when host is pingtimeout and CCP can not determine the host status, put the host to Alert status , no VM HA.

Author: Anthony Xu

api/src/com/cloud/host/Status.java

@@ -123,7 +123,7 @@ public static String[] toStrings(Status... states) {
         s_fsm.addTransition(Status.Connecting, Event.Ready, Status.Up);
         s_fsm.addTransition(Status.Connecting, Event.PingTimeout, Status.Alert);
         s_fsm.addTransition(Status.Connecting, Event.ShutdownRequested, Status.Disconnected);
-        s_fsm.addTransition(Status.Connecting, Event.HostDown, Status.Alert);
+        s_fsm.addTransition(Status.Connecting, Event.HostDown, Status.Down);
         s_fsm.addTransition(Status.Connecting, Event.Ping, Status.Connecting);
         s_fsm.addTransition(Status.Connecting, Event.ManagementServerDown, Status.Disconnected);
         s_fsm.addTransition(Status.Connecting, Event.AgentDisconnected, Status.Alert);

engine/orchestration/src/com/cloud/agent/manager/AgentManagerImpl.java

@@ -825,6 +825,10 @@ protected boolean handleDisconnectWithInvestigation(AgentAttache attache, Status
 
                 if (determinedState == Status.Down) {
                     s_logger.error("Host is down: " + host.getId() + "-" + host.getName() + ".  Starting HA on the VMs");
+                    if ((host.getType() != Host.Type.SecondaryStorage) && (host.getType() != Host.Type.ConsoleProxy)) {
+                        _alertMgr.sendAlert(AlertManager.AlertType.ALERT_TYPE_HOST, host.getDataCenterId(), host.getPodId(), "Host disconnected, " + host.getId(),
+                                "Host is down: " + host.getId() + "-" + host.getName() + ".  Starting HA on the VMs");
+                    }
                     event = Status.Event.HostDown;
                 } else if (determinedState == Status.Up) {
                     /* Got ping response from host, bring it back*/
@@ -857,20 +861,18 @@ protected boolean handleDisconnectWithInvestigation(AgentAttache attache, Status
                     HostPodVO podVO = _podDao.findById(host.getPodId());
                     String hostDesc = "name: " + host.getName() + " (id:" + host.getId() + "), availability zone: " + dcVO.getName() + ", pod: " + podVO.getName();
                     _alertMgr.sendAlert(AlertManager.AlertType.ALERT_TYPE_HOST, host.getDataCenterId(), host.getPodId(), "Host in ALERT state, " + hostDesc,
-                            "In availability zone " + host.getDataCenterId()
-                            + ", host is in alert state: " + host.getId() + "-" + host.getName());
+                            "In availability zone " + host.getDataCenterId() + ", " + host.getId() + "-" + host.getName()
+                            + " disconnect due to event " + event + ", ms can't determine the host status" );
                 }
             } else {
                 s_logger.debug("The next status of Agent " + host.getId() + " is not Alert, no need to investigate what happened");
             }
         }
-
         handleDisconnectWithoutInvestigation(attache, event, true, true);
         host = _hostDao.findById(hostId); // Maybe the host magically reappeared?
-        if (host != null && (host.getStatus() == Status.Alert || host.getStatus() == Status.Down)) {
+        if (host != null && host.getStatus() == Status.Down) {
             _haMgr.scheduleRestartForVmsOnHost(host, true);
         }
-
         return true;
     }
 

server/src/com/cloud/ha/HighAvailabilityManagerImpl.java

@@ -202,7 +202,7 @@ protected HighAvailabilityManagerImpl() {
     public Status investigate(final long hostId) {
         final HostVO host = _hostDao.findById(hostId);
         if (host == null) {
-            return null;
+            return Status.Alert;
         }
 
         Status hostState = null;
@@ -219,7 +219,7 @@ public Status investigate(final long hostId) {
             }
         }
 
-        return null;
+        return Status.Alert;
     }
 
     @Override