SSTables are immutable - once they're written to disk, they cannot be
changed.
On read C* checks *all* SSTables [1], but to make it faster, it uses
Bloom Filters, that can tell you if a row is *not* in a specific
SSTable, so you don't have to read it at all. However, *if* you read it
in case
I don't know Hector very much really, but I highly suspect that
ts.toString() is wrong, since composite column names are not strings. So
again, not a Hector expert, but I can't really see how converting the
composite into string could work.
--
Sylvain
On Wed, Jul 17, 2013 at 11:14 PM, Rahul
@Michal : look a this for the improvement of read performance :
https://issues.apache.org/jira/browse/CASSANDRA-2498
Best regards.
Jean Armel
2013/7/18 Michał Michalski mich...@opera.com
SSTables are immutable - once they're written to disk, they cannot be
changed.
On read C* checks *all*
Thanks! :-)
M.
W dniu 18.07.2013 08:42, Jean-Armel Luce pisze:
@Michal : look a this for the improvement of read performance :
https://issues.apache.org/jira/browse/CASSANDRA-2498
Best regards.
Jean Armel
2013/7/18 Michał Michalski mich...@opera.com
SSTables are immutable - once they're
+1 for Sylvain's answer.
This normally happens, if validation class for column value(s) differs.
-Vivek
On Thu, Jul 18, 2013 at 12:08 PM, Sylvain Lebresne sylv...@datastax.comwrote:
I don't know Hector very much really, but I highly suspect that
ts.toString() is wrong, since composite
Here are some posts about CQL and Thrift
http://thelastpickle.com/2013/01/11/primary-keys-in-cql/
http://www.datastax.com/dev/blog/cql3-for-cassandra-experts
http://www.datastax.com/dev/blog/thrift-to-cql3
Hope that helps.
-
Aaron Morton
Cassandra Consultant
New Zealand
IMHO you will want 4 cores and 4 to 8 GB for each VM to run both Cassandra and
Hadoop on the nodes.
For comparison people often use an EC2 m1.xlarge which has 4 cores and 16GB.
Also, I recommend anyone starting experiments with Cassandra and Hadoop use
DataStax Enterprise. So you can focus
This webinar I did a few months ago goes through the read and write path
http://www.youtube.com/watch?v=zFCjekgK7ZY
I get to that about 29 minutes in.
slides
http://www.slideshare.net/aaronmorton/cassandra-community-webinar-introduction-to-apache-cassandra-12-20353118
Cheers
Double check the stack size is to set 100K see
https://github.com/apache/cassandra/blob/cassandra-1.1/conf/cassandra-env.sh#L187
Sorry that was a late night typo, should have been 180K like in the link.
Cheers
-
Aaron Morton
Cassandra Consultant
New Zealand
@aaronmorton
Does this help ?
http://www.mail-archive.com/user@cassandra.apache.org/msg30973.html
Can you pull the data off the node so you can test it somewhere safe ?
Cheers
-
Aaron Morton
Cassandra Consultant
New Zealand
@aaronmorton
http://www.thelastpickle.com
On 18/07/2013, at 2:20
I saw that msg in the thread, I pulled the git files and it looks
like a suite of tools, do I install them on their own ? do I replace the
current ones ? its production data but I can copy the data to where
I want and experiment.
Jim
From: aaron morton
I'm not an expert, still learning C* but can tell something about your
questions.
1) You have to understand that CQL row is not the same as row that C* uses
to store data and which is accessible through the Trift interface. Primary
key in terms of CQL is not the same as Row key.
2) You have to
hi A:
Thank you for responding to my e-mail.
Sorry if i did not express my questions/doubts well.
I try to use the slice feature with CassandraStorage LOAD but I do not
know how to do it. I cannot find any detailed documentation about it.
I found only the references in my last mail.
Yet i
Thank you Aaron, your advice about a newer client it is really
interesting. We will take in account it!
Here, some numbers about our tests: we found that more or less that with
more than 500k elements (multiplying rows and columns requested) there was
the inflection point, and so asking for more
Thank You. Your solution worked...!!
Changed below line:
HCounterColumnString hColumn_ts = HFactory.createCounterColumn(ts.toString(),
value, StringSerializer.get());
To this:
HCounterColumnComposite hColumn_tsCG = HFactory.createCounterColumn(tsCG,
value, CompositeSerializer.get());
Rahul
Hi
I have an issue while inserting a composite column name to one of the Cassandra
column families. Below is a detailed description of what I had done and stuck
up at.
Please let me know where I had went wrong.
Requirement:
--
Rowkey- RowIdString
Many Thanks Aaron!!
As I work more with CQL and CLI as some other posting I have seen regarding
usage I am thinking that CLI for keyspace and Column Family setup and
maintenance is best
while CQL for queries/inserts ect is best. Mainly I am thinking this because
of better control over the
Many thanks Vladimir I am starting to see what you are talking about.
Yeah all I want to do is a simple Order By via SQL but having the column family
setup using CLI to support that is a bit of a challenge for me at the moment
since the two are at different levels but I prefer CLI for column
Looks like validation class for composite column value is different than
UTF8Type? Though code suggests it is:
composite.addComponent(TEXT1, StringSerializer.get());
Please validate.
-Vivek
On Thu, Jul 18, 2013 at 7:41 PM, anand_balara...@homedepot.com wrote:
Hi
I have an issue
I had been using the StringSerilaizer.get() for all UTF8Type fields so far. Do
not think I need to check the code.
Do you suspect the column family definition?
-Anand
From: Vivek Mishra [mailto:mishra.v...@gmail.com]
Sent: Thursday, July 18, 2013 11:29 AM
To: user@cassandra.apache.org
Subject:
Hello together,
today I experienced a problem while loading a snapshot from our
cassandra cluster to test cluster. The cluster has six nodes and I took
a snapshot from all nodes concurrently and tried to import them in the
other cluster.
From 5 out of 6 nodes importing went well with no
sstable might be corrupted due to bad disk. In that case, replication does
not matter.
On Thu, Jul 18, 2013 at 8:52 AM, Jan Kesten j.kes...@enercast.de wrote:
Hello together,
today I experienced a problem while loading a snapshot from our cassandra
cluster to test cluster. The cluster has
On Wed, Jul 17, 2013 at 7:23 PM, Langston, Jim
jim.langs...@compuware.comwrote:
As a follow up – I did upgrade the cluster to 1.2.6 and that
did take care of the issue. The upgrade went very smoothly,
the longest part was being thorough on the configuration
files, but I was able to able to
On Thu, Jul 18, 2013 at 8:12 AM, Tony Anecito adanec...@yahoo.com wrote:
As I work more with CQL and CLI as some other posting I have seen
regarding usage I am thinking that CLI for keyspace and Column Family setup
and maintenance is best
while CQL for queries/inserts ect is best. Mainly I
Hi,
I think it might be corrupted due to a poweroutage. Apart from this issue
reading the data with consistency level quorum (I have three replicas) did not
issue an error - only the import to a different cluster.
So, if I import all nodes except the one with the corrupted sstable - shoudn't
On Thu, Jul 18, 2013 at 10:17 AM, Jan Kesten j.kes...@enercast.de wrote:
I think it might be corrupted due to a poweroutage. Apart from this issue
reading the data with consistency level quorum (I have three replicas) did
not issue an error - only the import to a different cluster.
So, if I
Yes. can you please share output of describe keyspace which contains
CompositeColumnNameTest
What is the datatype for column values?
-Vive
On Thu, Jul 18, 2013 at 9:17 PM, anand_balara...@homedepot.com wrote:
I had been using the StringSerilaizer.get() for all UTF8Type fields so
far. Do
Output of describe command is:
[default@Test] describe CompositeColumnNameTest;
ColumnFamily: CompositeColumnNameTest
Key Validation Class: org.apache.cassandra.db.marshal.UTF8Type
Default column value validator: org.apache.cassandra.db.marshal.UTF8Type
Columns sorted by:
I'm seeing quite a few of these on pretty much all of the nodes of my 1.2.6
cluster. Is this something I should be worried about? If so, do I need to run
upgradesstables or run a scrub?
ERROR [CompactionExecutor:4] 2013-07-18 18:49:02,609 CassandraDaemon.java (line
192) Exception in thread
I have been looking at the stuff in the zip file, and also the
sstablesplit command script. This script is looking for a java
class StandaloneSplitter located in the package org.apache.cassandra.tools.
Where is this package located ? I looked in the lib directory but nothing
contains
the class.
https://github.com/pcmanus/cassandra/tree/sstable_split/src/java/org/apache/cassandra/tools
You'll have to clone Sylvain's 'sstable_split' branch and build from there.
(Commiter folks: this is helpful. @Sylvain - can you commit a patch
under this ticket (or wherever):
Thanks, was heading down that path .. after the build it
creates a 1.1.6 cassandra snapshot, I'm currently on 1.2.6 - will I
be able to use the tool ?
Jim
On 7/18/13 3:45 PM, Nate McCall zznat...@gmail.com wrote:
https://github.com/pcmanus/cassandra/tree/sstable_split/src/java/org/apach
Hi,
Is there a recommended data size for Reads/Writes in Cassandra? I tried
inserting 10 MB objects and the latency I got was pretty high. Also, I was
never able to insert larger objects (say 50 MB) since Cassandra kept
crashing when I tried that.
Here is my experiment setup:
I used two Large
Hi - We are planning on using Cassandra for an IMAP based implementation.
There are some questions that we are stuck with -
1) Each user will have a pre-defined mailbox size (say 10 MB). We need to
maintain a field to check if the mail-box size exceeds the predefined size.
Will using
there is a limit of thrift message ( thrift_max_message_length_in_mb), by
default it is 64m if I'm not mistaken. This is your limit.
On Thu, Jul 18, 2013 at 2:03 PM, hajjat haj...@purdue.edu wrote:
Hi,
Is there a recommended data size for Reads/Writes in Cassandra? I tried
inserting 10 MB
Without digging I'd so no - the SSTable versions will be pretty different.
You could test this pretty easily in isolation though just on a local
instance - I think the issue will be 1.1.6 reading the 1.2.6 SSTable
as the other way should be backwards compatible. Someone jump in if
i'm wrong?
On
The default limit is 16mb, but realistically you should try to keep writes
under 10mb, breaking up large values into multiple columns/rows if
necessary.
On Thu, Jul 18, 2013 at 4:31 PM, Andrey Ilinykh ailin...@gmail.com wrote:
there is a limit of thrift message (
Conter wont be updated when the old data is TTLed. I am not sure whether
you can use triggers https://issues.apache.org/jira/browse/CASSANDRA-1311
On Thu, Jul 18, 2013 at 2:30 PM, Kanwar Sangha kan...@mavenir.com wrote:
Hi - We are planning on using Cassandra for an IMAP based
Thanks Andrey and Tyler! That was useful :)
Do you guys have any idea why the 10 MB writes took a lot of time in my
case although I'm using Large VMs which have plenty of resources? Or do you
think this latency is expected?
I'm trying to see how much time is spent in the network versus processing
Large writes can sometimes put a lot of heap/GC pressure on the node, which
can be an additional source of latency. Use the query tracing in Cassandra
1.2+ to get a better picture of where the latency is.
On Thu, Jul 18, 2013 at 6:18 PM, Mohammad Hajjat haj...@purdue.edu wrote:
Thanks Andrey
Hi Rob,
Thanks for the feedback. I had heard about this in regards to CQL created table
not being visible to CLI but have not seen any examples of setting up CQL
table to be visible by CLI.
Best Regards,
-Tony
From: Robert Coli rc...@eventbrite.com
To:
Hi all,
I'm experiencing some problems after 3 years of cassandra in production (from
0.6 to 1.0.6) -- for 2 times in 3 weeks 2 nodes crashed with OutOfMemory
Exception.
In the log I can read the warn about the few heap available ... now I'm
increasing a little bit my RAM, my Java Heap (1/4 of
42 matches
Mail list logo
