Implement clean return from save/restore failure (so that original

domain can continue execution).
Signed-off-by: Andrei Petrov <andrei.petrov@xensource.com>

tools/python/xen/lowlevel/xc/xc.c

@@ -1064,9 +1064,9 @@ static PyMethodDef pyxc_methods[] = {
       "Destroy a domain.\n"
       " dom [int]:    Identifier of domain to be destroyed.\n\n"
       "Returns: [int] 0 on success; -1 on error.\n" },
-    
+
     { "domain_resume", 
-      (PyCFunction)pyxc_domain_resume, 
+      (PyCFunction)pyxc_domain_resume,
       METH_VARARGS, "\n"
       "Resume execution of a suspended domain.\n"
       " dom [int]: Identifier of domain to be resumed.\n\n"

tools/python/xen/xend/XendCheckpoint.py

@@ -122,6 +122,8 @@ def save(fd, dominfo, network, live, dst):
             os.remove("/tmp/xen.qemu-dm.%d" % dominfo.getDomid())
 
         dominfo.destroyDomain()
+        dominfo.testDeviceComplete()
+
         try:
             dominfo.setName(domain_name)
         except VmError:
@@ -134,11 +136,31 @@ def save(fd, dominfo, network, live, dst):
     except Exception, exn:
         log.exception("Save failed on domain %s (%s).", domain_name,
                       dominfo.getDomid())
+
+        dominfo._releaseDevices()
+        dominfo.testDeviceComplete()
+        dominfo.testvifsComplete()
+        log.debug("XendCheckpoint.save: devices released")
+
+        dominfo._resetChannels()
+
+        dominfo._removeDom('control/shutdown')
+        dominfo._removeDom('device-misc/vif/nextDeviceID')
+
+        dominfo._createChannels()
+        dominfo._introduceDomain()
+        dominfo._storeDomDetails()
+
+        dominfo._createDevices()
+        log.debug("XendCheckpoint.save: devices created")
+
+        dominfo.resumeDomain()
+        log.debug("XendCheckpoint.save: resumeDomain")
+
         try:
             dominfo.setName(domain_name)
         except:
             log.exception("Failed to reset the migrating domain's name")
-        raise Exception, exn
 
 
 def restore(xd, fd, dominfo = None, paused = False):

tools/python/xen/xend/XendDomain.py

@@ -1166,7 +1166,6 @@ class XendDomain:
         sock.send("receive\n")
         sock.recv(80)
         XendCheckpoint.save(sock.fileno(), dominfo, True, live, dst)
-        dominfo.testDeviceComplete()
         sock.close()
 
     def domain_save(self, domid, dst):

tools/python/xen/xend/XendDomainInfo.py

@@ -1580,6 +1580,16 @@ class XendDomainInfo:
             log.exception("Exception in alloc_unbound(%d)", self.domid)
             raise
 
+    def _resetChannels(self):
+        """Reset all event channels in the domain.
+        """
+        try:
+            return xc.evtchn_reset(dom=self.domid)
+        except:
+            log.exception("Exception in evtcnh_reset(%d)", self.domid)
+            raise
+
+
     #
     # Bootloader configuration
     #
@@ -1737,6 +1747,25 @@ class XendDomainInfo:
                 log.info("Dev still active but hit max loop timeout")
                 break
 
+    def testvifsComplete(self):
+        """ In case vifs are released and then created for the same
+        domain, we need to wait the device shut down.
+        """
+        start = time.time()
+        while True:
+            test = 0
+            diff = time.time() - start
+            for i in self.getDeviceController('vif').deviceIDs():
+                test = 1
+                log.info("Dev %s still active, looping...", i)
+                time.sleep(0.1)
+                
+            if test == 0:
+                break
+            if diff >= MIGRATE_TIMEOUT:
+                log.info("Dev still active but hit max loop timeout")
+                break
+
     def _storeVmDetails(self):
         to_store = {}