|
Problem IC78585
|
Status: Closed |
DATA CORRUPTION COULD OCCUR IF INPLACE TABLE REORG IS STOPPED DURING TRUNCATE PHASE |
| product: |
DB2 FOR LUW / DB2FORLUW / 970 - DB2 |
| Problem description: |
At the end of an inplace table reorg, the reorg enters a
truncate phase. The phase of an inplace reorg can be monitored
using the REORG_STATUS column of the SNAPTAB_REORG
administrative view or SNAP_GET_TAB_REORG table function. If an
inplace reorg "STOP" command is issued during this time,
corruption can occur on some number of pages on the table that
was being reorged. The symptom of this corruption can be seen if
a select performed on the table looks at one of the corrupted
pages. If the page was corrupted by this APAR, then DB2 will
abort the transaction and mark the page bad. In the db2diag.log,
a message similar to the following example will be shown:
2011-09-05-05.10.34.096691-240 I25758649E565 LEVEL: Severe
PID : 30201 TID : 140081973159680PROC :
db2sysc
INSTANCE: svtdbm2 NODE : 000 DB :
HADR975
APPHDL : 0-27511 APPID:
*LOCAL.svtdbm2.110905122458
AUTHID : SVTDBM2
EDUID : 440 EDUNAME: db2agent (HADR975)
FUNCTION: DB2 UDB, data management, sqldFetchOverflow,
probe:1797
MESSAGE : Followed REORGPOINTER but no OLR in progress!
DATA #1 : Hexdump, 8 bytes
0x00007F67603F6EB0 : 0A00 AC01 0000 0000
........
If such corruption occurs, the table data may still be
recoverable through the use of the DB2DART /DDEL utility.
Contact IBM support for assistance in recovering table data if
this problem occurs. |
| Problem Summary: |
****************************************************************
* USERS AFFECTED: *
* All *
****************************************************************
* PROBLEM DESCRIPTION: *
* Issuing a stop command during the truncate phase of inplace *
* reorg may leave behind some internal rows after the reorg is *
* stopped. When DB2 sees these rows, it will throw an error *
* and report corruption. User data should still be on the *
* pages and may be recovered using db2dart. *
****************************************************************
* RECOMMENDATION: *
* Upgrade to V9.7 Fix Pack 5. *
**************************************************************** |
| Local Fix: |
Instead of issuing an inplace reorg STOP, issue a PAUSE. A pause
behaves similarly to stop, but will not cause any corruption. A
stop command may be safely issued after a pause if required. |
| available fix packs: |
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
|
| Solution |
This problem was first fixed in V9.7 Fix Pack 5. |
| Workaround |
not known / see Local fix |
| Timestamps |
Date - problem reported :
Date - problem closed :
Date - last modified :
| 08.09.2011
23.12.2011
23.12.2011 |
| Problem solved at the following versions (IBM BugInfos) |
9.7.FP5
|
| Problem solved according to the fixlist(s) of the following version(s) |
| 9.7.0.5
|
 |
