Thinking about this a bit more, it seems that the only real disadvantage of doing this is that since it's DSL, the upload speed will be slow. This could make massive restore operations quite slow. We could offset this by continuing to do one weekly rsnapshot on the local disks of fyodor, and then have that be a starting point for the restore operation.
Let me know what you think. Justin On Fri, Jul 28, 2006 at 09:44:20AM -0700, Adam Chlipala wrote: > Justin S. Leitgeb wrote: > > >At any rate, I have 100 GB of unused disk space on a linux box with a DSL > >connection -- we could set up an account for HCoop rsync and take away this > >disk activity immediately, perhaps resulting in a usable system that would > >work for at least a couple of months while we stabilize membership a bit > >more on fyodor. I don't see how this would take more than a few minutes to > >set up. > > > > > I'm up for this, unless there are any objections. > > The main things to watch for will relate to exceeding our bandwidth > quota based on transferring too much of our hard drive to a far-off land > on a regular basis. We currently have daily, weekly, and monthly > back-ups. Perhaps we could start with that, changing the rsnapshot > configuration only to modify the destination and play it by ear to see > if this might manage to generate the ~1 TB/mo. of traffic we would need > to be in trouble? > > _______________________________________________ > HCoop-SysAdmin mailing list > [EMAIL PROTECTED] > http://hcoop.net/cgi-bin/mailman/listinfo/hcoop-sysadmin _______________________________________________ HCoop-SysAdmin mailing list [EMAIL PROTECTED] http://hcoop.net/cgi-bin/mailman/listinfo/hcoop-sysadmin
