[pylucene-dev] fatal error in GC : too many heap sections.

[Pete]

On Monday November 19 2007 1:49:57 pm Bill Peverill wrote:
> We would be grateful for any advice we can get. (Our gratitude could
> include beer contributions or other compensation if appropriate.) We'd also
> love to hear from people who have NOT had this problem.

Heh... we had the same problem at much the same point in development... pushed 
our rollout back by about a month.  Not fun.

I'll second what Andi said... try the new JCC implementation.  We had similar 
problems with PyLucene-GCJ, and while we were able to ameliorate them 
somewhat by fiddling GC/JVM settings, we never really solved the problem.  
Instead, we resorted to periodically restarting the process once an hour or 
so (which was desirable/acceptable for other reasons, but not great). The GCJ 
memory allocator/garbage collector just isn't that great/mature.

OTOMH, I'm pretty sure you want to build with -DLARGE_HEAP or somesuch. Don't 
remember what else we did. Really, try the JCC build. ;-)

> Note that because our codebase is not compatible with Python 2.5, we can't
> try the version of PyLucene that has been compiled with Python 2.5.

That shouldn't be true... AFAIK, there's nothing backwards incompatible in 
2.5.  You may get a few additional warnings here and there, but everything 
should still work.

-- 
Peter Fein   ||   773-575-0694   ||   pfein at pobox.com
http://www.pobox.com/~pfein/   ||   PGP: 0xCCF6AE6B
irc: pfein at freenode.net   ||   jabber: peter.fein at gmail.com