A good strategy for disaster recovery
Is it possible to snapshot an entire OS onto a SAN, replicate the OS to a DR site's SAN, then bring the system up identical to the first at the DR site including IPs? Additionally, can this be done for an entire network?
Yup, I'm doing this at a number of sites, for both Solaris and Windows. You will need a VLAN between the sites, your infrastructure servers -- DNS, Active Directory, PDC, BDC, DHCP, etc. -- need to span both sites. You will also need to boot into the SAN
and replicate the root volumes for all your application servers, along with all the data drives so that when the servers come up at the remote site they will assume the same personality.
This is a good strategy for DR. All data is replicated
between sites and both sides run production. During a failure a number of spare servers are used to bring up critical applications from the failed site and when they boot to the replicated root volumes, they assume the same personality of the server at the failed site.
Depending on the distance -- within SYNC
replication distance -- you can even create "stretched clusters" that automatically fail over resources to the surviving site. MSCS, VMS, VCS, GDPS, etc. clustering
solutions can be used in conjunction with data replication solutions like PPRC, SRDF, Truecopy, DRM, etc. to make this viable.
Editor's note: Do you agree with this expert's response? If you have more to share, post it in one of our
The views and opinions expressed by Christopher L. Poelker are his alone and not necessarily shared by Hitachi Data Systems.
This was first published in April 2004