Oracle Database Administration II key terms

A store of backup metadata such as backup pieces, archived logs, and scripts, kept either in the target database control file or in a dedicated recovery catalog schema.

A separate schema holding RMAN metadata for one or more target databases, surviving control file loss and enabling longer retention and centralized reporting.

An automatic RMAN backup of the control file and server parameter file taken after structural changes or backups, allowing recovery without a catalog.

A point-in-time copy of the control file that RMAN reads to get a consistent view during resynchronization and backup operations.

A rule stating how long backups must remain restorable, expressed as a recovery window or a redundancy count of backups per file.

A full backup of all data blocks in the chosen files; it forms the base that later differential or cumulative incrementals are applied to.

A level 1 backup containing blocks changed since the most recent level 0 or level 1 backup of the same file, so restores need the latest incremental only.

A level 1 backup containing all blocks changed since the last level 0 backup, reducing restore steps but increasing backup size.

A background feature that records changed block addresses in a bitmap file so incremental backups can skip reading unchanged blocks.

A byte-for-byte copy of a data file, archived log, or control file made by RMAN, usable directly by the database without a restore step.

An RMAN-generated binary file holding one or more backed-up files in a proprietary format that must be restored before use.

A managed disk location for archived redo logs, backups, flashback logs, and control file autobackups, with space governed by two initialization parameters.

The step that copies backed-up files such as data files or archived logs from backup media back to disk, without applying redo.

The step that applies archived and online redo to restored files to roll them forward to a chosen system change number or time.

Media recovery stopped at a specified time, SCN, or log sequence, producing a database that has lost transactions committed after that point.

An operation that opens the database after incomplete recovery, resets the log sequence to one, and starts a new incarnation of the database.

A distinct branch of a database's redo history created by each OPEN RESETLOGS, identified by a number that RMAN uses to pick valid backups.

Recovery of one or more tablespaces to an earlier time without affecting the rest of the database, typically done in a temporary auxiliary instance.

A temporary instance created with a backup control file to perform work such as TSPITR, duplicate database, or cross-platform conversion.

A diagnostic facility that detects data failures, assesses which are critical, and generates RMAN repair scripts the administrator can run.

A feature that rewinds the whole database to a past time using flashback logs, without restoring files, then opens it with RESETLOGS.

Special logs written to the fast recovery area that record before-images of changed blocks, enabling fast rewind of the whole database.

A named restore point that the database promises to keep reachable, making flashback logs or backups non-deletable until it is dropped.

An operation that returns a table to its state at a past time or SCN using undo data, avoiding a full restore.

A query that reads data as it appeared at an earlier time or SCN by using undo information, often with the AS OF clause.

A long-term history store for row changes in tracked tables, retaining versions beyond undo retention and enabling queries far into the past.

A multitenant database that holds a root container, a seed, and one or more pluggable databases sharing one instance and background processes.

A self-contained set of data files and metadata that appears as a normal database to applications but runs inside a container database.

The special container named CDB$ROOT that owns common users, common objects, and metadata shared by all pluggable databases in the CDB.

A read-only template pluggable database, PDB$SEED, used as the source when creating new PDBs, and which cannot be opened read-write.

A database user defined in the root whose name begins with C## and which can connect to any container where it has privileges.

A user created inside a single pluggable database, visible only there, typically used for application accounts in that PDB.

Moving a PDB between CDBs by generating an XML manifest, closing and unplugging it, then plugging it in elsewhere and opening it.

Creating a new PDB from an existing one, either from a local source, a remote CDB over a database link, or a snapshot copy.

A storage-efficient PDB whose data files share blocks with a source PDB until changes are made, requiring a storage system that supports it.

A rule inside a resource plan that maps a consumer group to CPU, parallelism, or session limits, or that switches sessions between groups.

A named collection of sessions that Resource Manager treats as one unit when allocating CPU, parallel servers, or active sessions.

A method of moving a set of tablespaces between databases by copying data files and importing their metadata from an export dump.

A set of accepted execution plans for a SQL statement that the optimizer is restricted to, protecting performance from plan regressions.

A built-in repository of performance statistics snapshots, used to compare time periods and feed advisory tools such as the SQL Tuning Advisor.

Scheduled jobs maintained by the database, including optimizer statistics gathering, Automatic Segment Advisor, and Automatic SQL Tuning Advisor runs.