DB2 - Problem description
| Problem IC74003 | Status: Closed |
ROLLFORWARD THROUGH REORG WITH COMPRESSION ON A TABLE MARKED UNAVAILABLE COULD FAIL | |
| product: | |
DB2 FOR LUW / DB2FORLUW / 970 - DB2 | |
| Problem description: | |
A table with XML compression enabled that has gone through
reorgs and nonrecoverable loads, could fail during rollforward.
It would fail in certain order and combination of reorgs with
longlobdata option and nonrecoverable load.
Example:
db2 -v update db cfg for xmldb using logretain recovery;
db2 -v backup db xmldb;
db2 -v connect to xmldb;
db2 -v drop table TAB1;
db2 -v "create table TAB1(a int, b int, c xml)";
db2 -v alter table TAB1 compress yes;
db2 -v "import from xmlrecords.del of del method p (1,1,5)
insert into TAB1";
db2 -v "import from xmlrecords.del of del method p (1,1,5)
insert into TAB1";
db2 -v "import from xmlrecords.del of del method p (1,1,5)
insert into TAB1";
db2 -v "import from xmlrecords.del of del method p (1,1,5)
insert into TAB1";
db2 -v "import from xmlrecords.del of del method p (1,1,5)
insert into TAB1";
db2 -v "import from xmlrecords.del of del method p (1,1,5)
insert into TAB1";
db2 -v "import from xmlrecords.del of del method p (1,1,5)
insert into TAB1";
# This reorg will build an XML dictionary
db2 -v reorg table TAB1 longlobdata resetdictionary;
# This nonrecoverable load will mark the object not available
db2 -v "load from xmlrecords.del of del method p(1,1,5) insert
into TAB1 nonrecoverable";
db2 -v reorg table TAB1 longlobdata keepdictionary;
# This load will remove the dictionary from the table
db2 -v "load from /dev/null of del replace resetdictionary into
TAB1 nonrecoverable";
# There should be no dictionary for the table after this reorg
db2 -v "reorg table TAB1 longlobdata keepdictionary";
db2 -v terminate;
db2 -v restore db xmldb from /home/singgih/simap_xml_compr
without prompting;
# Without the fix, this rollforward will fail and in
db2diag.log, there will be messages about
# sqldReorgCopyPhase complaining that a dictionary exists but
the dictionary RID is zero.
The following rollforward will fail:
db2 -v rollforward db $DB to end of logs and complete;
In the db2diag.log, the following message would be found:
FUNCTION: DB2 UDB, data management, sqldReorgCopyPhase,
probe:9785
MESSAGE : ZRC=0x82040001=-2113667071=SQLD_NONSEVERE_PRGERR
"non-severe dms programming error"
DIA8532C An internal processing error has occurred.
DATA #1 : String, 33 bytes
Dictionary built, but RID is zero
DATA #2 : String, 8 bytes
sqldreor
CALLSTCK:
[0] 0x00002AAAAC03D30C pdLog + 0x18C
[1] 0x00002AAAAFF0A3DD sqlzSetAndLog901 + 0x395
[2] 0x00002AAAAC7DA129 _Z18sqldReorgCopyPhaseP13SQLD_REORG_CB
+ 0xDE5
[3] 0x00002AAAAC7858C0 /home/singgih/sqllib/lib64/libdb2e.so.1
+ 0x1AA58C0
[4] 0x00002AAAAC7869B2 /home/singgih/sqllib/lib64/libdb2e.so.1
+ 0x1AA69B2
[5] 0x00002AAAAC78C3AC
_Z10sqldomRedoP8sqeAgentP10SQLDOM_LRHP9SQLP_LSN8P15SQLD_RECOV_IN
FO
+ 0x242
[6] 0x00002AAAAC773B4C
_Z8sqldmrdoP8sqeAgentP9SQLP_LSN8PcmmmmP8SQLP_TIDP15SQLD_RECOV_IN
FO
+ 0x61C
[7] 0x00002AAAAE7C84C9
_Z13sqlpRecDbRedoP8sqeAgentP8SQLP_ACBP9SQLP_DBCBP10SQLP_FRAPPP11
SQLP_TENTRYP16SQLPR_LOGREC_DISP10REDO_INPUT
+ 0x6FB
[8] 0x00002AAAAE7CC9DE
_Z15sqlprProcDPSrecP10SQLPR_PRCBjP20SQLPR_PR_QUEUE_ENTRY + 0x48A
[9] 0x00002AAAAE7D1C49
_Z15sqlpPRecReadLogP8sqeAgentP8SQLP_ACBP9SQLP_DBCB + 0x3D5 | |
| Problem Summary: | |
**************************************************************** * USERS AFFECTED: * * All users * **************************************************************** * PROBLEM DESCRIPTION: * * See APAR description * **************************************************************** * RECOMMENDATION: * * Upgrade to V97 fp4 * **************************************************************** | |
| Local Fix: | |
| available fix packs: | |
DB2 Version 9.7 Fix Pack 4 for Linux, UNIX, and Windows | |
| Solution | |
THe problem is fixed in V97 fp4. | |
| Workaround | |
not known / see Local fix | |
| Timestamps | |
Date - problem reported : Date - problem closed : Date - last modified : | 20.01.2011 28.04.2011 28.04.2011 |
| Problem solved at the following versions (IBM BugInfos) | |
9.7.FP4 | |
| Problem solved according to the fixlist(s) of the following version(s) | |
| 9.7.0.4 |
|