cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Stu Hood (JIRA)" <>
Subject [jira] Commented: (CASSANDRA-1733) get_range_slices doesn't return key from start_key in KeyRange any more
Date Fri, 12 Nov 2010 17:32:13 GMT


Stu Hood commented on CASSANDRA-1733:

The ranges to query are properly generated in StorageProxy:
[228...,0] are [[228...,333...], (333...,0]]

But ColumnFamilyStore.getRangeSlice creates DecoratedKeys from the tokens of the Range with
null keys: null keys now sort after valid keys, so the first key isn't properly included.

> get_range_slices doesn't return key from start_key in KeyRange any more
> -----------------------------------------------------------------------
>                 Key: CASSANDRA-1733
>                 URL:
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.7.0
>         Environment: Cent OS 5.4
> Cassandra Trunk from November 11th at 2pm
>            Reporter: Arya Goudarzi
>            Assignee: Stu Hood
> The following is a test case which used to work for me before, but after upgrading to
trunk it fails:
> -Insert 50 or so keys to StandardByUUID1 CF in Keyspace1
> -Try using get_range_slices with a KeyRange that has one of those keys you inserted earlier
as its start_key with a count of 20
> You will get 20 rows, but you won't get the row with specified start_key
> Expected Result: you should get 20 rows back with the specified start_key as one of the
> I am suing Random Partitioner with RF2 and CL1. I understand that Random Partitioner
will not give you keys in order, but the behavior before at least returned a random subset
including the row with specified start_key.
> Please investigate.
> -Arya

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message