properly deal with clients that talked to another campaign server before

A campaign server now tells all clients a unique run ID (the UNIX timestamp
when it was started).  This allows us to ignore results from "old" clients
that talked to another server before, and to tell them to die.

git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@1677 8c4709b5-6ec9-48aa-a5cd-a96041d1645a

src/core/cpn/JobServer.hpp

@@ -8,6 +8,7 @@
 #include "config/FailConfig.hpp"
 
 #include <list>
+#include <ctime>
 
 #ifndef __puma
 #include <boost/thread.hpp>
@@ -37,7 +38,7 @@ private:
 	unsigned m_maxThreads;
 	//! the maximal timeout per communication thread
 	int m_threadtimeout;
-	//! A of spawned threads
+	//! list of spawned threads
 #ifndef __puma
 	typedef std::list<boost::thread*> Tthreadlist;
 	Tthreadlist m_threadlist;
@@ -45,6 +46,9 @@ private:
 	boost::thread* m_serverThread;
 #endif // puma
 
+	//! unique server run ID
+	uint64_t m_runid;
+
 #ifdef SERVER_PERFORMANCE_MEASURE
 	static volatile unsigned m_DoneCount; //! the number of finished jobs
 #ifndef __puma
@@ -75,6 +79,7 @@ public:
 	JobServer(int port = SERVER_COMM_TCP_PORT) : m_port(port), m_finish(false), m_noMoreExps(false),
 		m_maxThreads(128), m_threadtimeout(0)
 	{ 
+		m_runid = std::time(0);
 #ifndef __puma
 		m_serverThread = new boost::thread(&JobServer::run, this); // run operator()() in a thread. 
 #ifdef SERVER_PERFORMANCE_MEASURE