IBM Support

PI99229: SELECT STATEMENTS ON A TABLE WITH XML DATA MAY FAIL FOR A CUSTOM SERDE SETUP

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as fixed if next.

Error description

  • Select statements on a table with XML data may fail with
    java.lang.NullPointerException.
    .
    In the bigsql.log, you may notice the following:
    .
    2018-06-06 15:10:05,357 ERROR
    com.ibm.biginsights.bigsql.dfsrw.reader.DfsBaseReader
    [Master-14-S:5.1001.1.0.0.356] : [BSL-XX-3d681902d] Exception
    raised by Reader at node: 14 Scan ID: S:5.1001.1.0.0.356 Table:
    <tablename> Spark: false VORC: false VPQ: false
    Exception Label: UNMAPPED(java.lang.NullPointerException)
    java.lang.NullPointerException
            at
    com.ibm.spss.hive.serde2.xml.XmlInputFormat$XmlRecordReader.<ini
    t>(XmlInputFormat.java:72)
            at
    com.ibm.spss.hive.serde2.xml.XmlInputFormat.getRecordReader(XmlI
    nputFormat.java:57)
            at
    com.ibm.biginsights.bigsql.dfsrw.reader.DfsHiveReader.initialize
    (DfsHiveReader.java:112)
            at
    com.ibm.biginsights.bigsql.dfsrw.jaro.DfsScan.init(DfsScan.java:
    133)
            at
    com.ibm.biginsights.bigsql.dfsrw.jaro.DfsScanManager.performScan
    (DfsScanManager.java:67)
            at
    com.ibm.biginsights.bigsql.dfsrw.reader.DfsBaseReader.performRea
    d(DfsBaseReader.java:687)
            at
    com.ibm.biginsights.bigsql.dfsrw.reader.DfsBaseReader.read(DfsBa
    seReader.java:631)
    .
    The same query through hive may work correctly.
    .
    The issue is caused by two properties, xmlinput.start and
    xmlinput.end, not being properly propagated to the Hive
    libraries by Big SQL.  This can be avoided by explicitly
    setting these properties before running the query.
    

Local fix

  • You can set the following properties prior to executing the
    query:
    set hadoop property xmlinput.start=
    set hadoop property xmlinput.end=
    .
    For instance:
    set hadoop property 'xmlinput.start'='<moment '
    set hadoop property 'xmlinput.end'='/'
    .
    The values that must be specified for these two properties
    depend on the table being queried.
    

Problem summary

  • See problem description
    

Problem conclusion

Temporary fix

Comments

APAR Information

  • APAR number

    PI99229

  • Reported component name

    IBM BIG SQL

  • Reported component ID

    5737E7400

  • Reported release

    502

  • Status

    CLOSED FIN

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2018-06-15

  • Closed date

    2020-08-24

  • Last modified date

    2020-08-24

  • 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

[{"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Product":{"code":"SSCRJT","label":"IBM Db2 Big SQL"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"502","Line of Business":{"code":"LOB10","label":"Data and AI"}}]

Document Information

Modified date:
25 August 2020