What RF and CL are you using?
On 2012/10/28, at 13:13, Andrew Bialecki wrote:
> Hey everyone,
>
> I'm trying to simulate what happens when a node goes down to make sure my
> cluster can gracefully handle node failures. For my setup I have a 3 node
> cluster running 1.1.5. I'm then using the
Oh sorry. It's pretty nice to know that.
On 2012/10/12, at 0:18, "B. Todd Burruss" wrote:
> as of 1.0 (CASSANDRA-2034) hints are generated for nodes that timeout.
>
> On Thu, Oct 11, 2012 at 3:55 AM, Watanabe Maki
> wrote:
>> Even if HH works fine, HH will
Perl never die... As Fortran
From iPhone
On 2012/10/12, at 3:09, Michael Kjellman wrote:
> Hi- A few months back I wrote a Perl client for Cassandra and I realized I
> never sent it out to this list. While I realize that while Perl is not the
> language du jour hopefully this will help someo
Even if HH works fine, HH will not be created until the failure detector marks
the node is dead.
HH will not be created for partially timeouted mutation request ( but meets CL
) also... In my understanding...
On 2012/10/11, at 5:55, Rob Coli wrote:
> On Tue, Oct 9, 2012 at 12:56 PM, Oleg Du
How much memory do you have on the box?
It seems you need more memory.
maki
On 2012/05/08, at 1:29, cyril auburtin wrote:
> well I uncommented lines96&97 in cassandra-env.sh, with lower values
>
> MAX_HEAP_SIZE="500M"
> HEAP_NEWSIZE="100M"
>
> seems to fix, it
>
> 2012/5/7 cyril auburtin
>
You should run repair. If the disk space is the problem, try to cleanup and
major compact before repair.
You can limit the streaming data by running repair for each column family
separately.
maki
On 2012/04/28, at 23:47, Raj N wrote:
> I have a 6 node cassandra cluster DC1=3, DC2=3 with 60 G
You can get some idea from reading
org.apache.cassandra.thrift.CassandraServer.java, but I wonder what kind of use
case will justify such effort.
From iPhone
On 2012/04/19, at 18:17, Tarun Gupta wrote:
> Hi,
>
> I am interesting in knowing what is the best way to create my Cassandra
> Clie
You need to install cql driver for python as it says.
% easy_install cql
If you don't have easy_install, you need to install it first. You will be able
to find easy_install by querying "easy_install python" on google.
maki
On 2012/04/17, at 20:18, Tamar Fraenkel wrote:
> Thanks for answering!
If you increase sstable_size_in_mb to 200MB, you will need more IO for each
compaction. For example, if your memtable will be flushed, and LCS needs to
compact it with 10 overwrapped L1 sstables, you will need almost 2GB read and
2GB write for the single compaction.
From iPhone
On 2012/04/11,
auto_bootstrap parameter has been removed and always enabled since 1.0.
maki
On 2012/04/12, at 6:10, Paolo Bernardi wrote:
> I think that setting auto_bootstrap = true or false into cassandra.yaml is
> enough (if it isn't there already just add it, for example, after
> initial_token)
>
> Pa
Plz refer to the following thread:
http://cassandra-user-incubator-apache-org.3065146.n2.nabble.com/cleanup-crashing-with-quot-java-util-concurrent-ExecutionException-java-lang-ArrayIndexOutOfBoundsEx-td7371682.html
maki
From iPhone
On 2012/04/10, at 17:21, Radim Kolar wrote:
> what is metho
1.0.9 is a maintenance release, so it's basically bug fixes with some minor
improvements.
If you plan to use LeveledCompaction, you should better to use 1.0.9+ or 1.1.0+.
maki
On 2012/04/07, at 6:49, Tim Dunphy wrote:
> Hello list,
>
>
> I just noticed that cassandra 1.0.9 was released. Wh
ow about update requests?
>
>
> Thanks,
>
>
> Zhiming
>
> On Thu, Apr 5, 2012 at 12:33 AM, Watanabe Maki
> wrote:
> I assume you are talking about nodes, rather than replicas.
> The data distribution over ring depends on Partitioner and Replica placemen
I assume you are talking about nodes, rather than replicas.
The data distribution over ring depends on Partitioner and Replica placement
strategy you use.
If you are using Random Partitioner and Simple Strategy, your data will be
automatically distributed over the nodes in the ring.
maki
On 20
LeveledCompaction will use less disk space(load), but need more IO.
If your traffic is too high for your disk, you will have many pending
compaction tasks, and large number of sstables which wait to be compacted.
Also the default sstable_size_in_mb (5MB) will be too small for large data
set. You
> - InternalResponseStage
Handles response to non client initiated messages, including bootstrap, schema
check, etc.
maki
On 2012/03/26, at 2:18, aaron morton wrote:
> Work is broken up into a series of stages.
>
> - ReadStage - performing a local read.
> - RequestResponseStage - handling
ight?
>
>
> Tamar Fraenkel
> Senior Software Engineer, TOK Media
>
>
>
> ta...@tok-media.com
> Tel: +972 2 6409736
> Mob: +972 54 8356490
> Fax: +972 2 5612956
>
>
>
>
>
> On Sun, Mar 18, 2012 at 9:57 AM, Watanabe Maki
> wrote:
&
Because your RF is 1, so you need all nodes up.
maki
On 2012/03/18, at 16:15, Tamar Fraenkel wrote:
> Hi!
> I have a 3 node cassandra cluster.
> I use Hector API.
>
> I give hecotr one of the node's IP address
> I call setAutoDiscoverHosts(true) and setRunAutoDiscoveryAtStartup(true).
>
> Th
How about to fill zeros before smaller digits?
Ex. 0001, 0002, etc
maki
On 2012/03/17, at 6:29, A J wrote:
> If I define my rowkeys to be Integer
> (key_validation_class=IntegerType) , how can I order the rows
> numerically ?
> ByteOrderedPartitioner orders lexically and retrieval usin
al Operations | p. +45 3231 6008 | x86008 | cell. +46 735 101
> 444 | jelme...@adobe.com
>
> On Thu, 2012-03-15 at 17:00 -0700, Watanabe Maki wrote:
>>
>> update column family with LCS option + upgradesstables should convert all of
>> your sstables.
>> Set lig4j
update column family with LCS option + upgradesstables should convert all of
your sstables.
Set lig4j config:
org.apache.cassandra.db.compaction=DEBUG
in conf/log4j-server.properties and retry your procedure to find what is happen.
maki
On 2012/03/16, at 7:05, Johan Elmerfjord wrote:
> Hi, I'
Typo
> 2. MKLINK /D LINK.TXT ORIG.TXT
MKLINK /H
On 2012/03/15, at 9:04, Watanabe Maki wrote:
> It's true on unix, but you can't delete hard links of opened files on windows.
> Try following:
> 1 Create a text file ORIG.TXT
> 2. MKLINK /D LINK.TXT ORIG.TXT
> Now
It's true on unix, but you can't delete hard links of opened files on windows.
Try following:
1 Create a text file ORIG.TXT
2. MKLINK /D LINK.TXT ORIG.TXT
Now you have hard link LINK.TXT .
3 Open ORIG.TXT with MS Word.
4. DEL LINK.TXT
It returns error.
Interesting thing is that you can del
How about to truncate HintsColumnFamily and then execute nodetool repair as
work around?
On 2012/03/02, at 18:11, Radim Kolar wrote:
> Dne 2.3.2012 9:49, Maki Watanabe napsal(a):
>> Fixed in 1.0?
>> https://issues.apache.org/jira/browse/CASSANDRA-3176
> that patch test if sstable is empty bef
I haven't use the packaged kit, but Cassandra uses half of physical memory on
your system by default.
You need to edit cassandra-env.sh to decrease heap size.
Update MAX_HEAP_SIZE and NEW_HEAP_SIZE and restart.
From iPhone
On 2012/02/18, at 20:40, Aditya Gupta wrote:
> I just installed Cassan
In general, write is faster than read in Cassandra.
maki
On 2011/12/13, at 13:54, Waqar Azeem wrote:
> Hi,
>
> 'threads' are nested in a 'forum', therefore, I decided to create a
> column-family 'thread' with a column named 'parent'.
>
>
> Is this idea matched with Cassandra philosophy? Be
It is expected behaviour and not relate on number of node.
After the failed node bringing back, the ring will be busy by Hinted Handoff
rewriting and Read Repair. If you run repair, all your 3 nodes need to build
Merkel Tree, compare the hash values, then transfer latest data to each other.
You
Repair doesn't compact. Those are different processes already.
maki
On 2011/07/01, at 7:21, A J wrote:
> Thanks all !
> In other words, I think it is safe to say that a node as a whole can
> be made consistent only on 'nodetool repair'.
>
> Has there been enough interest in providing anti-ent
iptables?
From iPhone
On 2011/06/25, at 12:47, Yang wrote:
> Thanks Jonathan.
>
> this provides a way to essentially get a copy of the outgoing messages,
> the messages onto the real connections still go through, but I would need a
> way
> to shut off the real connections too.
>
> shutting
I once built a 4 node ring on my laptop, with 64MB heap for each instances.
I could write and read on it, but nodetool repair caused OOM.
You should test essential operations with estimated data loaded, under expected
traffic.
Btw I'm using 96MBx4 node ring on my laptop now just for my private l
You can know endpoints which cassandra will store your key to with
getNaturalEndpoints, but you can't specify endpoint you want to use with this
API.
Partitioner decides which key will go to which node. With OPP, you may be able
to predicate which key range will be stored to a node, so you can c
It may not what you want, but please read about Network Topology Strategy and
DC_QUORUM.
http://www.datastax.com/dev/blog/deploying-cassandra-across-multiple-data-centers
You can configure your Cassandra "Data Center aware" . Your read and write will
be resolved in DC local, but will be replica
You can write if you use CL=QUORUM, but can't write with CL=ALL.
maki
On 2011/05/31, at 21:03, Flavio Baronti wrote:
> Suppose i have a cluster with 10 nodes and RF=5.
> Will every write succeed if one or two of my nodes are down, and I use
> ConsistencyLevel=ALL? Or will some of the writes f
I think I don't get your situation yet, but if you use RF=2, CL=QUORUM is
identical with CL=ALL.
Does it explain your experience?
maki
On 2011/05/25, at 19:39, Timo Nentwig wrote:
> Hi!
>
> 5 nodes, replication factor of 2, fifth node down.
>
> As long as I write a single column with hector
Cassandra is not a RDBMS. Only you can do is search on a key, or you need full
scan.
You need to design your schema carefully as your application needs.
On 2011/05/20, at 1:11, karim abbouh wrote:
> i'm new using cassandra database,
> i want to get data as in relationnel database:
> select * f
fferent
> volume. Currently if the data volume gets corrupted, the snapshots on it can
> also get corrupted.
>
> On Wed, May 18, 2011 at 5:44 PM, Watanabe Maki
> wrote:
> SSTables are immutable. Those won't changed once written to disk.
>
> From iPhone
>
>
SSTables are immutable. Those won't changed once written to disk.
From iPhone
On 2011/05/19, at 9:37, Sameer Farooqui wrote:
> As of 0.8.0, is it possible to take a Cassandra snapshot to a different
> volume (like a EBS volume dedicated for backups)?
>
> About a year ago, Jonathan Ellis said
It's interesting topic for me too.
How about to add measurement on static disk utilization (% used) and memory
utilization ( rss, JVM heap, JVM GC )?
maki
From iPhone
On 2011/05/12, at 0:49, Tomer B wrote:
> Hi
>
> I'm trying to predict when my cluster would soon be needing new nodes
> adde
How about to download the binary kit manually?
http://cassandra.apache.org/
From iPhone
On 2011/04/26, at 18:55, Luke Biddell wrote:
> Chaps,
>
> We're using 0.7.4 here and aren't ready to go to 0.8 just yet. If I do
> apt-get install cassandra=0.7.4 on a clean machine it appears to be
> un
Thanks!
From iPhone
On 2011/04/23, at 9:31, Jonathan Ellis wrote:
> No -- see SystemTable.incrementAndGetGeneration.
>
> On Fri, Apr 22, 2011 at 10:38 AM, Maki Watanabe
> wrote:
>> Hello,
>> I found Gossipper is initiated with seconds from Unix Epoch
>> (=System.currentTimeMillis() / 1000)
8081 is your mx4j port, isn't it? You need to connect jconsole to JMX_PORT
specified in cassandra-env.sh.
maki
From iPhone
On 2011/04/16, at 13:56, tinhuty he wrote:
> Maki, thanks for your reply. for the second question, I wasn't using the
> loopback address, I was using the actually IP ad
3.8E-4 means 3.8 * 10^-4, = 0.00038 = 0.038%, I think.
So your program must be using random enough keys against the key cache size.
maki
From iPhone
On 2011/04/16, at 15:17, mcasandra wrote:
> You mean read it like .00038880248833592535E? I didn't quite follow why? If
> it is 3.888024883359
Thanks a lot. It has became clear for me.
From iPhone
On 2011/04/06, at 23:51, Sylvain Lebresne wrote:
> On Tue, Apr 5, 2011 at 9:03 PM, Maki Watanabe wrote:
>> Thanks Sylvain, it's very clear.
>> But should I still need to force major compaction regularly to clear
>> tombstones?
>> I know t
With RF=2 & CL=ONE, take care on that you still have chance to read old data
which is not replicated yet.
Maki
From iPhone
On 2011/03/26, at 5:10, Jared Laprise wrote:
> No, what initially started it all was that I needed to increase my EC2 server
> instance size. So I removed said server f
44 matches
Mail list logo
