In this¹ debate, there seemed to be consensus on the following fact:
"In Cassandra, say you use N=3, W=3 & R=1. Let’s say you managed to
only write to replicas A & B, but not C. In this case Cassandra will
return an error to the application saying the write failed- which is
acceptable given than W
At Wed, 07 Apr 2010 13:19:26 -0700,
Mike Gallamore wrote:
>
> I have writes to cassandra that are failing, or at least a read shortly
> after a write is still getting an old value. I realize Cassandra is
> "eventually consistent" but this system is a single CPU single node with
> consistency le
Hi,
I am newbie. Downloaded Fauna and followed the instructions to run Fauna in
Centos.
But when i try to run cassandra_helper cassandra, i get the following
Error.
Can anyone help me solve this.
i have installed
1)Java
java version "1.6.0"
OpenJDK Runtime Environment (build 1.6.0-b09)
OpenJDK
On Thu, 08 Apr 2010 12:53:48 +0100 Philip Jackson
wrote:
PJ> At Wed, 07 Apr 2010 13:19:26 -0700,
PJ> Mike Gallamore wrote:
>>
>> I have writes to cassandra that are failing, or at least a read shortly
>> after a write is still getting an old value. I realize Cassandra is
>> "eventually consi
On Wed, 07 Apr 2010 13:19:26 -0700 Mike Gallamore
wrote:
MG> As an aside I motified some other code to use Net::Cassandra instead
MG> of Net::Cassandra::Easy and noticed that it seems to run 3-4X
MG> slower. Both aren't stunningly fast. The test clients are running on
MG> the same machine as Ca
On Thu, Apr 8, 2010 at 02:55, Paul Prescod wrote:
> In this¹ debate, there seemed to be consensus on the following fact:
>
> "In Cassandra, say you use N=3, W=3 & R=1. Let’s say you managed to
> only write to replicas A & B, but not C. In this case Cassandra will
> return an error to the applicati
On 7 April 2010 19:13, Jonathan Ellis wrote:
> One thing you can do is manually "randomize" keys for any CFs that
> don't need the OP by pre-pending their md5 to the key you send
> Cassandra. (This is all RP is doing under the hood anyway.)
>
Another possibility is to prepend some hash of somet
I don't see any way to increase the # of active Deserializers in
storage-conf.xml
Tpstats more than 8 hours after insert/read stop
Pool NameActive Pending Completed
FILEUTILS-DELETE-POOL 0 0227
STREAM-STAGE 0
Have you checked iostat -x ?
On Thu, Apr 8, 2010 at 9:45 AM, Mark Jones wrote:
> I don't see any way to increase the # of active Deserializers in
> storage-conf.xml
>
> Tpstats more than 8 hours after insert/read stop
>
> Pool Name Active Pending Completed
> FILEUTILS-D
The sawtooth ram graph is typical of normal GC activity, btw -- the GC
doesn't bother with a major collection until it reaches some percent
of the total available.
The tooth wave in memory utilization could be memtable dumps. I/O wait in
TCP happens when you are overwhelming the server with requests. Could you
run sar and find out how many bytes/sec you are receiving/transmitting?
Cheers
Avinash
On Thu, Apr 8, 2010 at 7:45 AM, Mark Jones wrote:
> I don't
Re "Why couldn't you walk though random-layout hash-keyed data by
token the same way? The hashes still have an order."
You can in 0.6.
On Wed, Apr 7, 2010 at 8:45 AM, Paul Prescod wrote:
> I'm working on a blog post that combines all of the information and
> ideas I can find relative to managing
Your first step should be upgrading to 0.6.
On Wed, Apr 7, 2010 at 10:38 AM, Philip Jackson wrote:
> At Wed, 7 Apr 2010 17:29:49 +0200,
> Sylvain Lebresne wrote:
>>
>> Use ConsistencyLevel.QUORUM when you write *and* when you read.
>
> I already do (plus, I only test with one node).
>
> BTW, I'm
On Thu, Apr 8, 2010 at 7:16 AM, Gary Dusbabek wrote:
> On Thu, Apr 8, 2010 at 02:55, Paul Prescod wrote:
>> In this¹ debate, there seemed to be consensus on the following fact:
>>
>> "In Cassandra, say you use N=3, W=3 & R=1. Let’s say you managed to
>> only write to replicas A & B, but not C. In
What your describing is a distributed transaction? Generally strong
consistency is always associated with doing transactional writes where you
never see the results of a failed write on a subsequent read no matter what
happens. Cassandra has no notion of rollback. That is why no combination
will gi
I restarted Cassandra on that node to clear out that queue, reduced the
available memory to java to 4GB, and now I'm able to read with 8 concurrent
threads, about 110/second
Running iostat -x I see a large amount of time in await, and a small amount of
time in svctm indicating the device is res
On 04/08/2010 04:53 AM, Philip Jackson wrote:
At Wed, 07 Apr 2010 13:19:26 -0700,
Mike Gallamore wrote:
I have writes to cassandra that are failing, or at least a read shortly
after a write is still getting an old value. I realize Cassandra is
"eventually consistent" but this system is a sin
is N:C:E possibly ignoring thrift exceptions?
On Thu, Apr 8, 2010 at 10:45 AM, Mike Gallamore
wrote:
> On 04/08/2010 04:53 AM, Philip Jackson wrote:
>>
>> At Wed, 07 Apr 2010 13:19:26 -0700,
>> Mike Gallamore wrote:
>>
>>>
>>> I have writes to cassandra that are failing, or at least a read shortl
On Wed, Apr 7, 2010 at 1:51 PM, Eric Evans wrote:
> On Tue, 2010-04-06 at 10:55 -0700, Tatu Saloranta wrote:
>> On Tue, Apr 6, 2010 at 12:15 AM, JKnight JKnight
>> wrote:
>> > When import, all data in json file will load in memory. So that, you
>> can not
>> > import large data.
>> > You need to
I stopped writing to the cluster more than 8 hours ago, at worst case, I could
only be getting a periodic memtable dump (I think)
Running 16 QUORUM read threads getting 600 records/second
Sar for all 3 nodes (collected almost simultaneously:
Average:CPU %user %nice %system %i
Can you keep this to one thread please? It is hard to follow when the
subject keeps changing.
On Thu, Apr 8, 2010 at 12:55 AM, Paul Prescod wrote:
>
> ¹ http://jsensarma.com/blog/2009/11/dynamo-part-i-a-followup-and-re-rebuttals/
>
Pay no attention to this disingenuous troll.
b
So unless you re-try the write, the previous stale write stays on the other
two nodes? Would a read repair fix this eventually?
On Thu, Apr 8, 2010 at 11:36 AM, Avinash Lakshman <
avinash.laksh...@gmail.com> wrote:
> What your describing is a distributed transaction? Generally strong
> consistenc
Yes. Or you would retry the write. Either way, the system achieves
consistency eventually, hence the name.
On Thu, Apr 8, 2010 at 9:36 AM, Mark Greene wrote:
> So unless you re-try the write, the previous stale write stays on the other
> two nodes? Would a read repair fix this eventually?
>
His arguments consistently (hah!) boil down to this: if you
misconfigure things for your intended application, you get undesirable
behavior. For example, the correct approach to the situation cited is
to use quorum reads and writes. W=3/R=1/N=3 might be appropriate for
situations in which you wan
A read repair will fix it immediately after the first read of the row.
On 2010-04-08 16:36, Mark Greene wrote:
> So unless you re-try the write, the previous stale write stays on the
> other two nodes? Would a read repair fix this eventually?
>
> On Thu, Apr 8, 2010 at 11:36 AM, Avinash Lakshman
cassandra_helper does a bunch of magic to set things up. looks like
the "extract a private copy of cassandra 0.6 beta2" part of the magic
is failing. you'll probably need to manually attempt the un-tar to
figure out why it is bailing.
On Thu, Apr 8, 2010 at 7:30 AM, Nirmala Agadgar wrote:
> Hi,
On Thu, Apr 8, 2010 at 9:49 AM, Jonathan Ellis wrote:
> cassandra_helper does a bunch of magic to set things up. looks like
> the "extract a private copy of cassandra 0.6 beta2" part of the magic
> is failing. you'll probably need to manually attempt the un-tar to
> figure out why it is bailing.
Retry is the best option. Because the read repair will fix it on a
subsequent read and it will actually fix it with a value that was actually
deemed a failed write to the client.
Avinash
On Thu, Apr 8, 2010 at 9:47 AM, David Strauss wrote:
> A read repair will fix it immediately after the first
Sounds like it's worth reporting on the github project then.
On Thu, Apr 8, 2010 at 11:53 AM, Paul Prescod wrote:
> On Thu, Apr 8, 2010 at 9:49 AM, Jonathan Ellis wrote:
>> cassandra_helper does a bunch of magic to set things up. looks like
>> the "extract a private copy of cassandra 0.6 beta2"
Yeah, this is a known issue, we're working on it today.
-ryan
On Thu, Apr 8, 2010 at 10:31 AM, Jonathan Ellis wrote:
> Sounds like it's worth reporting on the github project then.
>
> On Thu, Apr 8, 2010 at 11:53 AM, Paul Prescod wrote:
>> On Thu, Apr 8, 2010 at 9:49 AM, Jonathan Ellis wrote:
Sorry, I've read through the docs (although not too recently) and have
followed this mailing list for a bit. But I haven't seen how it's possible
to iterate with RP? Could you kindly point out to me where it shows how to
do this? TIA.
On Wed, Apr 7, 2010 at 4:37 PM, Benjamin Black wrote:
> It wa
On Thu, Apr 8, 2010 at 1:26 PM, Peter Chang wrote:
> Sorry, I've read through the docs (although not too recently) and have
> followed this mailing list for a bit. But I haven't seen how it's possible
> to iterate with RP? Could you kindly point out to me where it shows how to
> do this? TIA.
T
On 04/07/2010 01:31 PM, Eric Evans wrote:
On Wed, 2010-04-07 at 13:19 -0700, Mike Gallamore wrote:
I have writes to cassandra that are failing, or at least a read
shortly after a write is still getting an old value. I realize
Cassandra is "eventually consistent" but this system is a single C
On 04/08/2010 04:53 AM, Philip Jackson wrote:
At Wed, 07 Apr 2010 13:19:26 -0700,
Mike Gallamore wrote:
I have writes to cassandra that are failing, or at least a read shortly
after a write is still getting an old value. I realize Cassandra is
"eventually consistent" but this system is a sin
I have two boxes. One is a windows box running Cassandra .6, and the
other is an ubuntu box from which I'm trying to run the word count
program as in the readme.
The windows box seed is set to 127.0.0.1, and listen address to localhost.
The ubuntu box seed & listen is point to IP of the windows
On Thu, Apr 8, 2010 at 12:41 PM, Mike Gallamore
wrote:
> Hello. If you are doing exactly the same thing as N::C::Easy (ie a join on
> the gettimeofday). Then you should have the same problem I found a fix for.
> The problem is that the microseconds value isn't zero padded. So if you are
> at say 2
I'll work on making a benchmark sometime latter. But I don't think that
my changes would be batched. My rows only have one column and for this
test each row is only accessed once (when it is written), I pretty much
directly mapped over from a key value store that was using memcache before.
It
Strange setup, but, ok. What is your setting on the
Windows machine?
On Thu, Apr 8, 2010 at 11:44 AM, Sonny Heer wrote:
> I have two boxes. One is a windows box running Cassandra .6, and the
> other is an ubuntu box from which I'm trying to run the word count
> program as in the readme.
>
> Th
Are you actually trying to make the Ubuntu system another node in the
ring? While the first node is only listening on localhost? There's
your problem.
On Thu, Apr 8, 2010 at 11:44 AM, Sonny Heer wrote:
> I have two boxes. One is a windows box running Cassandra .6, and the
> other is an ubuntu
Yeah I agree it's strange, Windows is just my local box, and I'm
testing before setting up actual boxes :)
the thrift address is 'localhost' on the windows box.
On Thu, Apr 8, 2010 at 11:53 AM, Benjamin Black wrote:
> Strange setup, but, ok. What is your setting on the
> Windows machine?
>
> O
Single node cluster (the windows box). the Ubuntu box is only used to
run the word count
On Thu, Apr 8, 2010 at 11:54 AM, Benjamin Black wrote:
> Are you actually trying to make the Ubuntu system another node in the
> ring? While the first node is only listening on localhost? There's
> your pr
You are telling Windows to only listen on localhost, which is the
loopback, which is only accessible on the system itself, not from
external machines.
On Thu, Apr 8, 2010 at 11:56 AM, Sonny Heer wrote:
> Yeah I agree it's strange, Windows is just my local box, and I'm
> testing before setting up
On 04/08/2010 05:53 AM, Ted Zlatanov wrote:
On Thu, 08 Apr 2010 12:53:48 +0100 Philip Jackson
wrote:
PJ> At Wed, 07 Apr 2010 13:19:26 -0700,
PJ> Mike Gallamore wrote:
I have writes to cassandra that are failing, or at least a read shortly
after a write is still getting an old value. I r
"The ubuntu box seed & listen is point to IP of the windows box."
If you are setting storage-conf.xml parameters, you are trying to run
Cassandra on the Ubuntu system. Regardless, your earlier mail saying
you are setting ThriftAddress to localhost on the Windows machine
precludes anything connect
Right. The word_count program has a storage-conf.xml file which I'm
assuming it reads in order to discover the cluster.
I've changed the thrift listen address on the windows box to be the IP
instead. but still the same result. what is proper setup for easily
testing this?
On Thu, Apr 8, 2010 a
On 04/08/2010 11:46 AM, Jeremy Dunck wrote:
On Thu, Apr 8, 2010 at 12:41 PM, Mike Gallamore
wrote:
Hello. If you are doing exactly the same thing as N::C::Easy (ie a join on
the gettimeofday). Then you should have the same problem I found a fix for.
The problem is that the microseconds val
Okay I moved everything to the ubuntu box:
~/dev/cassandra-0.6.0-rc1/contrib/word_count$ bin/word_count_setup
10/04/08 11:15:10 INFO config.DatabaseDescriptor: Auto DiskAccessMode
determined to be standard
10/04/08 11:15:10 WARN config.DatabaseDescriptor: KeysCachedFraction
is deprecated: use Keys
Eagerly reading this post. One line here doesn't make sense to me.
"Out of the box, Cassandra does not support TimeUUIDs for sorting an
OrderPreservingPartitioner."
Does this mean you can't use Time UUIDs when using OPP? Or that the keys
will not have their order preserved? If it's the latter, pe
Is there other documentation on how to setup all the pieces?
Currently I'm simply trying to test the example word_count, but will
likely need to write other map/reduce programs over the cassandra data
set.
For this test I have one box (ubuntu) where i have moved cass .6 rc1
binary , and started
I thought OPP was required for get_range_slices. Is this no longer the case
for 0.6?
On Thu, Apr 8, 2010 at 11:29 AM, Brandon Williams wrote:
> On Thu, Apr 8, 2010 at 1:26 PM, Peter Chang wrote:
>
>> Sorry, I've read through the docs (although not too recently) and have
>> followed this mailing
On Thu, Apr 8, 2010 at 2:38 PM, Peter Chang wrote:
> I thought OPP was required for get_range_slices. Is this no longer the case
> for 0.6?
Right, get_range_slices works with either partitioner.
-Brandon
Missing the commons logging and commons httpclient jars. Must be
using the the wrong jdk?
On Thu, Apr 8, 2010 at 12:38 PM, Sonny Heer wrote:
> Is there other documentation on how to setup all the pieces?
>
> Currently I'm simply trying to test the example word_count, but will
> likely need to wr
Running the word_count example the hadoop job appears to be run internally.
If I have a Cassandra cluster of 10 nodes, how does the Hadoop cluster
get configured?
those aren't shipped with Cassandra.
On Thu, Apr 8, 2010 at 3:00 PM, Sonny Heer wrote:
> Missing the commons logging and commons httpclient jars. Must be
> using the the wrong jdk?
>
> On Thu, Apr 8, 2010 at 12:38 PM, Sonny Heer wrote:
>> Is there other documentation on how to setup all the pie
Code that uses Hadoop will look for mapred-site.xml, core-site.xml,
hdfs-site.xml etc on your CLASSPATH. If you add your Hadoop config directory to
CLASSPATH before running the script, Hadoop will use that configuration to
connect to your cluster.
-Original Message-
From: "Sonny Heer"
On 2010-04-08 18:29, Brandon Williams wrote:
> The method is the same for both partitioners: get_range_slices. The key
> difference is, with RP you don't really have a useful order, it's based
> on the hash of the row key.
That's fine; I don't care about the order. I just want the ability to
walk
Hi All,
I'm brand new to Cassandra and know absolutely nothing, so please forgive me
in advance.
A friend and I have each setup a few Cassandra stand alone nodes, completely
default.
His: Mac OSX Snow Leopard
Mac Book Pro
Intel Duo Core
4GB Ram
5400 rpm disk
Mine: debian 5.x
Where is it configured in the word_count example?
On Thu, Apr 8, 2010 at 1:29 PM, Stu Hood wrote:
> Code that uses Hadoop will look for mapred-site.xml, core-site.xml,
> hdfs-site.xml etc on your CLASSPATH. If you add your Hadoop config directory
> to CLASSPATH before running the script, Hadoop
Yeah I realized that shortly... :)
I'm still not able to point the word_count to a live cluster. If i
have a single node cluster and the thrift address is the IP of that
box, and it has seed value as the IP of itself as well.
How do i run the word_count remotely then? Sorry I must be missing
so
At Thu, 08 Apr 2010 11:41:30 -0700,
Mike Gallamore wrote:
>
> [1 ]
> On 04/08/2010 04:53 AM, Philip Jackson wrote:
> > At Wed, 07 Apr 2010 13:19:26 -0700,
> > Mike Gallamore wrote:
> >
> >> I have writes to cassandra that are failing, or at least a read shortly
> >> after a write is still get
While I wasn't able to reproduce the error, we did have another pop
up. I think I may have actually fixed your problem the other day. Pull
the latest master from fauna/cassandra and you should be good to go.
--
Jeff
On Thu, Apr 8, 2010 at 10:51 AM, Ryan King wrote:
> Yeah, this is a known issue,
On Thu, Apr 8, 2010 at 12:34 PM, Peter Chang wrote:
> Eagerly reading this post. One line here doesn't make sense to me.
> "Out of the box, Cassandra does not support TimeUUIDs for sorting an
> OrderPreservingPartitioner."
> Does this mean you can't use Time UUIDs when using OPP? Or that the keys
Not knowing know anything about the physical layout of the data on disk or how
it is accessed when it is read... Could someone who does help
estimate the worst case scenario(no caching at any level) for the number of
iops to read a row of modest size and modest number of columns in a
large col
64 matches
Mail list logo
