Re: Cassandra 1.0 - is disk seek required to access SSTable metadata

> Bloom filter is stored in RAM, but what about metadata? > Is disk seek required to access it? No, it's loaded in RAM when the sstable is loaded. -- Sylvain

Re: are asynchronous schema updates possible ?

To add to Aaron response, you can update a CF concurrently in 1.1 already. However, you cannot create multiple CF concurrently just yet, but that will be fixed in 1.2. -- Sylvain On Sun, Aug 26, 2012 at 11:04 PM, aaron morton wrote: > Concurrent schema changes are coming in 1.2. > > I could not

java.lang.NoClassDefFoundError when trying to do anything on one CF on one node

Hi I have a single node in a 6 node Cassandra 1.0.11 cluster that seems to have a single column family in a weird state. Repairs, upgradesstables, anything that touches this CF crashes. I've drained the node, removed every file for this CF from said node, removed the commit log, started it up and

Re: Cassandra API Library.

@Brian: you can add the Cassandra::Simple Perl client http://fmgoncalves.github.com/p5-cassandra-simple/ 2012/8/27 Paolo Bernardi > On 08/23/2012 01:40 PM, Thomas Spengler wrote: > >> 4) pelops (Thrift,Java) >> >> >> I've been using Pelops for quite some time with pretty good results; it > felt

Re: Cassandra and Apache Drill

Many queries on small portion of the data….sounds like playORM ;). As long as you partition your data with playOrm, you can do really fast queries into that data by partition using Scalabla SQL (SQL with the addition of a partition clause in front as to what partitions you are querying). Joins

Re: performance is drastically degraded after 0.7.8 --> 1.0.11 upgrade

it was good idea to have a look at StorageProxy :-) 1.0.10 Performance Tests StorageProxy RangeOperations: 546 ReadOperations: 694563 TotalHints: 0 TotalRangeLatencyMicros: 4469484 TotalReadLatencyMicros:245669679 TotalWriteLatencyMicros: 57819722 WriteOperations:208741 0.7.10 Performance Test

Re: find when minor compaction happened?

Tyler, yeah, i found those.thanks for your help. satoshi --- On Tue, 2012/9/4, Tyler Hobbs wrote: There are several logs associated with each minor compaction. Grep your logs for "Compacting". On Mon, Sep 3, 2012 at 7:41 AM, Satoshi Yamada wrote: hi, i know minor compaction is caused when

Re: Cassandra API Library.

You got it. (done) -brian On Tue, Sep 4, 2012 at 7:08 AM, Filipe Gonçalves wrote: > @Brian: you can add the Cassandra::Simple Perl client > http://fmgoncalves.github.com/p5-cassandra-simple/ > > > 2012/8/27 Paolo Bernardi >> >> On 08/23/2012 01:40 PM, Thomas Spengler wrote: >>> >>> 4) pelops (

Re: Cassandra API Library.

check also http://search.cpan.org/dist/perlcassa/ :) On Tue, Sep 4, 2012 at 10:05 AM, Brian O'Neill wrote: > You got it. (done) > > -brian > > On Tue, Sep 4, 2012 at 7:08 AM, Filipe Gonçalves > wrote: > > @Brian: you can add the Cassandra::Simple Perl client > > http://fmgoncalves.github.com/p

Re: Repair has now effect

embarrassing. Chef somehow ran in that box and updated the schema with a version of it that had RF=1. Sorry about that. On Mon, Sep 3, 2012 at 1:45 AM, Radim Kolar wrote: > > INFO [AntiEntropySessions:6] 2012-09-02 15:46:23,022 >> AntiEntropyService.java (line 663) [repair #%s] No neighbors

anyone know how to lookup non-continguous columns BUT for prefixes?

I have a row that is an index like so Index row -> ., ., ., . , ., ., . I would like to get all of the pks for which are pk32 and pk7 And which are pk54 This is a trimmed down example of course. I am thinking maybe I might just use the astyanax async to send out 500 requests instead. T

Re: anyone know how to lookup non-continguous columns BUT for prefixes?

You'd need to make n queries, or do a superset query from min>;-

Re: are asynchronous schema updates possible ?

Is it ok multiple servers will create/update the same CF at once ? I'm looking into dynamic schema update during application deploy/update. вторник, 4 сентября 2012 г. пользователь Sylvain Lebresne писал: > To add to Aaron response, you can update a CF concurrently in 1.1 > already. However, you

Re: are asynchronous schema updates possible ?

On Tue, Sep 4, 2012 at 8:23 PM, Илья Шипицин wrote: > Is it ok multiple servers will create/update the same CF at once ? I'm > looking into dynamic schema update during application deploy/update. As said above, it is ok for update the same CF concurrently in 1.1 but *not* for creation (if you cre

keyspace not propagated to a new node?

I let a new node join the ring, and giving it a new token by -Dinitial_token= both the new node and old nodes on the ring discover each other, that's good. I already defined a Keyspace on the old ring, but this KS is not propagated to the new node. Even after 10 minutes, the new node still does

Re: are asynchronous schema updates possible ?

what kind of problems? "nodes do not agree about schema" exception on later node ? something worse? 2012/9/5 Sylvain Lebresne > On Tue, Sep 4, 2012 at 8:23 PM, Илья Шипицин wrote: > > Is it ok multiple servers will create/update the same CF at once ? I'm > > looking into dynamic schema update

Re: Practical node size limits

I'm following up on this issue, which I've been monitoring for the last several weeks. I thought people might find my observations interesting. Ever since increasing the heap size to 64GB, we've had no OOM conditions that resulted in a JVM termination. Our nodes have around 2.5TB of data each, a

Re: are asynchronous schema updates possible ?

+1 What kinds of problems? Thanks, Dean From: Илья Шипицин mailto:chipits...@gmail.com>> Reply-To: "user@cassandra.apache.org" mailto:user@cassandra.apache.org>> Date: Tuesday, September 4, 2012 1:12 PM To: "user@cassandra.apache.org

Helenos - web based gui tool

Dear all, I'm happy to announce a first release of Helenos. This is a web based gui tool to manage your data stored in Cassandra. Project site: https://github.com/tomekkup/helenos Some screens: https://picasaweb.google.com/tomekkuprowski/Helenos Hope you'll find it usefull. I'll be grateful fo

playOrm now supports N-level joins on cassandra (no limitations on where clause)

There is no = or < limitations. Joins are in beta and currently can only do inner joins at this time….Also, queries return a Cursor so you can page as well and keep the cursor in a web server session if needed for paging. It also looks like joins may be faster with cassandra/playOrm vs. DBMS/h

Re: Order of the cyclic group of hashed partitioners

> I believe the question is why is the maximum 2**127 and not > 0x The maximum is the size of the digest created by MD5. Does that answer the question? - Aaron Morton Freelance Developer @aaronmorton http://www.thelastpickle.com On 3/09/2012, at 8:20 PM, Tim Win

Re: Text searches and free form queries

AFAIk if you want to keep it inside cassandra then DSE, roll your own from scratch or start with https://github.com/tjake/Solandra . Outside of Cassandra I've heard of people using Elastic Search or Solr which I *think* is now faster at updating the index. Hope that helps. ---

Re: configure KeyCahce to use Non-Heap memory ?

> Is there any way I can configure KeyCahce to use Non-Heap memory ? No. You could add a feature request here https://issues.apache.org/jira/browse/CASSANDRA Could you post some stats on the current key cache size and hit rate ? (from nodetool info) It would be interesting to know how many key

Re: java.lang.NoClassDefFoundError when trying to do anything on one CF on one node

Hmmm, this looks like an error in ctor for NodeId$LocalNodeIdHistory. Are there any other ERROR log messages? Do you see either of these two messages in the log: "No saved local node id, using newly generated: {}" or "Saved local node id: {}" Can you use cassandra-cli / cqlsh to print the conte

Re: performance is drastically degraded after 0.7.8 --> 1.0.11 upgrade

That's slower. the Recent* metrics are the best to look at. They recent each time you look at them. So read them, then run the test, then read them again. You'll need to narrow it down still. e.g. Is there a single test taking a very long time or are all tests running slower ? The Histogram

Re: anyone know how to lookup non-continguous columns BUT for prefixes?

It's not possible to get non contiguous column slices. > This is a trimmed down example of course. I am thinking maybe I might just > use the astyanax async to send out 500 requests instead. Keep an eye on what happens on the server side. Sometimes less is more. Cheers - Aar

Re: keyspace not propagated to a new node?

What version are you on ? Anything in the logs from MigrationManager ? You could try nodetool resetlocalschema . Cheers - Aaron Morton Freelance Developer @aaronmorton http://www.thelastpickle.com On 5/09/2012, at 7:02 AM, Yang wrote: > I let a new node join the ring, and giv

Re: why do I have to use internal IP for EC2 nodes?

See http://aws.amazon.com/articles/1145?_encoding=UTF8&jiveRedirect=1#12 The external dns will resolve to the internal IP when resolved internally. Using the internal IP means you are not charged for IO and it makes it clear you do not expect this service to be access from outside. Cheers

Re: performance is drastically degraded after 0.7.8 --> 1.0.11 upgrade

all tests use similar data access patterns, so every test on 1.0.11 is slower than 0.7.8 recent micros confirms that. 2012/9/5 aaron morton > That's slower. > > the Recent* metrics are the best to look at. They recent each time you > look at them. So read them, then run the test, then read them

Re: keyspace not propagated to a new node?

this is 1.1.2 unfortunately the logs does not show anything except for the initial startup sequences. on the siblings, somehow the /var/log/cassandra/system.log were wrongly deleted, so I can't see their responses either I'll try resetlocalschema next time thanks! yang On Tue, Sep 4, 2012 at 7:

Re: why do I have to use internal IP for EC2 nodes?

thanks, but if the communication between cluster nodes all resolve to "internal to internal", amazon will not charge the traffic as external traffic, right? On Tue, Sep 4, 2012 at 7:08 PM, aaron morton wrote: > See http://aws.amazon.com/articles/1145?_encoding=UTF8&jiveRedirect=1#12 > > The exter

Re: why do I have to use internal IP for EC2 nodes?

@Yang: Sounds legit, as internal is not the same as external. Beware of the fact that internal traffic is only free when it's in the same availability zone. In the same region is charged with a small amount (~ $0.01). With kind regards, Robin Verlangen *Software engineer* * * W http://www.robinve