ovn: Support a new Logical_Switch_Port.type - 'external'

In the case of OpenStack + OVN, when the VMs are booted on hypervisors supporting SR-IOV nics, there are no OVS ports for these VMs. When these VMs sends DHCPv4, DHPCv6 or IPv6 Router Solicitation requests, the local ovn-controller cannot reply to these packets. OpenStack Neutron dhcp agent service needs to be run to serve these requests. With the new logical port type - 'external', OVN itself can handle these requests avoiding the need to deploy any external services like neutron dhcp agent. To make use of this feature, CMS has to - create a logical port for such VMs - set the type to 'external' - create an HA chassis group and associate the logical port to it or associate an already existing HA chassis group. - create a localnet port for the logical switch - configure the ovn-bridge-mappings option in the OVS db. HA chassis with the highest priority becomes the master of the HA chassis group and the ovn-controller running in that 'chassis', claims the Port_Binding for that logical port and it adds the necessary DHCPv4/v6 OF flows. Since the packet enters the logical switch pipeline via the localnet port, the inport register (reg14) is set to the tunnel key of localnet port in the match conditions. In case the chassis goes down for some reason, next higher priority HA chassis becomes the master and claims the port. When the VM with the external port, sends an ARP request for the router ips, only the chassis which has claimed the port, will reply to the ARP requests. Rest of the chassis on receiving these packets drop them in the ingress switch datapath stage - S_SWITCH_IN_EXTERNAL_PORT which is just before S_SWITCH_IN_L2_LKUP. This would guarantee that only the chassis which has claimed the external ports will run the router datapath pipeline. Acked-by: Mark Michelson <[email protected]> Acked-by: Han Zhou <[email protected]> Signed-off-by: Numan Siddique <[email protected]> Signed-off-by: Ben Pfaff <[email protected]>
qihuaheng · Apr 16, 2019 · 9608008 · 9608008
1 parent bddb73d
commit 9608008
Show file tree

Hide file tree

Showing 11 changed files with 1,141 additions and 23 deletions.
diff --git a/NEWS b/NEWS
@@ -25,6 +25,7 @@ Post-v2.11.0
    - OVN:
      * Select IPAM mac_prefix in a random manner if not provided by the user
      * Added the HA chassis group support.
+     * Added 'external' logical port support.
    - New QoS type "linux-netem" on Linux.
 
 v2.11.0 - 19 Feb 2019

diff --git a/ovn/controller/binding.c b/ovn/controller/binding.c
@@ -502,6 +502,18 @@ consider_local_datapath(struct ovsdb_idl_txn *ovnsb_idl_txn,
          * for them. */
         sset_add(local_lports, binding_rec->logical_port);
         our_chassis = false;
+    } else if (!strcmp(binding_rec->type, "external")) {
+        if (ha_chassis_group_contains(binding_rec->ha_chassis_group,
+                                      chassis_rec)) {
+            our_chassis = ha_chassis_group_is_active(
+                binding_rec->ha_chassis_group,
+                active_tunnels, chassis_rec);
+
+            add_local_datapath(sbrec_datapath_binding_by_key,
+                               sbrec_port_binding_by_datapath,
+                               sbrec_port_binding_by_name,
+                               binding_rec->datapath, false, local_datapaths);
+        }
     }
 
     if (our_chassis

diff --git a/ovn/controller/ovn-controller.c b/ovn/controller/ovn-controller.c
@@ -145,6 +145,7 @@ update_sb_monitors(struct ovsdb_idl *ovnsb_idl,
      * ports that have a Gateway_Chassis that point's to our own
      * chassis */
     sbrec_port_binding_add_clause_type(&pb, OVSDB_F_EQ, "chassisredirect");
+    sbrec_port_binding_add_clause_type(&pb, OVSDB_F_EQ, "external");
     if (chassis) {
         /* This should be mostly redundant with the other clauses for port
          * bindings, but it allows us to catch any ports that are assigned to

diff --git a/ovn/lib/ovn-util.c b/ovn/lib/ovn-util.c
@@ -319,6 +319,7 @@ static const char *OVN_NB_LSP_TYPES[] = {
     "localport",
     "router",
     "vtep",
+    "external",
 };
 
 bool

diff --git a/ovn/northd/ovn-northd.8.xml b/ovn/northd/ovn-northd.8.xml
@@ -626,7 +626,8 @@ nd_na_router {
     <p>
       This table adds the DHCPv4 options to a DHCPv4 packet from the
       logical ports configured with IPv4 address(es) and DHCPv4 options,
-      and similarly for DHCPv6 options.
+      and similarly for DHCPv6 options. This table also adds flows for the
+      logical ports of type <code>external</code>.
     </p>
 
     <ul>
@@ -827,7 +828,39 @@ output;
       </li>
     </ul>
 
-    <h3>Ingress Table 16 Destination Lookup</h3>
+    <h3>Ingress table 16 External ports</h3>
+
+    <p>
+      Traffic from the <code>external</code> logical ports enter the ingress
+      datapath pipeline via the <code>localnet</code> port. This table adds the
+      below logical flows to handle the traffic from these ports.
+    </p>
+
+    <ul>
+      <li>
+        <p>
+          A priority-100 flow is added for each <code>external</code> logical
+          port which doesn't reside on a chassis to drop the ARP/IPv6 NS
+          request to the router IP(s) (of the logical switch) which matches
+          on the <code>inport</code> of the <code>external</code> logical port
+          and the valid <code>eth.src</code> address(es) of the
+          <code>external</code> logical port.
+        </p>
+
+        <p>
+          This flow guarantees that the ARP/NS request to the router IP
+          address from the external ports is responded by only the chassis
+          which has claimed these external ports. All the other chassis,
+          drops these packets.
+        </p>
+      </li>
+
+      <li>
+        A priority-0 flow that matches all packets to advances to table 17.
+      </li>
+    </ul>
+
+    <h3>Ingress Table 17 Destination Lookup</h3>
 
     <p>
       This table implements switching behavior.  It contains these logical

diff --git a/ovn/northd/ovn-northd.c b/ovn/northd/ovn-northd.c
@@ -119,7 +119,8 @@ enum ovn_stage {
     PIPELINE_STAGE(SWITCH, IN,  DHCP_RESPONSE, 13, "ls_in_dhcp_response") \
     PIPELINE_STAGE(SWITCH, IN,  DNS_LOOKUP,    14, "ls_in_dns_lookup")    \
     PIPELINE_STAGE(SWITCH, IN,  DNS_RESPONSE,  15, "ls_in_dns_response")  \
-    PIPELINE_STAGE(SWITCH, IN,  L2_LKUP,       16, "ls_in_l2_lkup")       \
+    PIPELINE_STAGE(SWITCH, IN,  EXTERNAL_PORT, 16, "ls_in_external_port") \
+    PIPELINE_STAGE(SWITCH, IN,  L2_LKUP,       17, "ls_in_l2_lkup")       \
                                                                           \
     /* Logical switch egress stages. */                                   \
     PIPELINE_STAGE(SWITCH, OUT, PRE_LB,       0, "ls_out_pre_lb")         \
@@ -2342,6 +2343,18 @@ ovn_port_update_sbrec(struct northd_context *ctx,
             }
 
             sbrec_port_binding_set_nat_addresses(op->sb, NULL, 0);
+
+            if (!strcmp(op->nbsp->type, "external")) {
+                if (op->nbsp->ha_chassis_group) {
+                    sync_ha_chassis_group_for_sbpb(
+                        ctx, op->nbsp->ha_chassis_group,
+                        sbrec_chassis_by_name, op->sb);
+                    sset_add(active_ha_chassis_grps,
+                             op->nbsp->ha_chassis_group->name);
+                } else {
+                    sbrec_port_binding_set_ha_chassis_group(op->sb, NULL);
+                }
+            }
         } else {
             const char *chassis = NULL;
             if (op->peer && op->peer->od && op->peer->od->nbr) {
@@ -3039,6 +3052,12 @@ lsp_is_up(const struct nbrec_logical_switch_port *lsp)
     return !lsp->up || *lsp->up;
 }
 
+static bool
+lsp_is_external(const struct nbrec_logical_switch_port *nbsp)
+{
+    return !strcmp(nbsp->type, "external");
+}
+
 static bool
 build_dhcpv4_action(struct ovn_port *op, ovs_be32 offer_ip,
                     struct ds *options_action, struct ds *response_action,
@@ -3936,6 +3955,10 @@ build_acls(struct ovn_datapath *od, struct hmap *lflows,
      * logical ports of the datapath if the CMS has configured DHCPv4 options.
      * */
     for (size_t i = 0; i < od->nbs->n_ports; i++) {
+        if (lsp_is_external(od->nbs->ports[i])) {
+            continue;
+        }
+
         if (od->nbs->ports[i]->dhcpv4_options) {
             const char *server_id = smap_get(
                 &od->nbs->ports[i]->dhcpv4_options->options, "server_id");
@@ -4322,6 +4345,10 @@ build_lswitch_flows(struct hmap *datapaths, struct hmap *ports,
             continue;
         }
 
+        if (lsp_is_external(op->nbsp)) {
+            continue;
+        }
+
         ds_clear(&match);
         ds_clear(&actions);
         ds_put_format(&match, "inport == %s", op->json_key);
@@ -4388,6 +4415,10 @@ build_lswitch_flows(struct hmap *datapaths, struct hmap *ports,
             continue;
         }
 
+        if (lsp_is_external(op->nbsp)) {
+            continue;
+        }
+
         for (size_t i = 0; i < op->n_lsp_addrs; i++) {
             for (size_t j = 0; j < op->lsp_addrs[i].n_ipv4_addrs; j++) {
                 ds_clear(&match);
@@ -4496,6 +4527,14 @@ build_lswitch_flows(struct hmap *datapaths, struct hmap *ports,
             continue;
         }
 
+        bool is_external = lsp_is_external(op->nbsp);
+        if (is_external && (!op->od->localnet_port ||
+                            !op->nbsp->ha_chassis_group)) {
+            /* If it's an external port and there is no localnet port
+             * and if it doesn't belong to an HA chassis group ignore it. */
+            continue;
+        }
+
         for (size_t i = 0; i < op->n_lsp_addrs; i++) {
             for (size_t j = 0; j < op->lsp_addrs[i].n_ipv4_addrs; j++) {
                 struct ds options_action = DS_EMPTY_INITIALIZER;
@@ -4508,9 +4547,16 @@ build_lswitch_flows(struct hmap *datapaths, struct hmap *ports,
                     ds_put_format(
                         &match, "inport == %s && eth.src == %s && "
                         "ip4.src == 0.0.0.0 && ip4.dst == 255.255.255.255 && "
-                        "udp.src == 68 && udp.dst == 67", op->json_key,
+                        "udp.src == 68 && udp.dst == 67",
+                        is_external ? op->od->localnet_port->json_key :
+                            op->json_key,
                         op->lsp_addrs[i].ea_s);
 
+                    if (is_external) {
+                        ds_put_format(&match, " && is_chassis_resident(%s)",
+                                      op->json_key);
+                    }
+
                     ovn_lflow_add(lflows, op->od, S_SWITCH_IN_DHCP_OPTIONS,
                                   100, ds_cstr(&match),
                                   ds_cstr(&options_action));
@@ -4525,9 +4571,16 @@ build_lswitch_flows(struct hmap *datapaths, struct hmap *ports,
                      */
                     ds_put_format(
                         &match, "inport == %s && eth.src == %s && "
-                        "%s && udp.src == 68 && udp.dst == 67", op->json_key,
+                        "%s && udp.src == 68 && udp.dst == 67",
+                        is_external ? op->od->localnet_port->json_key :
+                            op->json_key,
                         op->lsp_addrs[i].ea_s, ds_cstr(&ipv4_addr_match));
 
+                    if (is_external) {
+                        ds_put_format(&match, " && is_chassis_resident(%s)",
+                                      op->json_key);
+                    }
+
                     ovn_lflow_add(lflows, op->od, S_SWITCH_IN_DHCP_OPTIONS,
                                   100, ds_cstr(&match),
                                   ds_cstr(&options_action));
@@ -4538,8 +4591,16 @@ build_lswitch_flows(struct hmap *datapaths, struct hmap *ports,
                     ds_put_format(
                         &match, "inport == %s && eth.src == %s && "
                         "ip4 && udp.src == 68 && udp.dst == 67"
-                        " && "REGBIT_DHCP_OPTS_RESULT, op->json_key,
+                        " && "REGBIT_DHCP_OPTS_RESULT,
+                        is_external ? op->od->localnet_port->json_key :
+                            op->json_key,
                         op->lsp_addrs[i].ea_s);
+
+                    if (is_external) {
+                        ds_put_format(&match, " && is_chassis_resident(%s)",
+                                      op->json_key);
+                    }
+
                     ovn_lflow_add(lflows, op->od, S_SWITCH_IN_DHCP_RESPONSE,
                                   100, ds_cstr(&match),
                                   ds_cstr(&response_action));
@@ -4560,9 +4621,16 @@ build_lswitch_flows(struct hmap *datapaths, struct hmap *ports,
                     ds_put_format(
                         &match, "inport == %s && eth.src == %s"
                         " && ip6.dst == ff02::1:2 && udp.src == 546 &&"
-                        " udp.dst == 547", op->json_key,
+                        " udp.dst == 547",
+                        is_external ? op->od->localnet_port->json_key :
+                            op->json_key,
                         op->lsp_addrs[i].ea_s);
 
+                    if (is_external) {
+                        ds_put_format(&match, " && is_chassis_resident(%s)",
+                                      op->json_key);
+                    }
+
                     ovn_lflow_add(lflows, op->od, S_SWITCH_IN_DHCP_OPTIONS, 100,
                                   ds_cstr(&match), ds_cstr(&options_action));
 
@@ -4614,7 +4682,9 @@ build_lswitch_flows(struct hmap *datapaths, struct hmap *ports,
     /* Ingress table 12 and 13: DHCP options and response, by default goto
      * next. (priority 0).
      * Ingress table 14 and 15: DNS lookup and response, by default goto next.
-     * (priority 0).*/
+     * (priority 0).
+     * Ingress table 16 - External port handling, by default goto next.
+     * (priority 0). */
 
     HMAP_FOR_EACH (od, key_node, datapaths) {
         if (!od->nbs) {
@@ -4625,9 +4695,60 @@ build_lswitch_flows(struct hmap *datapaths, struct hmap *ports,
         ovn_lflow_add(lflows, od, S_SWITCH_IN_DHCP_RESPONSE, 0, "1", "next;");
         ovn_lflow_add(lflows, od, S_SWITCH_IN_DNS_LOOKUP, 0, "1", "next;");
         ovn_lflow_add(lflows, od, S_SWITCH_IN_DNS_RESPONSE, 0, "1", "next;");
+        ovn_lflow_add(lflows, od, S_SWITCH_IN_EXTERNAL_PORT, 0, "1", "next;");
     }
 
-    /* Ingress table 16: Destination lookup, broadcast and multicast handling
+    HMAP_FOR_EACH (op, key_node, ports) {
+        if (!op->nbsp || !lsp_is_external(op->nbsp) ||
+            !op->od->localnet_port) {
+           continue;
+        }
+
+        /* Table 16: External port. Drop ARP request for router ips from
+         * external ports  on chassis not binding those ports.
+         * This makes the router pipeline to be run only on the chassis
+         * binding the external ports. */
+
+        for (size_t i = 0; i < op->n_lsp_addrs; i++) {
+            for (size_t j = 0; j < op->od->n_router_ports; j++) {
+                struct ovn_port *rp = op->od->router_ports[j];
+                for (size_t k = 0; k < rp->n_lsp_addrs; k++) {
+                    for (size_t l = 0; l < rp->lsp_addrs[k].n_ipv4_addrs;
+                         l++) {
+                        ds_clear(&match);
+                        ds_put_format(
+                            &match, "inport == %s && eth.src == %s"
+                            " && !is_chassis_resident(%s)"
+                            " && arp.tpa == %s && arp.op == 1",
+                            op->od->localnet_port->json_key,
+                            op->lsp_addrs[i].ea_s, op->json_key,
+                            rp->lsp_addrs[k].ipv4_addrs[l].addr_s);
+                        ovn_lflow_add(lflows, op->od,
+                                      S_SWITCH_IN_EXTERNAL_PORT, 100,
+                                      ds_cstr(&match), "drop;");
+                    }
+                    for (size_t l = 0; l < rp->lsp_addrs[k].n_ipv6_addrs;
+                         l++) {
+                        ds_clear(&match);
+                        ds_put_format(
+                            &match, "inport == %s && eth.src == %s"
+                            " && !is_chassis_resident(%s)"
+                            " && nd_ns && ip6.dst == {%s, %s} && "
+                            "nd.target == %s",
+                            op->od->localnet_port->json_key,
+                            op->lsp_addrs[i].ea_s, op->json_key,
+                            rp->lsp_addrs[k].ipv6_addrs[l].addr_s,
+                            rp->lsp_addrs[k].ipv6_addrs[l].sn_addr_s,
+                            rp->lsp_addrs[k].ipv6_addrs[l].addr_s);
+                        ovn_lflow_add(lflows, op->od,
+                                      S_SWITCH_IN_EXTERNAL_PORT, 100,
+                                      ds_cstr(&match), "drop;");
+                    }
+                }
+            }
+        }
+    }
+    /* Ingress table 17: Destination lookup, broadcast and multicast handling
      * (priority 100). */
     HMAP_FOR_EACH (op, key_node, ports) {
         if (!op->nbsp) {
@@ -4647,9 +4768,9 @@ build_lswitch_flows(struct hmap *datapaths, struct hmap *ports,
                       "outport = \""MC_FLOOD"\"; output;");
     }
 
-    /* Ingress table 16: Destination lookup, unicast handling (priority 50), */
+    /* Ingress table 17: Destination lookup, unicast handling (priority 50), */
     HMAP_FOR_EACH (op, key_node, ports) {
-        if (!op->nbsp) {
+        if (!op->nbsp || lsp_is_external(op->nbsp)) {
             continue;
         }
 
@@ -4766,7 +4887,7 @@ build_lswitch_flows(struct hmap *datapaths, struct hmap *ports,
         }
     }
 
-    /* Ingress table 16: Destination lookup for unknown MACs (priority 0). */
+    /* Ingress table 17: Destination lookup for unknown MACs (priority 0). */
     HMAP_FOR_EACH (od, key_node, datapaths) {
         if (!od->nbs) {
             continue;
@@ -4801,7 +4922,7 @@ build_lswitch_flows(struct hmap *datapaths, struct hmap *ports,
      * Priority 150 rules drop packets to disabled logical ports, so that they
      * don't even receive multicast or broadcast packets. */
     HMAP_FOR_EACH (op, key_node, ports) {
-        if (!op->nbsp) {
+        if (!op->nbsp || lsp_is_external(op->nbsp)) {
             continue;
         }
-Original file line number
+Diff line change
@@ Expand Up / @@ -319,6 +319,7 @@ static const char *OVN_NB_LSP_TYPES[] = { @@
         "localport",
         "router",
         "vtep",
+        "external",
     };
     bool
@@ Expand Down @@