Data Replication Using SnapMirror16- 11
time. If the filer does not have enough resources to perform all scheduled mirror
updates, it postpones some updates until it has enough resources to do so. As a
result, the filer might need to perform subsequent updates at times that are dif-
ferent from those you specify in the /etc/snapmirror.conf file.
During data replication, the filer copies data from all snapshots from the source
volume. Therefore, the filer must preserve all snapshots when data replication is
in progress, even though the snapshot schedule might call for the deletion of
some snapshots. If you want the filer to delete snapshots at the exact time spec-
ified by the snap sched command, schedule the mirror updates to be different
from the snapshot deletion times.
The filer transfers data faster if the source volume and the mirror have the follow-
ing characteristics:
They contain disks of the same size.
They contain RAID groups of the same size.
They contain the same number of RAID groups.
If the source volume has quotas enabled and you want to apply the same quota
restrictions to the mirror after converting the mirror to a regular volume, make
sure that the filer on which the mirror resides contains a copy of the
/etc/quotas
file from the source filer. Whenever you make a change to the /etc/quotas file on
the source filer, make the same change to the copy of the /etc/quotas file on the
destination filer.
6WHSV
Follow these steps to replicate a volume:
1. Add the host name of the destination filer to the /etc/snapmirror.allow file on the
source filer.
2. Edit the /etc/snapmirror.conf file on the destination filer to specify the volume to
be replicated and the schedule at which the mirror is updated.
3. Enter the vol snapmirror on command on both the source filer and destina-
tion filer to enable SnapMirror.
Result: The filer reads the /etc/snapmirror.conf file. If the filer is a destination filer
specified in the /etc/snapmirror.conf file, it establishes a connection with the
source filer at the time of the scheduled mirror update.
If a baseline version of the mirror does not exist, the filer takes a snapshot of the
source volume at the time of the scheduled mirror update and transfers all data in
the snapshot from the source volume to the mirror.
The filer does not display any messages if it can transfer the data successfully.
You can check the data replication status following the instructions in Checking
Data Replication Status.
The filer makes subsequent updates to the mirror according to the schedule
specified in the /etc/snapmirror.conf file.