Oops, Sorry. My bad. Think I meant SolrIndex not SolrInjector. Best check though! (away from my 'puter right now) If you run bin/nutch you will get a list of all the commands you can use. Pfft, so much for my howto skills!

Andy



On 7 Mar 2009, at 07:09, Tony Wang <ivyt...@gmail.com> wrote:

Hi Andy,

Sorry to bum this message again. I think I did everything right except the SolrInjector part. However, from another thread, mentioned by Toby: Maybe I should run as below after crawling data in nutch (as indicated in your step
3):

bin/nutch *org.apache.nutch.indexer.solr.SolrIndexer*
http://localhost:8080/solr/ /opt/tomcat6/webapps/nutch/data/crawldb
/opt/tomcat6/webapps/nutch/data/linkdb
/opt/tomcat6/webapps/nutch/data/segments

Then I run this: './crawl.sh data' (data is the folder that I store my crawl data). According to Toby again, this should put my nutch index in Solr
instances. However, when I do a search in the admin of Solr (
http://localhost:8080/solr/admin), any term doesn't yield any result though.

I think I'm getting closer to get nutch/solr integration done, but still
need a little bit clarification. Thanks!!!

Tony



2009/3/6 Andrew Smith <theid...@idrinktoomuch.org>

This works, (with the dev trunk. if you're using the 0.9 release then
there's a full tutorial on the nutch wiki) but it's not particularly
refined. I'm assuming you've got a working nutch crawl or whatever already.

 1. Get yourself a working solr test server running using the Getting
 Started section of the tutorial :
 http://lucene.apache.org/solr/tutorial.html
 2. modify the solr schema using the one found at

http://blog.foofactory.fi/2007/02/online-indexing-integrating-nutch-with.html(There's <http://blog.foofactory.fi/2007/02/online-indexing-integrating-nutch-with.html%28There%27s >
a missing field called boost with a type of int. There may well be
a more up to date version of the schema somewhere which doesn't have this
 error)
3. run your inject/fetch (or crawl, which will index unnecessarily but
 hey ho, it's simple...)
 4. run bin/nutch SolrInjector

Sorry, that's kida vague. You'll have to work out what parameters the
SolrInjector job wants for yourself, but if I can work it out then anyone
can! (well OK. maybe not my mum but apart from her...)

Andy

2009/3/6 Tony Wang <ivyt...@gmail.com>

Hi all,

For those on the Solr user list who have already seen my question about Nutch/Solr integration, I want to apologize for the redundant messages
and
I
don't mean to spam the two mailing lists. I have been desperately seeking
information/documentation on making Nutch crawl for Solr indexing.

So my question here is: Was anyone on the list able to make Solr use
Nutch's
crawl data? If yes, could you please share it with me and the rest of us?
I
sincerely appreciate any input.

Thanks in advance!

Tony

--
Are you RCholic? www.RCholic.com
ΠÁ¼ ¹§ ¼ó Èà ÈÊ Òå Àñ ÖÇ ÐÅ





--
Are you RCholic? www.RCholic.com
ΠÁ¼ ¹§ ¼ó Èà ÈÊ Òå Àñ ÖÇ ÐÅ

Reply via email to