Handle wrong interface for locally-routed packets.

CHANGELOG

@@ -1,3 +1,33 @@
+version 2.66
+            Add the ability to act as an authoritative DNS
+            server. Dnsmasq can now answer queries from the wider 'net
+            with local data, as long as the correct NS records are set
+            up. Only local data is provided, to avoid creating an open
+            DNS relay. Zone transfer is supported, to allow secondary
+            servers to be configured.
+
+	    Add "constructed DHCP ranges" for DHCPv6. This is intended
+	    for IPv6 routers which get prefixes dynamically via prefix
+	    delegation. With suitable configuration, stateful DHCPv6
+	    and RA can happen automatically as prefixes are delegated
+	    and then deprecated, without having  to re-write the
+	    dnsmasq configuration file or restart the daemon. Thanks to
+	    Steven Barth for extensive testing and development work on
+	    this idea.
+
+	    Fix crash on startup on Solaris 11. Regression probably
+	    introduced in 2.61.  Thanks to Geoff Johnstone for the
+	    patch.
+
+	    Add code to make behaviour for TCP DNS requests that same
+	    as for UDP requests, when a request arrives for an allowed 
+	    address, but via a banned interface. This change is only
+	    active on Linux, since the relevant API is missing (AFAIK)
+	    on other platforms. Many thanks to Tomas Hozza for
+	    spotting the problem, and doing invaluable discovery of
+	    the obscure and undocumented API required for the solution.
+  
+
 version 2.65
 	    Fix regression which broke forwarding of queries sent via
 	    TCP which are not for A and AAAA and which were directed to

man/dnsmasq.8

@@ -75,6 +75,9 @@ the upstream DNS servers.
 .B --max-cache-ttl=<time>
 Set a maximum TTL value for entries in the cache.
 .TP
+.B --auth-ttl=<time>
+Set the TTL value returned in answers from the authoritative server.
+.TP
 .B \-k, --keep-in-foreground
 Do not go into the background at startup but otherwise run as
 normal. This is intended for use when dnsmasq is run under daemontools
@@ -541,8 +544,27 @@ needed for a client to do validation itself.
 .TP
 .B --auth-zone=<domain>[,<subnet>[,<subnet>.....]]
 Define a DNS zone for which dnsmasq acts as authoritative server. Locally defined DNS records which are in the domain
-will be served, except that A and AAAA records must be in one of the specified subnets, or in a subnet corresponding to a contructed DHCP range. The subnet(s) are also used to define in-addr.arpa and ipv6.arpa domains which are served for reverse-DNS queries.
-.TP  
+will be served, except that A and AAAA records must be in one of the
+specified subnets, or in a subnet corresponding to a contructed DHCP
+range. The subnet(s) are also used to define in-addr.arpa and
+ipv6.arpa domains which are served for reverse-DNS queries.
+.TP
+.B --auth-soa=<serial>[,<hostmaster>[,<refresh>[,<retry>[,<expiry>]]]]
+Specify fields in the SOA record associated with authoritative
+zones. Note that this is optional, all the values are set to sane defaults.
+.TP
+.B --auth-sec-servers=<domain>[,<domain>[,<domain>...]]
+Specify any secondary servers for a zone for which dnsmasq is
+authoritative. These servers must be configured to get zone data from
+dnsmasq by zone transfer, and answer queries for the same
+authoritative zones and dnsmasq.
+.TP
+.B --auth-peer=<ip-address>[,<ip-address>[,<ip-address>...]]
+Specify the addresses of secondary servers which are allowed to
+initiate zone transfer (AXFR) requests for zones for which dnsmasq is
+authoritative. If this option is not given, then AXFR requests wil be
+accepted from any secondary. 
+.TP 
 .B --conntrack
 Read the Linux connection track mark associated with incoming DNS
 queries and set the same mark value on upstream traffic used to answer
@@ -695,7 +717,13 @@ A single
 .B dhcp-host 
 may contain an IPv4 address or an IPv6 address, or both. IPv6 addresses must be bracketed by square brackets thus:
 .B --dhcp-host=laptop,[1234::56]
-Note that in IPv6 DHCP, the hardware address is not normally available, so a client must be identified by client-id (called client DUID in IPv6-land) or hostname.
+IPv6 addresses may contain only the host-identifier part:
+.B --dhcp-host=laptop,[::56]
+in which case thay act as wildcards in constructed dhcp ranges, with
+the appropriate network part inserted. 
+Note that in IPv6 DHCP, the hardware address is not normally
+available, so a client must be identified by client-id (called client
+DUID in IPv6-land) or hostname. 
 
 The special option id:* means "ignore any client-id 
 and use MAC addresses only." This is useful when a client presents a client-id sometimes 
@@ -1671,6 +1699,153 @@ parameter in a BOOTP request is used as a tag,
 as is the tag "bootp", allowing some control over the options returned to
 different classes of hosts.
 
+.SH AUTHORITATIVE CONFIGURATION
+.PP 
+Configuring dnsmasq to act as an authoritative DNS server is
+complicated by the fact that it involves configuration of external DNS
+servers to provide delegation. We will walk through three scenarios of
+increasing complexity. Prerequisites for all of these scenarios
+are a globally accesible IP address, an A or AAAA record pointing to that address,
+and an external DNS server capable of doing delegation of the zone in
+question. For the first part of this explanation, we will call the A (or AAAA) record
+for the globally accessible address server.example.com, and the zone
+for which dnsmasq is authoritative our.zone.com.
+
+The simplest configuration consists of two lines of dnsmasq configuration; something like
+
+.nf
+.B auth-server=server.example.com,eth0
+.B auth-zone=our.zone.com,1.2.3.0/24
+.fi
+
+and two records in the external DNS
+
+.nf
+server.example.com       A    192.0.43.10
+our.zone.com            NS    server.example.com
+.fi
+
+eth0 is the external network interface on which dnsmasq is listening,
+and has (globally accessible) address 192.0.43.10. 
+
+Note that the external IP address may well be dynamic (ie assigned
+from an ISP by DHCP or PPP) If so, the A record must be linked to this
+dynamic assignment by one of the usual dynamic-DNS systems.
+
+A more complex, but practically useful configuration has the address
+record for the globally accessible IP address residing in the
+authoritative zone which dnsmasq is serving, typically at the root. Now
+we have
+
+.nf
+.B auth-server=our.zone.com,eth0
+.B auth-zone=our.zone.com,1.2.3.0/24
+.fi
+
+.nf
+our.zone.com             A    192.0.43.10
+our.zone.com            NS    our.zone.com
+.fi
+
+The A record for our.zone.com has now become a glue record, it solves
+the chicken-and-egg problem of finding the IP address of the
+nameserver for our.zone.com when the A record is within that
+zone. Note that this is the only role of this record: as dnsmasq is
+now authoritative from our.zone.com it too must provide this
+record. If the external address is static, this can be done with an
+.B /etc/hosts 
+entry or 
+.B --host-record.
+If the external address is dynamic,
+then it must be done using something like
+
+.nf
+.B --interface-name=our.zone.com,eth0
+.fi
+
+Our final configuration builds on that above, but also adds a
+secondary DNS server. This is another DNS server which learns the DNS data
+for the zone by doing zones transfer, and acts as a backup should
+the primary server become inaccessible. The configuration of the
+secondary is beyond the scope of this man-page, but the extra
+configuration of dnsmasq is simple:
+
+.nf
+.B auth-sec-servers=secondary.myisp.com
+.fi
+
+and
+
+.nf
+our.zone.com           NS    secondary.myisp.com
+.fi
+
+Adding auth-sec-servers enables zone transfer in dnsmasq, to allow the
+secondary to collect the DNS data. If you wish to restrict this data
+to particular hosts then
+
+.nf
+.B auth-peer=<IP address of secondary>
+.fi
+
+will do so.
+
+Dnsmasq acts as an authoritative server for  in-addr.arpa and
+ipv6.arpa domains associated with the subnets given in auth-zone
+declarations, so reverse (address to name) lookups can be simply
+configured with a suitable NS record, for instance in this example,
+where we allow 1.2.3.0/24 addresses.
+
+.nf
+ 3.2.1.in-addr.arpa  NS    our.zone.com
+.fi
+
+Note that at present, reverse (in-addr.arpa and ip6.arpa) zones are
+not available in zone transfers, so there is no point arranging
+secondary servers for reverse lookups.
+
+.PP
+When dnsmasq is configured to act as an authoritative server, the
+following data is used to populate the authoritative zone.
+.PP
+.B --mx-host, --srv-host, --dns-rr, --txt-record, --naptr-record
+, as long as the record names are in the authoritative domain.
+.PP
+.B --cname
+as long as the record name is in  the authoritative domain. If the
+target of the CNAME is unqualified, then it  is qualified with the
+authoritative zone name.
+.PP
+IPv4 and IPv6 addresses from /etc/hosts (and 
+.B --addn-hosts
+) and
+.B --host-record
+provided the address falls into one of the subnets specified in the
+.B --auth-zone.
+.PP
+Addresses specified by 
+.B --interface-name.
+In this case, the address is not contrained to a subnet from
+.B --auth-zone. 
+
+.PP
+Addresses of DHCP leases, provided the address falls into one of the subnets specified in the
+.B --auth-zone
+OR a constructed DHCP range. In the default mode, where a DHCP lease
+has an unqualified name, and possibly a qualified name constructed
+using 
+.B --domain
+then the name in the authoritative zone is constructed from the
+unqualified name and the zone's domain. This may or may not equal
+that specified by 
+.B --domain.
+If 
+.B --dhcp-fqdn
+is set, then the fully qualified names associated with DHCP leases are
+used, and must match the zone's domain.
+ 
+
+
 .SH EXIT CODES
 .PP
 0 - Dnsmasq successfully forked into the background, or terminated

src/auth.c

@@ -376,35 +376,33 @@ size_t answer_auth(struct dns_header *header, char *limit, size_t qlen, time_t n
 	    }
       	  else if (qtype == T_AXFR)
 	    {
-	      if (daemon->auth_peers)
+	      struct iname *peers;
+	      
+	      if (peer_addr->sa.sa_family == AF_INET)
+		peer_addr->in.sin_port = 0;
+#ifdef HAVE_IPV6
+	      else
+		peer_addr->in6.sin6_port = 0; 
+#endif
+	      
+	      for (peers = daemon->auth_peers; peers; peers = peers->next)
+		if (sockaddr_isequal(peer_addr, &peers->addr))
+		  break;
+	      
+	      /* Refuse all AXFR unless --auth-sec-servers is set */
+	      if ((!peers && daemon->auth_peers) || !daemon->secondary_forward_server)
 		{
-		  struct iname *peers;
-		  
 		  if (peer_addr->sa.sa_family == AF_INET)
-		    peer_addr->in.sin_port = 0;
+		    inet_ntop(AF_INET, &peer_addr->in.sin_addr, daemon->addrbuff, ADDRSTRLEN);
 #ifdef HAVE_IPV6
 		  else
-		    peer_addr->in6.sin6_port = 0; 
+		    inet_ntop(AF_INET6, &peer_addr->in6.sin6_addr, daemon->addrbuff, ADDRSTRLEN); 
 #endif
 		  
-		  for (peers = daemon->auth_peers; peers; peers = peers->next)
-		    if (sockaddr_isequal(peer_addr, &peers->addr))
-		      break;
-		  
-		  if (!peers)
-		    {
-		      if (peer_addr->sa.sa_family == AF_INET)
-			inet_ntop(AF_INET, &peer_addr->in.sin_addr, daemon->addrbuff, ADDRSTRLEN);
-#ifdef HAVE_IPV6
-		      else
-			inet_ntop(AF_INET6, &peer_addr->in6.sin6_addr, daemon->addrbuff, ADDRSTRLEN); 
-#endif
-		      
-		      my_syslog(LOG_WARNING, _("ignoring zone transfer request from %s"), daemon->addrbuff);
-		      return 0;
-		    }
+		  my_syslog(LOG_WARNING, _("ignoring zone transfer request from %s"), daemon->addrbuff);
+		  return 0;
 		}
-	      
+	       	      
 	      soa = 1; /* inhibits auth section */
 	      ns = 1; /* ensure we include NS records! */
 	      axfr = 1;

src/bpf.c

@@ -111,7 +111,7 @@ int iface_enumerate(int family, void *parm, int (*callback)())
 	{
 	  int iface_index = if_nametoindex(addrs->ifa_name);
 
-	  if (iface_index == 0)
+	  if (iface_index == 0 || !addrs->ifa_addr || !addrs->ifa_netmask)
 	    continue;
 
 	  if (family == AF_INET)
@@ -119,7 +119,10 @@ int iface_enumerate(int family, void *parm, int (*callback)())
 	      struct in_addr addr, netmask, broadcast;
 	      addr = ((struct sockaddr_in *) addrs->ifa_addr)->sin_addr;
 	      netmask = ((struct sockaddr_in *) addrs->ifa_netmask)->sin_addr;
-	      broadcast = ((struct sockaddr_in *) addrs->ifa_broadaddr)->sin_addr; 
+	      if (addrs->ifa_broadaddr)
+		broadcast = ((struct sockaddr_in *) addrs->ifa_broadaddr)->sin_addr; 
+	      else 
+		broadcast.s_addr = 0;	      
 	      if (!((*callback)(addr, iface_index, netmask, broadcast, parm)))
 		goto err;
 	    }

src/dhcp-common.c

@@ -679,7 +679,7 @@ void log_context(int family, struct dhcp_context *context)
 
   void *start = &context->start;
   void *end = &context->end;
-  char *n = "", *m = "", *p = daemon->namebuff;
+  char *template = "", *p = daemon->namebuff;
   
   *p = 0;
     
@@ -694,67 +694,59 @@ void log_context(int family, struct dhcp_context *context)
       end = &context->end6;
     }
 #endif
-	
-  if ((context->flags & CONTEXT_DHCP) || 
-      !(context->flags & (CONTEXT_CONSTRUCTED | CONTEXT_TEMPLATE)))
+
+  if (family != AF_INET && (context->flags & CONTEXT_DEPRECATE))
+    strcpy(daemon->namebuff, _(", prefix deprecated"));
+  else
     {
-      if (family != AF_INET && (context->flags & CONTEXT_DEPRECATE))
-	strcpy(daemon->namebuff, _(", prefix deprecated"));
-      else
-	{
-	  p += sprintf(p, _(", lease time "));
-	  m = p;
-	  prettyprint_time(p, context->lease_time);
-	  p += strlen(p);
-	}
-    }
-    
+      p += sprintf(p, _(", lease time "));
+      prettyprint_time(p, context->lease_time);
+      p += strlen(p);
+    }	
+
 #ifdef HAVE_DHCP6
   if (context->flags & CONTEXT_CONSTRUCTED)
-   {
-     char ifrn_name[IFNAMSIZ];
-     if (indextoname(daemon->doing_dhcp6 ? daemon->dhcp6fd : daemon->icmp6fd, context->if_index, ifrn_name))
-       {
-	 n = p;
-	 p += sprintf(p, ", constructed for %s", ifrn_name);
-       }
-   }
-
- if (context->flags & CONTEXT_TEMPLATE)
-   {
-     n = p;
-     p += sprintf(p, ", template for %s%s", context->template_interface, 
-		  (context->flags & CONTEXT_WILDCARD) ? "*" : "");  
-   }
+    {
+      char ifrn_name[IFNAMSIZ];
+      
+      template = p;
+      p += sprintf(p, ", ");
+      
+      if (indextoname(daemon->doing_dhcp6 ? daemon->dhcp6fd : daemon->icmp6fd, context->if_index, ifrn_name))
+	sprintf(p, "constructed for %s", ifrn_name);
+    }
+  else if (context->flags & CONTEXT_TEMPLATE)
+    {
+      template = p;
+      p += sprintf(p, ", ");
+       
+      sprintf(p, "template for %s%s", context->template_interface, 
+	      (context->flags & CONTEXT_WILDCARD) ? "*" : "");  
+    }
 #endif
-
+     
   if ((context->flags & CONTEXT_DHCP) || family == AF_INET) 
     {
       inet_ntop(family, start, daemon->dhcp_buff, 256);
       inet_ntop(family, end, daemon->dhcp_buff3, 256);
       my_syslog(MS_DHCP | LOG_INFO, 
 	      (context->flags & CONTEXT_RA_STATELESS) ? 
-	      _("%s stateless on %s%.0s%.0s") :
+	      _("%s stateless on %s%.0s%.0s%s") :
 	      (context->flags & CONTEXT_STATIC) ? 
-	      _("%s, static leases only on %.0s%s%s") :
+	      _("%s, static leases only on %.0s%s%s%.0s") :
 	      (context->flags & CONTEXT_PROXY) ?
-	      _("%s, proxy on subnet %.0s%s%.0s") :
-	      _("%s, IP range %s -- %s%s"),
+	      _("%s, proxy on subnet %.0s%s%.0s%.0s") :
+	      _("%s, IP range %s -- %s%s%.0s"),
 	      (family != AF_INET) ? "DHCPv6" : "DHCP",
-	      daemon->dhcp_buff, daemon->dhcp_buff3, daemon->namebuff);
+		daemon->dhcp_buff, daemon->dhcp_buff3, daemon->namebuff, template);
     }
   
 #ifdef HAVE_DHCP6
   if (context->flags & CONTEXT_RA_NAME)
-    my_syslog(MS_DHCP | LOG_INFO, _("DHCPv4-derived IPv6 names on %s%s"), 
-	      daemon->addrbuff, n);
-  
+    my_syslog(MS_DHCP | LOG_INFO, _("DHCPv4-derived IPv6 names on %s%s"), daemon->addrbuff, template);
        
-  if (context->flags & CONTEXT_RA)
-    my_syslog(MS_DHCP | LOG_INFO, _("router advertisement on %s%s%s"), 
-	      daemon->addrbuff,
-	      (context->flags & (CONTEXT_CONSTRUCTED | CONTEXT_TEMPLATE)) ? "" : ", prefix valid ",
-	      (context->flags & (CONTEXT_CONSTRUCTED | CONTEXT_TEMPLATE)) ? n : m);
+  if ((context->flags & CONTEXT_RA) || (option_bool(OPT_RA) && (context->flags & CONTEXT_DHCP) && family == AF_INET6)) 
+    my_syslog(MS_DHCP | LOG_INFO, _("router advertisement on %s%s"), daemon->addrbuff, template);
 #endif
 
 }

src/dhcp6.c

@@ -211,7 +211,8 @@ static int complete_context6(struct in6_addr *local,  int prefix,
       
       for (context = daemon->dhcp6; context; context = context->next)
 	{
-	  if (!(context->flags & CONTEXT_TEMPLATE) &&
+	  if ((context->flags & CONTEXT_DHCP) &&
+	      !(context->flags & CONTEXT_TEMPLATE) &&
 	      prefix == context->prefix &&
 	      is_same_net6(local, &context->start6, prefix) &&
 	      is_same_net6(local, &context->end6, prefix))
@@ -328,6 +329,7 @@ int address6_allocate(struct dhcp_context *context,  unsigned char *clid, int cl
   return 0;
 }
 
+/* can dynamically allocate addr */
 struct dhcp_context *address6_available(struct dhcp_context *context, 
 					struct in6_addr *taddr,
 					struct dhcp_netid *netids)
@@ -352,6 +354,21 @@ struct dhcp_context *address6_available(struct dhcp_context *context,
   return NULL;
 }
 
+/* address OK if configured */
+struct dhcp_context *address6_valid(struct dhcp_context *context, 
+					struct in6_addr *taddr,
+					struct dhcp_netid *netids)
+{
+  struct dhcp_context *tmp;
+ 
+  for (tmp = context; tmp; tmp = tmp->current)
+    if (is_same_net6(&tmp->start6, taddr, tmp->prefix) &&
+	match_netid(tmp->filter, netids, 1))
+      return tmp;
+
+  return NULL;
+}
+
 struct dhcp_context *narrow_context6(struct dhcp_context *context, 
 				     struct in6_addr *taddr,
 				     struct dhcp_netid *netids)
@@ -366,21 +383,12 @@ struct dhcp_context *narrow_context6(struct dhcp_context *context,
   
   struct dhcp_context *tmp;
 
-  if (!(tmp = address6_available(context, taddr, netids)))
-    {
-      for (tmp = context; tmp; tmp = tmp->current)
-        if (match_netid(tmp->filter, netids, 1) &&
-            is_same_net6(taddr, &tmp->start6, tmp->prefix) && 
-            (tmp->flags & CONTEXT_STATIC))
-          break;
-      
-      if (!tmp)
-        for (tmp = context; tmp; tmp = tmp->current)
-          if (match_netid(tmp->filter, netids, 1) &&
-              is_same_net6(taddr, &tmp->start6, tmp->prefix) &&
-              !(tmp->flags & CONTEXT_PROXY))
-            break;
-    }
+  if (!(tmp = address6_available(context, taddr, netids)) &&
+      !(tmp = address6_valid(context, taddr, netids)))
+    for (tmp = context; tmp; tmp = tmp->current)
+      if (match_netid(tmp->filter, netids, 1) &&
+	  is_same_net6(taddr, &tmp->start6, tmp->prefix))
+	break;
   
   /* Only one context allowed now */
   if (tmp)
@@ -391,11 +399,19 @@ struct dhcp_context *narrow_context6(struct dhcp_context *context,
 
 static int is_config_in_context6(struct dhcp_context *context, struct dhcp_config *config)
 {
-  if (!context) /* called via find_config() from lease_update_from_configs() */
-    return 1; 
+  /* expand wildcard on contructed contexts */
+  if ((config->flags & CONFIG_WILDCARD) && 
+      (context->flags & CONTEXT_CONSTRUCTED))
+    {
+      u64 addrpart = addr6part(&config->addr6);
+      config->addr6 = context->start6;
+      setaddr6part(&config->addr6, addrpart);
+      return 1;
+    }
+  
   if (!(config->flags & CONFIG_ADDR6) || is_addr_in_context6(context, &config->addr6))
     return 1;
-  
+      
   return 0;
 }
 

src/dnsmasq.c

@@ -1337,7 +1337,7 @@ static void check_dns_listeners(fd_set *set, time_t now)
 
       if (listener->tcpfd != -1 && FD_ISSET(listener->tcpfd, set))
 	{
-	  int confd;
+	  int confd, client_ok = 1;
 	  struct irec *iface = NULL;
 	  pid_t p;
 	  union mysockaddr tcp_addr;
@@ -1348,25 +1348,58 @@ static void check_dns_listeners(fd_set *set, time_t now)
 	  if (confd == -1 ||
 	      getsockname(confd, (struct sockaddr *)&tcp_addr, &tcp_len) == -1)
 	    continue;
-	  
-	  if (option_bool(OPT_NOWILD) || option_bool(OPT_CLEVERBIND))
+
+	   if (option_bool(OPT_NOWILD))
 	    iface = listener->iface; /* May be NULL */
-	  else
-	    {
-	      /* Check for allowed interfaces when binding the wildcard address:
-		 we do this by looking for an interface with the same address as 
-		 the local address of the TCP connection, then looking to see if that's
-		 an allowed interface. As a side effect, we get the netmask of the
-		 interface too, for localisation. */
-	      
-	      /* interface may be new since startup */
-	      if (enumerate_interfaces())
-		for (iface = daemon->interfaces; iface; iface = iface->next)
-		  if (sockaddr_isequal(&iface->addr, &tcp_addr))
-		    break;
-	    }
-	  
-	  if (!iface && !(option_bool(OPT_NOWILD) || option_bool(OPT_CLEVERBIND)))
+	   else 
+	     {
+	       int if_index;
+	       char intr_name[IF_NAMESIZE];
+ 
+	       /* In full wildcard mode, need to refresh interface list.
+		  This happens automagically in CLEVERBIND */
+	       if (!option_bool(OPT_CLEVERBIND))
+		 enumerate_interfaces();
+	       
+	       /* if we can find the arrival interface, check it's one that's allowed */
+	       if ((if_index = tcp_interface(confd, tcp_addr.sa.sa_family)) != 0 &&
+		   indextoname(listener->tcpfd, if_index, intr_name))
+		 {
+		   struct all_addr addr;
+		   addr.addr.addr4 = tcp_addr.in.sin_addr;
+#ifdef HAVE_IPV6
+		   if (tcp_addr.sa.sa_family == AF_INET6)
+		     addr.addr.addr6 = tcp_addr.in6.sin6_addr;
+#endif
+		   
+		   for (iface = daemon->interfaces; iface; iface = iface->next)
+		     if (iface->index == if_index)
+		       break;
+		   
+		   if (!iface && !loopback_exception(listener->tcpfd, tcp_addr.sa.sa_family, &addr, intr_name))
+		     client_ok = 0;
+		 }
+	       
+	       if (option_bool(OPT_CLEVERBIND))
+		 iface = listener->iface; /* May be NULL */
+	       else
+		 {
+		    /* Check for allowed interfaces when binding the wildcard address:
+		       we do this by looking for an interface with the same address as 
+		       the local address of the TCP connection, then looking to see if that's
+		       an allowed interface. As a side effect, we get the netmask of the
+		      interface too, for localisation. */
+		   
+		   for (iface = daemon->interfaces; iface; iface = iface->next)
+		     if (sockaddr_isequal(&iface->addr, &tcp_addr))
+		       break;
+		   
+		   if (!iface)
+		     client_ok = 0;
+		 }
+	     }
+	  	  
+	  if (!client_ok)
 	    {
 	      shutdown(confd, SHUT_RDWR);
 	      close(confd);

src/dnsmasq.h

@@ -593,6 +593,7 @@ struct dhcp_config {
 #define CONFIG_DECLINED       1024    /* address declined by client */
 #define CONFIG_BANK           2048    /* from dhcp hosts file */
 #define CONFIG_ADDR6          4096
+#define CONFIG_WILDCARD       8192
 
 struct dhcp_opt {
   int opt, len, flags;
@@ -1001,7 +1002,9 @@ void create_wildcard_listeners(void);
 void create_bound_listeners(int die);
 int is_dad_listeners(void);
 int iface_check(int family, struct all_addr *addr, char *name, int *auth_dns);
+int loopback_exception(int fd, int family, struct all_addr *addr, char *name);
 int fix_fd(int fd);
+int tcp_interface(int fd, int af);
 struct in_addr get_ifaddr(char *intr);
 #ifdef HAVE_IPV6
 int set_ipv6pktinfo(int fd);
@@ -1149,6 +1152,9 @@ int is_addr_in_context6(struct dhcp_context *context, struct in6_addr *addr);
 struct dhcp_context *address6_available(struct dhcp_context *context, 
 					struct in6_addr *taddr,
 					struct dhcp_netid *netids);
+struct dhcp_context *address6_valid(struct dhcp_context *context, 
+				    struct in6_addr *taddr,
+				    struct dhcp_netid *netids);
 struct dhcp_context *narrow_context6(struct dhcp_context *context, 
 				     struct in6_addr *taddr,
 				     struct dhcp_netid *netids);

src/forward.c

@@ -763,10 +763,17 @@ void receive_query(struct listener *listen, time_t now)
       
       /* enforce available interface configuration */
       
-      if (!indextoname(listen->fd, if_index, ifr.ifr_name) ||
-	  !iface_check(listen->family, &dst_addr, ifr.ifr_name, &auth_dns))
+      if (!indextoname(listen->fd, if_index, ifr.ifr_name))
 	return;
       
+      if (!iface_check(listen->family, &dst_addr, ifr.ifr_name, &auth_dns))
+	{
+	   if (!option_bool(OPT_CLEVERBIND))
+	     enumerate_interfaces(); 
+	   if (!loopback_exception(listen->fd, listen->family, &dst_addr, ifr.ifr_name))
+	     return;
+	}
+
       if (listen->family == AF_INET && option_bool(OPT_LOCALISE))
 	{
 	  struct irec *iface;
@@ -780,7 +787,7 @@ void receive_query(struct listener *listen, time_t now)
 	      break;
 	  
 	  /* interface may be new */
-	  if (!iface)
+	  if (!iface && !option_bool(OPT_CLEVERBIND))
 	    enumerate_interfaces(); 
 	  
 	  for (iface = daemon->interfaces; iface; iface = iface->next)

src/network.c

@@ -170,7 +170,40 @@ int iface_check(int family, struct all_addr *addr, char *name, int *auth)
 
   return ret; 
 }
-      
+
+
+/* Fix for problem that the kernel sometimes reports the loopback inerface as the
+   arrival interface when a packet originates locally, even when sent to address of 
+   an interface other than the loopback. Accept packet if it arrived via a loopback 
+   interface, even when we're not accepting packets that way, as long as the destination
+   address is one we're believing. Interface list must be up-to-date before calling. */
+int loopback_exception(int fd, int family, struct all_addr *addr, char *name)    
+{
+  struct ifreq ifr;
+  struct irec *iface;
+
+  strncpy(ifr.ifr_name, name, IF_NAMESIZE);
+  if (ioctl(fd, SIOCGIFFLAGS, &ifr) != -1 &&
+      ifr.ifr_flags & IFF_LOOPBACK)
+    {
+      for (iface = daemon->interfaces; iface; iface = iface->next)
+	if (iface->addr.sa.sa_family == family)
+	  {
+	    if (family == AF_INET)
+	      {
+		if (iface->addr.in.sin_addr.s_addr == addr->addr.addr4.s_addr)
+		  return 1;
+	      }
+#ifdef HAVE_IPV6
+	    else if (IN6_ARE_ADDR_EQUAL(&iface->addr.in6.sin6_addr, &addr->addr.addr6))
+	      return 1;
+#endif
+	    
+	  }
+    }
+  return 0;
+}
+
 static int iface_allowed(struct irec **irecp, int if_index, 
 			 union mysockaddr *addr, struct in_addr netmask, int dad) 
 {
@@ -459,6 +492,77 @@ int set_ipv6pktinfo(int fd)
   return 0;
 }
 #endif
+
+
+/* Find the interface on which a TCP connection arrived, if possible, or zero otherwise. */
+int tcp_interface(int fd, int af)
+{ 
+  int if_index = 0;
+
+#ifdef HAVE_LINUX_NETWORK
+  int opt = 1;
+  struct cmsghdr *cmptr;
+  struct msghdr msg;
+  
+  /* use mshdr do that the CMSDG_* macros are available */
+  msg.msg_control = daemon->packet;
+  msg.msg_controllen = daemon->packet_buff_sz;
+  
+  /* we overwrote the buffer... */
+  daemon->srv_save = NULL;
+  
+  if (af == AF_INET)
+    {
+      if (setsockopt(fd, IPPROTO_IP, IP_PKTINFO, &opt, sizeof(opt)) != -1 &&
+	  getsockopt(fd, IPPROTO_IP, IP_PKTOPTIONS, msg.msg_control, (socklen_t *)&msg.msg_controllen) != -1)
+	for (cmptr = CMSG_FIRSTHDR(&msg); cmptr; cmptr = CMSG_NXTHDR(&msg, cmptr))
+	  if (cmptr->cmsg_level == IPPROTO_IP && cmptr->cmsg_type == IP_PKTINFO)
+            {
+              union {
+                unsigned char *c;
+                struct in_pktinfo *p;
+              } p;
+	      
+	      p.c = CMSG_DATA(cmptr);
+	      if_index = p.p->ipi_ifindex;
+	    }
+    }
+#ifdef HAVE_IPV6
+  else
+    {
+      /* Only the RFC-2292 API has the ability to find the interface for TCP connections,
+	 it was removed in RFC-3542 !!!! 
+
+	 Fortunately, Linux kept the 2292 ABI when it moved to 3542. The following code always
+	 uses the old ABI, and should work with pre- and post-3542 kernel headers */
+
+#ifdef IPV6_2292PKTOPTIONS   
+#  define PKTOPTIONS IPV6_2292PKTOPTIONS
+#else
+#  define PKTOPTIONS IPV6_PKTOPTIONS
+#endif
+
+      if (set_ipv6pktinfo(fd) &&
+	  getsockopt(fd, IPPROTO_IPV6, PKTOPTIONS, msg.msg_control, (socklen_t *)&msg.msg_controllen) != -1)
+	{
+          for (cmptr = CMSG_FIRSTHDR(&msg); cmptr; cmptr = CMSG_NXTHDR(&msg, cmptr))
+            if (cmptr->cmsg_level == IPPROTO_IPV6 && cmptr->cmsg_type == daemon->v6pktinfo)
+              {
+                union {
+                  unsigned char *c;
+                  struct in6_pktinfo *p;
+                } p;
+                p.c = CMSG_DATA(cmptr);
+		
+		if_index = p.p->ipi6_ifindex;
+              }
+	}
+    }
+#endif /* IPV6 */
+#endif /* Linux */
+ 
+  return if_index;
+}
       
 static struct listener *create_listeners(union mysockaddr *addr, int do_tftp, int dienow)
 {

src/option.c

@@ -392,7 +392,7 @@ static struct {
   { LOPT_RR, ARG_DUP, "<name>,<RR-number>,[<data>]", gettext_noop("Specify arbitrary DNS resource record"), NULL },
   { LOPT_CLVERBIND, OPT_CLEVERBIND, NULL, gettext_noop("Bind to interfaces in use - check for new interfaces"), NULL },
   { LOPT_AUTHSERV, ARG_ONE, "<NS>,<interface>", gettext_noop("Export local names to global DNS"), NULL },
-  { LOPT_AUTHZONE, ARG_DUP, "<domain>,<subnet>[,<subnet>]", gettext_noop("Domain to export to global DNS"), NULL },
+  { LOPT_AUTHZONE, ARG_DUP, "<domain>,[<subnet>...]", gettext_noop("Domain to export to global DNS"), NULL },
   { LOPT_AUTHTTL, ARG_ONE, "<integer>", gettext_noop("Set TTL for authoritative replies"), NULL },
   { LOPT_AUTHSOA, ARG_ONE, "<serial>[,...]", gettext_noop("Set authoritive zone information"), NULL },
   { LOPT_AUTHSFS, ARG_DUP, "<NS>[,<NS>...]", gettext_noop("Secondary authoritative nameservers for forward domains"), NULL },
@@ -2468,6 +2468,14 @@ static int one_opt(int option, char *arg, char *errstr, char *gen_err, int comma
 		  
 		  if (!inet_pton(AF_INET6, arg, &new->addr6))
 		    ret_err(_("bad IPv6 address"));
+
+		  for (i= 0; i < 8; i++)
+		    if (new->addr6.s6_addr[i] != 0)
+		      break;
+
+		  /* set WILDCARD if network part all zeros */
+		  if (i == 8)
+		    new->flags |= CONFIG_WILDCARD;
 		  
 		  new->flags |= CONFIG_ADDR6;
 		}

src/rfc3315.c

@@ -82,18 +82,20 @@ static int dhcp6_maybe_relay(struct in6_addr *link_address, struct dhcp_netid **
 	{
 	  struct dhcp_context *c;
 	  context = NULL;
+	   
+	  if (!IN6_IS_ADDR_LOOPBACK(link_address) &&
+	      !IN6_IS_ADDR_LINKLOCAL(link_address) &&
+	      !IN6_IS_ADDR_MULTICAST(link_address))
+	    for (c = daemon->dhcp6; c; c = c->next)
+	      if ((c->flags & CONTEXT_DHCP) &&
+		  !(c->flags & CONTEXT_TEMPLATE) &&
+		  is_same_net6(link_address, &c->start6, c->prefix) &&
+		  is_same_net6(link_address, &c->end6, c->prefix))
+		{
+		  c->current = context;
+		  context = c;
+		}
 	  
-	  for (c = daemon->dhcp6; c; c = c->next)
-	    if (!IN6_IS_ADDR_LOOPBACK(link_address) &&
-		!IN6_IS_ADDR_LINKLOCAL(link_address) &&
-		!IN6_IS_ADDR_MULTICAST(link_address) &&
-		is_same_net6(link_address, &c->start6, c->prefix) &&
-		is_same_net6(link_address, &c->end6, c->prefix))
-	      {
-		c->current = context;
-		context = c;
-	      }
-
 	  if (!context)
 	    {
 	      inet_ntop(AF_INET6, link_address, daemon->addrbuff, ADDRSTRLEN); 
@@ -525,32 +527,38 @@ static int dhcp6_no_relay(int msg_type, struct in6_addr *link_address, struct dh
 		  {
 		    struct in6_addr *req_addr = opt6_ptr(ia_option, 0);
 		    requested_time = opt6_uint(ia_option, 16, 4);
-		    
-		    if (!address6_available(context, req_addr, tags) &&
-			(!have_config(config, CONFIG_ADDR6) || memcmp(&config->addr6, req_addr, IN6ADDRSZ) != 0))
+		    struct dhcp_context *dynamic;
+
+		    if ((dynamic = address6_available(context, req_addr, tags)) || address6_valid(context, req_addr, tags))
 		      {
-			if (msg_type == DHCP6REQUEST)
+			if (!dynamic && !(have_config(config, CONFIG_ADDR6) && memcmp(&config->addr6, req_addr, IN6ADDRSZ) == 0))
 			  {
-			    /* host has a lease, but it's not on the correct link */
+			    /* Static range, not configured. */
 			    o1 = new_opt6(OPTION6_STATUS_CODE);
-			    put_opt6_short(DHCP6NOTONLINK);
-			    put_opt6_string("Not on link");
+			    put_opt6_short(DHCP6UNSPEC);
+			    put_opt6_string("Address unavailable");
 			    end_opt6(o1);
 			  }
+			else if (lease6_find_by_addr(req_addr, 128, 0) &&
+				 !(lease = lease6_find(clid, clid_len, ia_type == OPTION6_IA_NA ? LEASE_NA : LEASE_TA, iaid, req_addr)))
+			  {
+			    /* Address leased to another DUID/IAID */
+			    o1 = new_opt6(OPTION6_STATUS_CODE);
+			    put_opt6_short(DHCP6UNSPEC);
+			    put_opt6_string("Address in use");
+			    end_opt6(o1);
+			  } 
+			else 
+			  addrp = req_addr;
 		      }
-		    else if ((lease = lease6_find(NULL, 0, ia_type == OPTION6_IA_NA ? LEASE_NA : LEASE_TA, 
-						  iaid, req_addr)) &&
-			     (clid_len != lease->clid_len ||
-			      memcmp(clid, lease->clid, clid_len) != 0))
+		    else if (msg_type == DHCP6REQUEST)
 		      {
-			/* Address leased to another DUID */
+			/* requested address not on the correct link */
 			o1 = new_opt6(OPTION6_STATUS_CODE);
-			put_opt6_short(DHCP6UNSPEC);
-			put_opt6_string("Address in use");
+			put_opt6_short(DHCP6NOTONLINK);
+			put_opt6_string("Not on link");
 			end_opt6(o1);
-		      } 
-		    else
-		      addrp = req_addr;
+		      }
 		  }
 		else
 		  {
@@ -567,9 +575,9 @@ static int dhcp6_no_relay(int msg_type, struct in6_addr *link_address, struct dh
 			inet_ntop(AF_INET6, &config->addr6, daemon->addrbuff, ADDRSTRLEN);
 			
 			if (ltmp && ltmp->clid && 
-			    (ltmp->clid_len != clid_len || memcmp(ltmp->clid, clid, clid_len) != 0))
-			  my_syslog(MS_DHCP | LOG_WARNING, _("not using configured address %s because it is leased to %s"),
-				    daemon->addrbuff, print_mac(daemon->namebuff, ltmp->clid, ltmp->clid_len));
+			    (ltmp->clid_len != clid_len || memcmp(ltmp->clid, clid, clid_len) != 0 || ltmp->hwaddr_type != iaid))
+			  my_syslog(MS_DHCP | LOG_WARNING, _("not using configured address %s because it is leased to %s#%d"),
+				    daemon->addrbuff, print_mac(daemon->namebuff, ltmp->clid, ltmp->clid_len), ltmp->hwaddr_type);
 			else if (have_config(config, CONFIG_DECLINED) &&
 				 difftime(now, config->decline_time) < (float)DECLINE_BACKOFF)
 			  my_syslog(MS_DHCP | LOG_WARNING, _("not using configured address %s because it was previously declined"), 

src/tftp.c

@@ -61,6 +61,7 @@ void tftp_request(struct listener *listen, time_t now)
   char *name = NULL;
   char *prefix = daemon->tftp_prefix;
   struct tftp_prefix *pref;
+  struct all_addr addra;
 
   union {
     struct cmsghdr align; /* this ensures alignment */
@@ -189,18 +190,22 @@ void tftp_request(struct listener *listen, time_t now)
 	return;
 
       name = namebuff;
+      
+      addra.addr.addr4 = addr.in.sin_addr;
 
 #ifdef HAVE_IPV6
       if (listen->family == AF_INET6)
+	addra.addr.addr6 = addr.in6.sin6_addr;
+#endif
+
+      if (!iface_check(listen->family, &addra, name, NULL))
 	{
-	  if (!iface_check(AF_INET6, (struct all_addr *)&addr.in6.sin6_addr, name, NULL))
+	  if (!option_bool(OPT_CLEVERBIND))
+	    enumerate_interfaces(); 
+	  if (!loopback_exception(listen->tftpfd, listen->family, &addra, name))
 	    return;
 	}
-      else
-#endif
-        if (!iface_check(AF_INET, (struct all_addr *)&addr.in.sin_addr, name, NULL))
-	  return;
-
+      
 #ifdef HAVE_DHCP      
       /* allowed interfaces are the same as for DHCP */
       for (tmp = daemon->dhcp_except; tmp; tmp = tmp->next)