acfsutil repl init

Purpose

Initiates replication on all the files in an Oracle ACFS file system or only those files with a specified list of tags.

Syntax and Description

acfsutil repl init -h
acfsutil repl init primary -s standby_connect_string [tagname...]
     [-m  standby_mount_point] [-c primary_service]
     [-d trace_level] [-z {on|off}] mount_point
acfsutil repl init standby -p primary_connect_string
     [-c standby_service] [-d trace_level] mount_point

acfsutil repl init -h displays help text and exits.

Table 16-33 contains the options available with the acfsutil repl init command.


Table 16-33 Options for the acfsutil repl init command

Option Description

primary

Specifies replication of the primary file system.

standby

Specifies replication of the standby file system.

-s standby_connect_string

Specifies the connection string for the site hosting the standby file system. The user in the connection string must have SYSASM and SYSDBA privileges. Oracle wallets should be used to manage security credentials.

-p primary_connect_string

Specifies the connection string for the site hosting the primary file system. The user in the connection string must have SYSASM and SYSDBA privileges. Oracle wallets should be used to manage security credentials.

tagname

Specifies the tag names of the files to be replicated.

-m standby_mount_point

Specifies the directory where the standby file system is mounted if it is different than the primary file system. This is valid only for the primary file system initialization.

-c primary_service

Specifies the service name for the primary file system. Required if the primary file system is using a different service name than the standby file system.

-c standby_service

Specifies the service name for the standby file system. Required if the standby file system is using a different service name than the primary file system.

-d trace_level

Specifies the trace level setting (0..6) for replication logs.

-z on|off

Turn on or off compression on replication logs. Compression is off by default.

mount_point

Specifies the directory where the file system is mounted. For repl init primary, mount_point specifies the primary file system. For repl init standby, mount_point specifies the standby file system.


When the acfsutil repl init primary command completes successfully, processing begins to replicate initial copies of all specified files to the standby file system. In addition, any changes to these files and any new files subsequently created in the file system on the primary file system are captured and transported to the standby file system. However, successful completion of the command does not indicate that the specified files have been copied in full to the standby file system. To monitor the progress of the initial replication of all specified files to the standby file system, users run the acfsutil repl info -c command.

The first phase of the initial replication copy process begins with a complete copy of the file system directory tree structure, not the files within the directory tree, to the standby file system. This copy process includes directories without specified tags. During the copy process, attempts to rename directories may fail. After the directory tree structure has been successfully copied to the standby file system, replication initialization begins the second phase of the initial copy process: copying all files to be replicated to the standby file system. While a file is being copied to the standby file system during this initial replication phase, writes and truncates to that file are blocked.

Symbolic link files are replicated as is. If the symbolic link resolves to an absolute path name, and that path name does not exist on the standby file system, then referencing the symbolic link results in errors.

A new service name must be created for replication. You cannot use the +ASM service name. You must specify a unique service name for each file system that you want to replicate when there are multiple replicated file systems on a node or cluster.

A file system being replicated can only be mounted on one path for the primary or standby file system. Otherwise, dismounting any path on a node where it is mounted on multiple paths stops the replication background threads.

A replicated Oracle ACFS file system must be mounted on the same directory within a cluster. The file system does not have to be mounted on all of the nodes in a cluster, but if it is mounted, it must be mounted on the directory specified with the acfsutil repl init command. In addition, no other file system can be mounted on that directory on other nodes.

A single Oracle ACFS file system cannot be configured both as a primary and a standby file system. If replication is configured on the same host or within the same cluster, then the primary and standby file systems cannot use the same named mount point.

The primary file system must have a minimum size of 4 GB for each node that is mounting the file system. The standby file system must have a minimum size of 4 GB and should be sized appropriately for the amount of data being replicated and the space necessary for the replication logs sent from the primary file system.

The sites hosting the primary and standby file systems must be running the same operating system and must have the same system architecture.

Oracle wallets should be used to manage security credentials.

See Also:

You must have system administrator privileges to run this command.

Examples

Example 16-27 shows the use of the acfsutil repl init command to set up the primary and standby file systems. In these examples, the connect string searches in the user's wallet for the user name and password associated with the alias primary_repl_site or primary_repl_site.

Example 16-27 Using the acfsutil repl init command

$ /sbin/acfsutil repl init primary 
     -s /@standby_repl_site
     -m /standby/repl_data -c primary_repl_service 
     /acfsmounts/repl_data

$ /sbin/acfsutil repl init standby 
    -p /@primary_repl_site
    -c standby_repl_service /standby/repl_data