DB2 FOR LUW / DB2FORLUW / 970 - DB2

This problem happens when we redistribute a range partitioned 
table which has had partition range 0 detached, and we're 
collecting statistics during the redistribute. 
. 
Repro: 
db2 connect to testdb; 
db2 create database partition group g1 on "dbpartitionnum (0, 1, 
2) "; 
db2 create tablespace ts1 in database partition group g1 
pagesize 4 K extentsize 2; 
db2 "create table ts1.t1(i1 int, i2 int, i3 int) partition by 
range(i1) (starting 1 ending 3000 every 100) distribute by hash 
(i2)"; 
db2 "load from a.del of del insert into ts1.t1"; 
db2 "create unique index t1i1 on ts1.t1(i2)" 
db2 runstats on table ts1.t1  with distribution and detailed 
indexes all set profile only 
db2 "alter table ts1.t1 detach partition PART0 into ts1.td1" 
db2 "redistribute nodegroup g1 not rollforward recoverable 
uniform add dbpartitionnum(3) statistics use profile" 
SQL6064N  SQL error "-902" occurred during data redistribution. 
. 
db2diag.log: 
2010-09-21-13.45.40.718522-240 I394155E786         LEVEL: Error 
PID     : 10218                TID  : 47396079790400PROC : 
db2sysc 0 
INSTANCE: db2inst1             NODE : 000          DB   : TESTDB 
APPHDL  : 0-337                APPID: *N0.DB2.100921174613 
AUTHID  : DB2INST1 
EDUID   : 854                  EDUNAME: db2rwork 0 
FUNCTION: DB2 UDB, database utilities - Redistribute, 
sqlurCObjectManagerList::zFind, probe:952 
MESSAGE : 
ZRC=0x87B000AE=-2018508626=SQLUC_ZRC_ERROR_OBJECT_NOT_FOUND 
          "Object looking for does not exist." 
DATA #1 : RDSTID, PD_TYPE_RDSTID, 47 bytes 
RDSTID: U.56.20100921134530.0|643(65530;32768)| 
DATA #2 : File name, 24 bytes 
sqlurObjectManagerList.C 
DATA #3 : String, 34 bytes 
Error searching for object manager 
DATA #4 : unsigned integer, 2 bytes 
0 
. 
If any other ranges are detached this problem is not hit, only 
when we drop range 0.

This problem happens when we redistribute a range partitioned 
table which has had partition range 0 detached, and we're 
collecting statistics during the redistribute. 
. 
Repro: 
db2 connect to testdb; 
db2 create database partition group g1 on "dbpartitionnum (0, 1, 
2) "; 
db2 create tablespace ts1 in database partition group g1 
pagesize 4 K extentsize 2; 
db2 "create table ts1.t1(i1 int, i2 int, i3 int) partition by 
range(i1) (starting 1 ending 3000 every 100) distribute by hash 
(i2)"; 
db2 "load from a.del of del insert into ts1.t1"; 
db2 "create unique index t1i1 on ts1.t1(i2)" 
db2 runstats on table ts1.t1  with distribution and detailed 
indexes all set profile only 
db2 "alter table ts1.t1 detach partition PART0 into ts1.td1" 
db2 "redistribute nodegroup g1 not rollforward recoverable 
uniform add dbpartitionnum(3) statistics use profile" 
SQL6064N  SQL error "-902" occurred during data redistribution. 
. 
db2diag.log: 
2010-09-21-13.45.40.718522-240 I394155E786         LEVEL: Error 
PID     : 10218                TID  : 47396079790400PROC : 
db2sysc 0 
INSTANCE: db2inst1             NODE : 000          DB   : TESTDB 
APPHDL  : 0-337                APPID: *N0.DB2.100921174613 
AUTHID  : DB2INST1 
EDUID   : 854                  EDUNAME: db2rwork 0 
FUNCTION: DB2 UDB, database utilities - Redistribute, 
sqlurCObjectManagerList::zFind, probe:952 
MESSAGE : 
ZRC=0x87B000AE=-2018508626=SQLUC_ZRC_ERROR_OBJECT_NOT_FOUND 
          "Object looking for does not exist." 
DATA #1 : RDSTID, PD_TYPE_RDSTID, 47 bytes 
RDSTID: U.56.20100921134530.0|643(65530;32768)| 
DATA #2 : File name, 24 bytes 
sqlurObjectManagerList.C 
DATA #3 : String, 34 bytes 
Error searching for object manager 
DATA #4 : unsigned integer, 2 bytes 
0 
. 
If any other ranges are detached this problem is not hit, only 
when we drop range 0.

Redistributing with STATISTICS NONE bypasses this problem and 
allows redistribute to continue. Doing runstats after the detach 
does not prevent this problem. This can be hit every time we 
redistribute a range partitioned table with partition 0 
detached, and collect statistics during the redist.

DB2 Version 9.7 Fix Pack 4 for Linux, UNIX, and Windows
 DB2 Version 9.7 Fix Pack 5 for Linux, UNIX, and Windows
 DB2 Version 9.7 Fix Pack 6 for Linux, UNIX, and Windows
 DB2 Version 9.7 Fix Pack 7 for Linux, UNIX, and Windows
 DB2 Version 9.7 Fix Pack 8 for Linux, UNIX, and Windows
 DB2 Version 9.7 Fix Pack 9 for Linux, UNIX, and Windows
 DB2 Version 9.7 Fix Pack 9a for Linux, UNIX, and Windows
 DB2 Version 9.7 Fix Pack 10 for Linux, UNIX, and Windows
 

fixed >= v97 fpk4

Redistributing with STATISTICS NONE bypasses this problem and 
allows redistribute to continue. Doing runstats after the detach 
does not prevent this problem. This can be hit every time we 
redistribute a range partitioned table with partition 0 
detached, and collect statistics during the redist.

Vorgänger  : APAR is sysrouted TO one or more of the following: IC71947 IC72828 IC74574 
Nachfolger : 

Datum - Problem gemeldet    :
Datum - Problem geschlossen :
Datum - der letzten Änderung:

21.09.2010
29.04.2011
29.04.2011

9.7.FPk4