Hi, ----- Original Message ---- From: Ryan McKinley <[EMAIL PROTECTED]> To: solr-user@lucene.apache.org Sent: Friday, December 7, 2007 1:13:02 PM Subject: Re: spellchecker and multi-core index replication
> > 2) with multi-core stuff getting in, is Bill Au going to be adding support for multiple indices anyway? > As is, you can use all existing scripts for each index. You just need to run snapwhatever for each index. OG: Are you referring to the -d param? For example, snapshooter -d? In case of spellchecker index, I believe the index lives under the main 'data' directory and I think that is the problem. I think -d would work if each index/core had its own separate data directory, for example: solr/main/data/index -- main index solr/spell/data/index -- spellchecker index This is much like the core entries in multicore.xml now, no? Perhaps if distribution could be handled for a directory that is not necessarily the index (maybe a folder with stuff + index*) then you could have: <multicore adminPath="/admin/multicore"> <core name="core0" instanceDir="core0" dataDir="/path/to/data/core0" /> <core name="core1" instanceDir="core1" dataDir="/path/to/data/core1" /> </multicore> Then if /path/to/data was distributed, it could have: /path/to/data/core0/index/... /path/to/data/core0/boost.xml (see SOLR-418) /path/to/data/core1/index/... ... OG: Yes, I think that makes sense - distribute everything for a given core, not just its index. And the spellchecker could then also have its data dir (and only index/ underneath really) and be replicated in the same fashion. Right? Otis
