ep-engine segfault during heavy write load

pytests/performance/mcsoda.py default@localhost json=1 max-creates=1000000 ratio-creates=1.0 ratio-sets=1 json=0

Loading simple k/v items with the above command on a single node got "stuck" at 875K items. load was still running at 7K ops/sec, all sets, but nothing was changed, write queue remains empty.

(FULL BACKTRACE ATTACHED)

Thread 1 (Thread 18289):
#0 0x00002aaaaad773f9 in HashTable::resize(unsigned long) () from /opt/couchbase/lib/memcached/ep.so
#1 0x00002aaaaad7767b in HashTable::resize() () from /opt/couchbase/lib/memcached/ep.so
#2 0x00002aaaaad6a790 in ResizingVisitor::visitBucket(RCPtr<VBucket>&) () from /opt/couchbase/lib/memcached/ep.so
#3 0x00002aaaaad29e0f in VBCBAdaptor::callback(Dispatcher&, std::tr1::shared_ptr<Task>) () from /opt/couchbase/lib/memcached/ep.so
#4 0x00002aaaaad2683f in Task::run(Dispatcher&, std::tr1::shared_ptr<Task>) () from /opt/couchbase/lib/memcached/ep.so
#5 0x00002aaaaad2594d in Dispatcher::run() () from /opt/couchbase/lib/memcached/ep.so