IBM Sterling Transformation Extender

Sterling Transformation Extender

Come for answers, stay for best practices. All we're missing is you.


#Sterling
#Supplychain
 View Only

  • 1.  MERCATOR BUG in 6.7

    Posted 12/07/05 11:28 AM

    Originally posted by: SystemAdmin


    Hi,

    This is a very basic EDIFACT Validation map which run on our Production Environment.
    Today we got some bad data from one of our customers . They sent a DESADV version D97A with a bad QTY Segment as show below.
    QTY+3:18000?:C62'

    Basically the ? is causing the map to go into a infinite loop and is crashing our servers.
    I tried it on my desktop and same thing is happening on my Desktop also.

    I did a little digging on my own and here is my analysis,

    all the segments and composites have ? as the default release character and when we get the above QTY Segment
    where the second element is numeric once it reads the 18000 and the sees the ? and considers the rest of the data followed by the ? mark
    as still second element in QTY so basically it should be read 18000:C62 and as a result it should fail for type being wrong.

    But that's not happening it is just going into a infinite loop and crashing the system.

    When I make the Release Character as None on the Composite it starts failing which is what is supposed to happen.

    Below is excerpt from the trace file

    (Level 8: Offset 446, len 1, comp 1 of 3, #1, DI 00000090:)
    Data at offset 446 ('3') was found to be of TYPE
    X'00F1' (QtyQual'r Element VD97A EDIFACT EDI).

    (Level 8: Offset 448, len 10, comp 2 of 3, #1, DI 00000091:)
    Data at offset 448 ('18000?:C62') was found to be of TYPE
    X'00F2' (Qty Element VD97A EDIFACT EDI).

    (Level 7: Offset 446, len 12, comp 1 of 1, #1, DI 00000092:)
    Data at offset 446 ('3:18000?:C62') was found to be of TYPE
    X'00F0' (QuantityDtls Composite VD97A EDIFACT EDI).

    (Level 6: Offset 445, len 13, comp 5 of 21, #1, DI 00000093:)
    Data at offset 445 ('+3:18000?:C62') was found to be of TYPE
    X'00EF' (QTY Segment VD97A EDIFACT EDI).

    We do not understand why is it not failing invalid data type and most of all why is it going into infinite loop.

    Any help is greatly appreciated.
    If anyone wants to look at the actual files related to the issue please let me know I can zip them and send them to you.

    Thanks,
    Sai.
    #DataExchange
    #IBM-Websphere-Transformation-Extender
    #IBMSterlingTransformationExtender


Related Content