[ https://issues.apache.org/jira/browse/CASSANDRA-6309?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13816498#comment-13816498 ]
Alex Liu commented on CASSANDRA-6309: ------------------------------------- If we can't get the patch for LOCAL_ONE write path in this ticket, I can post it to CASSANDRA-6238 > Pig CqlStorage generates ERROR 1108: Duplicate schema alias > ------------------------------------------------------------ > > Key: CASSANDRA-6309 > URL: https://issues.apache.org/jira/browse/CASSANDRA-6309 > Project: Cassandra > Issue Type: Bug > Components: Hadoop > Reporter: Thunder Stumpges > Assignee: Alex Liu > Attachments: 6309-2.0.txt, LOCAL_ONE-write-for-all-strategies.txt > > > In Pig after loading a simple CQL3 table from Cassandra 2.0.1, and dumping > contents, I receive: > Caused by: org.apache.pig.impl.plan.PlanValidationException: ERROR 1108: > Duplicate schema alias: author in "cm" > cm = load 'cql://thunder_test/cassandra_messages' USING CqlStorage; > dump cm > ERROR org.apache.pig.tools.grunt.Grunt - > org.apache.pig.impl.logicalLayer.FrontendException: ERROR 1066: Unable to > open iterator for alias cm > ... > Caused by: org.apache.pig.impl.plan.PlanValidationException: ERROR 1108: > Duplicate schema alias: author in "cm" > at > org.apache.pig.newplan.logical.visitor.SchemaAliasVisitor.validate(SchemaAliasVisitor.java:75) > running 'describe cm' gives: > cm: {message_id: chararray,author: chararray,author: chararray,body: > chararray,message_id: chararray} > The original table schema in Cassandra is: > CREATE TABLE cassandra_messages ( > message_id text, > author text, > body text, > PRIMARY KEY (message_id, author) > ) WITH > bloom_filter_fp_chance=0.010000 AND > caching='KEYS_ONLY' AND > comment='null' AND > dclocal_read_repair_chance=0.000000 AND > gc_grace_seconds=864000 AND > index_interval=128 AND > read_repair_chance=0.100000 AND > replicate_on_write='true' AND > populate_io_cache_on_flush='false' AND > default_time_to_live=0 AND > speculative_retry='NONE' AND > memtable_flush_period_in_ms=0 AND > compaction={'class': 'SizeTieredCompactionStrategy'} AND > compression={'sstable_compression': 'LZ4Compressor'}; > it appears that the code in CqlStorage.getColumnMetadata at ~line 478 takes > the "keys" columns (in my case, message_id and author) and appends the > columns from getColumnMeta (which has all three columns). Thus the keys > columns are duplicated. -- This message was sent by Atlassian JIRA (v6.1#6144)