Description of problem:
If more subscribers are bound to a flow-to-disk (FTD) queue, the queue may be stuck after raising JERR__UNEXPRESPONSE / RHM_IORES_EMPTY exception.

Version-Release number of selected component (if applicable):
# rpm -qa | grep qpid
qpid-java-example-0.10-9.el6.noarch
qpid-cpp-client-devel-docs-0.10-8.el6_1.noarch
python-qpid-0.12-1.el6.noarch
qpid-tools-0.12-2.el6.noarch
qpid-cpp-client-ssl-0.12-6.el6.x86_64
qpid-cpp-client-0.12-6.el6.x86_64
qpid-qmf-0.12-6.el6.x86_64
python-qpid-qmf-0.12-6.el6.x86_64
qpid-cpp-server-devel-0.12-6.el6.x86_64
qpid-cpp-server-store-0.12-6.el6.x86_64
qpid-java-client-0.10-9.el6.noarch
qpid-java-common-0.10-9.el6.noarch
qpid-cpp-server-0.12-6.el6.x86_64
qpid-cpp-client-devel-0.12-6.el6.x86_64
qpid-cpp-server-xml-0.12-6.el6.x86_64
#

How reproducible:
100% within few minutes

Steps to Reproduce:
Run this script:
./drain "ftd_queue; { create:always, node:{ type:queue, durable:true, x-declare:{ arguments:{ qpid.policy_type:flow_to_disk, qpid.max_count:100, qpid.max_size:0, qpid.flow_stop_count:0, qpid.flow_stop_size:0, qpid.file_count:4, qpid.file_size:512 } } }}"
while true; do
        echo "$(date): starting next round"
        ./spout -c 10000 --content "some content some content some content some content some content some content some content some content some content some content some content some content some content some content some content some content some content some content" "ftd_queue" &
	./drain -f -c 5000 "ftd_queue" &
	./drain -f -c 5000 "ftd_queue" &
	wait
done

(the script creates a FTD queue, and in parallel sends there messages on 1 connection and drains messages on 2 parallel connections - every loop sends+drains 10k messages)

  
Actual results:
Few issues seen:
1) After a while, drain program raises exception:
2012-01-09 16:04:14 warning Broker closed connection: 320, Queue ftd_queue: loadContent() failed: jexception 0x0108 JournalImpl::loadMsgContent() threw JERR__UNEXPRESPONSE: Unexpected response to call or event. (read_data_record() returned RHM_IORES_EMPTY) (MessageStoreImpl.cpp:1249)
connection-forced: Queue ftd_queue: loadContent() failed: jexception 0x0108 JournalImpl::loadMsgContent() threw JERR__UNEXPRESPONSE: Unexpected response to call or event. (read_data_record() returned RHM_IORES_EMPTY) (MessageStoreImpl.cpp:1249)
2) Sometimes (easy to reproduce by repeating the cycle from the script), the queue becomes "stuck", having messages in it (per qpid-stat -q) but a drain fails with the same exception even during fetching a single message. The queue is then absolutely useless.
3) Rarely, the bug even segfaults qpidd process - coredump to be added

Expected results:
No such behavior (either 1, 2 or 3 from above)

Additional info: