Re: RandomPartitioner and new token allocation algorithm

[Romain Hardouin]

I targeted the dev list because I would like to know why the developer (patch 
by branimir and reviewed by benedict) mentions "Only supported with the 
Murmur3Partitioner" whereas his patch uses IPartitioner interface. (I will try 
to reach them on IRC if they don't see this message.)

Keyspaces not found in cqlsh

[kedar]

Dev Team,

Need some help with a burning cqlsh issue

I am using cqlsh 5.0.1 | Cassandra 2.1.2, recently we are unable to see
/ desc keyspaces and query tables through cqlsh on either of the two nodes

cqlsh> desc keyspaces



cqlsh> use user_index;
cqlsh:user_index> desc table list_1_10;

Keyspace 'user_index' not found.
cqlsh:user_index>
cqlsh>  select * from system.schema_keyspaces;
Keyspace 'system' not found.
cqlsh>
We are running a 2 node cluster. The Python - Django app that inserts
data is running without any failure and system logs show nothing abnormal.

./nodetool repair on one node hasn't helped ./nodetool cfstats shows all
the tables too

ls -l cassandra/data/*  on each node:

https://gist.github.com/anonymous/3dddbe728a52c07d7c52
https://gist.github.com/anonymous/302ade0875dd6410087b




 --
Thanks,
Kedar Parikh

Re: RandomPartitioner and new token allocation algorithm

[Brice Dutheil]

On Thu, Feb 11, 2016 at 12:28 PM, Sylvain Lebresne 
wrote:

> and honestly the difference with Murmur3 are not big enough that its worth
> going through a very painful migration if that's your case
>

Agreed, after some tests, we didn't found it was an actual issue for
our *reasonably
sized* clusters.


-- Brice

Re: Statistics.db file in Cassandra 3.0

[Rajath Subramanyam]

Thanks Tyler.


Rajath Subramanyam


On Wed, Feb 10, 2016 at 8:45 AM, Tyler Hobbs  wrote:

> Take a look at MetadataSerializer and the MetadataComponent subclasses.
>
> On Tue, Feb 9, 2016 at 4:34 PM, Rajath Subramanyam 
> wrote:
>
> > Hello Cassandra-Dev,
> >
> > I have noticed that in Cassandra 3.0 there is a new file in the
> > // called
> > ma--big-Statistics.db.
> >
> > What does this file contain ? Is it compressed ? How do I read it ?
> >
> > Thanks in advance for sharing some information on this.
> >
> > - Rajath
> > 
> > Rajath Subramanyam
> >
>
>
>
> --
> Tyler Hobbs
> DataStax 
>

Short column names

[Bhuvan Rawal]

Hi,

We are modelling schema for database revamp from mysql to Cassandra. It has
been recommended in several places that column names must be kept as small
as possible to optimise disk storage.

I have a doubt here, why can't we map column names and store it as an
index, say in memory. I mean, make column name really small human
unreadable and store it in disk  but map it with real column while
querying. That way one can go ahead with readable column names .

Let me know if I can go ahead and create a jira for the same

Regards,
Bhuvan

Re: Short column names

[Jeremy Hanna]

It has been talked about in the past, see 
https://issues.apache.org/jira/browse/CASSANDRA-4175 for example.  However with 
https://issues.apache.org/jira/browse/CASSANDRA-8099, the duplication of column 
names is gone.  So once you’re on Cassandra 3+, this optimization is a lot less 
valuable.

> On Feb 11, 2016, at 4:19 PM, Bhuvan Rawal  wrote:
> 
> Hi,
> 
> We are modelling schema for database revamp from mysql to Cassandra. It has
> been recommended in several places that column names must be kept as small
> as possible to optimise disk storage.
> 
> I have a doubt here, why can't we map column names and store it as an
> index, say in memory. I mean, make column name really small human
> unreadable and store it in disk  but map it with real column while
> querying. That way one can go ahead with readable column names .
> 
> Let me know if I can go ahead and create a jira for the same
> 
> Regards,
> Bhuvan



signature.asc
Description: Message signed with OpenPGP using GPGMail

Re: RandomPartitioner and new token allocation algorithm

[Sylvain Lebresne]

On Thu, Feb 11, 2016 at 11:56 AM, Romain Hardouin 
wrote:

> I targeted the dev list because I would like to know why the developer
> (patch by branimir and reviewed by benedict) mentions "Only supported with
> the Murmur3Partitioner" whereas his patch uses IPartitioner interface. (I
> will try to reach them on IRC if they don't see this message.)
>

While the patch uses the IPartitioner interface, it also requires that the
Token for the partitioner supported to implement the 2 new methods 'size'
and 'increaseSlightly', which only Murmur3Partition tokens does (it throws
UnsupportedOperationException for RandomPartitioner in particular).
So the documentation in the yaml is correct, only Murmur3Partitioner is
currently supported.

That said, I think the initial rational for only supporting
Murmur3Partitioner is that it's the default and the one recommended, but
it's certainly true that old installs are stuck to RandomPartitioner (and
honestly the difference with Murmur3 are not big enough that its worth
going through a very painful migration if that's your case) and it's not
immediately clear to me that adding support for it would be too hard. Feel
free to open an improvement ticket for that if you're interested.

--
Sylvain