Replication Pair

Overview

A Replication Pair is used to map a designated set of data on a source computer to a location on a destination computer. This monitors change on the source and replicate the changes to the destination computer.

Manage Replication Pair

The activity of Replication Pairs can be controlled individually, or when acting from the Replication Set level, multiple Replication Pairs are controlled at the same time. Activities related to a Replication Pair include:

Adding or Deleting a Replication Pair
Changing the state of a Replication Pair; Start, Suspend, Resume, or Abort data replication
Filtering content for all Replication Pairs in a Replication Set

For step-by-step instructions, see Add a Replication Pair

Common Base Folder

You can optionally specify a Common Base Folder, so that all destination paths for all Replication Pairs in the Replication Set will be created in the same folder on the destination host. This can be useful when you are adding multiple Replication Pairs at the same time, as in the case of replicating application data. For example:

If you specify C:\replication\dest as the base directory, your Replication Pairs might use the following as their destination directories:

C:\replication\dest\source_path_1\
C:\replication\dest\source_path_2\
C:\replication\dest\source_path_3\

If the source_path_1 is specified as C:\rep_pair_1 then the Replication Pair will use C:\replication\dest\C\rep_pair_1 as its destination directory.

Notes

If you change the settings for a Common Base Folder after any Replication Pairs have been created, the changes you make will affect only Replication Pairs created after that point; existing Replication Pairs will not be impacted by the change.
A Common Base Folder is specified in the Manage Pairs dialog. For instructions about how to specify a Common Base Folder, see Add a Replication Pair.
For more information about replicating application data, see Application Integration.

Replication Pair Conflicts

Certain Replication Pair combinations are not possible, or are only possible using different Replication Sets. The following table highlights the most common such cases, without enabling chaining for replication option in a control panel:

Type of Combination or Conflict	Same Replication Set	Different Replication Set	Examples Note that (A), (B), (C) in the examples represent different computers	Notes
Exact same destination path	No	No	(A) D:\test => (B) E:\test (A) F:\test => (B) E:\test
Child/parent destination paths	No	No	(A) D:\test => (B) E:\test (C) F:\test => (B) E:\test\test1 - - - - - - (A) D:\test => (B) E:\test (C) F:\ => (B) E:\
Source path same as destination (volumes are on different computers)	N/A	Yes	(A) D:\test => (B) D:\test (C) D:\test => (B) E:\test
Same destination path on multiple computers	N/A	Yes	(A) D:\test => (B) D:\test (A) D:\test => (C) D:\test
Same destination volume, different folders	Yes	Yes	(A) D:\test => (B) D:\test (C) E:\data => (B) D:\test1
Same source path in multiple pairs	No	Yes	(A) D:\test => (B) D:\test (A) D:\test => (C) E:\test
Source parent folder already used (Nested mount point is not supported.)	Yes	Yes	Existing: (A) D:\ => (B) D:\ New: (A) D:\data => (B) E:\data	The new pair will change to the parent path.
Source child folder already used	No	No	Existing: (A) D:\data => (B) E:\data New: (A) D:\ => (B) D:\
Destination parent folder already used	No	No	Existing: (A) E:\src => (B) G:\des New: (B) G:\des => (C) I:\ (B) G:\ => (C) I:\des1 (B) G:\data => (C) H:\test (B) G:\des\test => (C) I:

Notes

To enable chaining for replication to use destination paths as a source for replication pairs. see Enable Chaining for Replication.
For chaining, replication pair source and destination paths can be configured as follows:
(A) E:\test => (B) G:\test
(B) G:\test => (C) H:\data
(C) H:\data => (D) I:\test
Ensure that your destination data is not replicated back to any volume that is being used as a source as this may cause an endless loop. For example:
(A) E:\test => (B) G:\test
(B) G:\test => (C) H:\data
(C) H:\data => (D) I:\test
(D) I:\test => (A) E:\test or E:\
(OR)
(A) D:\test => (B) G:\data
(A) F:\test1 => (B) H:\data1
Note that when you create a new replication pair(s) as below then it will from an endless loop.
(B) G:\data => (A) F:\test1
(B) H:\data1 => (A) D:\test
Here (A), (B), (C), (D) are different computers

Important Considerations

When creating a Replication Pair, and selecting the source and destination paths for replication, consider the following:

Note

Attempting to use CDR to protect the system volume of a Windows computer is strongly discouraged. Due to the normally high rate of I/O on such a volume (e.g., the C: drive), it would prove very difficult to create a snapshot of it, since a sufficient period of no disk activity is required in order to create the snapshot. Similarly, a system volume is an inappropriate choice as a destination for data replication, and is also strongly discouraged.

Source

A specific directory or volume on a source computer can be replicated multiple times to different destination paths on either the same/different destination computer, or different destination computers, by creating a separate Replication Set and Pair for each different destination path.

For instance, Replication Set 1 could have a Replication Pair configured as follows:
- Source on Computer A: D:\
- Destination on Computer B: G:\A_to_B_1\
Replication Set 2 could have a Replication Pair configured as follows:
- Source on Computer A: D:\
- Destination on Computer B: H:\A_to_B_2\
When source data contains stubs from a data migration, the stubs themselves will be replicated, and not the actual data which has been migrated.
For a Replication Pair with the Automatically detect mount points option selected, for each mount point in the specified directory or volume, an additional, separate Replication Pair will be automatically created and its source path marked with an asterisk ("*") at the beginning to indicate it is a mount point.
It is possible to effect the initial transfer of data from a source to a destination without using the Baselining Phases. This can be useful when the connection between the source and the destination is constrained, such as a slow WAN connection. For step-by-step instructions to perform initial transfer of data, see Out Of Band Sync from a Replication Pair.

Destination

The source path cannot be a mapped network drive.
On Windows, a UNC path can be used as a destination. For data replication to succeed, the destination computer must have permissions to the location specified by the UNC path. Note that the UNC path can be specified to a NetApp file server. For more information, see ONTAP Snapshot for ContinuousDataReplicator.
When selecting a destination folder for a Replication Pair, ensure that the selected path does not contain any data as the existing data will be pruned during Orphan File processing.
It is recommended that, you do not include UNC path and iSCSI LUNs as destination path in the same Replication Set.