[dhcp_relay][telemetry] Update events test for dhcp6relay bind issue

tests/telemetry/events/dhcp-relay_events.py

@@ -56,21 +56,29 @@ def trigger_dhcp_relay_bind_failure(duthost):
 
     vlan = dhcp_test_info["vlan"]
     dhcp6_relay_process = dhcp_test_info["dhcp6relay_process"]
-    ipv6_ip = dhcp_test_info["ipv6_address"]
+    ipv6_global_unique_ip = dhcp_test_info["ipv6_address"]
 
     try:
         # Flush ipv6 address from vlan
-        duthost.shell("ip -6 address flush dev {}".format(vlan))
+        duthost.shell("ip -6 address del {} dev {}".format(ipv6_global_unique_ip, vlan))
 
         # Restart dhcrelay process
         duthost.shell("docker exec dhcp_relay supervisorctl restart {}".format(dhcp6_relay_process))
+        # Wait dhcp6relay to hit bind failure, dhcp6relay would try 6 times with interval 5s, hence wait 35s to hit
+        # bind failure
+        time.sleep(35)
 
     finally:
         # Add back ipv6 address to vlan
-        duthost.shell("ip address add {} dev {}".format(ipv6_ip, vlan))
-
-        # Restart dhcrelay process
-        duthost.shell("docker exec dhcp_relay supervisorctl restart {}".format(dhcp6_relay_process))
+        duthost.shell("ip address add {} dev {}".format(ipv6_global_unique_ip, vlan))
+
+        # After bind failure test, dhcp6relay would exit because fail to bind. It's critical process of dhcp_relay,
+        # hence maybe in that time dhcp_relay container has crashed, we need to restart whole dhcp_relay service to
+        # recover
+        duthost.shell("systemctl reset-failed dhcp_relay")
+        duthost.restart_service("dhcp_relay")
+        py_assert(wait_until(100, 10, 0, duthost.is_service_fully_started_per_asic_or_host, "dhcp_relay"),
+                  "dhcp_relay not started.")
 
 
 def send_dhcp_discover_packets(duthost, ptfadapter, packets_to_send=5, interval=1):