Defining source and destination through snapmirror.conf

If it does not already exist, use a text editor to create a file called snapmirror.conf in the /etc directory on the root volume of the destination storage system.

If you want to add comments to the /etc/snapmirror.conf file, precede the comment with a pound sign (#).

Example

# Replicating from systemA

For each destination volume or qtree that is to be located on this storage system, type an entry specifying the source, destination, characteristics and schedule of the data transfer on one line using the following syntax:

source_system:{source_volume | 
/vol/volume_name/qtree_name} dest_system:{dest_volume | 
/vol/volume_name/qtree_name} arguments schedule

For a full description of the snapmirror.conf file entry syntax see snapmirror.conf file entry syntax.

Note
When you use a qtree as a source, you do not create another qtree to serve as the destination. SnapMirror automatically creates one for you, using the name you specify. However, you must specify the name of the qtree destination in the volume where you want it to be placed, either in the /etc/snapmirror.conf file or in the snapmirror initialize command.

Note
Do not specify more than 254 destination qtrees for any one volume in your /etc/snapmirror.conf file entries.

If you want to combine all non-qtree data in a volume into a qtree to use as a SnapMirror source, use this syntax:

source_system:/vol/source_volume/- 
dest_system:/vol/dest_volume/qtree_name

The dash (-) character indicates all non-qtree data in the specified volume.

Note
The data in /vol/source_volume/- qtree can only be a SnapMirror source, never a destination. That is, after you create the /- qtree, you can copy data from it to another qtree you name, but you cannot copy data to it from another qtree.

Save edits to the file.

src_system

The name of the storage system from which you are copying data. SnapMirror uses the /etc/hosts file or the database used by DNS and NIS for name resolution. When SnapMirror searches for the source system name, it should find the IP address for the source system on the network over which you want the transfer to occur.

Example

Assume you created a private network connecting a source system and a destination system and you named the interface on the source system systemA-e0. The interface name systemA-e0 is what you enter in the source_system field.

src_vol
/src_qtree

The name of the volume or qtree that you are copying. Use the volume name alone for volumes. Use the full path name for qtrees.

Example

If the name of the volume is vol1, enter vol1 in the src_vol/src_qtree field. If the name of the qtree is qtree3, and it is contained in volume vol3, enter the full path, /vol/vol3/qtree3, in the src_vol/src_qtree field.

dest_system

The host name of the system to which the data is copied. The name you use must be the exact host name of the destination system. The dest_system field is case-sensitive.

You can use the hostname command on the destination system to determine what you enter for this field.

Example

If the name of the destination system is systemA, enter systemA in the dest_system field.

dest_vol
[/dest_qtree]

The name of the destination volume or qtree to which you are copying data. Use the volume name alone for volumes. Use the full path name for qtrees.

Example

If the name of the volume is vol1, enter vol1 in the dest_vol[/dest_qtree] field. If the name of the qtree is qtree4, and it is in vol2, enter the full path, /vol/vol2/qtree4, in the dest_vol[/dest_qtree] field.

kbs=kbs

Maximum transfer speed, in kilobytes per second, that Data ONTAP can use to transfer data.

The kbs and restart arguments are expressed as a comma-separated list of name=value pairs, with no spaces: for example, kbs=2000,restart=always.

restart=
{never | always | default}

Restart mode that SnapMirror uses to continue an incremental transfer from a checkpoint if it is interrupted.

There are three options:

• Never--Transfers are always restarted from the beginning of a transfer and never from where they were before an interruption. This mode is useful if you must have the latest data on the destination.
• Always--Transfers are always restarted if possible from where they were before an interruption. This mode is useful for copying large volumes.
• Default--Transfers are restarted if they do not conflict with a scheduled transfer. This is the recommended option.

  SnapMirror always restarts from where the transfer was before an interruption. However, it does not restart if the restart occurs after a scheduled transfer, because the scheduled transfer gives more up-to-date data than the restarted transfer.

The kbs and restart arguments are expressed as a comma-separated list of name=value pairs, with no spaces: for example, kbs=2000,restart=always.

cksum=
{none | crc32c}

Selects the checksum algorithm that is used to protect SnapMirror transmitted data.

outstanding=
{xops | xms | xs}

Determines the synchronicity level of the synchronous SnapMirror. The ops suffix allows x number of outstanding write operations before forcing the client to wait for an acknowledgement. The ms and s suffixes allow x number of milliseconds and seconds, respectively.

visibility_
interval={xs | xm | xh}

Determines the amount of time before an automatic Snapshot copy is created on the source volume that is synchronously mirrored. With synchronous SnapMirror, changes to the source volume do not show immediately on the destination volume, even though the changes have been mirrored. The changes are shown only after the source system takes an automatic Snapshot copy of the source volume. This happens every three minutes by default. You can change the interval for automatic Snapshot copies, but performance can degrade if you set smaller intervals because more Snapshot copies are taken more often. The smallest interval you can set is 30 seconds.

The s, m, and h suffixes specify seconds, minutes, and hours, respectively.

wsize=size

Determines the TCP window size used by a connection. Different from the snapmirror.window_size option which determines the TCP window size for allSnapMirror replications. Use the formula shown in Adjusting the TCP window size to calculate a TCP window size that works well for a particular connection. TCP window size is in bytes.

schedule

Determines the schedule the destination uses for updating data. The schedule is mandatory.

The schedule consists of four space-separated fields in order:

minute hour dayofmonth dayofweek

minute can be a value from 0 to 59.

hour can be a value from 0 (midnight) to 23 (11 p.m.).

dayofmonth can be a value from 1 to 31.

dayofweek can be a value from 0 (Sunday) to 6 (Saturday).

Multiple values, separated by commas, can be entered for any field.

All possible values for a field can be applied with an asterisk (*). If you specify an asterisk in each field of the schedule, SnapMirror updates the destination every minute.

A single dash (-) in any field means "never" and prevents this schedule entry from executing. (This is useful if you want the server to appear in the /etc/snapmirror.conf file so that snapmirror update can find it, but you do not want the SnapMirror scheduler to run automatically.)

A range of values for any field can be indicated with a low value and a high value separated by a dash. For example, you can indicate that you want an update every hour from 8:00 a.m. to 5:00 p.m by entering this value in the hour field:

8-17

A range of values followed by a slash and a number indicates the frequency of the update. For example, you can indicate that you want an update every five minutes by entering this value in the minutes field:

0-59/5

Typing sync instead of the four space-separated fields specifies synchronous replication. See the na_snapmirror.conf(5) man page for more information.