[DRBD-user] Delayed initial synchronisation

Derek Danks derek.danks at comodo.com
Thu Apr 29 17:36:16 CEST 2010

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


Hi,
	As a newbie to DRBD could anyone answer the following or make reasonable 
suggestions? 
	Is it possible to create a DRBD cluster with only 1 node initialised? If it 
is how long would it work before lack of resources (disk space, I guess) would 
break it? 
Scenario:
	We have a machine with 20TB of data that we need to clusterize in location A 
and another machine with enough space available in location B. The idea is to 
create multiple resources on the machine in location B and copy the data from 
the machine in location A to the machine in location B. Then access the data 
from the machine in location B. After all the data is copied to the machine in 
location B we would like to then create the resources on the machine in 
location A and start synchronisation between the 2. As the 2 locations are 
separated by a great distance the data copy will take quite some time hence 
the second question. The reason for the first question is that we don't have 
enough available space in location A to create the resources initially. 

TIA

Regards Derek



More information about the drbd-user mailing list