PI92342: CREATING A HADOOP TABLE WITH A LINE TERMINATOR OTHER THAN THE DEFAULT NEWLINE CHARACTER MAY RESULT IN UNEXPECTED BEHAVIOUR

APAR status

• Closed as fixed if next.

Error description

• Although the CREATE HADOOP TABLE syntax allows for the
  specification of a line terminator other than the default, the
  newline character \n is the only supported option as documented
  ("The only valid delimiter is the newline character (\n)").
  This limitation exists due to the same limitation in Hive.
  The code to restrict line terminators other than the \n is
  currently missing from Big SQL.
  .
  Data which contains a different line terminator cannot be read
  by the Big SQL Java reader, nor can it be properly processed by
  Analyze.  As a result, querying a table with a different line
  terminator may result in errors or wrong results or may cause
  performance issues due to incorrect statistics.
  .
  To prevent this, the specification of line terminators other
  than the default will now be blocked.  Any tables created with
  invalid line terminators will have to be recreated and the
  data corrected.
  

Local fix

• N/A
  

Problem summary

• Problem is fixed in IBM Db2 Big SQL version
  6.0.0.0
  

Problem conclusion

Temporary fix

• N/A
  Use the supported line termination character (\n)
  

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

Fix information

Applicable component levels

• R420 PSN

     UP