph...@apache.org]
Sent: Friday, October 23, 2009 1:23 AM
To: zookeeper-user@hadoop.apache.org; Mark Vigeant
Great to hear. Good luck.
Patrick
Mark Vigeant wrote:
Oh awesome! Thanks Patrick!
Everything works now. Thanks a lot everyone for all your help!
-Mark
-Original Message-
From: Patrick Hunt [mailto:ph...@apache.org]
Sent: Friday, October 23, 2009 11:58 AM
To: Mark Vigeant
Cc
Some users have asked what performance one should expect from a ZK
cluster based on number of cores, client load, disk performance, etc...
I put together the following survey that looks at performance on
hardware with varying core counts using a shared 7200RPM disk. Many of
our users don't wish
FYI - in Ted's scenario your application will most likely not be
effected (if that's what you mean by "start a round of consensus").
From your app's perspective your ZK clients will get disconnected from
one server and reconnected to another (as the server it is connected to
is restarted with
I'm really interested to know how ec2 compares wrt disk and network
performance to what I've documented here under the "hardware" section:
http://wiki.apache.org/hadoop/ZooKeeper/ServiceLatencyOverview#Hardware
Is it possible for someone to compare the network and disk performance
(scp, dd, md5
On Mon, Nov 9, 2009 at 4:54 PM, Patrick Hunt wrote:
I'm really interested to know how ec2 compares wrt disk and network
performance to what I've documented here under the "hardware" section:
http://wiki.apache.org/hadoop/ZooKeeper/ServiceLatencyOverview#Hardware
Is it possible
between host bandwidth is generally about what you saw. We have been able
to sustain 20-30MB/s into EC2 to a single node which should be harder than
moving data between nodes. I have heard rumors that others were able to get
double what I got for incoming transfer.
On Mon, Nov 9, 2009 at 9:47 PM, Pat
t, I think that we blamed EC2 for
some of our own GC misconfiguration.
I would not use our configuration here as canonical since we didn't apply a
whole lot of brainpower to this problem.
On Tue, Nov 10, 2009 at 9:29 AM, Patrick Hunt wrote:
Ted, could you provide your configuration info
Ben has been looking at using netty instead of our own direct NIO code.
My understanding is that it supports the encryption/auth between
servers, but I'm not 100% on that -- Ben, can you elaborate?
Patrick
Gustavo Niemeyer wrote:
Dear ZooKeepers,
I'm quite interested in the features related
I couldn't find a JIRA for removing the sys exits so I created one:
https://issues.apache.org/jira/browse/ZOOKEEPER-575
there's also this which seems like it should be easy for someone
who knows osgi container jar format requirements:
https://issues.apache.org/jira/browse/ZOOKEEPER-425
Now we ju
unning the zookeeper
server in OSGi and I'll report back...
David
2009/11/10 Patrick Hunt
I couldn't find a JIRA for removing the sys exits so I created one:
https://issues.apache.org/jira/browse/ZOOKEEPER-575
there's also this which seems like it should be easy for someone
who k
Tomorrow is the BA HUG, if anyone is interested to talk with Mahadev or
I f2f regarding ZooKeeper we'll both be in attendance.
Patrick
--- Begin Message ---
Hi all,
We are one week away from the next Bay Area Hadoop User Group - Yahoo!
Sunnyvale Campus, next Wednesday (Nov 18th) at 6PM
We ha
101 - 200 of 473 matches
Mail list logo
