incubator-cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jonathan Ellis <jbel...@gmail.com>
Subject Re: get_range_slice(s) question
Date Tue, 16 Mar 2010 20:59:41 GMT
Thanks for the patch!

Since we've had some jira notification whackiness -- I just attached a
v2 patch and commented.

On Tue, Mar 16, 2010 at 3:43 PM, Omer van der Horst Jansen
<omerhj@gmail.com> wrote:
> The ticket is here: https://issues.apache.org/jira/browse/CASSANDRA-884
>
> I've attached a patch there that seems to fix the problem.
>
> Has anybody else seen this behavior?
>
>
>> That would be a bug, not intended behavior.  Can you open a ticket?
>>
>> On Fri, Mar 12, 2010 at 11:48 AM, Omer van der Horst Jansen
>> <ome...@yahoo.com> wrote:
>> > I've noticed that both 0.5.1 and 0.6b2 return (ReplicationFactor)
>> > identical copies of the data stored in my keyspace whenever I make a
>> > call to get_range_slice or get_range_slices using
>> > ConsistencyLevel.QUORUM.
>> >
>> > So with ReplicationFactor set to 2 for my application's KeySpace I get
>> > double the number of KeySlices that I expect to get. When using
>> > ConsistencyLevel.ONE I get only one KeySlice for each row.
>> >
>> > The same routine running against the Standard1 keyspace with a
>> > ReplicationFactor of 1 returns only a single KeySlice for each row. A
>> > ReplicationFactor of three gives me three identical KeySlices when using
>> > ConsistencyLevel.QUORUM.
>> >
>> > Is this the intended behavior of get_range_slices? I remember reading in
>> > one of the Dynamo papers that applications (and not Dynamo) are required
>> > to sort out any discrepancies in the data, but in this case there aren't
>> > any discrepancies.
>> >
>> > Omer
>

Mime
View raw message