OK. Call me chicken little.
We must have had bad class files or something hanging out in our build
that had the issues. Having built from trunk, we're seeing perfectly
fine response times even at 500 requests a second.
Thank you for your help, and sorry to bring it up without testing trunk.
Thanks again for your time!
Matthew Runo
Software Engineer, Zappos.com
mr...@zappos.com - 702-943-7833
On Feb 27, 2009, at 8:11 AM, Matthew Runo wrote:
We're just using an SVN up, with no local modifications. It's
probably a formatting difference from having opened solr in an IDE.
We're building from lucene and solr trunk right now, and I'll let
you all know how that goes. We'll test it as best we can with
JMeter. The build we had up there was breaking with between 100 and
200 simultaneous threads (due to blocking on isDeleted).
Thanks for your time!
Matthew Runo
Software Engineer, Zappos.com
mr...@zappos.com - 702-943-7833
On Feb 27, 2009, at 8:01 AM, Chris Hostetter wrote:
: Solr Implementation Version: 1.4-dev 737141M - root - 2009-01-23
10:46:02
that M indicates there were local modifications (relative svn version
#737141) at the time of compilation.
Do you have some local patches?
anything that would have affected the way IndexReaders get opened?
-Hoss
