It was/is a client issue. The googlebot executed a request where the search term was CP1252 encoded, maybe the brokenn url got indexed from a third party site.
We solved the issue by filtering all non valid utf8 chars from the search input string because there is no proper way to check what encoding the incoming term has todo a correct conversion. -- View this message in context: http://lucene.472066.n3.nabble.com/URLDecoder-error-message-tp4039883p4040604.html Sent from the Solr - User mailing list archive at Nabble.com.
