[ 
https://issues.apache.org/jira/browse/CASSANDRA-13369?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Ariel Weisberg updated CASSANDRA-13369:
---------------------------------------
    Attachment: test_stdout.txt

bootstrap_test.TestBootstrap.consistent_range_movement_false_with_replica_down_should_succeed_test
bootstrap_test.TestBootstrap.simultaneous_bootstrap_test
cqlsh_tests.cqlsh_copy_tests.CqlshCopyTest.test_bulk_round_trip_blogposts
materialized_views_test.TestMaterializedViews.clustering_column_test
materialized_views_test.TestMaterializedViews.clustering_column_test
paxos_tests.TestPaxos.contention_test_many_threads
secondary_indexes_test.TestPreJoinCallback.write_survey_test
topology_test.TestTopology.size_estimates_multidc_test
topology_test.TestTopology.size_estimates_multidc_test

> If there are multiple values for a key, CQL grammar choses last value. This 
> should not be silent or should not be allowed.
> --------------------------------------------------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-13369
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-13369
>             Project: Cassandra
>          Issue Type: Bug
>          Components: CQL
>            Reporter: Nachiket Patil
>            Assignee: Nachiket Patil
>            Priority: Minor
>         Attachments: 3.X.diff, test_stdout.txt, trunk.diff
>
>
> If through CQL, multiple values are specified for a key, grammar parses the 
> map and last value for the key wins. This behavior is bad.
> e.g. 
> {code}
> CREATE KEYSPACE Excalibur WITH REPLICATION = {'class': 
> 'NetworkTopologyStrategy', 'dc1': 2, 'dc1': 5};
> {code}
> Parsing this statement, 'dc1' gets RF = 5. This can be catastrophic, may even 
> result in loss of data. This behavior should not be silent or not be allowed 
> at all.  



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org

Reply via email to