This is embarrassing. I just realized that in the experiments where I saw
Solr providing good service in the face of the overload requests, I was
actually sending requests at a rate of 30 requests per second, not 300
requests per second. Once I ratcheted up the rate a little bit, Solr
started to ov
Gian,
Even if you can't write a failing unit test (if you did it would be awesome),
please open a JIRA issue on this and attach your patch. Also, you may want to
try 4.0 as opposed to 3.6 as some of the 3.6 issues with DIH are resolved in
4.0.
https://issues.apache.org/jira/secure/Dashboard.j
You say you're using a "snapshot". So it all depends on when
the snapshot was taken. There were a bunch of times prior
to the 4.0 ALPHA where the index format changed requiring
re-indexing. So unless you've been tracking all the times the
index format changed relative to your snapshot, it's a guess
I believe the problem is not that you need BSF -- because you are using
java6 (you have to to compile Solr) so BSF isn't needed. The problem (i
think) is that FreeBSD's JDK doesn't include javascript by default - i
believe you just need to install the rhino "js.jar"
You are right. FreeBSD
