@23-JAN-01, rpowell.uk, Distribution justification: "Should not be used by @ customers unless told to by support so making noncustomer-viewable." - dbither.us Event: 10218 Text: dump uba of applied undo ------------------------------------------------------------------------------- Note: Events should NEVER be set by customers unless advised to do so by Oracle Support Services. See <Note:75713.1> for more details. Description: This event simply shows the undo block addresses for the undo applied during recovery. This can be useful to see the progress of the rolling out of transactions, and to help identify where a rollback has gotten stuck. Example output: uba: 80000ca.0d.12 uba: 80000ca.0d.11 uba: 80000ca.0d.10 uba: 80000ca.0d.0f uba: 80000ca.0d.0e uba: 80000ca.0d.0d uba: 80000ca.0d.0c uba: 80000ca.0d.0b
EVENT:10218 dump uba of applied undo
2006/07/22 by Leave a Comment
EVENT:10226 trace CR applications of undo for data
2006/07/22 by Leave a Comment
Error: ORA 10226
Text: trace CR applications of undo for data operations
-------------------------------------------------------------------------------
Explanation:
This is NOT an error but is a special EVENT code. It can be used
to signal Oracle to perform SQL_TRACE actions.
This event causes application of UNDO for producing CR copies of
blocks. It can produce a lot of output (in the form of REDO
application dump to rewind the block)
EVENT:10061 disable SMON from cleaning temp segment
2006/07/22 by Leave a Comment
Error: ORA 10061 Text: disable SMON from cleaning temp segment ------------------------------------------------------------------------------- Explanation: This is NOT an error but is a special EVENT code. It should *NOT* be used unless explicitly requested by RD support. It allows you to prevent SMON from cleaning up temporary segments. If set in: 7.0.X: This will prevent SMON performing cleanup of TEMP segments and from coalsecing extents. The cleanup action was performed every 3 minutes in 7.0.X releases. 7.1.X: The event behaviour is modified in that TEMP segment cleanup is skipped on TIMEOUTS but will be performed if SMON is posted. The cleanup only occurs every 12 hours or when posted in 7.1.X releases. You can POST SMON by performing a failed CREATE TABLE command. A new event 10269 is for coalescing extents. Usage: event="10061 trace name context forever, level 10" Articles: @Archived Managing SMON Activity on multiple OPS instances <Note:45856.1>
EVENT:10212 check cluster integrity
2006/07/22 by Leave a Comment
Error: ORA 10212
Text: check cluster integrity
-------------------------------------------------------------------------------
Explanation:
This is NOT an error but is a special EVENT code.
It should *NOT* be used unless explicitly requested by RD support.
Event 10212 enables cluster block integrity checking.
The causes a number of checks on the contents of a block.
If the block is found to be corrupt in some way it is marked as
SOFTWARE CORRUPT.
NB: Blocks are ONLY checked/corrupted when CHANGED. They are NOT checked
when accessed for read.
Take care with this as if there are blocks in the database which
are being accessed happily but have a minor corruption this will
mark the block as corrupt.
Once a block in an cluster is marked SOFTWARE CORRUPT it is best to
rebuild the entire cluster.
Usage: event="10212 trace name context forever, level 10"
Articles:
Data and Index block equivalents <Oerr:10210> <Oerr:10211>
EVENT:10228 trace application of redo by kcocbk
2006/07/22 by Leave a Comment
Error: ORA 10228
Text: trace application of redo by kcocbk
-------------------------------------------------------------------------------
Explanation:
This is NOT an error but is a special EVENT code.
It should *NOT* be used unless explicitly requested by RD support.
Event 10228 is potentially useful in recovery situations since it
traces the application of redo. Therefore, if cache recovery
appears to hang or corrupt blocks then this can help track the dba
and operations upon it.
This event is available in Oracle 6 and 7.
The code just calls the same routines used to dump a redo log using
'alter system dump logfile' so has the same output.
The benefit is that it'll stop when it hits the corruption.
Articles:
Interpreting DUMP LOGFILE Output <Note:29726.1>
EVENT:10210 check data block integrity
2006/07/22 by Leave a Comment
Event:10210 See for similar events / parameters
~~~~~~~~~~~
Version/Use:
7.0 - 8.1.7 Check data block integrity after each modification
9.0+ This event is no longer valid. Use <Parameter:DB_BLOCK_CHECKING>
instead.
7.0 - 8.1.7 "Check data block integrity after each modification"
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
NOTE: Events should NEVER be set by customers unless advised to do so by
Oracle Support Services. Read before setting any event.
Summary Syntax:
This event should be set at instance level:
EVENT="10210 trace name context forever, level 10"
Levels:
This event is either ON (level 1 or higher) or OFF (not set)
Note that in 8.1.6 onwards you should use the
instead of this event.
Description/Steps:
Event 10210 enables data block integrity checking.
This causes the data layer to perform a number of checks on the
contents of a data block. If the block is found to be corrupt
in some way it is marked as SOFTWARE CORRUPT.
NB: Blocks are ONLY checked/corrupted when CHANGED. They are NOT
checked when accessed for read.
Take care with this, as if there are blocks in the database which
are being accessed happily but have a minor corruption such as
an incorrect free space count, this event will mark the block as
corrupt. Once marked SOFTWARE CORRUPT the data in the block
is not accessible to Oracle.
See for more detail of various block checking features.
Example Output / Interpreting Output:
This event will cause an ORA-600 error to be signalled if a corruption
is detected after a data block has been modified. The exact content
of the trace file depends on the ORA-600 and the action at the time
of the corruption being noticed. Refer to the relevant ORA-600 article
for the error which is signalled.
Related:
Database block checking features
Index and Cluster block equivalents
See event 10231 which allows SOFTWARE CORRUPT blocks
to be skipped on full table scans
