Hi, i'm trying a new bulkload setup for old jobs after migration whole environment from z/Os to Linux. The jobs creates work table from from db2 table which are afterwards written in different lib on db2 Old bulkload looked like this: DATA DW_1.Tdw111 (BULKLOAD = YES BL_DB2TBLXST = YES BL_DB2LDEXT = GENONLY BL_DB2IN = "&Outputfiles..Tdw111.TEXT" BL_DB2REC = "&Outputfiles..Tdw111.DATA" BL_DB2RECSP = &DB2RecspTdw111 BL_DB2DEVT_PERM = &DB2DevtPerm BL_DB2UNITCOUNT = 50 ); SET DW_1.tdw111(OBS=0) WORK.tdw111; RUN; When i was testing new one, i simplified creation of the work table but still gets the same data as before but i used no extra options, to see if and how it's working: DATA DW_1.Tdw111 (BULKLOAD = YES) SET DW_1.tdw111(OBS=0) WORK.tdw111; RUN; No luck. Had to ask DB2 team to empty the target table as i had no auth to access so afterwards i used the same data step but without "DW_1.tdw111(OBS=0)". Still no luck and was getting error, that Db2 table is opened for output but isn't supporting REPLACE. So tried: proc append base = DW_1.Tdw111 (BULKLOAD=YES DBCOMMIT=100000 BL_DATAFILE= "/*random pathway*/BL_TDW111_VNT_LL.IXF" BL_LOG= "/*random pathway*/BL_TDW111.log" ) data = WORK.Tdw111(obs=1000000) force; RUN; Bulkload worked, created *.ixf file and log but couldn't write it in the target table with: ERROR: Bulk loader error: SQL3088N The source column specified to be loaded into database column "END_INTENT" is not compatible with the database column, but the database column is not nullable. I can't change the formats of source or target tables, so i tried modify via proc sql create table, with not null but still nothing. Would you have any suggestions either for the bulkload data step or how to deal with the last error written? Thank you. Martin
... View more