×
The internal search function is temporarily non-functional. The current search engine is no longer viable and we are researching alternatives.
As a stop gap measure, we are using Google's custom search engine service.
If you know of an easy to use, open source, search engine ... please contact support@midrange.com.
Mike,
An idea to ponder. I set up a CL program that would use SAVRST and
SAVRSTDLO to move only changed objects to a remote site periodically.
At this customer it happens to be every 2 hours. It pushes both IFS and
DLO objects that have been changed over to a remote site. Since the
rate of change in the IFS/DLO is reasonably low, this technique works
quite well. The way it works is it submits itself to a job queue to run
in a configurable amount of time into the future, then runs the back up
and restores.
This works really well with smaller objects in DB/2 as well, but
anything of any size might run into a problem.
The downside is you have to set up the communications using enterprise
extenders, which of course is free from IBM. The upside, your recovery
point is at what ever time you set in the program to run.
Then use a data replication tool to push most of the DB/2 data to a
remote site. It does not need to be a full HA if you don't need that.
By using these types of techniques, you'll have all the data ( DB/2 is
very few seconds, and IFS/DLO is a small time frame ) at your remote
site in the cloud, and recovery can be reasonably fast ( within a couple
of hours ) since the system on the other end is by definition a Power
Systems box running IBM i.
Does this scenario fit everyone, nope, but it's a nice half step between
an HA solution and nothing at all, with a cost that's reasonable as
well. There are several vendors that have the capability of putting
this together for you.
Jim Oberholtzer
Chief Technical Architect
Agile Technology Architects
On 7/26/2013 7:14 AM, Mike Cunningham wrote:
It's good to see we have considered the same issues as we talked here. We currently have around 750GB used space on a 1.2TB system. We do a full user library backup and a full IFS (we store scanned documents in the IFS) backup nightly. We do not do only changed objects. Makes restore so much easier. We do a full system save monthly. If we made this change I would still do the system save and have it local, off site in a vault. I would probably still keep a tape of the nightly backup locally for quick access for non-disaster restores and instead of driving the tapes to the vault daily, replicate the backup to the cloud.
midrange.com
