Add ONOS node failures to SRRouting

Change-Id: I29da63d59154d880839dca2297ed648a2f98e5f0
diff --git a/TestON/tests/USECASE/SegmentRouting/SRRouting/SRRouting.py b/TestON/tests/USECASE/SegmentRouting/SRRouting/SRRouting.py
index f808d4e..07f1260 100644
--- a/TestON/tests/USECASE/SegmentRouting/SRRouting/SRRouting.py
+++ b/TestON/tests/USECASE/SegmentRouting/SRRouting/SRRouting.py
@@ -558,4 +558,183 @@
                                staticRouteConfigure=True,
                                switchFailure=True )
 
+    def CASE301( self, main ):
+        """
+        Kill and recover ONOS nodes
+        Ping between all ipv4 hosts in the topology.
+        """
+
+        from tests.USECASE.SegmentRouting.SRRouting.dependencies.SRRoutingTest import SRRoutingTest
+
+        SRRoutingTest.runTest( main,
+                               test_idx=301,
+                               onosNodes=3,
+                               dhcp=1,
+                               routers=1,
+                               ipv4=1,
+                               ipv6=0,
+                               countFlowsGroups=False,
+                               description="Test node failures with IPv4 hosts",
+                               nodeFailure=True )
+
+    def CASE302( self, main ):
+        """
+        Kill and recover ONOS nodes
+        Ping between all ipv6 hosts in the topology.
+        """
+
+        from tests.USECASE.SegmentRouting.SRRouting.dependencies.SRRoutingTest import SRRoutingTest
+
+        SRRoutingTest.runTest( main,
+                               test_idx=302,
+                               onosNodes=3,
+                               dhcp=1,
+                               routers=1,
+                               ipv4=0,
+                               ipv6=1,
+                               countFlowsGroups=False,
+                               description="Test node failures with IPv6 hosts",
+                               nodeFailure=True )
+
+    def CASE303( self, main ):
+        """
+        Kill and recover ONOS nodes
+        Ping between all ipv4 and ipv6 hosts in the topology.
+        """
+
+        from tests.USECASE.SegmentRouting.SRRouting.dependencies.SRRoutingTest import SRRoutingTest
+
+        SRRoutingTest.runTest( main,
+                               test_idx=303,
+                               onosNodes=3,
+                               dhcp=1,
+                               routers=1,
+                               ipv4=1,
+                               ipv6=1,
+                               countFlowsGroups=False,
+                               description="Test node failures with IPv4 and IPv6 hosts",
+                               nodeFailure=True )
+
+    def CASE304( self, main ):
+        """
+        Kill and recover ONOS nodes
+        Ping between all ipv4 hosts in the topology and check connectivity to external ipv4 hosts
+        """
+
+        from tests.USECASE.SegmentRouting.SRRouting.dependencies.SRRoutingTest import SRRoutingTest
+
+        SRRoutingTest.runTest( main,
+                               test_idx=304,
+                               onosNodes=3,
+                               dhcp=1,
+                               routers=1,
+                               ipv4=1,
+                               ipv6=0,
+                               description="Test node failures with IPv4 hosts (including external hosts)",
+                               checkExternalHost=True,
+                               nodeFailure=True )
+
+    def CASE305( self, main ):
+        """
+        Kill and recover ONOS nodes
+        Ping between all ipv6 hosts in the topology and check connectivity to external ipv6 hosts
+        """
+
+        from tests.USECASE.SegmentRouting.SRRouting.dependencies.SRRoutingTest import SRRoutingTest
+
+        SRRoutingTest.runTest( main,
+                               test_idx=305,
+                               onosNodes=3,
+                               dhcp=1,
+                               routers=1,
+                               ipv4=0,
+                               ipv6=1,
+                               description="Test node failures with IPv6 hosts (including external hosts)",
+                               checkExternalHost=True,
+                               nodeFailure=True )
+
+    def CASE306( self, main ):
+        """
+        Kill and recover ONOS nodes
+        Ping between all ipv4 and ipv6 hosts in the topology and check connectivity to external ipv4 and ipv6 hosts
+        """
+
+        from tests.USECASE.SegmentRouting.SRRouting.dependencies.SRRoutingTest import SRRoutingTest
+
+        SRRoutingTest.runTest( main,
+                               test_idx=306,
+                               onosNodes=3,
+                               dhcp=1,
+                               routers=1,
+                               ipv4=1,
+                               ipv6=1,
+                               description="Test node failures with IPv4 and IPv6 hosts (including external hosts)",
+                               checkExternalHost=True,
+                               nodeFailure=True )
+
+    def CASE307( self, main ):
+        """
+        Kill and recover ONOS nodes
+        Ping between ipv4 hosts and an external host that is not configured in
+        external router config, but reachable through the use of route-add command.
+        """
+
+        from tests.USECASE.SegmentRouting.SRRouting.dependencies.SRRoutingTest import SRRoutingTest
+
+        SRRoutingTest.runTest( main,
+                               test_idx=307,
+                               onosNodes=3,
+                               dhcp=1,
+                               routers=1,
+                               ipv4=1,
+                               ipv6=0,
+                               description="Test node failures with IPv4 hosts (including external host configured with route-add command)",
+                               checkExternalHost=False,
+                               countFlowsGroups=False,
+                               staticRouteConfigure=True,
+                               nodeFailure=True )
+
+    def CASE308( self, main ):
+        """
+        Kill and recover ONOS nodes
+        Ping between ipv6 hosts and an external host that is not configured in
+        external router config, but reachable through the use of route-add command.
+        """
+
+        from tests.USECASE.SegmentRouting.SRRouting.dependencies.SRRoutingTest import SRRoutingTest
+
+        SRRoutingTest.runTest( main,
+                               test_idx=308,
+                               onosNodes=3,
+                               dhcp=1,
+                               routers=1,
+                               ipv4=0,
+                               ipv6=1,
+                               description="Test node failures with IPv6 hosts (including external host configured with route-add command)",
+                               checkExternalHost=False,
+                               countFlowsGroups=False,
+                               staticRouteConfigure=True,
+                               nodeFailure=True )
+
+    def CASE309( self, main ):
+        """
+        Kill and recover ONOS nodes
+        Ping between ipv4 and pv6 hosts and external hosts that is not configured in
+        external router config, but reachable through the use of route-add command.
+        """
+
+        from tests.USECASE.SegmentRouting.SRRouting.dependencies.SRRoutingTest import SRRoutingTest
+
+        SRRoutingTest.runTest( main,
+                               test_idx=309,
+                               onosNodes=3,
+                               dhcp=1,
+                               routers=1,
+                               ipv4=1,
+                               ipv6=1,
+                               description="Test node failures with IPv4 and IPv6 hosts (including external host configured with route-add command)",
+                               checkExternalHost=False,
+                               countFlowsGroups=False,
+                               staticRouteConfigure=True,
+                               nodeFailure=True )
 
diff --git a/TestON/tests/USECASE/SegmentRouting/SRRouting/dependencies/SRRoutingTest.py b/TestON/tests/USECASE/SegmentRouting/SRRouting/dependencies/SRRoutingTest.py
index 4b58d10..5707a7c 100644
--- a/TestON/tests/USECASE/SegmentRouting/SRRouting/dependencies/SRRoutingTest.py
+++ b/TestON/tests/USECASE/SegmentRouting/SRRouting/dependencies/SRRoutingTest.py
@@ -21,6 +21,7 @@
 
 from tests.USECASE.SegmentRouting.dependencies.Testcaselib import Testcaselib as run
 import time
+import json
 
 class SRRoutingTest ():
 
@@ -32,7 +33,8 @@
     @staticmethod
     def runTest( main, test_idx, onosNodes, dhcp, routers, ipv4, ipv6,
                  description, countFlowsGroups=False, checkExternalHost=False,
-                 staticRouteConfigure=False, switchFailure=False, linkFailure=False ):
+                 staticRouteConfigure=False, switchFailure=False, linkFailure=False,
+                 nodeFailure=False ):
 
         skipPackage = False
         init = False
@@ -53,7 +55,7 @@
         if staticRouteConfigure:
             main.cfgName += '_static=1'
 
-        main.resultFileName = 'CASE%02d' % test_idx
+        main.resultFileName = 'CASE%03d' % test_idx
         main.Cluster.setRunningNode( onosNodes )
 
         run.installOnos( main, skipPackage=skipPackage, cliSleep=5,
@@ -97,7 +99,7 @@
             time.sleep( 60 )
 
         # ping hosts
-        run.pingAll( main, 'CASE%02d' % test_idx, acceptableFailed=5, basedOnIp=True )
+        run.pingAll( main, 'CASE%03d' % test_idx, acceptableFailed=5, basedOnIp=True )
 
         # check flows / groups numbers
         if countFlowsGroups:
@@ -107,11 +109,11 @@
         if switchFailure:
             for switch, expected in main.switchFailureChart.items():
                 run.killSwitch( main, switch, expected['switches_after_failure'], expected['links_after_failure'] )
-                run.pingAll( main, 'CASE%02d' % test_idx, acceptableFailed=5, basedOnIp=True )
+                run.pingAll( main, 'CASE%03d' % test_idx, acceptableFailed=5, basedOnIp=True )
                 if countFlowsGroups:
                     run.checkFlowsGroupsFromFile(main)
                 run.recoverSwitch( main, switch, expected['switches_before_failure'], expected['links_before_failure'] )
-                run.pingAll( main, 'CASE%02d' % test_idx, acceptableFailed=5, basedOnIp=True )
+                run.pingAll( main, 'CASE%03d' % test_idx, acceptableFailed=5, basedOnIp=True )
                 if countFlowsGroups:
                     run.checkFlowsGroupsFromFile(main)
 
@@ -124,13 +126,32 @@
                 linksAfter = info['links_after']
 
                 run.killLinkBatch( main, linksToRemove, linksAfter )
-                run.pingAll( main, 'CASE%02d' % test_idx, acceptableFailed=5, basedOnIp=True )
+                run.pingAll( main, 'CASE%03d' % test_idx, acceptableFailed=5, basedOnIp=True )
 
                 run.restoreLinkBatch( main, linksToRemove, linksBefore )
-                run.pingAll( main, 'CASE%02d' % test_idx, acceptableFailed=5, basedOnIp=True )
+                run.pingAll( main, 'CASE%03d' % test_idx, acceptableFailed=5, basedOnIp=True )
                 if countFlowsGroups:
                     run.checkFlowsGroupsFromFile(main)
 
+        # Test node failures
+        if nodeFailure:
+            numCtrls = len( main.Cluster.runningNodes )
+            links = len( json.loads( main.Cluster.next().links() ) )
+            switches = len( json.loads( main.Cluster.next().devices() ) )
+            for ctrl in xrange( numCtrls ):
+                run.killOnos( main, [ ctrl ], switches, links, ( numCtrls - 1 ) )
+                main.Cluster.active(0).CLI.balanceMasters()
+                run.pingAll( main, 'CASE%03d' % test_idx, acceptableFailed=5, basedOnIp=True )
+                if countFlowsGroups:
+                    run.checkFlowsGroupsFromFile( main )
+
+                run.recoverOnos( main, [ ctrl ], switches, links, numCtrls )
+                main.Cluster.active(0).CLI.balanceMasters()
+                run.pingAll( main, 'CASE%03d' % test_idx, acceptableFailed=5, basedOnIp=True )
+                if countFlowsGroups:
+                    run.checkFlowsGroupsFromFile( main )
+
+        # Cleanup
         if hasattr( main, 'Mininet1' ):
             run.cleanup( main )
         else:
diff --git a/TestON/tests/USECASE/SegmentRouting/dependencies/Testcaselib.py b/TestON/tests/USECASE/SegmentRouting/dependencies/Testcaselib.py
index 0125da6..bd63a80 100644
--- a/TestON/tests/USECASE/SegmentRouting/dependencies/Testcaselib.py
+++ b/TestON/tests/USECASE/SegmentRouting/dependencies/Testcaselib.py
@@ -97,8 +97,8 @@
         # main.scale[ 0 ] determines the current number of ONOS controller
         if not main.apps:
             main.log.error( "App list is empty" )
-        main.log.info( "NODE COUNT = " + str( main.Cluster.numCtrls ) )
-        main.log.info( ''.join( main.Cluster.getIps() ) )
+        main.log.info( "Cluster size: " + str( main.Cluster.numCtrls ) )
+        main.log.info( "Cluster ips: " + ', '.join( main.Cluster.getIps() ) )
         main.dynamicHosts = [ 'in1', 'out1' ]
         main.testSetUp.ONOSSetUp( main.Cluster, newCell=True, cellName=main.cellName,
                                   skipPack=skipPackage,
@@ -280,7 +280,7 @@
     @staticmethod
     def checkFlows( main, minFlowCount, tag="", dumpflows=True, sleep=10 ):
         main.step(
-                " Check whether the flow count is bigger than %s" % minFlowCount )
+                "Check whether the flow count is bigger than %s" % minFlowCount )
         if tag == "":
             tag = 'CASE%d' % main.CurrentTestCaseNumber
         count = utilities.retry( main.Cluster.active( 0 ).CLI.checkFlowCount,
@@ -638,7 +638,7 @@
         switches, links, nodes: number of expected switches, links and nodes after KillOnos, ex.: '4', '6'
         Completely Kill an ONOS instance and verify the ONOS cluster can see the proper change
         """
-        main.step( "Killing ONOS instance" )
+        main.step( "Killing ONOS instances with index(es): {}".format( nodes ) )
 
         for i in nodes:
             killResult = main.ONOSbench.onosDie( main.Cluster.runningNodes[ i ].ipAddress )
@@ -685,7 +685,7 @@
         switches, links, nodes: number of expected switches, links and nodes after recoverOnos, ex.: '4', '6'
         Recover an ONOS instance and verify the ONOS cluster can see the proper change
         """
-        main.step( "Recovering ONOS instance" )
+        main.step( "Recovering ONOS instances with index(es): {}".format( nodes ) )
         [ main.ONOSbench.onosStart( main.Cluster.runningNodes[ i ].ipAddress ) for i in nodes ]
         for i in nodes:
             isUp = main.ONOSbench.isup( main.Cluster.runningNodes[ i ].ipAddress )