cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jonathan Ellis (JIRA)" <j...@apache.org>
Subject [jira] Issue Comment Edited: (CASSANDRA-161) occasinal CME in getKeyRange
Date Tue, 12 May 2009 15:51:45 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-161?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12708474#action_12708474
] 

Jonathan Ellis edited comment on CASSANDRA-161 at 5/12/09 8:51 AM:
-------------------------------------------------------------------

02
    fix race condition in compaction -- it was possible for a read thread to "snapshot" ssTables_,
then have
    the compactor thread delete those (after merging them into a new file) before the read
thread checked   
    them.  Since the read thread's "snapshot" doesn't include the new merged sstable, it incorrectly
tells  
    the caller that the key does not exist.                                              
                  

01
    fix race condition between when memtable is replaced as the active one and when it's added
to
    the set of historical (pending flush) memtables                                      
                     

these patches also fix the heisenbugs in NameSort and TimeSort tests.

      was (Author: jbellis):
    02
    fix race condition in compaction -- it was possible for a read thread to "snapshot" ssTables_,
then have
    the compactor thread delete those (after merging them into a new file) before the read
thread checked   
    them.  Since the read thread's "snapshot" doesn't include the new merged sstable, it incorrectly
tells  
    the caller that the key does not exist.                                              
                  

01
    CASSANDRA-159 fix race condition between when memtable is replaced as the active one and
when it's added to
    the set of historical (pending flush) memtables                                      
                     

these patches also fix the heisenbugs in NameSort and TimeSort tests.
  
> occasinal CME in getKeyRange
> ----------------------------
>
>                 Key: CASSANDRA-161
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-161
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Jonathan Ellis
>            Assignee: Jonathan Ellis
>         Attachments: 0001-CASSANDRA-161-fix-race-condition-between-when-memtable.txt,
0002-fix-race-condition-in-compaction-it-was-possible-fo.txt
>
>
>    [testng] java.util.ConcurrentModificationException
>    [testng]     at java.util.HashMap$HashIterator.nextEntry(HashMap.java:793)
>    [testng]     at java.util.HashMap$KeyIterator.next(HashMap.java:828)
>    [testng]     at java.util.Collections$UnmodifiableCollection$1.next(Collections.java:1010)
>    [testng]     at org.apache.cassandra.db.Table.getKeyRange(Table.java:903)
>    [testng]     at org.apache.cassandra.db.ColumnFamilyStoreTest.testCompactions(ColumnFamilyStoreTest.java:426)
>    [testng] ... Removed 22 stack frames

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


Mime
View raw message