After restore from a split mirror backup ("offloaded backup"),
orphaned extents and extents used but marked as free might be
found on the target system if extent movement was active on the
source system when snapshot backup was taken. This might result
in a runtime BadPage error during or after the rollforward
indicating a zeroed-out page.
Offloaded backup is scenario when snapshot backup is initialized
using db2inidb:
$ db2inidb <db_name> as mirror
and then initialized copy used to create a regular DB2 backup,
during which following message is being logged:
FUNCTION: DB2 UDB, data protection services, sqlpgint,
probe:5110
DATA #1 : <preformatted>
Backup after split mirror.
Db2dart run on the target system (after the restore) will
indicate orphaned extents, e.g.:
Checking for orphaned extents:
Error: Extent number 72183 in tablespace 5 is
orphaned
or extents used but marked as free, e.g.:
Traversing extent map for object type: 0
Error: Extent used but marked FREE in SMP (# 0, pool page 32)!
Page: 4928
Typical runtime error that might be encountered as the result is
DIA8400C (SQLB_BADP) after database reads page with blank header
(all fields equal to zero)
FUNCTION: DB2 UDB, buffer pool services, sqlb_verify_page,
probe:2
MESSAGE : ZRC=0x86020001=-2046689279=SQLB_BADP "page is bad"
DIA8400C A bad page was encountered.
DATA #1 : String, 64 bytes
Error encountered trying to read a page - information follows :
DATA #2 : String, 23 bytes
Page verification error
...
DATA #6 : Page header, PD_TYPE_SQLB_PAGE_HEAD, 48 bytes
pageHead: {tbspSeed:0;obj:0;type:0} PPNum:0 OPNum:0
begoff: 0
datlen: 0
pagebinx: 0
revnum: 0
pagelsn: 0000000000000000
flag: 0
future32: 0
cbits1to31: 0
cbits32to63: 0
...
Extent movement operations on the source system (one, from which
snapshot backup is taken) are:
ALTER TABLESPACE REDUCE |
