traffic: add option to reanimate dying conversations

The traffic model is generated from a window in time, which makes
conversations appear to start and stop unnaturally at the window
boundaries. When the window is short compared to the traffic replay
time and the true expected conversation length, this has a significant
distorting effect, leading to more conversations than would be
expected to generate a given number of packets.

To offset this slightly we add the --conversation-persistence option
which tries to convert apparent death into a longish wait.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

3 files changed, 39 insertions, 4 deletions

diff --git a/python/samba/emulate/traffic.py b/python/samba/emulate/traffic.py
index 448cacfda41..76c39efba56 100644
--- a/python/samba/emulate/traffic.py
+++ b/python/samba/emulate/traffic.py
@@ -1226,7 +1226,8 @@ class TrafficModel(object):
     def construct_conversation_sequence(self, timestamp=0.0,
                                         hard_stop=None,
                                         replay_speed=1,
-                                        ignore_before=0):
+                                        ignore_before=0,
+                                        persistence=0):
         """Construct an individual conversation packet sequence from the
         model.
         """
@@ -1238,7 +1239,15 @@ class TrafficModel(object):
         while True:
             p = random.choice(self.ngrams.get(key, (NON_PACKET,)))
             if p == NON_PACKET:
-                break
+                if timestamp < ignore_before:
+                    break
+                if random.random() > persistence:
+                    print("ending after %s (persistence %.1f)" % (key, persistence),
+                          file=sys.stderr)
+                    break
+
+                p = 'wait:%d' % random.randrange(5, 12)
+                print("trying %s instead of end" % p, file=sys.stderr)
 
             if p in self.query_details:
                 extra = random.choice(self.query_details[p])
@@ -1260,10 +1269,16 @@ class TrafficModel(object):
                     c.append((timestamp, protocol, opcode, extra))
 
             key = key[1:] + (p,)
+            if key[-2][:5] == 'wait:' and key[-1][:5] == 'wait:':
+                # two waits in a row can only be caused by "persistence"
+                # tricks, and will not result in any packets being found.
+                # Instead we pretend this is a fresh start.
+                key = (NON_PACKET,) * (self.n - 1)
 
         return c
 
-    def generate_conversation_sequences(self, scale, duration, replay_speed=1):
+    def generate_conversation_sequences(self, scale, duration, replay_speed=1,
+                                        persistence=0):
         """Generate a list of conversation descriptions from the model."""
 
         # We run the simulation for ten times as long as our desired
@@ -1280,7 +1295,8 @@ class TrafficModel(object):
             c = self.construct_conversation_sequence(start,
                                                      hard_stop=duration,
                                                      replay_speed=replay_speed,
-                                                     ignore_before=0)
+                                                     ignore_before=0,
+                                                     persistence=persistence)
             # will these "packets" generate actual traffic?
             # some (e.g. ldap unbind) will not generate anything
             # if the previous packets are not there, and if the
diff --git a/python/samba/tests/blackbox/testdata/traffic_replay-2.expected b/python/samba/tests/blackbox/testdata/traffic_replay-2.expected
new file mode 100644
index 00000000000..7850a255043
--- /dev/null
+++ b/python/samba/tests/blackbox/testdata/traffic_replay-2.expected
@@ -0,0 +1,17 @@
+0.011388	06		2	1	ldap	3	searchRequest	2	DC,DC		cn			
+0.221447	06		2	1	ldap	2	unbindRequest							
+0.460878	06		3	1	ldap	3	searchRequest	2	DC,DC		cn			
+0.581933	11		4	1	cldap	3	searchRequest				Netlogon			
+0.596977	11		4	1	cldap	3	searchRequest				Netlogon			
+0.611184	11		4	1	cldap	3	searchRequest				Netlogon			
+0.666808	06		3	1	ldap	2	unbindRequest							
+0.692730	11		5	1	cldap	3	searchRequest				Netlogon			
+0.692879	11		5	1	cldap	3	searchRequest				Netlogon			
+0.692946	11		5	1	cldap	3	searchRequest				Netlogon			
+0.744297	06		4	1	rpc_netlogon	29	NetrLogonGetDomainInfo	
+0.768994	06		4	1	kerberos			
+0.772476	06		4	1	ldap	3	searchRequest	2	DC,DC		cn			
+0.827760	06		5	1	rpc_netlogon	29	NetrLogonGetDomainInfo	
+0.828419	06		5	1	kerberos			
+0.862850	06		5	1	ldap	3	searchRequest				subschemaSubentry,dsServiceName,namingContexts,defaultNamingContext,schemaNamingContext,configurationNamingContext,rootDomainNamingContext,supportedControl,supportedLDAPVersion,supportedLDAPPolicies,supportedSASLMechanisms,dnsHostName,ldapServiceName,serverName,supportedCapabilities			
+0.865384	06		6	1	ldap	3	searchRequest				subschemaSubentry,dsServiceName,namingContexts,defaultNamingContext,schemaNamingContext,configurationNamingContext,rootDomainNamingContext,supportedControl,supportedLDAPVersion,supportedLDAPPolicies,supportedSASLMechanisms,dnsHostName,ldapServiceName,serverName,supportedCapabilities			
diff --git a/python/samba/tests/blackbox/traffic_replay.py b/python/samba/tests/blackbox/traffic_replay.py
index 4e9783ec515..8370939d243 100644
--- a/python/samba/tests/blackbox/traffic_replay.py
+++ b/python/samba/tests/blackbox/traffic_replay.py
@@ -70,6 +70,8 @@ class TrafficLearnerTests(BlackboxTestCase):
 
         for i, opts in enumerate((["--random-seed=3"],
                                   ["--random-seed=4"],
+                                  ["--random-seed=3",
+                                   "--conversation-persistence=0.5"],
                                   )):
             with temp_file(self.tempdir) as output:
                 command = ([SCRIPT, MODEL,