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] [Commented] (CASSANDRA-2867) Starting 0.8.1 after upgrade from 0.7.6-2 fails
Date Sun, 24 Jul 2011 20:57:09 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-2867?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13070251#comment-13070251
] 

Jonathan Ellis commented on CASSANDRA-2867:
-------------------------------------------

You have an index name set on at least one CF.  If you don't actually have an index on it,
Cassandra shouldn't have let you set the index name, but it's possible an older version was
not as rigorous about that.

If you can't figure out where the index name is, you can also drop your schema and migrations
system CFs and rebuild your schema from scratch.

> Starting 0.8.1 after upgrade from 0.7.6-2 fails
> -----------------------------------------------
>
>                 Key: CASSANDRA-2867
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-2867
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.8.1
>         Environment: CentOS 5.6
>            Reporter: Yaniv Kunda
>              Labels: exception, index, starting
>
> After upgrading the binaries to 0.8.1 I get an exception when starting cassandra:
> {noformat}
> [root@bserv2 local]#  INFO 12:51:04,512 Logging initialized
>  INFO 12:51:04,523 Heap size: 8329887744/8329887744
>  INFO 12:51:04,524 JNA not found. Native methods will be disabled.
>  INFO 12:51:04,531 Loading settings from file:/usr/local/apache-cassandra-0.8.1/conf/cassandra.yaml
>  INFO 12:51:04,621 DiskAccessMode 'auto' determined to be mmap, indexAccessMode is mmap
>  INFO 12:51:04,707 Global memtable threshold is enabled at 2648MB
>  INFO 12:51:04,708 Removing compacted SSTable files (see http://wiki.apache.org/cassandra/MemtableSSTable)
>  INFO 12:51:04,713 Removing compacted SSTable files (see http://wiki.apache.org/cassandra/MemtableSSTable)
>  INFO 12:51:04,714 Removing compacted SSTable files (see http://wiki.apache.org/cassandra/MemtableSSTable)
>  INFO 12:51:04,716 Removing compacted SSTable files (see http://wiki.apache.org/cassandra/MemtableSSTable)
>  INFO 12:51:04,717 Removing compacted SSTable files (see http://wiki.apache.org/cassandra/MemtableSSTable)
>  INFO 12:51:04,719 Removing compacted SSTable files (see http://wiki.apache.org/cassandra/MemtableSSTable)
>  INFO 12:51:04,770 reading saved cache /vm1/cassandraDB/saved_caches/system-IndexInfo-KeyCache
>  INFO 12:51:04,776 Opening /vm1/cassandraDB/data/system/IndexInfo-f-9
>  INFO 12:51:04,792 reading saved cache /vm1/cassandraDB/saved_caches/system-Schema-KeyCache
>  INFO 12:51:04,794 Opening /vm1/cassandraDB/data/system/Schema-f-194
>  INFO 12:51:04,797 Opening /vm1/cassandraDB/data/system/Schema-f-195
>  INFO 12:51:04,802 Opening /vm1/cassandraDB/data/system/Schema-f-193
>  INFO 12:51:04,811 Opening /vm1/cassandraDB/data/system/Migrations-f-193
>  INFO 12:51:04,814 reading saved cache /vm1/cassandraDB/saved_caches/system-LocationInfo-KeyCache
>  INFO 12:51:04,815 Opening /vm1/cassandraDB/data/system/LocationInfo-f-292
>  INFO 12:51:04,843 Loading schema version 586e70fd-a332-11e0-828e-34b74a661156
> ERROR 12:51:04,996 Exception encountered during startup.
> org.apache.cassandra.db.marshal.MarshalException: A long is exactly 8 bytes: 15
>         at org.apache.cassandra.db.marshal.LongType.getString(LongType.java:72)
>         at org.apache.cassandra.config.CFMetaData.getDefaultIndexName(CFMetaData.java:971)
>         at org.apache.cassandra.config.CFMetaData.inflate(CFMetaData.java:381)
>         at org.apache.cassandra.config.KSMetaData.inflate(KSMetaData.java:172)
>         at org.apache.cassandra.db.DefsTable.loadFromStorage(DefsTable.java:99)
>         at org.apache.cassandra.config.DatabaseDescriptor.loadSchemas(DatabaseDescriptor.java:479)
>         at org.apache.cassandra.service.AbstractCassandraDaemon.setup(AbstractCassandraDaemon.java:139)
>         at org.apache.cassandra.service.AbstractCassandraDaemon.activate(AbstractCassandraDaemon.java:315)
>         at org.apache.cassandra.thrift.CassandraDaemon.main(CassandraDaemon.java:80)
> Exception encountered during startup.
> org.apache.cassandra.db.marshal.MarshalException: A long is exactly 8 bytes: 15
>         at org.apache.cassandra.db.marshal.LongType.getString(LongType.java:72)
>         at org.apache.cassandra.config.CFMetaData.getDefaultIndexName(CFMetaData.java:971)
>         at org.apache.cassandra.config.CFMetaData.inflate(CFMetaData.java:381)
>         at org.apache.cassandra.config.KSMetaData.inflate(KSMetaData.java:172)
>         at org.apache.cassandra.db.DefsTable.loadFromStorage(DefsTable.java:99)
>         at org.apache.cassandra.config.DatabaseDescriptor.loadSchemas(DatabaseDescriptor.java:479)
>         at org.apache.cassandra.service.AbstractCassandraDaemon.setup(AbstractCassandraDaemon.java:139)
>         at org.apache.cassandra.service.AbstractCassandraDaemon.activate(AbstractCassandraDaemon.java:315)
>         at org.apache.cassandra.thrift.CassandraDaemon.main(CassandraDaemon.java:80)
> {noformat}
> It seems this has something to do with indexes, and I do have a CF with an index on it,
but it is not used.
> I can try and remove the index with 0.7.x binaries, but I will wait a bit to see if anyone
needs it to reproduce the bug.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message