2
0
mirror of https://github.com/openvswitch/ovs synced 2025-09-05 00:35:33 +00:00

ovn: l3ha, make is_chassis_active aware of gateway_chassis

is_chassis_active now is only true for redirect-chassis ports
in the case of the specific lport being active on the
local chassis.

This will naturally make the ARP responder / redirection openflow
rules automatically inserted/removed when a router goes active/backup
in a way that BACKUP routers won't respond to ARP on gateway ports,
and they won't route packets that arrive on the wrong gateway
chassis (that can happen until all hypervisors converge in the
new MASTER thanks to the BFD monitoring of the tunnel endpoints).

Signed-off-by: Miguel Angel Ajo <majopela@redhat.com>
Signed-off-by: Russell Bryant <russell@ovn.org>
This commit is contained in:
majopela@redhat.com
2017-07-13 17:02:39 +00:00
committed by Russell Bryant
parent 3475695ea6
commit 508b7f961b
11 changed files with 193 additions and 54 deletions

View File

@@ -411,20 +411,10 @@ consider_local_datapath(struct controller_ctx *ctx,
chassis_index);
if (gateway_chassis &&
gateway_chassis_contains(gateway_chassis, chassis_rec)) {
struct gateway_chassis *gwc;
LIST_FOR_EACH (gwc, node, gateway_chassis) {
if (!gwc->db->chassis) {
continue;
}
if (!strcmp(gwc->db->chassis->name, chassis_rec->name)) {
/* sb_rec_port_binding->chassis should reflect master */
our_chassis = true;
break;
}
if (sset_contains(active_tunnels, gwc->db->chassis->name)) {
break;
}
}
our_chassis = gateway_chassis_is_active(
gateway_chassis, chassis_rec, active_tunnels);
add_local_datapath(ldatapaths, lports, binding_rec->datapath,
false, local_datapaths);
}

View File

@@ -17,6 +17,7 @@
#include "gchassis.h"
#include "lport.h"
#include "lib/sset.h"
#include "openvswitch/vlog.h"
#include "ovn/lib/chassis-index.h"
#include "ovn/lib/ovn-sb-idl.h"
@@ -110,7 +111,7 @@ gateway_chassis_get_ordered(const struct sbrec_port_binding *binding,
}
bool
gateway_chassis_contains(struct ovs_list *gateway_chassis,
gateway_chassis_contains(const struct ovs_list *gateway_chassis,
const struct sbrec_chassis *chassis) {
struct gateway_chassis *chassis_item;
if (gateway_chassis) {
@@ -174,3 +175,48 @@ gateway_chassis_in_pb_contains(const struct sbrec_port_binding *binding,
return false;
}
bool
gateway_chassis_is_active(const struct ovs_list *gateway_chassis,
const struct sbrec_chassis *local_chassis,
const struct sset *active_tunnels)
{
struct gateway_chassis *gwc;
if (!gateway_chassis
|| (gateway_chassis && ovs_list_is_empty(gateway_chassis))) {
return false;
}
/* if there's only one chassis, and local chassis is on the list
* it's not HA and it's the equivalent of being active */
if (ovs_list_is_singleton(gateway_chassis) &&
gateway_chassis_contains(gateway_chassis, local_chassis)) {
return true;
}
/* if there are no other tunnels active, we assume that the
* connection providing tunneling is down, hence we're down */
if (sset_is_empty(active_tunnels)) {
return false;
}
/* gateway_chassis is an ordered list, by priority, of chassis
* hosting the redirect of the port */
LIST_FOR_EACH (gwc, node, gateway_chassis) {
if (!gwc->db->chassis) {
continue;
}
/* if we found the chassis on the list, and we didn't exit before
* on the active_tunnels check for other higher priority chassis
* being active, then this chassis is master. */
if (!strcmp(gwc->db->chassis->name, local_chassis->name)) {
return true;
}
/* if we find this specific chassis on the list to have an active
* tunnel, then 'local_chassis' is not master */
if (sset_contains(active_tunnels, gwc->db->chassis->name)) {
return false;
}
}
return false;
}

View File

@@ -26,6 +26,7 @@ struct ovsdb_idl;
struct sbrec_chassis;
struct sbrec_gateway_chassis;
struct sbrec_port_binding;
struct sset;
/* Gateway_Chassis management
@@ -48,7 +49,7 @@ struct ovs_list *gateway_chassis_get_ordered(
/* Checks if an specific chassis is contained in the gateway_chassis
* list */
bool gateway_chassis_contains(struct ovs_list *gateway_chassis,
bool gateway_chassis_contains(const struct ovs_list *gateway_chassis,
const struct sbrec_chassis *chassis);
/* Destroy a gateway_chassis list from memory */
@@ -60,4 +61,11 @@ bool gateway_chassis_in_pb_contains(
const struct sbrec_port_binding *binding,
const struct sbrec_chassis *chassis);
/* Returns true if the local chassis is the active gateway among a set
* of gateway_chassis. Return false if the local chassis is currently a
* backup in a set of multiple gateway_chassis. */
bool gateway_chassis_is_active(
const struct ovs_list *gateway_chassis,
const struct sbrec_chassis *local_chassis,
const struct sset *active_tunnels);
#endif /* ovn/controller/gchassis.h */

View File

@@ -54,10 +54,13 @@ struct lookup_port_aux {
struct condition_aux {
const struct lport_index *lports;
const struct sbrec_chassis *chassis;
const struct sset *active_tunnels;
const struct chassis_index *chassis_index;
};
static void consider_logical_flow(const struct lport_index *lports,
const struct mcgroup_index *mcgroups,
const struct chassis_index *chassis_index,
const struct sbrec_logical_flow *lflow,
const struct hmap *local_datapaths,
struct group_table *group_table,
@@ -66,7 +69,8 @@ static void consider_logical_flow(const struct lport_index *lports,
struct hmap *dhcpv6_opts,
uint32_t *conj_id_ofs,
const struct shash *addr_sets,
struct hmap *flow_table);
struct hmap *flow_table,
struct sset *active_tunnels);
static bool
lookup_port_cb(const void *aux_, const char *port_name, unsigned int *portp)
@@ -97,10 +101,24 @@ is_chassis_resident_cb(const void *c_aux_, const char *port_name)
const struct sbrec_port_binding *pb
= lport_lookup_by_name(c_aux->lports, port_name);
if (pb && pb->chassis && pb->chassis == c_aux->chassis) {
return true;
if (!pb) {
return false;
}
if (strcmp(pb->type, "chassisredirect")) {
/* for non-chassisredirect ports */
return pb->chassis && pb->chassis == c_aux->chassis;
} else {
return gateway_chassis_in_pb_contains(pb, c_aux->chassis);
struct ovs_list *gateway_chassis;
gateway_chassis = gateway_chassis_get_ordered(pb,
c_aux->chassis_index);
if (gateway_chassis) {
bool active = gateway_chassis_is_active(gateway_chassis,
c_aux->chassis,
c_aux->active_tunnels);
gateway_chassis_destroy(gateway_chassis);
return active;
}
return false;
}
}
@@ -124,11 +142,13 @@ is_gateway_router(const struct sbrec_datapath_binding *ldp,
static void
add_logical_flows(struct controller_ctx *ctx, const struct lport_index *lports,
const struct mcgroup_index *mcgroups,
const struct chassis_index *chassis_index,
const struct hmap *local_datapaths,
struct group_table *group_table,
const struct sbrec_chassis *chassis,
const struct shash *addr_sets,
struct hmap *flow_table)
struct hmap *flow_table,
struct sset *active_tunnels)
{
uint32_t conj_id_ofs = 1;
const struct sbrec_logical_flow *lflow;
@@ -149,10 +169,11 @@ add_logical_flows(struct controller_ctx *ctx, const struct lport_index *lports,
}
SBREC_LOGICAL_FLOW_FOR_EACH (lflow, ctx->ovnsb_idl) {
consider_logical_flow(lports, mcgroups, lflow, local_datapaths,
consider_logical_flow(lports, mcgroups, chassis_index,
lflow, local_datapaths,
group_table, chassis,
&dhcp_opts, &dhcpv6_opts, &conj_id_ofs,
addr_sets, flow_table);
addr_sets, flow_table, active_tunnels);
}
dhcp_opts_destroy(&dhcp_opts);
@@ -162,6 +183,7 @@ add_logical_flows(struct controller_ctx *ctx, const struct lport_index *lports,
static void
consider_logical_flow(const struct lport_index *lports,
const struct mcgroup_index *mcgroups,
const struct chassis_index *chassis_index,
const struct sbrec_logical_flow *lflow,
const struct hmap *local_datapaths,
struct group_table *group_table,
@@ -170,7 +192,8 @@ consider_logical_flow(const struct lport_index *lports,
struct hmap *dhcpv6_opts,
uint32_t *conj_id_ofs,
const struct shash *addr_sets,
struct hmap *flow_table)
struct hmap *flow_table,
struct sset *active_tunnels)
{
/* Determine translation of logical table IDs to physical table IDs. */
bool ingress = !strcmp(lflow->pipeline, "ingress");
@@ -267,7 +290,8 @@ consider_logical_flow(const struct lport_index *lports,
return;
}
struct condition_aux cond_aux = { lports, chassis };
struct condition_aux cond_aux = { lports, chassis, active_tunnels,
chassis_index};
expr = expr_simplify(expr, is_chassis_resident_cb, &cond_aux);
expr = expr_normalize(expr);
uint32_t n_conjs = expr_to_matches(expr, lookup_port_cb, &aux,
@@ -393,13 +417,16 @@ lflow_run(struct controller_ctx *ctx,
const struct sbrec_chassis *chassis,
const struct lport_index *lports,
const struct mcgroup_index *mcgroups,
const struct chassis_index *chassis_index,
const struct hmap *local_datapaths,
struct group_table *group_table,
const struct shash *addr_sets,
struct hmap *flow_table)
struct hmap *flow_table,
struct sset *active_tunnels)
{
add_logical_flows(ctx, lports, mcgroups, local_datapaths, group_table,
chassis, addr_sets, flow_table);
add_logical_flows(ctx, lports, mcgroups, chassis_index, local_datapaths,
group_table, chassis, addr_sets, flow_table,
active_tunnels);
add_neighbor_flows(ctx, lports, flow_table);
}

View File

@@ -35,6 +35,7 @@
#include <stdint.h>
struct chassis_index;
struct controller_ctx;
struct group_table;
struct hmap;
@@ -42,6 +43,7 @@ struct lport_index;
struct mcgroup_index;
struct sbrec_chassis;
struct simap;
struct sset;
struct uuid;
/* OpenFlow table numbers.
@@ -66,10 +68,12 @@ void lflow_run(struct controller_ctx *,
const struct sbrec_chassis *chassis,
const struct lport_index *,
const struct mcgroup_index *,
const struct chassis_index *,
const struct hmap *local_datapaths,
struct group_table *group_table,
const struct shash *addr_sets,
struct hmap *flow_table);
struct hmap *flow_table,
struct sset *active_tunnels);
void lflow_destroy(void);
#endif /* ovn/lflow.h */

View File

@@ -15,11 +15,11 @@
#include <config.h>
#include "lib/sset.h"
#include "lport.h"
#include "hash.h"
#include "openvswitch/vlog.h"
#include "ovn/lib/ovn-sb-idl.h"
VLOG_DEFINE_THIS_MODULE(lport);
static struct ldatapath *ldatapath_lookup_by_key__(

View File

@@ -26,6 +26,7 @@ struct sbrec_chassis;
struct sbrec_datapath_binding;
struct sbrec_port_binding;
/* Database indexes.
* =================
*

View File

@@ -649,7 +649,6 @@ main(int argc, char *argv[])
&chassis_index, &active_tunnels, &local_datapaths,
&local_lports);
}
if (br_int && chassis) {
struct shash addr_sets = SHASH_INITIALIZER(&addr_sets);
addr_sets_init(&ctx, &addr_sets);
@@ -668,8 +667,8 @@ main(int argc, char *argv[])
struct hmap flow_table = HMAP_INITIALIZER(&flow_table);
lflow_run(&ctx, chassis, &lports, &mcgroups,
&local_datapaths, &group_table,
&addr_sets, &flow_table);
&chassis_index, &local_datapaths, &group_table,
&addr_sets, &flow_table, &active_tunnels);
if (chassis_id) {
bfd_run(&ctx, br_int, chassis, &local_datapaths,
@@ -678,7 +677,7 @@ main(int argc, char *argv[])
physical_run(&ctx, mff_ovn_geneve,
br_int, chassis, &ct_zones, &lports,
&flow_table, &local_datapaths, &local_lports,
&chassis_index);
&chassis_index, &active_tunnels);
ofctrl_put(&flow_table, &pending_ct_zones,
get_nb_cfg(ctx.ovnsb_idl));

View File

@@ -295,6 +295,7 @@ consider_port_binding(enum mf_field_id mff_ovn_geneve,
const struct simap *ct_zones,
const struct lport_index *lports,
const struct chassis_index *chassis_index,
struct sset *active_tunnels,
struct hmap *local_datapaths,
const struct sbrec_port_binding *binding,
const struct sbrec_chassis *chassis,
@@ -362,15 +363,10 @@ consider_port_binding(enum mf_field_id mff_ovn_geneve,
struct ovs_list *gateway_chassis
= gateway_chassis_get_ordered(binding, chassis_index);
/* XXX: later in the series we should insert the next flows only
* on the active chassis, and not on all of them. This is useful to
* check that the BFD implementation on following patches has
* an effect and routes packet by the chassis which is responding,
* but later on we should not create those flows on all the
* chassis of the gateway_chassis list */
if (!strcmp(binding->type, "chassisredirect")
&& (binding->chassis == chassis
|| gateway_chassis_contains(gateway_chassis, chassis))) {
|| gateway_chassis_is_active(gateway_chassis, chassis,
active_tunnels))) {
/* Table 33, priority 100.
* =======================
@@ -865,7 +861,8 @@ physical_run(struct controller_ctx *ctx, enum mf_field_id mff_ovn_geneve,
const struct simap *ct_zones, struct lport_index *lports,
struct hmap *flow_table, struct hmap *local_datapaths,
const struct sset *local_lports,
struct chassis_index *chassis_index)
struct chassis_index *chassis_index,
struct sset *active_tunnels)
{
/* This bool tracks physical mapping changes. */
@@ -987,7 +984,7 @@ physical_run(struct controller_ctx *ctx, enum mf_field_id mff_ovn_geneve,
const struct sbrec_port_binding *binding;
SBREC_PORT_BINDING_FOR_EACH (binding, ctx->ovnsb_idl) {
consider_port_binding(mff_ovn_geneve, ct_zones, lports,
chassis_index,
chassis_index, active_tunnels,
local_datapaths, binding, chassis,
&ofpacts, flow_table);
}

View File

@@ -49,6 +49,7 @@ void physical_run(struct controller_ctx *, enum mf_field_id mff_ovn_geneve,
const struct simap *ct_zones, struct lport_index *,
struct hmap *flow_table, struct hmap *local_datapaths,
const struct sset *local_lports,
struct chassis_index *chassis_index);
struct chassis_index *chassis_index,
struct sset *active_tunnels);
#endif /* ovn/physical.h */

View File

@@ -7792,17 +7792,25 @@ for chassis in gw1 gw2 hv1 hv2; do
echo "------ $chassis dump ----------"
ovs-ofctl show br-int
ovs-ofctl dump-flows br-int
echo ""
echo "BFD (from $chassis):"
# dump BFD config and status to the other chassis
for chassis2 in gw1 gw2 hv1 hv2; do
if [[ "$chassis" != "$chassis2" ]]; then
echo " -> $chassis2:"
echo " $(ovs-vsctl --bare --columns bfd,bfd_status find Interface name=ovn-$chassis2-0)"
fi
done
echo "--------------------------"
done
function bfd_dump() {
for chassis in gw1 gw2 hv1 hv2; do
as $chassis
echo "------ $chassis dump (BFD)----"
echo "BFD (from $chassis):"
# dump BFD config and status to the other chassis
for chassis2 in gw1 gw2 hv1 hv2; do
if [[ "$chassis" != "$chassis2" ]]; then
echo " -> $chassis2:"
echo " $(ovs-vsctl --bare --columns bfd,bfd_status find Interface name=ovn-$chassis2-0)"
fi
done
echo "--------------------------"
done
}
bfd_dump
hv1_gw1_ofport=$(as hv1 ovs-vsctl --bare --columns ofport find Interface name=ovn-gw1-0)
hv1_gw2_ofport=$(as hv1 ovs-vsctl --bare --columns ofport find Interface name=ovn-gw2-0)
@@ -7820,13 +7828,38 @@ as hv1 ovs-ofctl dump-flows br-int table=32
echo "--- hv2 ---"
as hv2 ovs-ofctl dump-flows br-int table=32
gw1_chassis=$(ovn-sbctl --bare --columns=_uuid find Chassis name=gw1)
gw2_chassis=$(ovn-sbctl --bare --columns=_uuid find Chassis name=gw2)
AT_CHECK([as hv1 ovs-ofctl dump-flows br-int table=32 | grep active_backup | grep slaves:$hv1_gw1_ofport,$hv1_gw2_ofport | wc -l], [0], [1
])
AT_CHECK([as hv2 ovs-ofctl dump-flows br-int table=32 | grep active_backup | grep slaves:$hv2_gw1_ofport,$hv2_gw2_ofport | wc -l], [0], [1
])
# set higher priority to gw2 instead of gw1, and check for changes
sleep 3 # let BFD sessions settle so we get the right flows on the right chassis
# make sure that flows for handling the outside router port reside on gw1
AT_CHECK([as gw1 ovs-ofctl dump-flows br-int table=23 | grep 00:00:02:01:02:04 | wc -l], [0], [[1
]])
AT_CHECK([as gw2 ovs-ofctl dump-flows br-int table=23 | grep 00:00:02:01:02:04 | wc -l], [0], [[0
]])
# make sure ARP responder flows for outside router port reside on gw1 too
AT_CHECK([as gw1 ovs-ofctl dump-flows br-int table=9 | grep arp_tpa=192.168.0.101 | wc -l], [0], [[1
]])
AT_CHECK([as gw2 ovs-ofctl dump-flows br-int table=9 | grep arp_tpa=192.168.0.101 | wc -l], [0], [[0
]])
# check that the chassis redirect port has been claimed by the gw1 chassis
AT_CHECK([ovn-sbctl --columns chassis --bare find Port_Binding logical_port=cr-outside | grep $gw1_chassis | wc -l],
[0],[[1
]])
# at this point, we invert the priority of the gw chassis between gw1 and gw2
ovn-nbctl --id=@gc0 create Gateway_Chassis \
name=outside_gw1 chassis_name=gw1 priority=10 -- \
@@ -7834,15 +7867,21 @@ ovn-nbctl --id=@gc0 create Gateway_Chassis \
name=outside_gw2 chassis_name=gw2 priority=20 -- \
set Logical_Router_Port outside 'gateway_chassis=[@gc0,@gc1]'
# XXX: Let the change propagate down to the ovn-controllers
ovn-nbctl --wait=hv sync
# we make sure that the hypervisors noticed, and inverted the slave ports
AT_CHECK([as hv1 ovs-ofctl dump-flows br-int table=32 | grep active_backup | grep slaves:$hv1_gw2_ofport,$hv1_gw1_ofport | wc -l], [0], [1
])
AT_CHECK([as hv2 ovs-ofctl dump-flows br-int table=32 | grep active_backup | grep slaves:$hv2_gw2_ofport,$hv2_gw1_ofport | wc -l], [0], [1
])
# check that the chassis redirect port has been reclaimed by the gw2 chassis
AT_CHECK([ovn-sbctl --columns chassis --bare find Port_Binding logical_port=cr-outside | grep $gw2_chassis | wc -l],
[0],[[1
]])
# check BFD enablement on tunnel ports from gw1 #########
as gw1
@@ -7890,6 +7929,33 @@ AT_CHECK([ovs-vsctl --bare --columns bfd find Interface name=ovn-hv1-0],[0],
[[enable=false
]])
sleep 3 # let BFD sessions settle so we get the right flows on the right chassis
# make sure that flows for handling the outside router port reside on gw2 now
AT_CHECK([as gw2 ovs-ofctl dump-flows br-int table=23 | grep 00:00:02:01:02:04 | wc -l], [0], [[1
]])
AT_CHECK([as gw1 ovs-ofctl dump-flows br-int table=23 | grep 00:00:02:01:02:04 | wc -l], [0], [[0
]])
# disconnect GW2 from the network, GW1 should take over
as gw2
port=${sandbox}_br-phys
as main ovs-vsctl del-port n1 $port
sleep 4
bfd_dump
# make sure that flows for handling the outside router port reside on gw2 now
AT_CHECK([as gw1 ovs-ofctl dump-flows br-int table=23 | grep 00:00:02:01:02:04 | wc -l], [0], [[1
]])
AT_CHECK([as gw2 ovs-ofctl dump-flows br-int table=23 | grep 00:00:02:01:02:04 | wc -l], [0], [[0
]])
# check that the chassis redirect port has been reclaimed by the gw1 chassis
AT_CHECK([ovn-sbctl --columns chassis --bare find Port_Binding logical_port=cr-outside | grep $gw1_chassis | wc -l],
[0],[[1
]])
OVN_CLEANUP([gw1],[gw2],[hv1],[hv2])
AT_CLEANUP