You can do that, but its better you copy them to each webserver.
Wikipedia uses Lucene for their searching, and they have a huge rsync job that goes on to keep all their search stuff uptodate. At least thats how they did it quite a few years ago when i spoke to Doug Cutting (the lead on Lucene/Hadoop). At that time we were supporting Lucene from SpikeSource
So when some one updated a wiki page, it took 30mins before it would appear in the search results.
For you it all depends on the size of your dataset and how many machines you are looking to sync with. But basically, only __ONE__ server should ever be updating the index.
Jason King wrote:
on my current webserver, the catalogs are in C:\openbd\webapps\website\WEB-INF\bluedragon\work\cfcollection Could I make that cfcollection folder a share on a NAS? and have each webserver mapped to the same catalogs? I know I can actually make the folder and link and all that, but would their be any locking issues i might have? For instance, if one webserver is making an update to the catalog, and another tries to update it, would it error out or just be patient and wait for it to unlock? Please advise.
-- Open BlueDragon Public Mailing List http://www.openbluedragon.org/ http://twitter.com/OpenBlueDragon mailing list - http://groups.google.com/group/openbd?hl=en !! save a network - please trim replies before posting !!
