I'm running boost 1.43 on linux and seeing an issue where the scoped_lock<named_recursive_mutex> timed lock is returning immediately when another thread holds the lock.    The lock is being stored in shared memory Am I doing something wrong?  The code in question is the azLog function below, (I included some other relevant snips as well). thanks! Dan #define LOCK_WAIT_TIME 5; //seconds static named_recursive_mutex *globalLock = create_logger_lock(); static named_recursive_mutex *create_logger_lock () {     named_recursive_mutex *ptr = NULL;     if (globalLock) {         return(globalLock);     }     deadlock_count = 0;     lock_timeout   = LOCK_WAIT_TIME;     try {         ptr = new named_recursive_mutex(open_or_create,                                         (getShareName() + GLOBAL_LOCK).c_str());     } catch (const interprocess_exception &ie) {         cerr << "FATAL: Couldn't initialize the logging subsytem, IE err:" <<                 ie.what() << endl;         throw;     } catch (const std::exception &se) {         cerr << "FATAL: Couldn't initialize the logging subsytem, std err:" <<                 se.what() << endl;         throw;     }     return(ptr); } void azLog(AZ_LOG_SEVERITY sev, AZ_LOG_COMPONENT comp, const string& log_str) {     ptime to = second_clock::local_time() + seconds(lock_timeout);     scoped_lock<named_recursive_mutex> lock(*globalLock, to);     while(!lock.owns() && (second_clock::local_time() < to)) {         cerr << pthread_self() << ": SPURIOUS WAKE!!" << to_simple_string(second_clock::local_time()) << " <=:=> " << to_simple_string(to) << endl;         lock.timed_lock(to);     } .....