On Wed, 2026-01-28 at 15:01 +0100, Dumitru Ceara wrote:
> On 1/26/26 10:39 AM, Martin Kalcok via dev wrote:
> > Load Balancer Health Checks require specifying "source IP". This IP
> > has to be from the subnet of the monitored LB backend and should
> > not
> > be used by any other port in the LSP. If the "source IP" is also
> > used
> > by some other port, the host with the LB backend won't be able to
> > communicate
> > with this port due to the ARP responder rule installed by the
> > controller.
> >
> > This limitation forces CMSes to reserve an IP per LSP with LB
> > backend, which
> > is not always practical or possible.
> >
> > This change allows usage of LRP IPs as the source of LB health
> > check
> > probes. It introduces following changes for such scenario:
> >
> > * service_monitor (SBDB) will use MAC address of the LRP
> > * ARP responder rule is not necessary in this case
> > * Destination IP, "inport" and source port will be used to
> > determine that a
> > packet is a response to a health check probe. These packets
> > will be
> > redirected to the controller.
> >
> > Behavior is unchanged for the LB health checks that use
> > reserved/unused
> > "source IPs".
> >
> > Signed-off-by: Martin Kalcok <[email protected]>
> > ---
> >
>
> Hi Martin,
>
> Thanks for this new revision!
>
Sorry, I forgot to reply to one of your questions regarding the tests
in my previous mail. I'm adding it below.
> > RFC -> v2
> > * Reduce number of iterations when searching for monitor source
> > IP in ports
> > by searching only in lrp_networks of LRPs, using
> > `lrp_find_member_ip`.
> > * Narrowed down match for health check replies by matching
> > "inport == <monitored_lb_backend_port>"
> > * ICMP unreachable messages are also forwarded to the controller
> > to allow
> > for monitoring UDP ports and detecting when they are down.
> > * Flows that send monitor replies to the controller are still
> > inserted into
> > LS datapath, but the logic is tied to the LR. i.e.: Whether to
> > add the
> > flows is evaluated when LB is attached to LR.
> > It removes the necessity to add LB to the LS as well. It is
> > enough to add
> > it to the LR.
> > * IPv6 support added
> > * System and Unit tests added
> > * existing LB and Service Monitor checks were already quite
> > involved, so
> > I opted to add new set of tests that focus specifically on a
> > setup
> > with LRP IPs as a source for service checks.
> > * Documentation of "ip_port_mappings" field in "Load_Balancer"
> > was extended
> > to explicitly state requirements for the source_ip.
> >
> > v2 -> v3
> > * rebased on `main`
> > * `vector_get` replaced with `sparse_array_get`
> > * `ovn_lflow_metered` replace with WITH_CTRL_METER macro
> > * Addressed review comments
> > * NEWS entry added
> > * Formatting and comments fixed
> > * Formatting of "match" rules in
> > `build_lb_health_check_response_lflows`
> > split for ipv4 and ipv6 cases.
> > * [tests] added `check` to ovs-vsctl calls
> > * [tests] dropped unnecessary --wait=sb
> > * Bug found: Function `build_lb_health_check_response_lflows`
> > was, on each
> > call, iterating over backends of the same `ovn_northd_lb_vip`
> > instance
> > (see in v2: `&lb->vips_nb->backends_nb[j++];`). This caused
> > occasional
> > segfaults and was replaced by passing "current"
> > `ovn_northd_lb_vip` instance
> > as an argument to the function.
> >
> > NEWS | 5 +
> > northd/northd.c | 143 +++++++++++++++-
> > ovn-nb.xml | 5 +-
> > tests/ovn.at | 325 +++++++++++++++++++++++++++++++++++
> > tests/system-ovn.at | 400
> > ++++++++++++++++++++++++++++++++++++++++++++
> > 5 files changed, 875 insertions(+), 3 deletions(-)
> >
> > diff --git a/NEWS b/NEWS
> > index b7268c835..2a2b5e12d 100644
> > --- a/NEWS
> > +++ b/NEWS
> > @@ -93,6 +93,11 @@ Post v25.09.0
> > other_config column.
> > - Introduce the capability to specify multiple ips for ovn-
> > evpn-local-ip
> > option.
> > + - Load balancer health checks can now use Logical Router Port
> > IPs as the
> > + source IP for health check probes. Previously, health checks
> > required
> > + reserving an unused IP from the backend's subnet. This change
> > allows
> > + using LRP IPs directly, eliminating the need to reserve
> > additional IPs
> > + per backend port.
> >
> > OVN v25.09.0 - xxx xx xxxx
> > --------------------------
> > diff --git a/northd/northd.c b/northd/northd.c
> > index adaa94e85..1e4e24216 100644
> > --- a/northd/northd.c
> > +++ b/northd/northd.c
> > @@ -3260,6 +3260,28 @@ ovn_lb_svc_create(struct ovsdb_idl_txn
> > *ovnsb_txn,
> > continue;
> > }
> >
> > + /* If the service monitor is backed by a real port,
> > use its MAC
> > + * address instead of the default service check MAC.
> > */
> > + const char *source_mac = svc_monitor_mac;
> > + const struct eth_addr *source_mac_ea =
> > svc_monitor_mac_ea;
>
> Nit: I'd swap these two around.
>
> > + if (op) {
> > + struct ovn_port *svc_mon_op;
> > + VECTOR_FOR_EACH (&op->od->router_ports,
> > svc_mon_op) {
> > + if (!svc_mon_op->peer) {
> > + continue;
> > + }
> > + const char *lrp_ip = lrp_find_member_ip(
> > + svc_mon_op->peer,
> > + backend_nb->svc_mon_src_ip);
>
> Nit: the indentation is still a bit weird.
>
> > + if (lrp_ip && !strcmp(lrp_ip,
> > + backend_nb-
> > >svc_mon_src_ip)) {
> > + source_mac = svc_mon_op->peer-
> > >lrp_networks.ea_s;
> > + source_mac_ea = &svc_mon_op->peer-
> > >lrp_networks.ea;
>
> Nit: I'd swap these two around.
>
> > + break;
> > + }
> > + }
> > + }
> > +
> > const char *protocol = lb->nlb->protocol;
> > if (!protocol || !protocol[0]) {
> > protocol = "tcp";
> > @@ -3289,9 +3311,9 @@ ovn_lb_svc_create(struct ovsdb_idl_txn
> > *ovnsb_txn,
> > struct eth_addr ea;
> > if (!mon_info->sbrec_mon->src_mac ||
> > !eth_addr_from_string(mon_info->sbrec_mon-
> > >src_mac, &ea) ||
> > - !eth_addr_equals(ea, *svc_monitor_mac_ea)) {
> > + !eth_addr_equals(ea, *source_mac_ea)) {
> > sbrec_service_monitor_set_src_mac(mon_info-
> > >sbrec_mon,
> > -
> > svc_monitor_mac);
> > + source_mac);
> > }
> >
> > if (!mon_info->sbrec_mon->src_ip ||
> > @@ -8499,6 +8521,99 @@ build_lb_rules_for_stateless_acl(struct
> > lflow_table *lflows,
> > }
> > }
> >
> > +static void
> > +build_lb_health_check_response_lflows(
> > + struct lflow_table *lflows,
> > + const struct ovn_northd_lb *lb,
> > + const struct ovn_lb_vip *lb_vip,
> > + const struct ovn_northd_lb_vip *lb_vip_nb,
> > + const struct ovn_lb_datapaths *lb_dps,
> > + const struct ovn_datapaths *lr_datapaths,
> > + const struct shash *meter_groups,
> > + struct ds *match,
> > + struct ds *action)
> > +{
> > + /* For each LB backend that is monitored by a source_ip
> > belonging
> > + * to a real LRP, install rule that punts service check
> > replies to the
> > + * controller. */
> > + size_t j = 0;
> > + const struct ovn_lb_backend *backend;
>
> Nit: reverse xmas tree and I'd add an empty line here.
>
> > + VECTOR_FOR_EACH_PTR (&lb_vip->backends, backend) {
> > + struct ovn_northd_lb_backend *backend_nb =
> > + &lb_vip_nb->backends_nb[j++];
> > +
> > + if (!backend_nb->health_check) {
> > + continue;
> > + }
> > +
> > + const char *protocol = lb->nlb->protocol;
> > + if (!protocol || !protocol[0]) {
> > + protocol = "tcp";
> > + }
> > +
> > + size_t index;
> > + DYNAMIC_BITMAP_FOR_EACH_1 (index, &lb_dps->nb_lr_map) {
> > + struct ovn_datapath *od =
> > sparse_array_get(&lr_datapaths->dps,
> > + index);
>
> Nit: indentation is one space off.
>
> > + /* Only install the rule if the datapath has a port
> > with
> > + * monitor source IP. */
> > + struct ovn_datapath *peer_switch_od = NULL;
> > + struct ovn_port *svc_mon_op;
> > +
> > + HMAP_FOR_EACH (svc_mon_op, dp_node, &od->ports) {
> > + if (!svc_mon_op->peer) {
> > + continue;
> > + }
> > + const char *lrp_ip = lrp_find_member_ip(
> > + svc_mon_op,
> > + backend_nb->svc_mon_src_ip);
> > + if (lrp_ip &&
> > + !strcmp(lrp_ip, backend_nb-
> > >svc_mon_src_ip)) {
> > + peer_switch_od = svc_mon_op->peer->od;
>
> What if svc_mon_op->peer->od is not a switch datapath? That is, what
> if
> svc_mon_op is a router port connecting this router to another router?
>
> I know the topology would be a bit weird, but I think we should check
> that and maybe just skip those ports. Otherwise the lflows we add
> below
> might cause issues.
>
> > + break;
> > + }
> > + }
> > + if (!peer_switch_od) {
> > + continue;
> > + }
> > +
> > + ds_clear(match);
> > + ds_clear(action);
> > +
> > + /* icmp6 type 1 and icmp4 type 3 are included as well,
> > because
> > + * the controller is using them to detect unreachable
> > ports. */
> > + if (addr_is_ipv6(backend_nb->svc_mon_src_ip)) {
> > + ds_put_format(match,
> > + "inport == \"%s\" && ip6.dst == %s
> > && "
> > + "(%s.src == %s || icmp6.type == 1)",
> > + backend_nb->logical_port,
> > + backend_nb->svc_mon_src_ip,
> > + protocol,
> > + backend->port_str);
> > + } else {
> > + ds_put_format(match,
> > + "inport == \"%s\" && ip4.dst == %s
> > && "
> > + "(%s.src == %s || icmp4.type == 3)",
> > + backend_nb->logical_port,
> > + backend_nb->svc_mon_src_ip,
> > + protocol,
> > + backend->port_str);
> > + }
> > +
>
> Should we be a little more strict here and match on eth.dst ==
> $svc_monitor_mac (i.e., svc_mon_op->peer.lport_addresses.ea)?
>
> > + /* ovn-controller expects health check responses from
> > the LS
> > + * datapath in which the backend is located. That's
> > why we
> > + * install the response lflow into the peer's
> > datapath. */
> > + const char *meter = copp_meter_get(COPP_SVC_MONITOR,
> > + peer_switch_od-
> > >nbs->copp,
> > + meter_groups);
>
> Nit: indentation, the two lines above need to be one space to the
> left.
>
> > + ovn_lflow_add(lflows, peer_switch_od, S_SWITCH_IN_LB,
> > 160,
>
> I'm not sure if this was discussed in previous versions but for
> regular
> LB healthchecks (when we don't use the router IP) we
> handle_svc_check()
> in S_SWITCH_IN_L2_LKUP. Why do we do it in S_SWITCH_IN_LB for router
> IPs?
>
> > + ds_cstr(match),
> > "handle_svc_check(inport);",
> > + lb_dps->lflow_ref,
> > + WITH_CTRL_METER(meter));
>
> Nit: this fits on a single line.
>
> > + }
> > + }
> > +}
> > +
> > static void
> > build_lb_rules(struct lflow_table *lflows, struct ovn_lb_datapaths
> > *lb_dps,
> > const struct ovn_datapaths *ls_datapaths,
> > @@ -10349,6 +10464,26 @@ build_lswitch_arp_nd_local_svc_mon(const
> > struct ovn_lb_datapaths *lb_dps,
> > continue;
> > }
> >
> > + /* ARP responder is necessary only if the service
> > check is not
> > + * backed by a real port and an IP. */
> > + struct ovn_port *svc_mon_op;
> > + bool port_found = false;
> > + VECTOR_FOR_EACH (&op->od->router_ports, svc_mon_op) {
> > + if (!svc_mon_op->peer) {
> > + continue;
> > + }
> > + const char *lrp_ip = lrp_find_member_ip(
> > + svc_mon_op->peer,
> > + backend_nb->svc_mon_src_ip);
> > + if (lrp_ip && !strcmp(lrp_ip, backend_nb-
> > >svc_mon_src_ip)) {
> > + port_found = true;
> > + break;
> > + }
> > + }
> > + if (port_found) {
> > + continue;
> > + }
> > +
> > ds_clear(match);
> > ds_clear(actions);
> >
> > @@ -13136,6 +13271,10 @@ build_lrouter_flows_for_lb(struct
> > ovn_lb_datapaths *lb_dps,
> > build_lrouter_allow_vip_traffic_template(lflows, lb_dps,
> > lb_vip, lb,
> > lr_datapaths);
> >
> > + build_lb_health_check_response_lflows(
> > + lflows, lb, lb_vip, &lb->vips_nb[i], lb_dps,
> > lr_datapaths,
> > + meter_groups, match, action);
> > +
> > if (!build_empty_lb_event_flow(lb_vip, lb, match, action))
> > {
> > continue;
> > }
> > diff --git a/ovn-nb.xml b/ovn-nb.xml
> > index e9ca27413..2ddc3bc20 100644
> > --- a/ovn-nb.xml
> > +++ b/ovn-nb.xml
> > @@ -2402,7 +2402,10 @@
> > e.g.
> > <code><var>port_name</var>:<var>sourc_ip</var></code> for IPv4.
> > Health checks are sent to this port with the specified
> > source IP.
> > For IPv6 square brackets must be used around IP address,
> > e.g:
> > - <code><var>port_name</var>:<var>[sourc_ip]</var></code>
> > + <code><var>port_name</var>:<var>[sourc_ip]</var></code>.
> > The source
> > + IP must be from the subnet of the monitored endpoint. It
> > can be
>
> Nit: we don't always enforce it but our coding style say that
> sentences
> in comments should be separated by two spaces. We try to do that for
> docs too.
>
> > + either an unused IP from the subnet, or an IP of one of
> > the Logical
> > + Router Ports connected to the same switch.
> > Remote endpoint:
> > Specify :target_zone_name at the end of the above syntax
> > to create
> > remote health checks in a specific zone.
> > diff --git a/tests/ovn.at b/tests/ovn.at
> > index 4d15d4a62..782fd0256 100644
> > --- a/tests/ovn.at
> > +++ b/tests/ovn.at
> > @@ -26972,6 +26972,331 @@
> > ignored_tables=OFTABLE_CHK_LB_HAIRPIN,OFTABLE_CT_SNAT_HAIRPIN])
> > AT_CLEANUP
> > ])
> >
> > +OVN_FOR_EACH_NORTHD([
> > +AT_SETUP([Load balancer health checks - LRP source IP - IPv4])
> > +AT_KEYWORDS([lb])
> > +ovn_start
> > +
> > +net_add n1
> > +
> > +sim_add hv1
> > +as hv1
> > +ovs-vsctl add-br br-phys
> > +ovn_attach n1 br-phys 192.168.0.1
> > +check ovs-vsctl -- add-port br-int hv1-vif1 -- \
> > + set interface hv1-vif1 external-ids:iface-id=sw0-p1 \
> > + options:tx_pcap=hv1/vif1-tx.pcap \
> > + options:rxq_pcap=hv1/vif1-rx.pcap \
> > + ofport-request=1
> > +check ovs-vsctl -- add-port br-int hv1-vif2 -- \
> > + set interface hv1-vif2 external-ids:iface-id=sw0-p2 \
> > + options:tx_pcap=hv1/vif2-tx.pcap \
> > + options:rxq_pcap=hv1/vif2-rx.pcap \
> > + ofport-request=2
> > +
> > +sim_add hv2
> > +as hv2
> > +check ovs-vsctl add-br br-phys
> > +ovn_attach n1 br-phys 192.168.0.2
> > +check ovs-vsctl -- add-port br-int hv2-vif1 -- \
> > + set interface hv2-vif1 external-ids:iface-id=sw1-p1 \
> > + options:tx_pcap=hv2/vif1-tx.pcap \
> > + options:rxq_pcap=hv2/vif1-rx.pcap \
> > + ofport-request=1
> > +
> > +check ovn-nbctl ls-add sw0
> > +check ovn-nbctl lsp-add sw0 sw0-p1
> > +check ovn-nbctl lsp-set-addresses sw0-p1 "50:54:00:00:00:03
> > 10.0.0.3"
> > +check ovn-nbctl lsp-set-port-security sw0-p1 "50:54:00:00:00:03
> > 10.0.0.3"
> > +
> > +check ovn-nbctl lsp-add sw0 sw0-p2
> > +check ovn-nbctl lsp-set-addresses sw0-p2 "50:54:00:00:00:04
> > 10.0.0.4"
> > +check ovn-nbctl lsp-set-port-security sw0-p2 "50:54:00:00:00:04
> > 10.0.0.4"
> > +
> > +check ovn-nbctl ls-add sw1
> > +check ovn-nbctl lsp-add sw1 sw1-p1
> > +check ovn-nbctl lsp-set-addresses sw1-p1 "40:54:00:00:00:03
> > 20.0.0.3"
> > +check ovn-nbctl lsp-set-port-security sw1-p1 "40:54:00:00:00:03
> > 20.0.0.3"
> > +
> > +# Create a logical router and attach both logical switches.
> > +check ovn-nbctl lr-add lr0
> > +check ovn-nbctl lrp-add lr0 lr0-sw0 00:00:00:00:ff:01 10.0.0.1/24
> > +check ovn-nbctl lsp-add-router-port sw0 sw0-lr0 lr0-sw0
> > +
> > +check ovn-nbctl lrp-add lr0 lr0-sw1 00:00:00:00:ff:02 20.0.0.1/24
> > +check ovn-nbctl lsp-add-router-port sw1 sw1-lr0 lr0-sw1
> > +
> > +# Create TCP LB and configure health checks using router port IPs
> > as source.
> > +check ovn-nbctl lb-add lb1 10.0.0.10:80 10.0.0.3:80,20.0.0.3:80
> > +check ovn-nbctl set load_balancer lb1
> > ip_port_mappings:10.0.0.3=sw0-p1:10.0.0.1
> > +check ovn-nbctl set load_balancer lb1
> > ip_port_mappings:20.0.0.3=sw1-p1:20.0.0.1
> > +
> > +AT_CHECK([ovn-nbctl --wait=sb \
> > + -- --id=@hc create Load_Balancer_Health_Check
> > vip="10.0.0.10\:80" \
> > + -- add Load_Balancer lb1 health_check @hc | uuidfilt],
> > [0], [<0>
> > +])
> > +
> > +# Create UDP LB with health check using router port IP.
> > +check ovn-nbctl lb-add lb2 10.0.0.11:53 10.0.0.4:53 udp
> > +check ovn-nbctl set load_balancer lb2
> > ip_port_mappings:10.0.0.4=sw0-p2:10.0.0.1
> > +
> > +AT_CHECK([ovn-nbctl --wait=sb \
> > + -- --id=@hc create Load_Balancer_Health_Check
> > vip="10.0.0.11\:53" \
> > + -- add Load_Balancer lb2 health_check @hc | uuidfilt],
> > [0], [<0>
> > +])
> > +
> > +check ovn-nbctl --wait=sb lr-lb-add lr0 lb1
> > +check ovn-nbctl --wait=sb lr-lb-add lr0 lb2
> > +
> > +OVN_POPULATE_ARP
> > +wait_for_ports_up
> > +check ovn-nbctl --wait=hv sync
> > +
> > +wait_row_count Service_Monitor 3
> > +
> > +# Verify that Service_Monitor uses LRP MAC address, not
> > svc_monitor_mac.
> > +lrp_sw0_mac=$(ovn-nbctl get Logical_Router_Port lr0-sw0 mac | sed
> > 's/"//g')
> > +lrp_sw1_mac=$(ovn-nbctl get Logical_Router_Port lr0-sw1 mac | sed
> > 's/"//g')
> > +
> > +svc_mon_sw0_p1_mac=$(ovn-sbctl --bare --columns src_mac find
> > Service_Monitor logical_port=sw0-p1)
> > +svc_mon_sw0_p2_mac=$(ovn-sbctl --bare --columns src_mac find
> > Service_Monitor logical_port=sw0-p2)
> > +svc_mon_sw1_mac=$(ovn-sbctl --bare --columns src_mac find
> > Service_Monitor logical_port=sw1-p1)
> > +
> > +AT_CHECK([test "$svc_mon_sw0_p1_mac" = "$lrp_sw0_mac"])
> > +AT_CHECK([test "$svc_mon_sw0_p2_mac" = "$lrp_sw0_mac"])
> > +AT_CHECK([test "$svc_mon_sw1_mac" = "$lrp_sw1_mac"])
> > +
> > +# Verify priority 160 flows exist for punting health check replies
> > to controller.
> > +AT_CAPTURE_FILE([sbflows1])
> > +ovn-sbctl dump-flows sw0 > sbflows1
> > +AT_CAPTURE_FILE([sbflows2])
> > +ovn-sbctl dump-flows sw1 > sbflows2
> > +
> > +## TCP Load balancer health check flows.
> > +# TCP backend - verify flow matches on inport, destination IP, and
> > TCP source port.
> > +AT_CHECK([grep "priority=160" sbflows1 | grep "handle_svc_check" |
> > \
> > + grep 'inport == "sw0-p1"' | grep "ip4.dst == 10.0.0.1" |
> > \
> > + grep -q "tcp.src == 80" ])
> > +
> > +AT_CHECK([grep "priority=160" sbflows2 | grep "handle_svc_check" |
> > \
> > + grep 'inport == "sw1-p1"' | grep "ip4.dst == 20.0.0.1" |
> > \
> > + grep -q "tcp.src == 80" ])
> > +
> > +
>
> Is this extra newline on purpose?
>
> > +# Verify flow also matches ICMP unreachable (type 3 for IPv4).
> > +AT_CHECK([grep "priority=160" sbflows1 | grep "handle_svc_check" |
> > \
> > + grep 'inport == "sw0-p1"' | grep "ip4.dst == 10.0.0.1" |
> > \
> > + grep -q "icmp4.type == 3"])
> > +
> > +AT_CHECK([grep "priority=160" sbflows2 | grep "handle_svc_check" |
> > \
> > + grep 'inport == "sw1-p1"' | grep "ip4.dst == 20.0.0.1" |
> > \
> > + grep -q "icmp4.type == 3"])
> > +
> > +## UDP Load balancer health check flows.
> > +# UDP backend - verify flow uses UDP instead of TCP.
> > +AT_CHECK([grep "priority=160" sbflows1 | grep "handle_svc_check" |
> > \
> > + grep 'inport == "sw0-p2"' | grep "ip4.dst == 10.0.0.1"
> > |\
>
> Nit: missing space before \
>
> > + grep -q "udp.src == 53"])
>
> Can all the AT_CHECKs above be replaced with a plain "check"?
>
> > +
> > +# Verify NO TCP match in the UDP backend flow.
> > +AT_CHECK([grep "priority=160" sbflows1 | grep "handle_svc_check" |
> > \
> > + grep 'inport == "sw0-p2"' | grep "tcp.src"], [1])
> > +
> > +# Verify NO ARP responder flows using svc_monitor_mac for the
> > router port IPs.
> > +# The router port IPs should use the LRP MAC, not the global
> > svc_monitor_mac.
> > +svc_monitor_mac=$(ovn-nbctl get NB_Global .
> > options:svc_monitor_mac | sed 's/"//g')
> > +AT_CHECK([test -n "$svc_monitor_mac"])
> > +
> > +AT_CHECK([grep "arp.tpa == 10.0.0.1" sbflows1 | \
> > + grep "arp.sha = $svc_monitor_mac"], [1])
> > +AT_CHECK([grep "arp.tpa == 20.0.0.1" sbflows2 | \
> > + grep "arp.sha = $svc_monitor_mac"], [1])
> > +
> > +# Verify health check packets are sent with LRP MAC as source.
> > +lrp_sw0_mac_hex=$(echo $lrp_sw0_mac | sed 's/://g')
> > +lrp_sw1_mac_hex=$(echo $lrp_sw1_mac | sed 's/://g')
> > +
> > +OVS_WAIT_UNTIL(
> > + [test 1 -le `$PYTHON "$ovs_srcdir/utilities/ovs-pcap.in"
> > hv1/vif1-tx.pcap | \
> > + grep "505400000003${lrp_sw0_mac_hex}" | wc -l`]
> > +)
> > +
> > +OVS_WAIT_UNTIL(
> > + [test 1 -le `$PYTHON "$ovs_srcdir/utilities/ovs-pcap.in"
> > hv1/vif2-tx.pcap | \
> > + grep "505400000004${lrp_sw0_mac_hex}" | wc -l`]
> > +)
> > +
> > +OVS_WAIT_UNTIL(
> > + [test 1 -le `$PYTHON "$ovs_srcdir/utilities/ovs-pcap.in"
> > hv2/vif1-tx.pcap | \
> > + grep "405400000003${lrp_sw1_mac_hex}" | wc -l`]
> > +)
> > +
> > +OVN_CLEANUP([hv1], [hv2])
> > +AT_CLEANUP
> > +])
> > +
> > +OVN_FOR_EACH_NORTHD([
> > +AT_SETUP([Load balancer health checks - LRP source IP - IPv6])
> > +AT_KEYWORDS([lb])
> > +ovn_start
> > +
> > +net_add n1
> > +
> > +sim_add hv1
> > +as hv1
> > +ovs-vsctl add-br br-phys
> > +ovn_attach n1 br-phys 192.168.0.1
> > +check ovs-vsctl -- add-port br-int hv1-vif1 -- \
> > + set interface hv1-vif1 external-ids:iface-id=sw0-p1 \
> > + options:tx_pcap=hv1/vif1-tx.pcap \
> > + options:rxq_pcap=hv1/vif1-rx.pcap \
> > + ofport-request=1
> > +check ovs-vsctl -- add-port br-int hv1-vif2 -- \
> > + set interface hv1-vif2 external-ids:iface-id=sw0-p2 \
> > + options:tx_pcap=hv1/vif2-tx.pcap \
> > + options:rxq_pcap=hv1/vif2-rx.pcap \
> > + ofport-request=2
> > +
> > +sim_add hv2
> > +as hv2
> > +check ovs-vsctl add-br br-phys
> > +ovn_attach n1 br-phys 192.168.0.2
> > +check ovs-vsctl -- add-port br-int hv2-vif1 -- \
> > + set interface hv2-vif1 external-ids:iface-id=sw1-p1 \
> > + options:tx_pcap=hv2/vif1-tx.pcap \
> > + options:rxq_pcap=hv2/vif1-rx.pcap \
> > + ofport-request=1
> > +
> > +check ovn-nbctl ls-add sw0
> > +check ovn-nbctl lsp-add sw0 sw0-p1
> > +check ovn-nbctl lsp-set-addresses sw0-p1 "50:54:00:00:00:03
> > 2001::3"
> > +check ovn-nbctl lsp-set-port-security sw0-p1 "50:54:00:00:00:03
> > 2001::3"
> > +
> > +check ovn-nbctl lsp-add sw0 sw0-p2
> > +check ovn-nbctl lsp-set-addresses sw0-p2 "50:54:00:00:00:04
> > 2001::4"
> > +check ovn-nbctl lsp-set-port-security sw0-p2 "50:54:00:00:00:04
> > 2001::4"
> > +
> > +check ovn-nbctl ls-add sw1
> > +check ovn-nbctl lsp-add sw1 sw1-p1
> > +check ovn-nbctl lsp-set-addresses sw1-p1 "40:54:00:00:00:03
> > 2002::3"
> > +check ovn-nbctl lsp-set-port-security sw1-p1 "40:54:00:00:00:03
> > 2002::3"
> > +
> > +# Create a logical router and attach both logical switches.
> > +check ovn-nbctl lr-add lr0
> > +check ovn-nbctl lrp-add lr0 lr0-sw0 00:00:00:00:ff:01 2001::1/64
> > +check ovn-nbctl lsp-add-router-port sw0 sw0-lr0 lr0-sw0
> > +
> > +check ovn-nbctl lrp-add lr0 lr0-sw1 00:00:00:00:ff:02 2002::1/64
> > +check ovn-nbctl lsp-add-router-port sw1 sw1-lr0 lr0-sw1
> > +
> > +# Create TCP LB and configure health checks using router port IPs
> > as source.
> > +check ovn-nbctl lb-add lb1 [[2001::a]]:80
> > [[2001::3]]:80,[[2002::3]]:80
> > +check ovn-nbctl set load_balancer lb1
> > ip_port_mappings:\"[[2001::3]]\"=\"sw0-p1:[[2001::1]]\"
> > +check ovn-nbctl set load_balancer lb1
> > ip_port_mappings:\"[[2002::3]]\"=\"sw1-p1:[[2002::1]]\"
> > +
> > +AT_CHECK([ovn-nbctl --wait=sb \
> > + -- --id=@hc create Load_Balancer_Health_Check
> > vip="\[\[2001\:\:a\]\]\:80" \
> > + -- add Load_Balancer lb1 health_check @hc | uuidfilt],
> > [0], [<0>
> > +])
> > +
> > +# Create UDP LB with health check using router port IP.
> > +check ovn-nbctl lb-add lb2 [[2001::b]]:53 [[2001::4]]:53 udp
> > +check ovn-nbctl set load_balancer lb2
> > ip_port_mappings:\"[[2001::4]]\"=\"sw0-p2:[[2001::1]]\"
> > +
> > +AT_CHECK([ovn-nbctl --wait=sb \
> > + -- --id=@hc create Load_Balancer_Health_Check
> > vip="\[\[2001\:\:b\]\]\:53" \
> > + -- add Load_Balancer lb2 health_check @hc | uuidfilt],
> > [0], [<0>
> > +])
> > +
> > +check ovn-nbctl --wait=sb lr-lb-add lr0 lb1
> > +check ovn-nbctl --wait=sb lr-lb-add lr0 lb2
> > +
> > +OVN_POPULATE_ARP
> > +wait_for_ports_up
> > +check ovn-nbctl --wait=hv sync
> > +
> > +wait_row_count Service_Monitor 3
> > +
> > +# Verify that Service_Monitor uses LRP MAC address.
> > +lrp_sw0_mac=$(ovn-nbctl get Logical_Router_Port lr0-sw0 mac | sed
> > 's/"//g')
> > +lrp_sw1_mac=$(ovn-nbctl get Logical_Router_Port lr0-sw1 mac | sed
> > 's/"//g')
> > +
> > +svc_mon_sw0_p1_mac=$(ovn-sbctl --bare --columns src_mac find
> > Service_Monitor logical_port=sw0-p1)
> > +svc_mon_sw0_p2_mac=$(ovn-sbctl --bare --columns src_mac find
> > Service_Monitor logical_port=sw0-p2)
> > +svc_mon_sw1_mac=$(ovn-sbctl --bare --columns src_mac find
> > Service_Monitor logical_port=sw1-p1)
> > +
> > +AT_CHECK([test "$svc_mon_sw0_p1_mac" = "$lrp_sw0_mac"])
> > +AT_CHECK([test "$svc_mon_sw0_p2_mac" = "$lrp_sw0_mac"])
> > +AT_CHECK([test "$svc_mon_sw1_mac" = "$lrp_sw1_mac"])
> > +
> > +# Verify priority 160 flows exist for punting health check replies
> > to controller.
> > +AT_CAPTURE_FILE([sbflows1])
> > +ovn-sbctl dump-flows sw0 > sbflows1
> > +AT_CAPTURE_FILE([sbflows2])
> > +ovn-sbctl dump-flows sw1 > sbflows2
> > +
> > +## TCP Load balancer health check flows.
> > +# TCP backend - verify flow matches on inport, destination IP, and
> > TCP source port.
> > +AT_CHECK([grep "priority=160" sbflows1 | grep "handle_svc_check" |
> > \
> > + grep 'inport == "sw0-p1"' | grep "ip6.dst == 2001::1" |
> > \
> > + grep -q "tcp.src == 80" ])
> > +
> > +AT_CHECK([grep "priority=160" sbflows2 | grep "handle_svc_check" |
> > \
> > + grep 'inport == "sw1-p1"' | grep "ip6.dst == 2002::1" |
> > \
> > + grep -q "tcp.src == 80" ])
> > +
> > +# Verify flow also matches ICMP unreachable (type 1 for IPv6).
> > +AT_CHECK([grep "priority=160" sbflows1 | grep "handle_svc_check" |
> > \
> > + grep 'inport == "sw0-p1"' | grep "ip6.dst == 2001::1" |
> > \
> > + grep -q "icmp6.type == 1"])
> > +
> > +AT_CHECK([grep "priority=160" sbflows2 | grep "handle_svc_check" |
> > \
> > + grep 'inport == "sw1-p1"' | grep "ip6.dst == 2002::1" |
> > \
> > + grep -q "icmp6.type == 1"])
> > +
> > +## UDP Load balancer health check flows.
> > +# UDP backend - verify flow uses UDP instead of TCP.
> > +AT_CHECK([grep "priority=160" sbflows1 | grep "handle_svc_check" |
> > \
> > + grep 'inport == "sw0-p2"' | grep "ip6.dst == 2001::1" |
> > \
> > + grep -q "udp.src == 53"])
>
> Similar to the v4 case, can we use "check" instead of AT_CHECK for
> all
> these?
>
> > +
> > +# Verify NO TCP match in the UDP backend flow.
> > +AT_CHECK([grep "priority=160" sbflows1 | grep "handle_svc_check" |
> > \
> > + grep 'inport == "sw0-p2"' | grep "tcp.src"], [1])
> > +
> > +# Verify NO IPv6 ND responder flows using svc_monitor_mac for the
> > router port IPs.
> > +# The router port IPs should use the LRP MAC, not the global
> > svc_monitor_mac.
> > +svc_monitor_mac=$(ovn-nbctl get NB_Global .
> > options:svc_monitor_mac | sed 's/"//g')
> > +AT_CHECK([test -n "$svc_monitor_mac"])
> > +
> > +AT_CHECK([ovn-sbctl dump-flows sw0 | grep "nd.target == 2001::1" |
> > \
> > + grep "nd.tll = $svc_monitor_mac"], [1])
> > +AT_CHECK([ovn-sbctl dump-flows sw1 | grep "nd.target == 2002::1" |
> > \
> > + grep "nd.tll = $svc_monitor_mac"], [1])
> > +
> > +# Verify health check packets are sent with LRP MAC as source.
> > +lrp_sw0_mac_hex=$(echo $lrp_sw0_mac | sed 's/://g')
> > +lrp_sw1_mac_hex=$(echo $lrp_sw1_mac | sed 's/://g')
> > +
> > +OVS_WAIT_UNTIL(
> > + [test 1 -le `$PYTHON "$ovs_srcdir/utilities/ovs-pcap.in"
> > hv1/vif1-tx.pcap | \
> > + grep "505400000003${lrp_sw0_mac_hex}" | wc -l`]
> > +)
> > +
> > +OVS_WAIT_UNTIL(
> > + [test 1 -le `$PYTHON "$ovs_srcdir/utilities/ovs-pcap.in"
> > hv1/vif2-tx.pcap | \
> > + grep "505400000004${lrp_sw0_mac_hex}" | wc -l`]
> > +)
> > +
> > +OVS_WAIT_UNTIL(
> > + [test 1 -le `$PYTHON "$ovs_srcdir/utilities/ovs-pcap.in"
> > hv2/vif1-tx.pcap | \
> > + grep "405400000003${lrp_sw1_mac_hex}" | wc -l`]
> > +)
> > +
> > +OVN_CLEANUP([hv1], [hv2])
> > +AT_CLEANUP
> > +])
> > +
> > OVN_FOR_EACH_NORTHD([
> > AT_SETUP([SCTP Load balancer health checks])
> > AT_KEYWORDS([lb sctp])
> > diff --git a/tests/system-ovn.at b/tests/system-ovn.at
> > index 6eaf38a4c..a372d3676 100644
> > --- a/tests/system-ovn.at
> > +++ b/tests/system-ovn.at
> > @@ -4422,6 +4422,406 @@ OVS_TRAFFIC_VSWITCHD_STOP(["/failed to
> > query port patch-.*/d
> > AT_CLEANUP
> > ])
> >
> > +OVN_FOR_EACH_NORTHD([
> > +AT_SETUP([Load balancer health checks with LRP IP - IPv4])
> > +AT_KEYWORDS([lb])
> > +
> > +# This is a variant of a 'Load balancer health checks - IPv4' test
> > +# that uses LRP IPs as a source IP for the health checks. As such,
> > it
> > +# doesn't go so deep into testing and focuses only on "health
> > checks"
> > +# having proper status and the backend services being reachable.
> > +#
> > +# Topology:
> > +#
> > +# LB VIP: 172.16.0.42
> > +# Backends: 10.0.0.3 10.0.0.4
> > +#
> > +#
> > .3 +--------------+
> > +# +------------
> > + +------ | sw0-p1 |
> > +# .42 |
> > | | +--------------+
> > +# +--------------+ .2 .1 | |
> > .1 |
> > +# | sw-ext-p1 | ------- sw-ext ------- | lr0 | -------
> > sw0 10.0.0.0/24
> > +# +--------------+ |
> > | |
> > +# 172.16.0.0/24 |
> > | | .4 +--------------+
> > +# +------------
> > + +------ | sw0-p2 |
> > +#
> > +--------------+
> > +#
>
> Nice!
>
> > +ovn_start
> > +
> > +OVS_TRAFFIC_VSWITCHD_START()
> > +ADD_BR([br-int])
> > +
> > +# 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
> > +
> > +# Start ovn-controller.
> > +start_daemon ovn-controller
> > +
> > +# Create internal network with LB backends.
> > +check ovn-nbctl ls-add sw0
> > +
> > +check ovn-nbctl lsp-add sw0 sw0-p1
> > +check ovn-nbctl lsp-set-addresses sw0-p1 "50:54:00:00:00:03
> > 10.0.0.3"
> > +
> > +check ovn-nbctl lsp-add sw0 sw0-p2
> > +check ovn-nbctl lsp-set-addresses sw0-p2 "50:54:00:00:00:04
> > 10.0.0.4"
> > +
> > +# Create external network with LB VIP.
> > +check ovn-nbctl ls-add sw-ext
> > +
> > +check ovn-nbctl lsp-add sw-ext sw-ext-p1
> > +check ovn-nbctl lsp-set-addresses sw-ext-p1 "60:64:00:00:00:02
> > 172.16.0.2"
> > +
> > +# Create a logical router and attach both logical switches.
> > +check ovn-nbctl lr-add lr0
> > +check ovn-nbctl set logical_router lr0 options:chassis=hv1
> > +check ovn-nbctl lrp-add lr0 lr0-sw0 00:00:00:00:ff:01 10.0.0.1/24
> > +check ovn-nbctl lrp-add lr0 lr0-sw-ext 00:00:00:00:ee:01
> > 172.16.0.1/24
> > +check ovn-nbctl lsp-add-router-port sw0 sw0-lr0 lr0-sw0
> > +check ovn-nbctl lsp-add-router-port sw-ext sw-ext-lr0 lr0-sw-ext
> > +
> > +check ovn-nbctl lb-add lb1 172.16.0.42:80 10.0.0.3:80,10.0.0.4:80
> > +
> > +# Load balancer health check will be using LR IP as a source of
> > the health check.
> > +check ovn-nbctl set load_balancer . ip_port_mappings:10.0.0.3=sw0-
> > p1:10.0.0.1
> > +check ovn-nbctl set load_balancer . ip_port_mappings:10.0.0.4=sw0-
> > p2:10.0.0.1
> > +
> > +check_uuid ovn-nbctl --id=@hc create \
> > +Load_Balancer_Health_Check vip="172.16.0.42\:80" -- add
> > Load_Balancer . \
> > +health_check @hc
> > +
> > +# LB is bound only to the LR. It shouldn't be necessary to bind it
> > to the
> > +# LS as well.
> > +check ovn-nbctl lr-lb-add lr0 lb1
> > +
> > +OVN_POPULATE_ARP
> > +check ovn-nbctl --wait=hv sync
> > +
> > +ADD_NAMESPACES(sw0-p1)
> > +ADD_VETH(sw0-p1, sw0-p1, br-int, "10.0.0.3/24",
> > "50:54:00:00:00:03", \
> > + "10.0.0.1")
> > +
> > +ADD_NAMESPACES(sw0-p2)
> > +ADD_VETH(sw0-p2, sw0-p2, br-int, "10.0.0.4/24",
> > "50:54:00:00:00:04", \
> > + "10.0.0.1")
> > +
> > +ADD_NAMESPACES(sw-ext-p1)
> > +ADD_VETH(sw-ext-p1, sw-ext-p1, br-int, "172.16.0.2/24",
> > "60:64:00:00:00:02", \
> > + "172.16.0.1")
> > +
> > +wait_row_count Service_Monitor 2 status=offline
> > +
> > +# Start webservers in 'sw0-p1' and 'sw1-p1'.
> > +OVS_START_L7([sw0-p1], [http])
> > +sw0_p1_pid_file=`cat l7_pid_file`
> > +OVS_START_L7([sw0-p2], [http])
> > +
> > +wait_row_count Service_Monitor 2 status=online
> > +
> > +for i in `seq 1 10`; do
> > + NS_CHECK_EXEC([sw-ext-p1], [curl --connect-timeout 5
> > http://172.16.0.42 > curl$i.log 2>&1 ])
> > +done
> > +
> > +# Stop webserver in sw0-p1.
> > +kill `cat $sw0_p1_pid_file`
> > +
> > +wait_row_count Service_Monitor 1 logical_port=sw0-p1
> > status=offline
> > +wait_row_count Service_Monitor 1 logical_port=sw0-p2 status=online
> > +
> > +for i in `seq 11 20`; do
> > + NS_CHECK_EXEC([sw-ext-p1], [curl --connect-timeout 5
> > http://172.16.0.42 > curl$i.log 2>&1 ])
> > +done
> > +
> > +# Trigger port binding release and check if status changed to
> > offline.
> > +check ovs-vsctl remove interface ovs-sw0-p2 external_ids iface-id
> > +wait_row_count Service_Monitor 2
> > +wait_row_count Service_Monitor 2 status=offline
> > +
> > +check ovs-vsctl set interface ovs-sw0-p2 external_ids:iface-
> > id=sw0-p2
> > +wait_row_count Service_Monitor 2
> > +wait_row_count Service_Monitor 1 status=online
> > +
> > +# Remove TCP Load Balancer.
> > +check ovn-nbctl lb-del lb1
> > +
> > +# Create UDP load balancer.
> > +check ovn-nbctl lb-add lb2 172.16.0.42:53 10.0.0.3:53,10.0.0.4:53
> > udp
> > +lb_udp=`ovn-nbctl lb-list | grep udp | awk '{print $1}'`
> > +
> > +check ovn-nbctl set load_balancer $lb_udp
> > ip_port_mappings:10.0.0.3=sw0-p1:10.0.0.1
> > +check ovn-nbctl set load_balancer $lb_udp
> > ip_port_mappings:10.0.0.4=sw0-p2:10.0.0.1
> > +
> > +check_uuid ovn-nbctl --id=@hc create \
> > +Load_Balancer_Health_Check vip="172.16.0.42\:53" -- add
> > Load_Balancer $lb_udp \
> > +health_check @hc
> > +
> > +check ovn-nbctl --wait=hv lr-lb-add lr0 lb2
> > +
> > +# Wait until udp service_monitor are set to offline.
> > +wait_row_count Service_Monitor 2 status=offline protocol=udp
> > +
> > +# We need to use "-k" argument because we keep receiving health
> > check probes.
> > +NETNS_DAEMONIZE([sw0-p1], [nc -4 -k -l -u -p 53 -d 0.1 -c
> > '/bin/cat'], [udp1.pid])
> > +NETNS_DAEMONIZE([sw0-p2], [nc -4 -k -l -u -p 53 -d 0.1 -c
> > '/bin/cat'], [udp2.pid])
> > +
> > +# UDP nc with "-k" is bit tricky to cleanup, especially on test
> > failures, due to its
> > +# child processes keeping the listening port open. If they are not
> > cleaned up properly,
> > +# the whole test cleanup hangs.
>
> Is this something we need to fix in other tests too? Outside of this
> patch I mean.
I briefly looked and I didn't find any other tests that would use `nc`
"udp server" with the option `-k`. So it doesn't seem like we need to
add this to any other test.
Usually you can get away without using the '-k' if you want to test
just a single connection, or if you can re-use source port on the
client side for multiple "connections".
However in this tests, the "server" gets probed repeatedly by the
controller's service check. Without the '-k' option, only the first
probe would show success, and following probes would fail.
Martin.
>
> > +udp1_pid=$(cat udp1.pid)
> > +udp2_pid=$(cat udp2.pid)
> > +echo "pkill -P $udp1_pid && kill $udp1_pid" >> cleanup
> > +echo "pkill -P $udp2_pid && kill $udp2_pid" >> cleanup
> > +:> udp1.pid
> > +:> udp2.pid
> > +
> > +wait_row_count Service_Monitor 2 status=online
> > +
> > +NETNS_DAEMONIZE(sw-ext-p1, [nc -4 -u 172.16.0.42 53 -d 0.1 -c 'for
> > i in $(seq 1 5); do echo "udp_data $i"; read msg; echo "$msg"
> > >>./udp_data_1; done'], [udp1_client.pid])
> > +
> > +OVS_WAIT_FOR_OUTPUT([cat ./udp_data_1], [0], [dnl
> > +udp_data 1
> > +udp_data 2
> > +udp_data 3
> > +udp_data 4
> > +udp_data 5
> > +])
> > +
> > +# Stop UDP service in sw0-p1.
> > +# Due to the way UDP nc with "-k" forks, we need to kill all
> > +# children processes and then then the parent.
> > +NS_CHECK_EXEC([sw0-p1], [pkill -P $udp1_pid && kill $udp1_pid])
> > +
> > +wait_row_count Service_Monitor 1 logical_port=sw0-p1
> > status=offline
> > +wait_row_count Service_Monitor 1 logical_port=sw0-p2 status=online
> > +
> > +NETNS_DAEMONIZE(sw-ext-p1, [nc -4 -u 172.16.0.42 53 -d 0.1 -c 'for
> > i in $(seq 1 5); do echo "udp_data $i"; read msg; echo "$msg"
> > >>./udp_data_2; done'], [udp2_client.pid])
> > +
> > +OVS_WAIT_FOR_OUTPUT([cat ./udp_data_2], [0], [dnl
> > +udp_data 1
> > +udp_data 2
> > +udp_data 3
> > +udp_data 4
> > +udp_data 5
> > +])
> > +
> > +OVN_CLEANUP_CONTROLLER([hv1])
> > +
> > +OVN_CLEANUP_NORTHD
> > +
> > +as
> > +OVS_TRAFFIC_VSWITCHD_STOP(["/failed to query port patch-.*/d
> > +/connection dropped.*/d
> > +/Service monitor not found.*/d
> > +/handle service check: Unsupported protocol*/d"])
> > +
> > +AT_CLEANUP
> > +])
> > +
> > +OVN_FOR_EACH_NORTHD([
> > +AT_SETUP([Load balancer health checks with LRP IP - IPv6])
> > +AT_KEYWORDS([lb])
> > +
> > +# This is a variant of a 'Load balancer health checks - IPv6' test
> > +# that uses LRP IPs as a source IP for the health checks. As such,
> > it
> > +# doesn't go so deep into testing and focuses only on "health
> > checks"
> > +# having proper status and the backend services being reachable.
> > +#
> > +# Topology:
> > +#
> > +# LB VIP: fd20::42
> > +# Backends: fd11::3 fd11::4
> > +#
> > +#
> > ::3 +--------------+
> > +# +------------
> > + +------ | sw0-p1 |
> > +# ::42 |
> > | | +--------------+
> > +# +--------------+ ::2 ::1 | |
> > ::1 |
> > +# | sw-ext-p1 | ------- sw-ext ------- | lr0 | -------
> > sw0 fd11::/64
> > +# +--------------+ |
> > | |
> > +# fd20::/64 |
> > | | ::4 +--------------+
> > +# +------------
> > + +------ | sw0-p2 |
> > +#
> > +--------------+
> > +#
> > +ovn_start
> > +
> > +OVS_TRAFFIC_VSWITCHD_START()
> > +ADD_BR([br-int])
> > +
> > +# 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
> > +
> > +# Start ovn-controller.
> > +start_daemon ovn-controller
> > +
> > +# Create internal network with LB backends.
> > +check ovn-nbctl ls-add sw0
> > +
> > +check ovn-nbctl lsp-add sw0 sw0-p1
> > +check ovn-nbctl lsp-set-addresses sw0-p1 "50:54:00:00:00:03
> > fd11::3"
> > +
> > +check ovn-nbctl lsp-add sw0 sw0-p2
> > +check ovn-nbctl lsp-set-addresses sw0-p2 "50:54:00:00:00:04
> > fd11::4"
> > +
> > +# Create external network with LB VIP.
> > +check ovn-nbctl ls-add sw-ext
> > +
> > +check ovn-nbctl lsp-add sw-ext sw-ext-p1
> > +check ovn-nbctl lsp-set-addresses sw-ext-p1 "60:64:00:00:00:02
> > fd20::2"
> > +
> > +# Create a logical router and attach both logical switches.
> > +check ovn-nbctl lr-add lr0
> > +check ovn-nbctl set logical_router lr0 options:chassis=hv1
> > +check ovn-nbctl lrp-add lr0 lr0-sw0 00:00:00:00:ff:01 fd11::1/64
> > +check ovn-nbctl lrp-add lr0 lr0-sw-ext 00:00:00:00:ee:01
> > fd20::1/64
> > +check ovn-nbctl lsp-add-router-port sw0 sw0-lr0 lr0-sw0
> > +check ovn-nbctl lsp-add-router-port sw-ext sw-ext-lr0 lr0-sw-ext
> > +
> > +check ovn-nbctl lb-add lb1 [[fd20::42]]:80
> > [[fd11::3]]:80,[[fd11::4]]:80
> > +
> > +# Load balancer health check will be using LR IP as a source of
> > the health check.
> > +check ovn-nbctl set load_balancer .
> > ip_port_mappings:\"[[fd11::3]]\"=\"sw0-p1:[[fd11::1]]\"
> > +check ovn-nbctl set load_balancer .
> > ip_port_mappings:\"[[fd11::4]]\"=\"sw0-p2:[[fd11::1]]\"
> > +
> > +check_uuid ovn-nbctl --id=@hc create \
> > +Load_Balancer_Health_Check vip="\[\[fd20\:\:42\]\]\:80" -- add
> > Load_Balancer . \
> > +health_check @hc
> > +
> > +# LB is bound only to the LR. It shouldn't be necessary to bind it
> > to the
> > +# LS as well.
> > +check ovn-nbctl lr-lb-add lr0 lb1
> > +
> > +OVN_POPULATE_ARP
> > +check ovn-nbctl --wait=hv sync
> > +
> > +ADD_NAMESPACES(sw0-p1)
> > +ADD_VETH(sw0-p1, sw0-p1, br-int, "fd11::3/64",
> > "50:54:00:00:00:03", \
> > + "fd11::1")
> > +
> > +ADD_NAMESPACES(sw0-p2)
> > +ADD_VETH(sw0-p2, sw0-p2, br-int, "fd11::4/64",
> > "50:54:00:00:00:04", \
> > + "fd11::1")
> > +
> > +ADD_NAMESPACES(sw-ext-p1)
> > +ADD_VETH(sw-ext-p1, sw-ext-p1, br-int, "fd20::2/64",
> > "60:64:00:00:00:02", \
> > + "fd20::1")
> > +
> > +wait_row_count Service_Monitor 2 status=offline
> > +
> > +# Start webservers in 'sw0-p1' and 'sw0-p2'.
> > +OVS_START_L7([sw0-p1], [http6])
> > +sw0_p1_pid_file=`cat l7_pid_file`
> > +OVS_START_L7([sw0-p2], [http6])
> > +
> > +wait_row_count Service_Monitor 2 status=online
> > +
> > +for i in `seq 1 10`; do
> > + NS_CHECK_EXEC([sw-ext-p1], [curl --connect-timeout 5
> > http://[[fd20::42]] > curl$i.log 2>&1 ])
> > +done
> > +
> > +# Stop webserver in sw0-p1.
> > +kill `cat $sw0_p1_pid_file`
> > +
> > +wait_row_count Service_Monitor 1 logical_port=sw0-p1
> > status=offline
> > +wait_row_count Service_Monitor 1 logical_port=sw0-p2 status=online
> > +
> > +for i in `seq 11 20`; do
> > + NS_CHECK_EXEC([sw-ext-p1], [curl --connect-timeout 5
> > http://[[fd20::42]] > curl$i.log 2>&1 ])
> > +done
> > +
> > +# Trigger port binding release and check if status changed to
> > offline.
> > +check ovs-vsctl remove interface ovs-sw0-p2 external_ids iface-id
> > +wait_row_count Service_Monitor 2
> > +wait_row_count Service_Monitor 2 status=offline
> > +
> > +check ovs-vsctl set interface ovs-sw0-p2 external_ids:iface-
> > id=sw0-p2
> > +wait_row_count Service_Monitor 2
> > +wait_row_count Service_Monitor 1 status=online
> > +
> > +# Remove TCP Load Balancer.
> > +check ovn-nbctl lb-del lb1
> > +
> > +# Create udp load balancer.
> > +check ovn-nbctl lb-add lb2 [[fd20::42]]:53
> > [[fd11::3]]:53,[[fd11::4]]:53 udp
> > +lb_udp=`ovn-nbctl lb-list | grep udp | awk '{print $1}'`
> > +
> > +check ovn-nbctl set load_balancer $lb_udp
> > ip_port_mappings:\"[[fd11::3]]\"=\"sw0-p1:[[fd11::1]]\"
> > +check ovn-nbctl set load_balancer $lb_udp
> > ip_port_mappings:\"[[fd11::4]]\"=\"sw0-p2:[[fd11::1]]\"
> > +
> > +check_uuid ovn-nbctl --id=@hc create \
> > +Load_Balancer_Health_Check vip="\[\[fd20\:\:42\]\]\:53" -- add
> > Load_Balancer $lb_udp \
> > +health_check @hc
> > +
> > +check ovn-nbctl --wait=hv lr-lb-add lr0 lb2
> > +
> > +# Wait until udp service_monitor are set to offline.
> > +wait_row_count Service_Monitor 2 status=offline protocol=udp
> > +
> > +# We need to use "-k" argument because we keep receiving health
> > check probes.
> > +NETNS_DAEMONIZE([sw0-p1], [nc -6 -k -l -u -p 53 -d 0.1 -c
> > '/bin/cat'], [udp1.pid])
> > +NETNS_DAEMONIZE([sw0-p2], [nc -6 -k -l -u -p 53 -d 0.1 -c
> > '/bin/cat'], [udp2.pid])
> > +
> > +# UDP nc with "-k" is bit tricky to cleanup, especially on test
> > failures, due to its
> > +# child processes keeping the listening port open. If they are not
> > cleaned up properly,
> > +# the whole test cleanup hangs.
> > +udp1_pid=$(cat udp1.pid)
> > +udp2_pid=$(cat udp2.pid)
> > +echo "pkill -P $udp1_pid && kill $udp1_pid" >> cleanup
> > +echo "pkill -P $udp2_pid && kill $udp2_pid" >> cleanup
> > +:> udp1.pid
> > +:> udp2.pid
> > +
> > +wait_row_count Service_Monitor 2 status=online
> > +
> > +NETNS_DAEMONIZE(sw-ext-p1, [nc -6 -u fd20::42 53 -d 0.1 -c 'for i
> > in $(seq 1 5); do echo "udp_data $i"; read msg; echo "$msg"
> > >>./udp_data_1; done'], [udp1_client.pid])
> > +
> > +OVS_WAIT_FOR_OUTPUT([cat ./udp_data_1], [0], [dnl
> > +udp_data 1
> > +udp_data 2
> > +udp_data 3
> > +udp_data 4
> > +udp_data 5
> > +])
> > +
> > +# Stop UDP service in sw0-p1.
> > +# Due to the way UDP nc with "-k" forks, we need to kill all
> > +# children processes and then then the parent.
> > +NS_CHECK_EXEC([sw0-p1], [pkill -P $udp1_pid && kill $udp1_pid])
> > +
> > +wait_row_count Service_Monitor 1 logical_port=sw0-p1
> > status=offline
> > +wait_row_count Service_Monitor 1 logical_port=sw0-p2 status=online
> > +
> > +NETNS_DAEMONIZE(sw-ext-p1, [nc -6 -u fd20::42 53 -d 0.1 -c 'for i
> > in $(seq 1 5); do echo "udp_data $i"; read msg; echo "$msg"
> > >>./udp_data_2; done'], [udp2_client.pid])
> > +
> > +OVS_WAIT_FOR_OUTPUT([cat ./udp_data_2], [0], [dnl
> > +udp_data 1
> > +udp_data 2
> > +udp_data 3
> > +udp_data 4
> > +udp_data 5
> > +])
> > +
> > +OVN_CLEANUP_CONTROLLER([hv1])
> > +
> > +OVN_CLEANUP_NORTHD
> > +
> > +as
> > +OVS_TRAFFIC_VSWITCHD_STOP(["/failed to query port patch-.*/d
> > +/connection dropped.*/d
> > +/Service monitor not found.*/d
> > +/handle service check: Unsupported protocol*/d"])
> > +
> > +AT_CLEANUP
> > +])
> > +
> > OVN_FOR_EACH_NORTHD([
> > AT_SETUP([Load balancer health checks - IPv4])
> > AT_KEYWORDS([lb])
>
> Regards,
> Dumitru
_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
