33 Posts

April 20th, 2020 01:00

Your theory is great, but I have a bad new.

The replication from D1 to D3, still needs to read and split in chunks of information all the data.

For every chunk it will ask D3... Do you have this fingerprint?

If target has the reference, the data will not be sent over the network. If it does not have the reference it will sent it over the network.

In few words... even when the data is already at destination (as you stated), it is not going to be very fast, as it still needs to read the data.

Bear in mind also, that an mtree replication cannot be configured against an mtree that exists, so you probably will have to rename the mtree at D3, before establishing the new replication context.

Cheers, 

Gonzalo

No Events found!

Top