White Papers

45 Using Dell PS Series Asynchronous Replication | TR1052
6 Troubleshooting
This section addresses common problems with replication.
6.1.1 Replication fails due to lack of local replication reserve
Local replication reserve is the amount of space needed on the original group to keep track of changes
that occur to the volume while it is being replicated. A volume can run out of local replication reserve if
the volume grows in size (for example, a thin-provisioned volume that increases automatically), or if more
data is written to the volume than originally planned.
If a replication fails because there is not enough local replication reserve, do one or both of the following:
 Increase the replication reserve if there is enough free space on the group
 Allow the volume to borrow from free pool space (free space must be available). For more
information on space borrowing, refer to Space Borrowing for Snapshots and Replicas .
6.1.2 Replication times out or cannot complete before the next scheduled
replication
If you have a very large, busy volume with data that changes frequently, a lot of data must be replicated at
each scheduled interval. If your replication schedule is set to replicate too often, the previous replication
may not have had time to complete before the next one is scheduled to begin. You can try any or all of
the following:
 Increase the interval between replications to allow more time for transfers to complete.
 If the network is too congested at the times the schedule is trying to run, adjust the run time to a
less-busy time of day.
 Increase network bandwidth, or add a WAN optimizer.
 For very large transfers, or if the network bandwidth is insufficient, use the Manual Transfer Utility
to transfer the data using removable media, taking the replication completely off the network.
Note: You cannot create a schedule for manual transfers. For additional information on the Manual
Transfer Utility, please refer to the Manual Transfer Utility Version 1.2.3 Installation and User's Guide on
eqlsupport.dell.com.
6.1.3 Not enough replicas are being kept
If you have adjusted the frequency of a replication schedule and are now replicating more often but
transferring less data each time, you may want to increase the number of replicas being stored on the
partner.
For example, if you originally set up a schedule to replicate once per day and keep a week’s worth of
replicas (7), but then increase the frequency to twice per day, the replica set will still contain only 7