[GeoNode-users] Resilient hosting of GeoNode

[Jac Steiner]

I'm working with a government that presently has a GeoNode instance
installed locally and accessible publicly from the internet. This is
fragile due to infrastructure weaknesses on premises (e.g. unreliable
power). I'm considering recommending they move to an instance of GeoNode
installed in the cloud (perhaps AWS, but a managed solution would be
ideal). This instance would be locally mirrored on site using their current
hardware. The local mirror is essential redundancy to accommodate scenarios
where the internet may go down (e.g. emergency management). The local
mirror should remain relatively in-sync with the cloud master, but
infrequent synching, perhaps once a day, should sufficiently cover the
anticipated usage patterns.

I figure the devil is in the details here, as in a genuine emergency, if
the internet is down and changes are being made locally directly on the
mirror, once internet is back, the synching strategy with the master
suddenly gets complicated.

I'm not able to find examples of this type of setup in the docs or
otherwise. As I'm not a sys admin, I may be using the wrong search terms.
Can someone point me toward resources that provide further details on
implementing this type of strategy and best practices? If anyone has
additional relevant insights they can share, I'd greatly appreciate it.

Appreciate your help.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/geonode-users/attachments/20180525/6a63c3c7/attachment.html>