Tuning Recovery - Expert Oracle RAC Performance Diagnostics and Tuning - page 328

kjfmPriCheck: cannot make further changes to LMS priorities - single-inst mode
.....
.....
.....
*** 2014-04-02 22:27:21.439
global enqueue service detaching from CM:pid=22953
2014-04-02 22:27:21.479573 : kjxgrdisable: IMR recording device closed, terminating IMR
Crash Recovery
Crash recovery is the failure of all instances in a RAC environment or one instance in a single-instance environment.
Operations are identical to an instance recovery; however, recovery is performed by the first instance that is started
following a crash.
Thread Recovery
Thread recovery is used in both the crash recovery and instance recovery situations. The aim of a thread recovery is to
restore the data block images that were in the cache of the failed instance but had not been checkpointed back to disk.
One redo thread is processed at a time, beginning from the last checkpoint and continuing to the end of the
thread. All redo is applied from each thread before the next thread is started.
Online Block Level Recovery
Online block recovery is unique to the RAC implementation. Online block recovery occurs when a data buffer
becomes corrupt in an instance's cache. Block recovery will occur if either a foreground process dies while applying
changes or if an error is generated during redo application. If the block recovery is to be performed because of
the foreground process dying, then PMON initiates online block recovery. However, if this is not the case, then the
foreground process attempts to make an online recovery of the block.
Under normal circumstances, this involves finding the block's predecessor and applying redo records to this
predecessor from the online logs of the local instance. However, under the cache fusion architecture, copies of blocks
are available in the cache of other instances, and therefore, the predecessor is the most recent past image (PI) for that
buffer that exists in the cache of another instance. If, under certain circumstances, there is no PI for the corrupted
buffer, the block image from the disk data is used as the predecessor image.
Media Recovery
Any database is prone to failures; and during such failures, there could be situations when there is loss of data
either due to data corruption or human error or unforeseen disaster. In the case of the initial two situations, the
database is normally restored either completely, for example, when a disk goes bad or partially (point in time) when
a specific object needs to be restored. In the third situation, “unforeseen disaster,” a new database will need to be
configured and the data restored to it (if the external media is available); or a disaster recovery strategy will need to
be implemented. This strategy will require using tools such as Data Guard or Oracle Streams that will allow users to
connect to this disaster recovery location when the primary database is down.