Hello list, I'm trying to create a test scenario to understand what RMAN construes to be a "corrupted" block. The initial assumption is that its "backup validate check logical" command should identify such blocks. =20 Here's the test I ran. Environment: HP-UX 11.11 Oracle 9.2.0.6 Test Sequence: 1 Run whole database backup with RMAN (ARCHIVELOG) 2 Identify a table segment, block to corrupt: =09 SQL> select file_id,block_id,tablespace_name from dba_extents where = segment_name =3D 'STATUS'; FILE_ID BLOCK_ID TABLESPACE_NAME ---------- ---------- ------------------------------ 12 265 REACTOR5_1 SQL> select file#,name from v$Datafile where file#=3D12; FILE# NAME ---------- --------------------------------------------------- 12 /PERF/agentdb/oradata/PERF/REACTOR5_1.dbf 3 Overwrite identified block ./corrupt.ksh /PERF/agentdb/oradata/PERF/REACTOR5_1.dbf 265 0+1 records in 0+1 records out more corrupt.ksh #!/usr/bin/ksh FILE=3D${1:?'Parameter 1 should be set to file name'} BLOCK=3D${2:?'Parameter 2 should be set to the block to be corrupted'} dd of=3D$FILE bs=3D8192 conv=3Dnotrunc oseek=3D$BLOCK << EOF CORRUPT_by_intent EOF 4 Shutdown immediate, then startup 5 Run select on the corrupted segment: SQL> select count(*) from reactor5.status; COUNT(*) ---------- 44 6 "Backup validate" affected tablespace with RMAN: RMAN> backup validate check logical tablespace REACTOR5_1; Starting backup at 04-APR-2005 10:31:13 using channel ORA_DISK_1 channel ORA_DISK_1: starting full datafile backupset channel ORA_DISK_1: specifying datafile(s) in backupset input datafile fno=3D00012 = name=3D/PERF/agentdb/oradata/PERF/REACTOR5_1.dbf channel ORA_DISK_1: backup set complete, elapsed time: 00:00:01 Finished backup at 04-APR-2005 10:31:15 RMAN>=20 7 Check v$database_block_corruption view: SQL> select * from v$database_block_corruption; FILE# BLOCK# BLOCKS CORRUPTION_CHANGE# CORRUPTIO ---------- ---------- ---------- ------------------ --------- 12 265 1 0 FRACTURED QUESTION:=20 Why doesn't RMAN identify block corruption with "backup validate" while still registering block corruption in v$database_block_corruption? Cheers, Slava Zayarny Oracle DBA SAS Institute Inc -- //www.freelists.org/webpage/oracle-l