The user data gets pushed from the remote nodes back to the central node (which is Internet connected). The ddoc changes get pushed from the central node to the remote nodes.
Stefan, that's what I was looking into. I have filters for user data and system data (ddocs and other dependent documents). I may go with a two step approach to just send the user data with a capture of the _rev_diffs output. As for ddocs, I might just need to use git bundles and kanso on all the nodes. On Mon, Jan 27, 2014 at 12:49 PM, Stefan Klein <[email protected]>wrote: > 2014-01-27 Aaron Huslage <[email protected]> > > > > > These records need to be replicated back to a central office for > reporting > > purposes. > > > > These reports are so complex you have to have all the data on the central > server? > Can't you reduce the data on the local servers using a view, and just > transfer the view result to the central server? > If it's just for reports, maybe you can capture a (filtered) _changes > stream into a file and transfer this? > -- Aaron Huslage +1-919-600-1712 IM: GTalk - [email protected]; Skype - huslage PGP: 0x360DE590 / 30D4 B15B 74B2 F0E9 35D4 ABE0 2CB7 AC60 360D E590
