LI73822: Tablespace file corruption during backup/restore or connect to db with automatic storage.

DB2 Version 9.1 Fix Pack 7  for Linux, UNIX and Windows
DB2 Version 9.1 Fix Pack 7a  for Linux, UNIX and Windows
DB2 Version 9.1 Fix Pack 8  for Linux, UNIX and Windows
DB2 Version 9.1 Fix Pack 9  for Linux, UNIX and Windows
DB2 Version 9.1 Fix Pack 10  for Linux, UNIX and Windows
DB2 Version 9.1 Fix Pack 11  for Linux, UNIX and Windows
DB2 Version 9.1 Fix Pack 12  for Linux, UNIX and Windows

APAR status

• Closed as program error.

Error description

• <scenario #1>
  After restoring the database from backup, rollforward query
  status could fail with SQL0980 disk error and db2diag.log shows
  SQLB_CSUM "Bad Page, Checksum" error
  
  For example,
  
  db2 rollforward db PROD query status
  
  SQL0980C  A disk error occurred.  Subsequent SQL statements
  cannot be processed.  SQLSTATE=58005
  
  db2diag.log entry -
  
  2009-xx-xx-xx.xx.xx.xxxxxx-xxx xxxxxxxxxxxx       LEVEL: Error
  PID     : xxxxxx               TID  : 1           PROC :
  db2agent (PROD) 0
  INSTANCE: xxxxxxxx             NODE : 000         DB   : PROD
  APPHDL  : 0-473                APPID:
  *LOCAL.xxxxxxxx.xxxxxxxxxxxx
  FUNCTION: DB2 UDB, buffer pool services, sqlbReadPoolTable,
  probe:910
  RETCODE : ZRC=0x86020019=-2046689255=SQLB_CSUM "Bad Page,
  Checksum
  
  If restoring the same database backup the second time and issue
  the rollforward query status again, it will complete
  successfully.
  
  <scenario #2>
  If users changes permissions to /tmp readable only for instance
  owner, backup of database would fail with SQL2048 Error and
  corrupt SQLSPCS file. This would result in SQL0980C error while
  connecting to database and users will not be able to connect to
  database.
  
  Steps to recreate issue :
  ***********************************************************
  1. db2 "create database TEST"
  
  2. db2 connect to TEST
  
  3. As root ,CHANGE permissions ON /tmp [root@lenexa /]# chmod
  000 tmp
   d---------   26 root root         4096 Aug 27 11:40 tmp
  4. db2 backup db TEST
      SQL2048N  An error occurred while accessing object "1".
  Reason code: "6".
  
  5. db2 connect to TEST
      SQL0980C  A disk error occurred.  Subsequent SQL statements
  cannot be processed.  SQLSTATE=58005
  
  <scenario #3>
  This not only affects backup, but normal connect as well.  It is
  specific to automatic storage tablespace and temp tablespace.
  When you try to connect to an automatic storage database, db2
  will attempt to redefine the containers for any SMS tempspace
  (under certain conditions).  The problem is that when db2 tried
  to free the containers, "cont_mem_used" field in SQLSPCS files
  will be modified unconditionally even if the freeing of the
  container may fail.  This could result in cont_mem_used = 0 and
  checksum not recaculated. You will see following messages in
  db2diag.log.
  
  xxxx-xx-xx-xx.xx.xx.xxxxxx+xxx I9084G474          LEVEL: Error
  PID     : 18390                TID  : 3086894784  PROC :
  db2agent (SAMPLE) 0
  INSTANCE: xxxxxx               NODE : 000
  APPHDL  : 0-7                  APPID: *LOCAL.xxxxxxx.xxxxxxxxx
  AUTHID  : xxxxxx
  FUNCTION: DB2 UDB, buffer pool services, sqlbReadPoolOneFile,
  probe:955
  MESSAGE : ZRC=0x86020019=-2046689255=SQLB_CSUM "Bad Page,
  Checksum Error"
            DIA8426C A invalid page checksum was found for page
  "".
  
  xxxx-xx-xx-xx.xx.xx.xxxxxx+xxx I9084G474          LEVEL: Error
  PID     : 18390                TID  : 3086894784  PROC :
  db2agent (SAMPLE) 0
  INSTANCE: xxxxxx               NODE : 000
  APPHDL  : 0-7                  APPID: *LOCAL.xxxxxxx.xxxxxxxxx
  AUTHID  : xxxxxx
  FUNCTION: DB2 UDB, buffer pool services, sqlbReadPool, probe:920
  MESSAGE : ZRC=0x86020019=-2046689255=SQLB_CSUM "Bad Page,
  Checksum Error"
            DIA8426C A invalid page checksum was found for page
  "".
  DATA #1 : <preformatted>
  An error was encounterd while reading pool 1from offset 262144
  in both tablespace files.
  

Local fix

• <for scenario #1>
  The workaround is to restore the database again and rollforward
  the second time.
  
  <for scenario #2 and #3>
  Contact DB2 technical support for patching the corrupted SQLSPCS
  files.
  

Problem summary

• Users Affected: All
  

Problem conclusion

• First fixed in DB2 UDB Version 9.1 , FixPak 7
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

  LI73823 LI74528

Fix information

• Fixed component name

  DB2 UDE ESE LIN

• Fixed component ID

  5765F4104

Applicable component levels

• R910 PSY

     UP

• R950 PSY

     UP