Free 1Z0-078 Practice Test

ACFS asynchronous replication copies both file data and associated metadata, including audit logs, from the primary to the standby file system. Starting with 12.2, the SYSREPL privilege enables administrators to execute acfsutil replication and disaster-recovery commands without full SYSASM rights. Because replication is configured per-file-system pair, a single site can simultaneously act as primary for one ACFS file system and standby for another, allowing flexible hub-and-spoke or peer configurations.

A. Replication stops only when free space is exhausted; no fixed 2 GB threshold is defined in documentation.
C. ACFS replication is independent of Redo transport; standby redo logs are a Data Guard requirement, not needed here.

1. Oracle Automatic Storage Management 19c Administrator’s Guide, Sec. 16.5 “About ACFS Replication”, paragraphs “Audit Information Replication” & “Role-based Security” (PDF pp. 16-12 – 16-13).

2. Oracle Automatic Storage Management 19c Administrator’s Guide, Sec. 16.7 “Configuring Multiple Replications”, Example 16-4 (PDF p. 16-19).

3. Oracle Automatic Storage Management 19c Administrator’s Guide, Sec. 16.6 “ACFS Replication Space Management” (PDF p. 16-14).

Oracle ADVM is a volume-manager layer built on ASM. Each ADVM volume has a unique name and can hold exactly one ACFS (or raw) file system. ADVM supplies operating-system volume devices that map to space inside an ASM disk group, thereby extending ASM with a standard block-device interface.

C. ACFS inside an ADVM volume can store any POSIX files, not only database files.
D. Multiple ADVM volumes (up to 63 per disk group) are allowed.
E. COMPATIBLE.ASM and COMPATIBLE.ADVM need be ≥11.2; 12.1 is not required.

1. Oracle® Automatic Storage Management Administrator’s Guide 19c, Sec. “Overview of Oracle ADVM” & “Creating an ADVM Volume”, p.4-24, 4-29

2. Oracle® ACFS and ADVM White Paper, July-2019, p.7 (“one file system per volume”)

3. Oracle® Automatic Storage Management Administrator’s Guide 19c, Table 4-2 (“COMPATIBLE.ADVM attribute must be 11.2 or higher”)

In a Flex Cluster, hub nodes function like traditional cluster nodes: they run Clusterware, access shared storage, and hold the voting disks and OCR. Therefore every hub node must see the common shared storage, and each hub node must have direct access to both the OCR and the voting files. Leaf nodes obtain required metadata through their associated hub node and never access shared storage directly.

B. Leaf nodes have no direct connection to any shared storage; they access it only through hub nodes.
C. Same reason as B; no leaf-to-leaf shared-storage requirement exists.
D. Voting disks are located on storage only hub nodes can access; leaf nodes do not see them.

1. Oracle Clusterware Administration and Deployment Guide 19c, Sec. 2.3.3 “Oracle Flex Clusters—Hub and Leaf Nodes” (Doc ID F15640-07).

2. Oracle Clusterware Admin Guide 19c, Sec. 5.2 “Oracle Clusterware Files (OCR and Voting Files)”—requires direct storage access only for hub nodes.

Connect-time load balancing can be performed either by the client or by the listener, but it occurs only when the initial connection is being made. Client-side CTLB (loadbalance=yes) randomly selects an address before contacting the listener (D). Server-side CTLB lets the listener choose an instance for the new session during the first handshake (B).

A – CTLB also works with EZCONNECT; it is not limited to tnsnames.ora entries. C – Server-side runtime load balancing (not connect-time) uses service response-time metrics; CTLB does not. E – The SCAN listener’s advisory-based routing is part of server-side load balancing, not client-side CTLB.

1. Oracle Net Services Administrator’s Guide 19c, ch. 12 “Load Balancing and Failover”, sections “Client-Side Connect-Time Load Balancing” and “Server-Side Connect-Time Load Balancing” (E96270-10).

2. Oracle Real Application Clusters Administration and Deployment Guide 19c, 7.2 “Using Services for Load Balancing” (F32770-10).

HANFS exports reside on ACFS volumes that in turn sit on ADVM volumes; hence I/O performance ultimately depends on the underlying ASM disks. Because the HANFS service is managed by Oracle Clusterware with a virtual IP, NFS shares automatically fail over to a surviving node when a cluster node crashes, providing high availability.

A. Client failures are outside the scope of HANFS high-availability design.
C. HANFS relies on the host OS NFS kernel modules to service requests.
E. Clients mount a HANFS export through its virtual IP, not via DNS round-robin.

1. Oracle® ACFS, ADVM, and HANFS Guide 19c, Chap. 6 “High Availability NFS (HANFS)”, p.6-3 – 6-5

2. Oracle® Grid Infrastructure 19c Administrator’s Guide, Sec. “Virtual IP addresses for HANFS”, p.14-12

On HOST01 the root user first disables autostart of Oracle Clusterware locally with crsctl disable crs, then stops the entire cluster with crsctl stop cluster -all, and finally restarts it with crsctl start cluster -all.
1. crsctl stop cluster -all halts the Clusterware stack on every node (A).
2. crsctl disable crs places a flag file (/etc/oracle/sclsdisabled) on HOST01 only, preventing the stack from starting automatically there after the next reboot (B).
3. crsctl start cluster -all brings the stack back up immediately on all nodes (F).

C – disable crs affects only the local node, not every node. D – stop cluster -all stops the stack on HOST01 as well, not “all except HOST01”. E – after the final start command the stack is running on HOST01 too, so not “all except HOST01”.

1. Oracle Clusterware 19c Admin & Deployment Guide, Sec. 5.2.1 “crsctl stop cluster”, Example 5-2 (stops on all nodes).

2. Ibid., Sec. 5.2.3 “crsctl start cluster”, Example 5-5 (starts on all nodes).

3. Ibid., Sec. 5.3 “Enabling and Disabling Oracle Clusterware”, paragraph “crsctl disable crs creates the sclsdisabled file on the local node only”.

CREATE PLUGGABLE DATABASE … FROM performs a local clone.
1. Because no OPEN READ ONLY clause is specified, the source PDB (pdb1) must be closed in the instance that issues the command (A).
2. Only a common user connected to the root with the CREATE PLUGGABLE DATABASE privilege can run the statement (D).
3. Oracle must know where to place the cloned files; either Oracle Managed Files must be enabled or PDBFILENAMECONVERT supplied (E). Restricted mode for all instances or special temporary tablespaces are not required, and the read-only requirement in NOARCHIVELOG mode is irrelevant to this syntax.

B – Read-only source is required only when CDB is in ARCHIVELOG; statement does not mandate NOARCHIVELOG. C – No requirement exists that the source PDB use a global temporary tablespace. F – Instances need not be in restricted mode; normal open is acceptable.

1. Oracle Multitenant Admin Guide 19c, Sec. 3.3.1 “Cloning a PDB” – prerequisites list (source PDB closed unless using READ ONLY clause).

2. Ibid., Sec. 3.2 “About Creating a PDB” – requirement for common user with CREATE PLUGGABLE DATABASE.

3. Ibid., Example 3-8 – need for OMF or PDBFILENAMECONVERT when cloning in the same CDB.

ASM stripes and mirrors files evenly across all disks in a disk group. Using disks of uniform size, rotational speed, interface, and latency ensures even I/O distribution and prevents the slowest or largest disks from becoming bottlenecks. Oracle best-practice documentation therefore recommends putting disks of identical capacity, performance characteristics, and storage attributes (sector size, redundancy capabilities) in the same ASM disk group for optimum throughput and balanced load.

A. Files of different types can reside in the same disk group; separation is optional, not a performance requirement.
C. Double mirroring (hardware + ASM) may add write overhead; it does not inherently “guarantee” better performance.
E. Having two disk groups for data and FRA is a design choice, not a prerequisite for performance.

1. Oracle Automatic Storage Management 19c Administrator’s Guide, Sec. 13.2 “ASM Best Practices”, bullet “Use disks that have similar performance and size” (PDF p. 13-4).

2. Oracle Database Storage Administrator’s Guide 19c, Sec. 2.6.2 “Striping and Mirroring Considerations” (PDF p. 2-12).

Oracle ACFS allows a writable snapshot (–w option) to be created only from a read-only parent snapshot. Therefore, before the second command— which uses –w—can succeed, the parent snapshot must be read-only.

A. ADVM compatibility 12.1 is needed only for features like tagged snapshots, not basic writable snapshots.
C. Snapshot creation is independent of redundancy level of the volume.
D. Writable parent snapshots cannot be the source for another writable snapshot.

1. Oracle® Automatic Storage Management Administrator’s Guide 19c, Sec. 16-5-2 “Creating Read-Write Snapshots” (E96273-11): “Read-write snapshots can be created only from a read-only snapshot.”

2. Oracle® Automatic Storage Management Administrator’s Guide 19c, Table 16-3 ADVM/ACFS Compatibility Requirements – no special compatibility needed for writable snapshots.

Secure Cluster Communication (SCC) is automatically enabled in Oracle Grid Infrastructure 19c. During either a fresh installation or an upgrade, Oracle Clusterware generates the required TLS certificates and keys, stores them in the cluster wallet, and begins protecting all inter-instance traffic without any manual intervention. No parameter setting, wallet path, or user-supplied key or certificate is required from database or cluster administrators.

B – Wallet path need not be provided; wallet is auto-created. C – Certificates are auto-generated but administrators need not invoke SCC explicitly. D – User-supplied encryption keys are not requested. E – User-supplied server certificates or wallets are not required.

1. Oracle Grid Infrastructure Security Guide 19c, Ch. 5 “Secure Cluster Communication”, first paragraph (“enabled by default… no administrator action is required”).

2. Oracle Release 19c New Features for GI, Sec. “Secure Cluster Communication Automatically Enabled”.