On May 25, 2009, at 8:51 AM, Mark Miller wrote:
Erik Hatcher wrote:
I saw Peter Wolanin mention this on IRC's #solr the other day (Hi
Peter, nice meeting you at the meetup the other night!), and I just
encountered it too:
HTTP ERROR: 500
org.apache.lucene.index.ReadOnlySegmentReader cannot be cast to
org.apache.solr.search.SolrIndexReader
java.lang.ClassCastException:
org.apache.lucene.index.ReadOnlySegmentReader cannot be cast to
org.apache.solr.search.SolrIndexReader
at org.apache.solr.search.DocSetBase$1.getDocIdSet(DocSet.java:
254)
at
org.apache.lucene.search.IndexSearcher.doSearch(IndexSearcher.java:
237)
at
org.apache.lucene.search.IndexSearcher.search(IndexSearcher.java:221)
This comes from using an fq parameter:
http://localhost:8983/solr/select?q=*:*&fq=ipod
Works fine without fq.
Yikes!
Erik
Hmmm - what did you do leading up to it? I just tried a trunk
example, load docs, try search, and that doesn't seem to tickle it.
It appears that somehow, some code is passing in a lucene reader
that didnt come from SolrIndexSearcher (which wraps indexreader to
solrindexreader). I can't find the code that would be doing that
though. Was looking for a deeper stacktrace or something, but havn't
been able to tickle the bug yet.
Hmmm.... I guess I shouldn't have left out a couple of details, I'm
launching Solr like this: java -Dsolr.solr.home=../contrib/velocity/
src/main/solr/ -jar start.jar and I'm using the May 24 nightly build.
So it is using a different configuration, which is mostly a clone of
the example config, but apparently different in some respect. I'll
try with trunk and see. I mentioned trunk in the subject, sorry about
the misdirection.
I'll try on trunk now.
Erik
