Implement an option for setting static routes with gateway
[project/relayd.git] / route.c
diff --git a/route.c b/route.c
index f436691..6124a7e 100644 (file)
--- a/route.c
+++ b/route.c
 #include <string.h>
 #include <errno.h>
 #include <unistd.h>
+#include <fcntl.h>
+
+#include <linux/fib_rules.h>
 
 #include "relayd.h"
 
 static struct uloop_fd rtnl_sock;
 static unsigned int rtnl_seq, rtnl_dump_seq;
+int route_table = 16800;
+
+static void rtnl_flush(void)
+{
+       int fd;
+
+       fd = open("/proc/sys/net/ipv4/route/flush", O_WRONLY);
+       if (fd < 0)
+               return;
+
+       write(fd, "-1", 2);
+       close(fd);
+}
 
-static void rtnl_route_set(struct relayd_host *host, bool add)
+static void
+rtnl_rule_request(struct relayd_interface *rif, bool add, bool defgw_workaround)
 {
        static struct {
                struct nlmsghdr nl;
                struct rtmsg rt;
                struct {
                        struct rtattr rta;
-                       uint8_t ipaddr[4];
-               } __packed dst;
+                       int table;
+               } __packed table;
                struct {
                        struct rtattr rta;
-                       int ifindex;
+                       char ifname[IFNAMSIZ + 1];
                } __packed dev;
-       } __packed req;
+       } __packed req = {
+               .rt = {
+                       .rtm_family = AF_INET,
+                       .rtm_table = RT_TABLE_UNSPEC,
+                       .rtm_scope = RT_SCOPE_UNIVERSE,
+                       .rtm_protocol = RTPROT_BOOT,
+               },
+               .table.rta = {
+                       .rta_type = FRA_TABLE,
+                       .rta_len = sizeof(req.table),
+               },
+       };
 
-       memset(&req, 0, sizeof(req));
+       int padding = sizeof(req.dev.ifname);
 
-       req.nl.nlmsg_len = sizeof(req);
-       req.rt.rtm_family = AF_INET;
-       req.rt.rtm_dst_len = 32;
+       if (!defgw_workaround) {
+               req.dev.rta.rta_type = FRA_IFNAME;
+               padding -= strlen(rif->ifname) + 1;
+               strcpy(req.dev.ifname, rif->ifname);
+               req.dev.rta.rta_len = sizeof(req.dev.rta) + strlen(rif->ifname) + 1;
+       } else {
+               req.dev.rta.rta_type = FRA_PRIORITY;
+               req.dev.rta.rta_len = sizeof(req.dev.rta) + sizeof(uint32_t);
+               padding -= sizeof(uint32_t);
+               *((uint32_t *) &req.dev.ifname) = 1;
+       }
+       req.table.table = rif->rt_table;
+       req.nl.nlmsg_len = sizeof(req) - padding;
 
-       req.dst.rta.rta_type = RTA_DST;
-       req.dst.rta.rta_len = sizeof(req.dst);
-       memcpy(req.dst.ipaddr, host->ipaddr, sizeof(req.dst.ipaddr));
+       req.nl.nlmsg_flags = NLM_F_REQUEST;
+       if (add) {
+               req.nl.nlmsg_type = RTM_NEWRULE;
+               req.nl.nlmsg_flags |= NLM_F_CREATE | NLM_F_EXCL;
+
+               req.rt.rtm_type = RTN_UNICAST;
+       } else {
+               req.nl.nlmsg_type = RTM_DELRULE;
+               req.rt.rtm_type = RTN_UNSPEC;
+       }
+
+       send(rtnl_sock.fd, &req, req.nl.nlmsg_len, 0);
+       rtnl_flush();
+}
+
+
+static void
+rtnl_route_request(struct relayd_interface *rif, struct relayd_host *host,
+                  struct relayd_route *route, bool add)
+{
+       static struct {
+               struct nlmsghdr nl;
+               struct rtmsg rt;
+               struct {
+                       struct rtattr rta;
+                       uint8_t ipaddr[4];
+               } __packed dst;
+               struct {
+                       struct rtattr rta;
+                       int table;
+               } __packed table;
+               struct {
+                       struct rtattr rta;
+                       int ifindex;
+               } __packed dev;
+               struct {
+                       struct rtattr rta;
+                       uint8_t ipaddr[4];
+               } __packed gw;
+       } __packed req = {
+               .rt = {
+                       .rtm_family = AF_INET,
+                       .rtm_dst_len = 32,
+                       .rtm_table = RT_TABLE_MAIN,
+               },
+               .table.rta = {
+                       .rta_type = RTA_TABLE,
+                       .rta_len = sizeof(req.table),
+               },
+               .dst.rta = {
+                       .rta_len = sizeof(req.dst),
+               },
+               .dev.rta = {
+                       .rta_type = RTA_OIF,
+                       .rta_len = sizeof(req.dev),
+               },
+               .gw.rta = {
+                       .rta_type = RTA_GATEWAY,
+                       .rta_len = sizeof(req.gw),
+               },
+       };
+       int pktlen = sizeof(req);
 
-       req.dev.rta.rta_type = RTA_OIF;
-       req.dev.rta.rta_len = sizeof(req.dev);
        req.dev.ifindex = host->rif->sll.sll_ifindex;
+       req.table.table = rif->rt_table;
 
        req.nl.nlmsg_flags = NLM_F_REQUEST;
-       req.rt.rtm_table = RT_TABLE_MAIN;
        if (add) {
                req.nl.nlmsg_type = RTM_NEWROUTE;
                req.nl.nlmsg_flags |= NLM_F_CREATE | NLM_F_REPLACE;
 
                req.rt.rtm_protocol = RTPROT_BOOT;
-               req.rt.rtm_scope = RT_SCOPE_LINK;
+               if (route) {
+                       req.rt.rtm_scope = RT_SCOPE_UNIVERSE;
+               } else {
+                       req.rt.rtm_scope = RT_SCOPE_LINK;
+               }
                req.rt.rtm_type = RTN_UNICAST;
        } else {
                req.nl.nlmsg_type = RTM_DELROUTE;
                req.rt.rtm_scope = RT_SCOPE_NOWHERE;
        }
 
-       send(rtnl_sock.fd, &req, sizeof(req), 0);
+       req.dst.rta.rta_type = RTA_DST;
+       if (route) {
+               DPRINTF(2, "%s: add route to "IP_FMT"/%d via "IP_FMT"\n",
+                       host->rif->ifname, IP_BUF(route->dest), route->mask, IP_BUF(host->ipaddr));
+               if (!route->mask) {
+                       /* No DST -> reuse the DST attribute as the gateway attribute */
+                       memcpy(req.dst.ipaddr, host->ipaddr, sizeof(req.dst.ipaddr));
+                       req.dst.rta.rta_type = RTA_GATEWAY;
+                       pktlen -= sizeof(req.gw);
+                       req.rt.rtm_dst_len = 0;
+               } else {
+                       memcpy(req.gw.ipaddr, host->ipaddr, sizeof(req.gw.ipaddr));
+                       memcpy(req.dst.ipaddr, route->dest, sizeof(req.dst.ipaddr));
+                       req.rt.rtm_dst_len = route->mask;
+               }
+       } else {
+               pktlen -= sizeof(req.gw);
+               req.rt.rtm_dst_len = 32;
+               memcpy(req.dst.ipaddr, host->ipaddr, sizeof(req.dst.ipaddr));
+       }
+
+       req.nl.nlmsg_len = pktlen;
+       if (route)
+               rtnl_rule_request(rif, true, true);
+       send(rtnl_sock.fd, &req, pktlen, 0);
+       if (route)
+               rtnl_rule_request(rif, false, true);
+       rtnl_flush();
+}
+
+void
+rtnl_route_set(struct relayd_host *host, struct relayd_route *route, bool add)
+{
+       struct relayd_interface *rif;
+
+       list_for_each_entry(rif, &interfaces, list) {
+               if (rif == host->rif)
+                       continue;
+
+               rtnl_route_request(rif, host, route, add);
+       }
 }
 
-void relayd_add_route(struct relayd_host *host)
+void relayd_add_interface_routes(struct relayd_interface *rif)
 {
-       rtnl_route_set(host, true);
+       rif->rt_table = route_table++;
+       rtnl_rule_request(rif, true, false);
 }
 
-void relayd_del_route(struct relayd_host *host)
+void relayd_del_interface_routes(struct relayd_interface *rif)
 {
-       rtnl_route_set(host, false);
+       rtnl_rule_request(rif, false, false);
 }
 
 #ifndef NDA_RTA