You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Tyler L. Hobbs (JIRA)" <ji...@apache.org> on 2010/08/31 19:14:55 UTC
[jira] Created: (CASSANDRA-1450) Memtable flush causes bad
"reversed" get_slice
Memtable flush causes bad "reversed" get_slice
----------------------------------------------
Key: CASSANDRA-1450
URL: https://issues.apache.org/jira/browse/CASSANDRA-1450
Project: Cassandra
Issue Type: Bug
Components: Core
Affects Versions: 0.7 beta 1
Environment: Cassandra Trunk
Reporter: Tyler L. Hobbs
Attachments: reversed_slice_reproduce.txt
If columns are inserted into a row before and after a memtable flush, a get_slice() after the flush with reversed=True will return incorrect results. See attached patch to reproduce.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Resolved: (CASSANDRA-1450) Memtable flush causes bad
"reversed" get_slice
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-1450?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis resolved CASSANDRA-1450.
---------------------------------------
Reviewer: brandon.williams
Resolution: Fixed
committed, w/ test adjusted as suggested
> Memtable flush causes bad "reversed" get_slice
> ----------------------------------------------
>
> Key: CASSANDRA-1450
> URL: https://issues.apache.org/jira/browse/CASSANDRA-1450
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Affects Versions: 0.7 beta 1
> Environment: Cassandra Trunk
> Reporter: Tyler L. Hobbs
> Assignee: Jonathan Ellis
> Fix For: 0.7 beta 2
>
> Attachments: 1450.txt, reversed_slice_reproduce.txt
>
>
> If columns are inserted into a row before and after a memtable flush, a get_slice() after the flush with reversed=True will return incorrect results. See attached patch to reproduce.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-1450) Memtable flush causes bad
"reversed" get_slice
Posted by "Brandon Williams (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-1450?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12905945#action_12905945 ]
Brandon Williams commented on CASSANDRA-1450:
---------------------------------------------
+1, though maybe the flush in the test should be explicit since it's not immediately obvious it will occur after 20 ops.
> Memtable flush causes bad "reversed" get_slice
> ----------------------------------------------
>
> Key: CASSANDRA-1450
> URL: https://issues.apache.org/jira/browse/CASSANDRA-1450
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Affects Versions: 0.7 beta 1
> Environment: Cassandra Trunk
> Reporter: Tyler L. Hobbs
> Assignee: Jonathan Ellis
> Fix For: 0.7 beta 2
>
> Attachments: 1450.txt, reversed_slice_reproduce.txt
>
>
> If columns are inserted into a row before and after a memtable flush, a get_slice() after the flush with reversed=True will return incorrect results. See attached patch to reproduce.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (CASSANDRA-1450) Memtable flush causes bad
"reversed" get_slice
Posted by "Brandon Williams (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-1450?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Brandon Williams updated CASSANDRA-1450:
----------------------------------------
Fix Version/s: 0.7 beta 2
> Memtable flush causes bad "reversed" get_slice
> ----------------------------------------------
>
> Key: CASSANDRA-1450
> URL: https://issues.apache.org/jira/browse/CASSANDRA-1450
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Affects Versions: 0.7 beta 1
> Environment: Cassandra Trunk
> Reporter: Tyler L. Hobbs
> Fix For: 0.7 beta 2
>
> Attachments: reversed_slice_reproduce.txt
>
>
> If columns are inserted into a row before and after a memtable flush, a get_slice() after the flush with reversed=True will return incorrect results. See attached patch to reproduce.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Assigned: (CASSANDRA-1450) Memtable flush causes bad
"reversed" get_slice
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-1450?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis reassigned CASSANDRA-1450:
-----------------------------------------
Assignee: Jonathan Ellis
> Memtable flush causes bad "reversed" get_slice
> ----------------------------------------------
>
> Key: CASSANDRA-1450
> URL: https://issues.apache.org/jira/browse/CASSANDRA-1450
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Affects Versions: 0.7 beta 1
> Environment: Cassandra Trunk
> Reporter: Tyler L. Hobbs
> Assignee: Jonathan Ellis
> Fix For: 0.7 beta 2
>
> Attachments: reversed_slice_reproduce.txt
>
>
> If columns are inserted into a row before and after a memtable flush, a get_slice() after the flush with reversed=True will return incorrect results. See attached patch to reproduce.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (CASSANDRA-1450) Memtable flush causes bad
"reversed" get_slice
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-1450?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis updated CASSANDRA-1450:
--------------------------------------
Attachment: 1450.txt
patch applies reversed flag during collation from different data sources and makes QF.getColumnComparator package-private to make it more difficult to make the same mistake again. unit test added to TableTest.
> Memtable flush causes bad "reversed" get_slice
> ----------------------------------------------
>
> Key: CASSANDRA-1450
> URL: https://issues.apache.org/jira/browse/CASSANDRA-1450
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Affects Versions: 0.7 beta 1
> Environment: Cassandra Trunk
> Reporter: Tyler L. Hobbs
> Assignee: Jonathan Ellis
> Fix For: 0.7 beta 2
>
> Attachments: 1450.txt, reversed_slice_reproduce.txt
>
>
> If columns are inserted into a row before and after a memtable flush, a get_slice() after the flush with reversed=True will return incorrect results. See attached patch to reproduce.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (CASSANDRA-1450) Memtable flush causes bad
"reversed" get_slice
Posted by "Tyler L. Hobbs (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-1450?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Tyler L. Hobbs updated CASSANDRA-1450:
--------------------------------------
Attachment: reversed_slice_reproduce.txt
> Memtable flush causes bad "reversed" get_slice
> ----------------------------------------------
>
> Key: CASSANDRA-1450
> URL: https://issues.apache.org/jira/browse/CASSANDRA-1450
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Affects Versions: 0.7 beta 1
> Environment: Cassandra Trunk
> Reporter: Tyler L. Hobbs
> Attachments: reversed_slice_reproduce.txt
>
>
> If columns are inserted into a row before and after a memtable flush, a get_slice() after the flush with reversed=True will return incorrect results. See attached patch to reproduce.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.