1 Nickel

IP/ VNX Replicator Update SYNC

Hi There,

I currently have this situation in hand.We are now in the midst of working out a solution to refresh customer existing NS-120 – one in Egypt and one in Malaysia with one VNX5200 each. Currently the NS-120 is replicating to the other NS-120 using replicator (file only configuration – both NFS and CIFS shares). We are now trying to finalize the migration approach and we want to seek your guidance on this. Constraints that the customer has is that the bandwidth between Egypt and Malaysia is very small (to do a full initial sync between sites is difficult) and tape silvering is out of the question (they are using Backup Exec which is not supported).

Initial plan was to ship two VNX5200s to Egypt, do the sync locally on site in Egypt and ship back one of the VNX5200s to Malaysia but obviously it is a very expensive way logistic wise to do so

There was another suggestion  to

  1. move the data from NS-120 to VNX5200 would be via  IP/VNX Replicator at each site.
  2. Once both sites has been replicated and cut over to the new VNX5200, we will then create a new  replication set and replicate  back from Egypt’s VNX5200 to Malaysia’s VNX5200 incrementally. Is this even possible?  With data already in the VNX5200 filesystem now, would we be able to perform a incremental SYNC update from Egypt's VNX5200  to Malaysia's VNX5200 or will it be a Full SYNC?

0 Kudos
2 Replies
3 Zinc

Re: IP/ VNX Replicator Update SYNC

The VNX Replicator feature you might be interested in is "Incremental Attach".  You can use this to sync a common base checkpoint, then incrementally sync from there.  There's a good discussion that lays out the features here.

There's also a bit of an overview in the Using VNX Replicator documentation as well on this topic.

Let us know if this helps!


0 Kudos
5 Rhenium

Re: IP/ VNX Replicator Update SYNC

Its a bit difficult to find and can also be used for other purposes.

Basically it allows to refresh two checkpoints that aren’t directly connected

i.e. if you are replicating A -> B-> C then you can use it to make user checkpoints on A and C in sync

so that they can be used as a common base ckpt for a later rep between A and C without a baseline transfer

the command is nas_replicate – refresh A –source A_ckpt –destination C_ckpt

0 Kudos