Hang in 70_balance test

[GoogleCodeExporter]

I just ran the 70_balance test and it failed to complete.  The shell was hung 
here:

Thread 1 (Thread 0x7f46576f8720 (LWP 5748)):
#0  0x000000388600b3cc in pthread_cond_wait@@GLIBC_2.3.2 () from 
/lib64/libpthread.so.0
#1  0x00000000004f6cba in void 
boost::condition_variable_any::wait<boost::unique_lock<boost::mutex> 
>(boost::unique_lock<boost::mutex>&) ()
#2  0x00000000005d7665 in 
Hypertable::TableScannerQueue::next_result(boost::intrusive_ptr<Hypertable::Scan
Cells>&, int*, std::basic_string<char, std::char_traits<char>, 
std::allocator<char> >&) ()
#3  0x00000000005d7b7d in Hypertable::TableScanner::~TableScanner() ()
#4  0x00000000005d7fd9 in Hypertable::TableScanner::~TableScanner() ()
#5  0x0000000000501744 in (anonymous 
namespace)::cmd_select(boost::intrusive_ptr<Hypertable::Namespace>&, 
boost::intrusive_ptr<Hypertable::ConnectionManager>&, 
boost::intrusive_ptr<Hypertable::DfsBroker::Client>&, 
Hypertable::Hql::ParserState&, Hypertable::HqlInterpreter::Callback&) ()
#6  0x00000000005020b4 in 
Hypertable::HqlInterpreter::execute(std::basic_string<char, 
std::char_traits<char>, std::allocator<char> > const&, 
Hypertable::HqlInterpreter::Callback&) ()
#7  0x00000000004fad42 in 
Hypertable::HqlCommandInterpreter::execute_line(std::basic_string<char, 
std::char_traits<char>, std::allocator<char> > const&) ()
#8  0x0000000000641ec5 in Hypertable::CommandShell::run() ()
#9  0x00000000004e6c30 in main ()

Original issue reported on code.google.com by nuggetwh...@gmail.com on 14 Mar 2012 at 11:00

[GoogleCodeExporter]

It looks like one of the RangeServer crashed while it was being shut down.  
I've attached the stack trace.  We should make sure the TableScanner logic 
works properly in the face of a RangeServer crash.

Original comment by nuggetwh...@gmail.com on 14 Mar 2012 at 11:13

Attachments:

• rs-crash.txt

[GoogleCodeExporter]

Original comment by nuggetwh...@gmail.com on 18 Mar 2012 at 5:15

• Added labels: Milestone-Release0.9.5.0