On Fri, Apr 5, 2024 at 10:35 PM Numan Siddique <num...@ovn.org> wrote:
>
>
> On Tue, Apr 2, 2024 at 2:28 AM Ales Musil <amu...@redhat.com> wrote:
>
>> The current packet injection loses ct_state in the process. When
>> the ct_state is lost we might commit to DNAT zone and perform
>> zero SNAT after the packet injection. This causes the first session
>> to be wrong as the reply packets are not unDNATted.
>>
>> Instead of re-injecting the packet back into the pipeline when
>> we get the MAC binding, use paused controller action. The paused
>> controller action stores ct_state, which is important for the behavior
>> of the resumed packet.
>>
>> At the same time bump the OvS submodule latest branch-3.3. This is
>> mainly for [0], which fixes metering for paused controller actions.
>>
>> [0] c560f6ca3257 ("ofproto-dpif-xlate: Fix continuations with associated
>> metering.")
>>
>> Reported-at: https://issues.redhat.com/browse/FDP-439
>> Signed-off-by: Ales Musil <amu...@redhat.com>
>> Acked-by: Mark Michelson <mmich...@redhat.com>
>> ---
>> v2: Fix the Jira link and add ack from Mark.
>>
>
> Hi Ales,
>
> I see one problem during upgrades. The packet buffering functionality
> would be broken when ovn-controllers are updated to the version
> which has your patch but ovn-northd is still yet to.
>
> In this case, the below logical flow will be present (without the outer
> "output" action)
>
> ---
> table=21(lr_in_arp_request ), priority=100 , match=(eth.dst ==
> 00:00:00:00:00:00 && ip4), action=(arp { eth.dst = ff:ff:ff:ff:ff:ff;
> arp.spa = reg1; arp.tpa = reg0; arp.op = 1; output; };)
> ---
>
> The updated ovn-controller will now resume the packet after learning the
> nexthop. But this resumed packet will be dropped because there is no
> "next" or "output" action following arp {}.
>
> Is this behavior acceptable until ovn-northd is also updated/upgraded ?
>
> Thanks
> Numan
>
>
Hi Numan,
this is a very good point, thank you for raising this. It probably depends
how long the upgrade takes, for ovn-kubernetes with ic it is all done per
node so that should be fine. However it might be more concerning for
OpenStack. In the end we would be losing only a small amount of packets (in
most cases only 1) for connections without MAC binding, on the other hand
the scenario it tries to fix is causing the first connection (whole
connection) to be completely broken.
I've tried to think about a possible solution to avoid this problem and I
see only one way out of this. That would involve keeping the old arp
unchanged, adding new action that would result in the pause instead. I'm
not sure if that's worth the effort, it's hard to estimate how big of an
impact it might have during the upgrade. One unfortunate thing is that we
cannot modify the actions of the continuation, if we could there wouldn't
be any change required in northd.
Let me know if you have other ideas on how to avoid this issue, maybe the
solution might be easier.
> ---
>
>> controller/mac-learn.c | 30 +++++++++----
>> controller/mac-learn.h | 9 ++--
>> controller/pinctrl.c | 64 ++++++++++------------------
>> lib/actions.c | 41 +++++++++++++++++-
>> northd/northd.c | 6 +--
>> ovs | 2 +-
>> tests/ovn-northd.at | 3 ++
>> tests/ovn.at | 8 ++--
>> tests/system-ovn.at | 95 ++++++++++++++++++++++++++++++++++++++++++
>> 9 files changed, 195 insertions(+), 63 deletions(-)
>>
>> diff --git a/controller/mac-learn.c b/controller/mac-learn.c
>> index 071f01b4f..0c3b60c23 100644
>> --- a/controller/mac-learn.c
>> +++ b/controller/mac-learn.c
>> @@ -199,15 +199,24 @@ ovn_fdb_add(struct hmap *fdbs, uint32_t dp_key,
>> struct eth_addr mac,
>> /* packet buffering functions */
>>
>> struct packet_data *
>> -ovn_packet_data_create(struct ofpbuf ofpacts,
>> - const struct dp_packet *original_packet)
>> +ovn_packet_data_create(const struct ofputil_packet_in *pin,
>> + const struct ofpbuf *continuation)
>> {
>> struct packet_data *pd = xmalloc(sizeof *pd);
>>
>> - pd->ofpacts = ofpacts;
>> - /* clone the packet to send it later with correct L2 address */
>> - pd->p = dp_packet_clone_data(dp_packet_data(original_packet),
>> - dp_packet_size(original_packet));
>> + pd->pin = (struct ofputil_packet_in) {
>> + .packet = xmemdup(pin->packet, pin->packet_len),
>> + .packet_len = pin->packet_len,
>> + .flow_metadata = pin->flow_metadata,
>> + .reason = pin->reason,
>> + .table_id = pin->table_id,
>> + .cookie = pin->cookie,
>> + /* Userdata are empty on purpose,
>> + * it is not needed for the continuation. */
>> + .userdata = NULL,
>> + .userdata_len = 0,
>> + };
>> + pd->continuation = ofpbuf_clone(continuation);
>>
>> return pd;
>> }
>> @@ -216,8 +225,8 @@ ovn_packet_data_create(struct ofpbuf ofpacts,
>> void
>> ovn_packet_data_destroy(struct packet_data *pd)
>> {
>> - dp_packet_delete(pd->p);
>> - ofpbuf_uninit(&pd->ofpacts);
>> + free(pd->pin.packet);
>> + ofpbuf_delete(pd->continuation);
>> free(pd);
>> }
>>
>> @@ -307,7 +316,10 @@ ovn_buffered_packets_ctx_run(struct
>> buffered_packets_ctx *ctx,
>>
>> struct packet_data *pd;
>> LIST_FOR_EACH_POP (pd, node, &bp->queue) {
>> - struct eth_header *eth = dp_packet_data(pd->p);
>> + struct dp_packet packet;
>> + dp_packet_use_const(&packet, pd->pin.packet,
>> pd->pin.packet_len);
>> +
>> + struct eth_header *eth = dp_packet_data(&packet);
>> eth->eth_dst = mac;
>>
>> ovs_list_push_back(&ctx->ready_packets_data, &pd->node);
>> diff --git a/controller/mac-learn.h b/controller/mac-learn.h
>> index e0fd6a8d1..20a015e1a 100644
>> --- a/controller/mac-learn.h
>> +++ b/controller/mac-learn.h
>> @@ -24,6 +24,7 @@
>> #include "openvswitch/hmap.h"
>> #include "openvswitch/list.h"
>> #include "openvswitch/ofpbuf.h"
>> +#include "openvswitch/ofp-packet.h"
>>
>> struct ovsdb_idl_index;
>>
>> @@ -91,8 +92,8 @@ struct fdb_entry *ovn_fdb_add(struct hmap *fdbs,
>> struct packet_data {
>> struct ovs_list node;
>>
>> - struct ofpbuf ofpacts;
>> - struct dp_packet *p;
>> + struct ofpbuf *continuation;
>> + struct ofputil_packet_in pin;
>> };
>>
>> struct buffered_packets {
>> @@ -120,8 +121,8 @@ struct buffered_packets_ctx {
>> };
>>
>> struct packet_data *
>> -ovn_packet_data_create(struct ofpbuf ofpacts,
>> - const struct dp_packet *original_packet);
>> +ovn_packet_data_create(const struct ofputil_packet_in *pin,
>> + const struct ofpbuf *continuation);
>> void ovn_packet_data_destroy(struct packet_data *pd);
>> struct buffered_packets *
>> ovn_buffered_packets_add(struct buffered_packets_ctx *ctx, uint64_t
>> dp_key,
>> diff --git a/controller/pinctrl.c b/controller/pinctrl.c
>> index 9a4deb80a..2c0ecdb4c 100644
>> --- a/controller/pinctrl.c
>> +++ b/controller/pinctrl.c
>> @@ -258,9 +258,9 @@ static void pinctrl_handle_put_nd_ra_opts(
>> struct ofpbuf *continuation);
>> static void pinctrl_handle_nd_ns(struct rconn *swconn,
>> const struct flow *ip_flow,
>> - struct dp_packet *pkt_in,
>> - const struct match *md,
>> - struct ofpbuf *userdata);
>> + const struct ofputil_packet_in *pin,
>> + struct ofpbuf *userdata,
>> + const struct ofpbuf *continuation);
>> static void pinctrl_handle_put_icmp_frag_mtu(struct rconn *swconn,
>> const struct flow *in_flow,
>> struct dp_packet *pkt_in,
>> @@ -1534,11 +1534,13 @@ destroy_buffered_packets_ctx(void)
>>
>> /* Called with in the pinctrl_handler thread context. */
>> static void
>> -pinctrl_handle_buffered_packets(struct dp_packet *pkt_in,
>> - const struct match *md, bool is_arp)
>> +pinctrl_handle_buffered_packets(const struct ofputil_packet_in *pin,
>> + const struct ofpbuf *continuation,
>> + bool is_arp)
>> OVS_REQUIRES(pinctrl_mutex)
>> {
>> struct in6_addr ip;
>> + const struct match *md = &pin->flow_metadata;
>> uint64_t dp_key = ntohll(md->flow.metadata);
>> uint64_t oport_key = md->flow.regs[MFF_LOG_OUTPORT - MFF_REG0];
>>
>> @@ -1556,20 +1558,7 @@ OVS_REQUIRES(pinctrl_mutex)
>> return;
>> }
>>
>> - struct ofpbuf ofpacts;
>> - ofpbuf_init(&ofpacts, 4096);
>> - reload_metadata(&ofpacts, md);
>> - /* reload pkt_mark field */
>> - const struct mf_field *pkt_mark_field = mf_from_id(MFF_PKT_MARK);
>> - union mf_value pkt_mark_value;
>> - mf_get_value(pkt_mark_field, &md->flow, &pkt_mark_value);
>> - ofpact_put_set_field(&ofpacts, pkt_mark_field, &pkt_mark_value,
>> NULL);
>> -
>> - struct ofpact_resubmit *resubmit = ofpact_put_RESUBMIT(&ofpacts);
>> - resubmit->in_port = OFPP_CONTROLLER;
>> - resubmit->table_id = OFTABLE_OUTPUT_INIT;
>> -
>> - struct packet_data *pd = ovn_packet_data_create(ofpacts, pkt_in);
>> + struct packet_data *pd = ovn_packet_data_create(pin, continuation);
>> ovn_buffered_packets_packet_data_enqueue(bp, pd);
>>
>> /* There is a chance that the MAC binding was already created. */
>> @@ -1579,8 +1568,8 @@ OVS_REQUIRES(pinctrl_mutex)
>> /* Called with in the pinctrl_handler thread context. */
>> static void
>> pinctrl_handle_arp(struct rconn *swconn, const struct flow *ip_flow,
>> - struct dp_packet *pkt_in,
>> - const struct match *md, struct ofpbuf *userdata)
>> + const struct ofputil_packet_in *pin,
>> + struct ofpbuf *userdata, const struct ofpbuf
>> *continuation)
>> {
>> /* This action only works for IP packets, and the switch should only
>> send
>> * us IP packets this way, but check here just to be sure. */
>> @@ -1592,7 +1581,7 @@ pinctrl_handle_arp(struct rconn *swconn, const
>> struct flow *ip_flow,
>> }
>>
>> ovs_mutex_lock(&pinctrl_mutex);
>> - pinctrl_handle_buffered_packets(pkt_in, md, true);
>> + pinctrl_handle_buffered_packets(pin, continuation, true);
>> ovs_mutex_unlock(&pinctrl_mutex);
>>
>> /* Compose an ARP packet. */
>> @@ -1617,7 +1606,8 @@ pinctrl_handle_arp(struct rconn *swconn, const
>> struct flow *ip_flow,
>> ip_flow->vlans[0].tci);
>> }
>>
>> - set_actions_and_enqueue_msg(swconn, &packet, md, userdata);
>> + set_actions_and_enqueue_msg(swconn, &packet,
>> + &pin->flow_metadata, userdata);
>> dp_packet_uninit(&packet);
>> }
>>
>> @@ -3294,8 +3284,7 @@ process_packet_in(struct rconn *swconn, const
>> struct ofp_header *msg)
>>
>> switch (ntohl(ah->opcode)) {
>> case ACTION_OPCODE_ARP:
>> - pinctrl_handle_arp(swconn, &headers, &packet, &pin.flow_metadata,
>> - &userdata);
>> + pinctrl_handle_arp(swconn, &headers, &pin, &userdata,
>> &continuation);
>> break;
>> case ACTION_OPCODE_IGMP:
>> pinctrl_ip_mcast_handle(swconn, &headers, &packet,
>> &pin.flow_metadata,
>> @@ -3361,8 +3350,7 @@ process_packet_in(struct rconn *swconn, const
>> struct ofp_header *msg)
>> break;
>>
>> case ACTION_OPCODE_ND_NS:
>> - pinctrl_handle_nd_ns(swconn, &headers, &packet,
>> &pin.flow_metadata,
>> - &userdata);
>> + pinctrl_handle_nd_ns(swconn, &headers, &pin, &userdata,
>> &continuation);
>> break;
>>
>> case ACTION_OPCODE_ICMP:
>> @@ -4374,16 +4362,8 @@ send_mac_binding_buffered_pkts(struct rconn
>> *swconn)
>>
>> struct packet_data *pd;
>> LIST_FOR_EACH_POP (pd, node,
>> &buffered_packets_ctx.ready_packets_data) {
>> - struct ofputil_packet_out po = {
>> - .packet = dp_packet_data(pd->p),
>> - .packet_len = dp_packet_size(pd->p),
>> - .buffer_id = UINT32_MAX,
>> - .ofpacts = pd->ofpacts.data,
>> - .ofpacts_len = pd->ofpacts.size,
>> - };
>> - match_set_in_port(&po.flow_metadata, OFPP_CONTROLLER);
>> - queue_msg(swconn, ofputil_encode_packet_out(&po, proto));
>> -
>> + queue_msg(swconn, ofputil_encode_resume(&pd->pin,
>> pd->continuation,
>> + proto));
>> ovn_packet_data_destroy(pd);
>> }
>>
>> @@ -6317,8 +6297,9 @@ pinctrl_handle_nd_na(struct rconn *swconn, const
>> struct flow *ip_flow,
>> /* Called with in the pinctrl_handler thread context. */
>> static void
>> pinctrl_handle_nd_ns(struct rconn *swconn, const struct flow *ip_flow,
>> - struct dp_packet *pkt_in,
>> - const struct match *md, struct ofpbuf *userdata)
>> + const struct ofputil_packet_in *pin,
>> + struct ofpbuf *userdata,
>> + const struct ofpbuf *continuation)
>> {
>> /* This action only works for IPv6 packets. */
>> if (get_dl_type(ip_flow) != htons(ETH_TYPE_IPV6)) {
>> @@ -6328,7 +6309,7 @@ pinctrl_handle_nd_ns(struct rconn *swconn, const
>> struct flow *ip_flow,
>> }
>>
>> ovs_mutex_lock(&pinctrl_mutex);
>> - pinctrl_handle_buffered_packets(pkt_in, md, false);
>> + pinctrl_handle_buffered_packets(pin, continuation, false);
>> ovs_mutex_unlock(&pinctrl_mutex);
>>
>> uint64_t packet_stub[128 / 8];
>> @@ -6341,7 +6322,8 @@ pinctrl_handle_nd_ns(struct rconn *swconn, const
>> struct flow *ip_flow,
>> &ip_flow->ipv6_dst);
>>
>> /* Reload previous packet metadata and set actions from userdata. */
>> - set_actions_and_enqueue_msg(swconn, &packet, md, userdata);
>> + set_actions_and_enqueue_msg(swconn, &packet,
>> + &pin->flow_metadata, userdata);
>> dp_packet_uninit(&packet);
>> }
>>
>> diff --git a/lib/actions.c b/lib/actions.c
>> index 71615fc53..004b77a89 100644
>> --- a/lib/actions.c
>> +++ b/lib/actions.c
>> @@ -1977,6 +1977,44 @@ format_REJECT(const struct ovnact_nest *nest,
>> struct ds *s)
>> format_nested_action(nest, "reject", s);
>> }
>>
>> +static bool
>> +is_paused_nested_action(enum action_opcode opcode)
>> +{
>> + switch (opcode) {
>> + case ACTION_OPCODE_ARP:
>> + case ACTION_OPCODE_ND_NS:
>> + return true;
>> + case ACTION_OPCODE_IGMP:
>> + case ACTION_OPCODE_PUT_ARP:
>> + case ACTION_OPCODE_PUT_DHCP_OPTS:
>> + case ACTION_OPCODE_ND_NA:
>> + case ACTION_OPCODE_ND_NA_ROUTER:
>> + case ACTION_OPCODE_PUT_ND:
>> + case ACTION_OPCODE_PUT_FDB:
>> + case ACTION_OPCODE_PUT_DHCPV6_OPTS:
>> + case ACTION_OPCODE_DNS_LOOKUP:
>> + case ACTION_OPCODE_LOG:
>> + case ACTION_OPCODE_PUT_ND_RA_OPTS:
>> + case ACTION_OPCODE_ICMP:
>> + case ACTION_OPCODE_ICMP4_ERROR:
>> + case ACTION_OPCODE_ICMP6_ERROR:
>> + case ACTION_OPCODE_TCP_RESET:
>> + case ACTION_OPCODE_SCTP_ABORT:
>> + case ACTION_OPCODE_REJECT:
>> + case ACTION_OPCODE_PUT_ICMP4_FRAG_MTU:
>> + case ACTION_OPCODE_PUT_ICMP6_FRAG_MTU:
>> + case ACTION_OPCODE_EVENT:
>> + case ACTION_OPCODE_BIND_VPORT:
>> + case ACTION_OPCODE_DHCP6_SERVER:
>> + case ACTION_OPCODE_HANDLE_SVC_CHECK:
>> + case ACTION_OPCODE_BFD_MSG:
>> + case ACTION_OPCODE_ACTIVATION_STRATEGY_RARP:
>> + case ACTION_OPCODE_MG_SPLIT_BUF:
>> + default:
>> + return false;
>> + }
>> +}
>> +
>> static void
>> encode_nested_actions(const struct ovnact_nest *on,
>> const struct ovnact_encode_params *ep,
>> @@ -1992,7 +2030,8 @@ encode_nested_actions(const struct ovnact_nest *on,
>> * converted to OpenFlow, as its userdata. ovn-controller will
>> convert the
>> * packet to ARP or NA and then send the packet and actions back to
>> the
>> * switch inside an OFPT_PACKET_OUT message. */
>> - size_t oc_offset = encode_start_controller_op(opcode, false,
>> + bool pause = is_paused_nested_action(opcode);
>> + size_t oc_offset = encode_start_controller_op(opcode, pause,
>> ep->ctrl_meter_id,
>> ofpacts);
>> ofpacts_put_openflow_actions(inner_ofpacts.data, inner_ofpacts.size,
>> ofpacts, OFP15_VERSION);
>> diff --git a/northd/northd.c b/northd/northd.c
>> index c568f6360..f65f0d1d9 100644
>> --- a/northd/northd.c
>> +++ b/northd/northd.c
>> @@ -13427,7 +13427,7 @@ build_arp_request_flows_for_lrouter(
>> "ip6.dst = %s; "
>> "nd.target = %s; "
>> "output; "
>> - "};", ETH_ADDR_ARGS(eth_dst), sn_addr_s,
>> + "}; output;", ETH_ADDR_ARGS(eth_dst), sn_addr_s,
>> route->nexthop);
>>
>> ovn_lflow_add_with_hint__(lflows, od, S_ROUTER_IN_ARP_REQUEST,
>> 200,
>> @@ -13447,7 +13447,7 @@ build_arp_request_flows_for_lrouter(
>> "arp.tpa = " REG_NEXT_HOP_IPV4 "; "
>> "arp.op = 1; " /* ARP request */
>> "output; "
>> - "};",
>> + "}; output;",
>> copp_meter_get(COPP_ARP_RESOLVE, od->nbr->copp,
>> meter_groups),
>> lflow_ref);
>> @@ -13456,7 +13456,7 @@ build_arp_request_flows_for_lrouter(
>> "nd_ns { "
>> "nd.target = " REG_NEXT_HOP_IPV6 "; "
>> "output; "
>> - "};",
>> + "}; output;",
>> copp_meter_get(COPP_ND_NS_RESOLVE, od->nbr->copp,
>> meter_groups),
>> lflow_ref);
>> diff --git a/ovs b/ovs
>> index fe55ce37a..43db93787 160000
>> --- a/ovs
>> +++ b/ovs
>> @@ -1 +1 @@
>> -Subproject commit fe55ce37a7b090d09dee5c01ae0797320ad678f6
>> +Subproject commit 43db937876fcec57947634d6e642429673d1ec06
>> diff --git a/tests/ovn-northd.at b/tests/ovn-northd.at
>> index cd53755b2..b3f8bd158 100644
>> --- a/tests/ovn-northd.at
>> +++ b/tests/ovn-northd.at
>> @@ -6947,6 +6947,9 @@ ct_dnat /* assuming no un-dnat entry, so no change
>> */ {
>> arp.op = 1;
>> output("rtr-ls");
>> };
>> + ct_dnat /* assuming no un-dnat entry, so no change */ {
>> + output("rtr-ls");
>> + };
>> };
>> };
>> ])
>> diff --git a/tests/ovn.at b/tests/ovn.at
>> index dd2ebce98..471cafcf9 100644
>> --- a/tests/ovn.at
>> +++ b/tests/ovn.at
>> @@ -1540,11 +1540,11 @@ clone { ip4.dst = 255.255.255.255; output; };
>> next;
>>
>> # arp
>> arp { eth.dst = ff:ff:ff:ff:ff:ff; output; }; output;
>> - encodes as
>> controller(userdata=00.00.00.00.00.00.00.00.00.19.00.10.80.00.06.06.ff.ff.ff.ff.ff.ff.00.00.ff.ff.00.10.00.00.23.20.00.0e.ff.f8.OFTABLE_SAVE_INPORT_HEX.00.00.00),resubmit(,OFTABLE_SAVE_INPORT)
>> + encodes as
>> controller(userdata=00.00.00.00.00.00.00.00.00.19.00.10.80.00.06.06.ff.ff.ff.ff.ff.ff.00.00.ff.ff.00.10.00.00.23.20.00.0e.ff.f8.OFTABLE_SAVE_INPORT_HEX.00.00.00,pause),resubmit(,OFTABLE_SAVE_INPORT)
>> has prereqs ip4
>> arp { };
>> formats as arp { drop; };
>> - encodes as controller(userdata=00.00.00.00.00.00.00.00)
>> + encodes as controller(userdata=00.00.00.00.00.00.00.00,pause)
>> has prereqs ip4
>>
>> # get_arp
>> @@ -1674,12 +1674,12 @@ reg1[[0]] = put_dhcp_opts(offerip=1.2.3.4,
>> domain_search_list=1.2.3.4);
>>
>> # nd_ns
>> nd_ns { nd.target = xxreg0; output; };
>> - encodes as
>> controller(userdata=00.00.00.09.00.00.00.00.00.1c.00.18.00.
>> 80.00.00.00.00.00.00.00.01.de
>> .10.80.00.3e.10.00.00.00.00.ff.ff.00.10.00.00.23.20.00.0e.ff.f8.OFTABLE_SAVE_INPORT_HEX.00.00.00)
>> + encodes as
>> controller(userdata=00.00.00.09.00.00.00.00.00.1c.00.18.00.
>> 80.00.00.00.00.00.00.00.01.de
>> .10.80.00.3e.10.00.00.00.00.ff.ff.00.10.00.00.23.20.00.0e.ff.f8.OFTABLE_SAVE_INPORT_HEX.00.00.00,pause)
>> has prereqs ip6
>>
>> nd_ns { };
>> formats as nd_ns { drop; };
>> - encodes as controller(userdata=00.00.00.09.00.00.00.00)
>> + encodes as controller(userdata=00.00.00.09.00.00.00.00,pause)
>> has prereqs ip6
>>
>> # nd_na
>> diff --git a/tests/system-ovn.at b/tests/system-ovn.at
>> index fc69facb9..c69e649a6 100644
>> --- a/tests/system-ovn.at
>> +++ b/tests/system-ovn.at
>> @@ -12298,3 +12298,98 @@ OVS_TRAFFIC_VSWITCHD_STOP(["/failed to query
>> port patch-.*/d
>> /connection dropped.*/d"])
>> AT_CLEANUP
>> ])
>> +
>> +OVN_FOR_EACH_NORTHD([
>> +AT_SETUP([LB with first packet buffered])
>> +AT_KEYWORDS([ovnlb])
>> +
>> +CHECK_CONNTRACK()
>> +CHECK_CONNTRACK_NAT()
>> +
>> +ovn_start
>> +OVS_TRAFFIC_VSWITCHD_START()
>> +ADD_BR([br-int])
>> +ADD_BR([br-ext])
>> +
>> +check ovs-ofctl add-flow br-ext action=normal
>> +# Set external-ids in br-int needed for ovn-controller
>> +check ovs-vsctl \
>> + -- set Open_vSwitch . external-ids:system-id=hv1 \
>> + -- set Open_vSwitch .
>> external-ids:ovn-remote=unix:$ovs_base/ovn-sb/ovn-sb.sock \
>> + -- set Open_vSwitch . external-ids:ovn-encap-type=geneve \
>> + -- set Open_vSwitch . external-ids:ovn-encap-ip=169.0.0.1 \
>> + -- set bridge br-int fail-mode=secure
>> other-config:disable-in-band=true \
>> + -- set Open_vSwitch .
>> external-ids:ovn-bridge-mappings=phynet:br-ext
>> +
>> +# Start ovn-controller
>> +start_daemon ovn-controller
>> +
>> +check ovn-nbctl lr-add lr
>> +check ovn-nbctl ls-add internal
>> +check ovn-nbctl ls-add public
>> +
>> +check ovn-nbctl lrp-add lr lr-pub 00:00:01:01:02:03 192.168.100.1/24
>> 1000::1/64
>> +check ovn-nbctl lsp-add public pub-lr -- set Logical_Switch_Port pub-lr
>> \
>> + type=router options:router-port=lr-pub
>> addresses=\"00:00:01:01:02:03\"
>> +
>> +check ovn-nbctl lrp-add lr lr-internal 00:00:01:01:02:04
>> 192.168.200.1/24 2000::1/64
>> +check ovn-nbctl lsp-add internal internal-lr -- set Logical_Switch_Port
>> internal-lr \
>> + type=router options:router-port=lr-internal
>> addresses=\"00:00:01:01:02:04\"
>> +
>> +check ovn-nbctl lsp-add internal server -- lsp-set-addresses server
>> "unknown"
>> +
>> +ovn-nbctl lsp-add public ln_port \
>> + -- lsp-set-addresses ln_port unknown \
>> + -- lsp-set-type ln_port localnet \
>> + -- lsp-set-options ln_port network_name=phynet
>> +
>> +check ovn-nbctl set logical_router lr options:chassis=hv1
>> +
>> +check ovn-nbctl lb-add lb1 192.168.100.20 192.168.200.10
>> +check ovn-nbctl lb-add lb2 1000::20 2000::10
>> +check ovn-nbctl lr-lb-add lr lb1
>> +check ovn-nbctl lr-lb-add lr lb2
>> +check ovn-nbctl --wait=hv sync
>> +
>> +ADD_NAMESPACES(client)
>> +ADD_VETH(client, client, br-ext, "1000::10/64", "f0:00:00:01:02:03", \
>> + "1000::1", "nodad", "192.168.100.10/24", "192.168.100.1")
>> +
>> +ADD_NAMESPACES(server)
>> +ADD_VETH(server, server, br-int, "2000::10/64", "f0:00:0f:01:02:03", \
>> + "2000::1", "nodad", "192.168.200.10/24", "192.168.200.1")
>> +
>> +NETNS_DAEMONIZE([server], [nc -l -u 192.168.200.10 4242 > /dev/null],
>> [serverv4.pid])
>> +NETNS_DAEMONIZE([server], [nc -l -u 2000::10 4243 > /dev/null],
>> [serverv6.pid])
>> +
>> +NETNS_START_TCPDUMP([client], [-l -U -i client -vnne udp], [client])
>> +NETNS_START_TCPDUMP([server], [-l -U -i server -vnne udp], [server])
>> +
>> +check ovs-appctl dpctl/flush-conntrack
>> +
>> +NS_CHECK_EXEC([client], [nc -z -u 192.168.100.20 4242], [ignore],
>> [ignore], [ignore])
>> +OVS_WAIT_UNTIL([grep -q "192.168.200.10" server.tcpdump])
>> +
>> +NS_CHECK_EXEC([client], [nc -z -u 1000::20 4243])
>> +OVS_WAIT_UNTIL([grep -q "2000::10" server.tcpdump])
>> +
>> +zone_id=$(ovn-appctl -t ovn-controller ct-zone-list | grep lr_dnat | cut
>> -d ' ' -f2)
>> +AT_CHECK([ovs-appctl dpctl/dump-conntrack | grep -c "zone=$zone_id"],
>> [0], [2
>> +])
>> +
>> +OVS_APP_EXIT_AND_WAIT([ovn-controller])
>> +
>> +as ovn-sb
>> +OVS_APP_EXIT_AND_WAIT([ovsdb-server])
>> +
>> +as ovn-nb
>> +OVS_APP_EXIT_AND_WAIT([ovsdb-server])
>> +
>> +as northd
>> +OVS_APP_EXIT_AND_WAIT([ovn-northd])
>> +
>> +as
>> +OVS_TRAFFIC_VSWITCHD_STOP(["/failed to query port patch-.*/d
>> +/connection dropped.*/d"])
>> +AT_CLEANUP
>> +])
>> --
>> 2.44.0
>>
>> _______________________________________________
>> dev mailing list
>> d...@openvswitch.org
>> https://mail.openvswitch.org/mailman/listinfo/ovs-dev
>>
>>
Thanks,
Ales
--
Ales Musil
Senior Software Engineer - OVN Core
Red Hat EMEA <https://www.redhat.com>
amu...@redhat.com
<https://red.ht/sig>
_______________________________________________
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
