IT28881: TABLE SPACE RESTORE UNABLE TO USE ALL CONTAINER PATHS AND FAILS WITH DISK FULL

APAR status

• Closed as program error.

Error description

• A user running an offline table space restore could see a "Disk
  Full" message in db2diag.log when a table space is using more
  than one storage path and the table space size is larger than
  the first storage path. If restoring only one table space the
  user will be returned SQL2549N.
  
  db2 "restore db wsdb tablespace (tbsp1) from <PATH> taken at
  <TIMESTAMP>"
  SQL2549N The database was not restored because either all of the
  table spaces
  in the backup image are inaccessible, or one or more table space
  names in list
  of table space names to restore are invalid.
  ?
  db2diag.log shows:
  ==============
  2019-02-28-11.28.50.696243-480 I6850A849 LEVEL: Error
  PID : 11927598 TID : 33518 PROC : db2sysc 0
  INSTANCE: db2inst1 NODE : 000 DB : WSDB
  APPHDL : <>
  AUTHID : <>  HOSTNAME: <>
  EDUID : 33518 EDUNAME: db2bm.34938.0 (TEST) 0
  FUNCTION: DB2 UDB, buffer pool services,
  sqlbASDefineContainersForDMS, probe:10
  MESSAGE : ZRC=0x850F000C=-2062614516=SQLO_DISK "Disk full."
  DIA8312C Disk was full.
  
  The problem is only reproduced when:
  - the container 0 does not exist (deleted manually or from prior
  issues). If the container already exists it will be reused and
  will avoid the problem.
  - the free space in the first storage path's file system is less
  than the total size needed for the table space.
  
  Note :-
  A failed table space restore can remove the container
  paths/files. So, in use case, if the container file was not
  manually removed, it must be removed by a previous failed
  restore.
  

Local fix

• The following could help to avoid the issue and make the restore
  complete.
  1. restore the whole db
  2. increase the file-system of the first container path to have
  enough free space to hold the whole table space.
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * Users  running tablespace restore                            *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  * See Error Description                                        *
  ****************************************************************
  * RECOMMENDATION:                                              *
  * Upgrade to DB2 Version 11.1 m4fp5                            *
  ****************************************************************
  

Problem conclusion

• Problem first fixed in DB2 Version 11.1 m4fp5
  

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:

  IT30798

Fix information

• Fixed component name

  DB2 FOR LUW

• Fixed component ID

  DB2FORLUW

Applicable component levels

• RB10 PSN

     UP