Retry failed rebind attempts

When we receive a migration, if rebinding to the new listen address and
port fails for a reason which might be fixable, rather than killing the
server we retry once a second.  Also in this patch: non-overlapping log
messages and a fix for the client going away halfway through a sendfile
loop.

tests/acceptance/fakes/source/successful_transfer.rb

@@ -0,0 +1,29 @@
+#!/usr/bin/env ruby
+
+# Successfully send a migration, but squat on the IP and port which
+# the destination wants to rebind to.  The destination should retry
+# every second, so we give it up then attempt to connect to the new
+# server.
+
+require 'flexnbd/fake_source'
+include FlexNBD
+
+addr, port, srv_pid, newaddr, newport = *ARGV
+
+squatter = TCPServer.open( newaddr, newport.to_i )
+
+client  = FakeSource.new( addr, port, "Timed out connecting" )
+client.send_mirror()
+
+sleep(1)
+
+squatter.close()
+
+sleep(1)
+
+client2 = FakeSource.new( newaddr, newport.to_i, "Timed out reconnecting" )
+client2.read_hello
+client2.read( 0, 8 )
+client2.close
+
+exit( 0 )