Dump on DBIF_RSQL_INVALID_RSQL

Question

Hi,
I am getting a dump on execution of a SQL statement, where the data is huge.
The error description...
Possible errors:
o The maximum size of an SQL statement has been exceeded.
o The statement contains too many input variables.
o The space needed for the input data exceeds the available memory.
The  Statement raising the error ...
LOOP AT pp_agreement.
    TRANSLATE pp_agreement-fname TO UPPER CASE.
    CONDENSE pp_agreement-fname NO-GAPS.
    CASE pp_agreement-fname.
      WHEN 'AGREEMENT'.
        MOVE-CORRESPONDING pp_agreement TO r_agreement.
        APPEND r_agreement.
      WHEN OTHERS.
        CLEAR pp_error.
        pp_error-type = 'W'.
        CONCATENATE 'Invalid Agreement --' pp_agreement-fname
               INTO pp_error-message_v1.
        APPEND pp_error.
    ENDCASE.
    CLEAR  : r_agreement.
  ENDLOOP.
  IF pp_error IS INITIAL.
Get agreements effective as of validity date or in the future.
Inclue the customer related agreements.
    SELECT * FROM z9ars_agreement
             INTO CORRESPONDING FIELDS OF TABLE pp_z9p_generic_agreement
             WHERE agreement IN r_agreement
             AND   customer = space
            and   orderbegindate le pp_validity_date    "v003
            and   orderenddate ge pp_validity_date      "v003
             AND   orderbegindate LE l_to_date              "v003
             AND   orderenddate GE l_from_date.             "v003
    DESCRIBE TABLE pp_z9p_generic_agreement LINES g_line_count.
    IF g_line_count = 0.
      CLEAR pp_error.
      pp_error-type = 'W'.
      MOVE 'No agreements retrieved --'
             TO pp_error-message_v1.
      APPEND pp_error.
    ENDIF.
  ENDIF.
  REFRESH: r_agreement.
The data in the IT r_agreement is Huge causing the dump.
The Z program is calling a Function Module where this select statement is executed.
This FM is being used by lots of other programs.
Kindly let know how to rectify the Error.
Do let know if any other details are required.
Thanks in advance.
Regards,
Syed

Answer 1

This is what we did for a similar issue.
Here we kept a limit of 5000 records in ranges. Hope it helps
  SORT ra_objnr BY low.
  DELETE ADJACENT DUPLICATES FROM ra_objnr COMPARING low.
DO.
    APPEND LINES OF ra_objnr FROM l_count_start TO l_count_end TO ra_objnr2 .
    l_last = sy-tabix.
    l_count_start = l_count_end + 1.
    l_count_end = l_count_end + 5000.
    IF ra_objnr2[] IS NOT INITIAL.
      SELECT kokrs
             belnr
             buzei
             kstar
             wkgbtr
             objnr
             vrgng
             beknz  FROM coep
                    APPENDING TABLE it_coep_over
                    WHERE kokrs EQ p_kokrs
                      AND perio EQ p_perio
                      AND gjahr EQ p_gjahr
                      AND objnr IN ra_objnr2.
    ENDIF.
    REFRESH ra_objnr2.
    IF l_last NE 5000.
      EXIT.
    ENDIF.
  ENDDO.
Edited by: Romit Kewalramani on Jul 24, 2008 7:33 PM