[mapguide-users] MG Scale Out - isolating MG Repository out of MG server process

[Jackie Ng]

Replication of resources is a can of worms. Repeating resource API calls
across multiple MapGuide Servers is easy. But:

 1. What's the proper strategy for handling CopyResource/MoveResource? What
if the source resource being copied/moved doesn't exist in one or more of
the servers?
 2. Following that, if a resource service API fails on one of the servers,
do we consider that success or failure (throw exception)? The current API
cannot model a partial success/failure.
 3. Or should we give up trying to make MgResourceService multi-server aware
for library resources and just make a new service that is a subset of
MgResourceService that targets this particular use case?

I'd like some further discussion on this instead of you making a ticket. An
enhancement ticket without some "meat" in how to achieve the end goal would
be sort of pointless right now.

- Jackie



--
View this message in context: http://osgeo-org.1560.x6.nabble.com/MG-Scale-Out-isolating-MG-Repository-out-of-MG-server-process-tp5207766p5208356.html
Sent from the MapGuide Users mailing list archive at Nabble.com.