Troubleshooting EVPN
This section provides various commands to help you examine your EVPN configuration and provides troubleshooting tips.
General Commands
You can use various NVUE or Linux commands to examine interfaces, VLAN mappings and the bridge MAC forwarding database known to the Linux kernel. You can also use these commands to examine the neighbor cache and the routing table (for the underlay or for a specific tenant VRF). Some of the key commands are:
ip [-d] link show type vxlan
(Linux)nv show bridge domain <domain> mac-table
(NVUE) orbridge [-s] fdb show
(Linux)nv show bridge domain <domain> vlan
(NVUE) orbridge vlan show
(Linux)ip neighbor show
(Linux)ip route show [table <vrf-name>]
(Linux)
The sample output below shows ip -d link show type vxlan
command output for one VXLAN interface. Relevant parameters are the VNI value, the state, the local IP address for the VXLAN tunnel, the UDP port number (4789) and the bridge of which the interface is part (bridge in the example below). The output also shows that MAC learning is off on the VXLAN interface.
cumulus@leaf01:~$ ip -d link show type vxlan
14: vni10: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 9216 qdisc noqueue master bridge state UP mode DEFAULT group default qlen 1000
link/ether 42:83:73:20:46:ba brd ff:ff:ff:ff:ff:ff promiscuity 1 minmtu 68 maxmtu 65535
vxlan id 10 local 10.0.1.1 srcport 0 0 dstport 4789 nolearning ttl 64 ageing 300 udpcsum noudp6zerocsumtx noudp6zerocsumrx
bridge_slave state forwarding priority 8 cost 100 hairpin off guard off root_block off fastleave off learning off flood on port_id 0x8005 port_no 0x5 designated_port 32773 designated_cost 0 designated_bridge 8000.76:ed:2a:8a:67:24 designated_root 8000.76:ed:2a:8a:67:24 hold_timer 0.00 message_age_timer 0.00 forward_delay_timer 0.00 topology_change_ack 0 config_pending 0 proxy_arp off proxy_arp_wifi off mcast_router 1 mcast_fast_leave off mcast_flood on neigh_suppress on group_fwd_mask 0x0 group_fwd_mask_str 0x0 group_fwd_maskhi 0x0 group_fwd_maskhi_str 0x0 vlan_tunnel off isolated off addrgenmode eui64 numtxqueues 1 numrxqueues 1 gso_max_size 65536 gso_max_segs 65535
...
The following shows example output for the nv show bridge domain <domain> mac-table
command:
Examine Remote Router MAC Addresses
To examine the router MAC addresses corresponding to all remote VTEPs for symmetric routing, run the NVUE nv show vrf <vrf> evpn remote-router-mac
command or the vtysh show evpn rmac vni all
command. This command is only relevant for a layer 3 VNI:
cumulus@border01:mgmt:~$ nv show vrf RED evpn remote-router-mac
MAC address remote-vtep
----------------- -----------
44:38:39:22:01:7a 10.10.10.1
44:38:39:22:01:7c 10.10.10.64
44:38:39:22:01:8a 10.10.10.4
44:38:39:22:01:78 10.10.10.2
44:38:39:22:01:84 10.10.10.3
44:38:39:be:ef:aa 10.0.1.12
Examine Gateway Next Hops
To examine the gateway next hops for symmetric routing, run the NVUE nv show vrf <vrf> evpn nexthop-vtep
command or the vtysh show evpn next-hops vni all
command. This command is only relevant for a layer 3 VNI. The gateway next hop IP addresses correspond to the remote VTEP IP addresses. Cumulus Linux installs the remote host and prefix routes using these next hops.
cumulus@border01:mgmt:~$ nv show vrf RED evpn nexthop-vtep
Nexthop router-mac
----------- -----------------
10.0.1.12 44:38:39:be:ef:aa
10.10.10.1 44:38:39:22:01:7a
10.10.10.2 44:38:39:22:01:78
10.10.10.3 44:38:39:22:01:84
10.10.10.4 44:38:39:22:01:8a
10.10.10.64 44:38:39:22:01:7c
To show the router MAC address for a specific next hop, run the NVUE nv show vrf <vrf> evpn nexthop-vtep <ip-address>
command:
cumulus@leaf01:mgmt:~$ nv show vrf RED evpn nexthop-vtep 10.10.10.2
operational applied
---------- ----------------- -------
router-mac 44:38:39:22:01:78
To show the remote host and prefix routes through a specific next hop, run the vtysh show evpn next-hops vni <vni> ip <ip-address>
command:
cumulus@leaf01:mgmt:~$ sudo vtysh
...
leaf01# show evpn next-hops vni 4001 ip 10.0.1.2
Ip: 10.0.1.2
RMAC: 44:38:39:be:ef:bb
Refcount: 2
Prefixes:
10.1.10.104/32
10.1.20.105/32
To show the VTEP IP addresses for the next hop groups, run the nv show evpn l2-nhg vtep-ip
command.
Show Access VLANs
To show access VLANs on the switch and their corresponding VNI, run the NVUE nv show evpn access-vlan-info vlan
command or the vtysh show evpn access-vlan
command.
cumulus@border01:mgmt:~$ nv show evpn access-vlan-info vlan
Vlan-id member-interface-count vni vni-count vxlan-interface Summary
------- ---------------------- --- --------- --------------- --------------------------
1 2 member-interface: bond3
member-interface: peerlink
101 2 member-interface: bond3
member-interface: peerlink
102 2 member-interface: bond3
member-interface: peerlink
4024 1 vxlan48
4036 1 vxlan48
You can drill down and show information about a specific vlan with the nv show evpn access-vlan-info vlan <vlan>
command.
Show the VRF Routing Table in FRR
Run the vtysh show ip route vrf <vrf-name>
command or the net show route vrf <vrf-name>
command to examine the VRF routing table. Use this command for symmetric routing to verify that remote host and prefix routes are in the VRF routing table and point to the appropriate gateway next hop.
cumulus@leaf01:mgmt:~$ sudo vtysh
...
leaf01# show ip route vrf RED
show ip route vrf RED
======================
Codes: K - kernel route, C - connected, S - static, R - RIP,
O - OSPF, I - IS-IS, B - BGP, E - EIGRP, N - NHRP,
T - Table, v - VNC, V - VNC-Direct, A - Babel, D - SHARP,
F - PBR, f - OpenFabric,
> - selected route, * - FIB route, q - queued route, r - rejected route
VRF RED:
K>* 0.0.0.0/0 [255/8192] unreachable (ICMP unreachable), 00:53:46
C * 10.1.10.0/24 [0/1024] is directly connected, vlan10-v0, 00:53:46
C>* 10.1.10.0/24 is directly connected, vlan10, 00:53:46
B>* 10.1.10.104/32 [20/0] via 10.0.1.2, vlan4001 onlink, weight 1, 00:43:55
C * 10.1.20.0/24 [0/1024] is directly connected, vlan20-v0, 00:53:46
C>* 10.1.20.0/24 is directly connected, vlan20, 00:53:46
B>* 10.1.20.105/32 [20/0] via 10.0.1.2, vlan4001 onlink, weight 1, 00:20:07
...
In the output above, EVPN specifies the next hops for these routes to be onlink, or reachable over the specified SVI. This is necessary because this interface does not need to have an IP address. Even if the interface has an IP address, the next hop is not on the same subnet as it is typically the IP address of the remote VTEP (part of the underlay IP network).
Show the Global BGP EVPN Routing Table
Run the vtysh show bgp l2vpn evpn route
command or the net show bgp l2vpn evpn route
command to display all EVPN routes, both local and remote. Cumulus Linux bases the routes on the RD as they are across VNIs and VRFs:
cumulus@leaf01:mgmt:~$ sudo vtysh
...
leaf01# show bgp l2vpn evpn route
BGP table version is 6, local router ID is 10.10.10.1
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal
Origin codes: i - IGP, e - EGP, ? - incomplete
EVPN type-1 prefix: [1]:[ESI]:[EthTag]:[IPlen]:[VTEP-IP]
EVPN type-2 prefix: [2]:[EthTag]:[MAClen]:[MAC]:[IPlen]:[IP]
EVPN type-3 prefix: [3]:[EthTag]:[IPlen]:[OrigIP]
EVPN type-4 prefix: [4]:[ESI]:[IPlen]:[OrigIP]
EVPN type-5 prefix: [5]:[EthTag]:[IPlen]:[IP]
Network Next Hop Metric LocPrf Weight Path
Extended Community
Route Distinguisher: 10.10.10.1:3
*> [2]:[0]:[48]:[00:60:08:69:97:ef]
10.0.1.1 32768 i
ET:8 RT:65101:10 RT:65101:4001 Rmac:44:38:39:be:ef:aa
*> [2]:[0]:[48]:[26:76:e6:93:32:78]
10.0.1.1 32768 i
ET:8 RT:65101:10 RT:65101:4001 Rmac:44:38:39:be:ef:aa
*> [2]:[0]:[48]:[26:76:e6:93:32:78]:[32]:[10.1.10.101]
10.0.1.1 32768 i
ET:8 RT:65101:10 RT:65101:4001 Rmac:44:38:39:be:ef:aa
*> [2]:[0]:[48]:[26:76:e6:93:32:78]:[128]:[fe80::9465:45ff:fe6d:4890]
10.0.1.1 32768 i
ET:8 RT:65101:10
*> [2]:[0]:[48]:[c0:8a:e6:03:96:d0]
10.0.1.1 32768 i
ET:8 RT:65101:10 RT:65101:4001 MM:0, sticky MAC Rmac:44:38:39:be:ef:aa
*> [3]:[0]:[32]:[10.0.1.1]
10.0.1.1 32768 i
ET:8 RT:65101:10
Route Distinguisher: 10.10.10.1:4
*> [2]:[0]:[48]:[c0:8a:e6:03:96:d0]
10.0.1.1 32768 i
ET:8 RT:65101:20 RT:65101:4001 MM:0, sticky MAC Rmac:44:38:39:be:ef:aa
*> [2]:[0]:[48]:[cc:6e:fa:8d:ff:92]
10.0.1.1 32768 i
ET:8 RT:65101:20 RT:65101:4001 Rmac:44:38:39:be:ef:aa
*> [2]:[0]:[48]:[f0:9d:d0:59:60:5d]
10.0.1.1 32768 i
ET:8 RT:65101:20 RT:65101:4001 Rmac:44:38:39:be:ef:aa
*> [2]:[0]:[48]:[f0:9d:d0:59:60:5d]:[128]:[fe80::ce6e:faff:fe8d:ff92]
10.0.1.1 32768 i
ET:8 RT:65101:20
*> [3]:[0]:[32]:[10.0.1.1]
10.0.1.1 32768 i
ET:8 RT:65101:20
Route Distinguisher: 10.10.10.1:6
*> [2]:[0]:[48]:[c0:8a:e6:03:96:d0]
10.0.1.1 32768 i
ET:8 RT:65101:30 RT:65101:4002 MM:0, sticky MAC Rmac:44:38:39:be:ef:aa
*> [2]:[0]:[48]:[de:02:3b:17:c9:6d]
10.0.1.1 32768 i
ET:8 RT:65101:30 RT:65101:4002 Rmac:44:38:39:be:ef:aa
*> [2]:[0]:[48]:[de:02:3b:17:c9:6d]:[128]:[fe80::dc02:3bff:fe17:c96d]
10.0.1.1 32768 i
ET:8 RT:65101:30
*> [2]:[0]:[48]:[ea:77:bb:f1:a7:ca]
10.0.1.1 32768 i
ET:8 RT:65101:30 RT:65101:4002 Rmac:44:38:39:be:ef:aa
*> [3]:[0]:[32]:[10.0.1.1]
10.0.1.1 32768 i
ET:8 RT:65101:30
Route Distinguisher: 10.10.10.3:3
*> [2]:[0]:[48]:[12:15:9a:9c:f2:e1]
10.0.1.2 0 65199 65102 i
RT:65102:20 RT:65102:4001 ET:8 Rmac:44:38:39:be:ef:bb
* [2]:[0]:[48]:[12:15:9a:9c:f2:e1]
10.0.1.2 0 65199 65102 i
RT:65102:20 RT:65102:4001 ET:8 Rmac:44:38:39:be:ef:bb
...
You can filter the routing table based on EVPN route type. The available options are: ead: EAD (Type-1) route es: Ethernet Segment (type-4) route macip: MAC-IP (Type-2) route multicast: Multicast prefix: An IPv4 or IPv6 prefix
Show a Specific EVPN Route
To drill down on a specific route for more information, run the vtysh show bgp l2vpn evpn route rd <rd-value>
command or the net show bgp l2vpn evpn route rd <rd-value>
command. This command displays all EVPN routes with that RD and with the path attribute details for each path. Additional filtering is possible based on route type or by specifying the MAC and/or IP address. The following example shows the specific MAC/IP route of server05. The output shows that this remote host is behind VTEP 10.10.10.3 and is reachable through four paths; one through each spine switch. This example is from a symmetric routing configuration, so the route shows both the layer 2 VNI (20) and the layer 3 VNI (4001), as well as the EVPN route target attributes corresponding to each and the associated router MAC address.
cumulus@leaf01:mgmt:~$ sudo vtysh
leaf01# show bgp l2vpn evpn route rd 10.10.10.3:3 mac 12:15:9a:9c:f2:e1 ip 10.1.20.105
BGP routing table entry for 10.10.10.3:3:[2]:[0]:[48]:[12:15:9a:9c:f2:e1]:[32]:[10.1.20.105]
Paths: (4 available, best #1)
Advertised to non peer-group peers:
spine01(swp51) spine02(swp52) spine03(swp53) spine04(swp54)
Route [2]:[0]:[48]:[12:15:9a:9c:f2:e1]:[32]:[10.1.20.105] VNI 20/4001
65199 65102
10.0.1.2 from spine01(swp51) (10.10.10.101)
Origin IGP, valid, external, bestpath-from-AS 65199, best (Router ID)
Extended Community: RT:65102:20 RT:65102:4001 ET:8 Rmac:44:38:39:be:ef:bb
Last update: Fri Jan 15 08:16:24 2021
Route [2]:[0]:[48]:[12:15:9a:9c:f2:e1]:[32]:[10.1.20.105] VNI 20/4001
65199 65102
10.0.1.2 from spine04(swp54) (10.10.10.104)
Origin IGP, valid, external
Extended Community: RT:65102:20 RT:65102:4001 ET:8 Rmac:44:38:39:be:ef:bb
Last update: Fri Jan 15 08:16:24 2021
Route [2]:[0]:[48]:[12:15:9a:9c:f2:e1]:[32]:[10.1.20.105] VNI 20/4001
65199 65102
10.0.1.2 from spine02(swp52) (10.10.10.102)
Origin IGP, valid, external
Extended Community: RT:65102:20 RT:65102:4001 ET:8 Rmac:44:38:39:be:ef:bb
Last update: Fri Jan 15 08:16:24 2021
Route [2]:[0]:[48]:[12:15:9a:9c:f2:e1]:[32]:[10.1.20.105] VNI 20/4001
65199 65102
10.0.1.2 from spine03(swp53) (10.10.10.103)
Origin IGP, valid, external
Extended Community: RT:65102:20 RT:65102:4001 ET:8 Rmac:44:38:39:be:ef:bb
Last update: Fri Jan 15 08:16:24 2021
Displayed 4 paths for requested prefix
- Only use global VNIs. Even though the switch exchanges VNI values in the type-2 and type-5 routes, Cumulus Linux does not use the received values when installing the routes into the forwarding plane but uses the local configuration instead. Ensure that the VLAN to VNI mappings and the layer 3 VNI assignment for a tenant VRF are the same throughout the network.
- If the remote host is dual attached, the next hop for the EVPN route is the anycast IP address of the remote MLAG pair when MLAG is active.
Show the VNI EVPN Routing Table
The switch maintains the received EVPN routes in the global EVPN routing table (described above), even if there are no appropriate local VNIs to import them into. For example, a spine maintains the global EVPN routing table even though there are no VNIs present in the table. When local VNIs are present, the switch imports received EVPN routes into the per-VNI routing tables according to the route target attributes. You can examine the per-VNI routing table with the vtysh show bgp vni <vni>
command:
leaf01# show bgp vni 10
BGP table version is 351, local router ID is 10.10.10.1
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal
Origin codes: i - IGP, e - EGP, ? - incomplete
EVPN type-1 prefix: [1]:[ESI]:[EthTag]:[IPlen]:[VTEP-IP]:[Frag-id]
EVPN type-2 prefix: [2]:[EthTag]:[MAClen]:[MAC]:[IPlen]:[IP]
EVPN type-3 prefix: [3]:[EthTag]:[IPlen]:[OrigIP]
EVPN type-4 prefix: [4]:[ESI]:[IPlen]:[OrigIP]
EVPN type-5 prefix: [5]:[EthTag]:[IPlen]:[IP]
Network Next Hop Metric LocPrf Weight Path
*> [2]:[0]:[48]:[44:38:39:00:00:32]:[32]:[10.1.10.101]
10.0.1.12 (leaf01)
32768 i
ET:8 RT:65101:10 RT:65101:4001 Rmac:44:38:39:be:ef:aa
*> [2]:[0]:[48]:[44:38:39:00:00:32]:[128]:[fe80::4638:39ff:fe00:32]
10.0.1.12 (leaf01)
32768 i
ET:8 RT:65101:10
* [2]:[0]:[48]:[44:38:39:00:00:3e]:[128]:[fe80::4638:39ff:fe00:3e]
10.0.1.34 (leaf02)
0 65102 65199 65104 i
RT:65104:10 ET:8
* [2]:[0]:[48]:[44:38:39:00:00:3e]:[128]:[fe80::4638:39ff:fe00:3e]
10.0.1.34 (leaf02)
0 65102 65199 65103 i
RT:65103:10 ET:8
* [2]:[0]:[48]:[44:38:39:00:00:3e]:[128]:[fe80::4638:39ff:fe00:3e]
10.0.1.34 (spine02)
0 65199 65104 i
RT:65104:10 ET:8
* [2]:[0]:[48]:[44:38:39:00:00:3e]:[128]:[fe80::4638:39ff:fe00:3e]
10.0.1.34 (spine02)
0 65199 65103 i
RT:65103:10 ET:8
* [2]:[0]:[48]:[44:38:39:00:00:3e]:[128]:[fe80::4638:39ff:fe00:3e]
10.0.1.34 (spine04)
0 65199 65104 i
RT:65104:10 ET:8
* [2]:[0]:[48]:[44:38:39:00:00:3e]:[128]:[fe80::4638:39ff:fe00:3e]
10.0.1.34 (spine04)
0 65199 65103 i
RT:65103:10 ET:8
* [2]:[0]:[48]:[44:38:39:00:00:3e]:[128]:[fe80::4638:39ff:fe00:3e]
10.0.1.34 (spine03)
0 65199 65104 i
RT:65104:10 ET:8
* [2]:[0]:[48]:[44:38:39:00:00:3e]:[128]:[fe80::4638:39ff:fe00:3e]
10.0.1.34 (spine03)
0 65199 65103 i
RT:65103:10 ET:8
* [2]:[0]:[48]:[44:38:39:00:00:3e]:[128]:[fe80::4638:39ff:fe00:3e]
10.0.1.34 (spine01)
0 65199 65104 i
RT:65104:10 ET:8
*> [2]:[0]:[48]:[44:38:39:00:00:3e]:[128]:[fe80::4638:39ff:fe00:3e]
10.0.1.34 (spine01)
0 65199 65103 i
RT:65103:10 ET:8
...
To display the VNI routing table for all VNIs, run the vtysh show bgp l2vpn evpn route vni all
command.
To view the EVPN RIB with NVUE, run the nv show vrf <vrf> router bgp address-family l2vpn-evpn loc-rib rd <rd> route-type <type> route
command.
Show the VRF BGP Routing Table
For symmetric routing, the switch imports received type-2 and type-5 routes into the VRF routing table (according to address family: IPv4 unicast or IPv6 unicast) based on a match on the route target attributes. To examine the BGP VRF routing table, run the vtysh show bgp vrf <vrf-name> ipv4 unicast
and show bgp vrf <vrf-name> ipv6 unicast
command. You can also run the net show bgp vrf <vrf-name> ipv4 unicast
command or the net show bgp vrf <vrf-name> ipv6 unicast
command.
cumulus@leaf01:mgmt:~$ sudo vtysh
...
leaf01# show bgp vrf RED ipv4 unicast
BGP table version is 2, local router ID is 10.1.20.2, vrf id 24
Default local pref 100, local AS 65101
Status codes: s suppressed, d damped, h history, * valid, > best, = multipath,
i internal, r RIB-failure, S Stale, R Removed
Nexthop codes: @NNN nexthop's vrf id, < announce-nh-self
Origin codes: i - IGP, e - EGP, ? - incomplete
Network Next Hop Metric LocPrf Weight Path
* 10.1.10.104/32 10.0.1.2< 0 65199 65102 i
* 10.0.1.2< 0 65199 65102 i
* 10.0.1.2< 0 65199 65102 i
* 10.0.1.2< 0 65199 65102 i
*> 10.0.1.2< 0 65199 65102 i
* 10.0.1.2< 0 65199 65102 i
* 10.0.1.2< 0 65199 65102 i
* 10.0.1.2< 0 65199 65102 i
* 10.1.20.105/32 10.0.1.2< 0 65199 65102 i
*> 10.0.1.2< 0 65199 65102 i
* 10.0.1.2< 0 65199 65102 i
* 10.0.1.2< 0 65199 65102 i
* 10.0.1.2< 0 65199 65102 i
* 10.0.1.2< 0 65199 65102 i
* 10.0.1.2< 0 65199 65102 i
* 10.0.1.2< 0 65199 65102 i
Displayed 2 routes and 16 total paths
Support for EVPN Neighbor Discovery (ND) Extended Community
In EVPN VXLAN with ARP and ND suppression where you only configure the VTEPs for layer 2, EVPN needs to carry additional information for the attached devices so proxy ND can provide the correct information to attached hosts. Without this information, hosts cannot configure their default routers or lose their existing default router information. Cumulus Linux supports the EVPN Neighbor Discovery (ND) Extended Community with a type field value of 0x06, a subtype field value of 0x08 (ND Extended Community), and a router flag; this enables the switch to determine if a particular IPv6-MAC pair belongs to a host or a router.
The following configurations use the router flag (R-bit):
- Centralized VXLAN routing with a gateway router.
- A layer 2 switch with ARP and ND suppression.
When the MAC/IP (type-2) route contains the IPv6-MAC pair with the R-bit flag, the route belongs to a router. If the R-bit is zero, the route belongs to a host. If the router is in a local LAN segment, the switch implementing the proxy ND function learns of this information by snooping on neighbor advertisement messages for the associated IPv6 address. Other EVPN peers exchange this information by using the ND extended community in BGP updates.
To show that the neighbor table includes the EVPN arp-cache and that the IPv6-MAC entry belongs to a router, run the vtysh show evpn arp-cache vni <vni> ip <address>
command or the net show evpn arp-cache vni <vni> ip <address>
command. For example:
cumulus@leaf01:mgmt:~$ sudo vtysh
...
leaf01# show evpn arp-cache vni 20 ip 10.1.20.105
IP: 10.1.20.105
Type: remote
State: active
MAC: 12:15:9a:9c:f2:e1
Sync-info: -
Remote VTEP: 10.0.1.2
Local Seq: 0 Remote Seq: 0
Examine MAC Moves
The first time a MAC moves from behind one VTEP to behind another, BGP associates a MAC Mobility (MM) extended community attribute of sequence number 1, with the type-2 route for that MAC. From there, each time this MAC moves to a new VTEP, the MM sequence number increments by 1. You can examine the MM sequence number associated with a MAC’s type-2 route with the vtysh show bgp l2vpn evpn route vni <vni> mac <mac>
command or the net show bgp l2vpn evpn route vni <vni> mac <mac>
command. The example output below shows the type-2 route for a MAC that has moved three times:
cumulus@switch:~$ sudo vtysh
...
switch# show bgp l2vpn evpn route vni 10109 mac 00:02:22:22:22:02
BGP routing table entry for [2]:[0]:[0]:[48]:[00:02:22:22:22:02]
Paths: (1 available, best #1)
Not advertised to any peer
Route [2]:[0]:[0]:[48]:[00:02:22:22:22:02] VNI 10109
Local
6.0.0.184 from 0.0.0.0 (6.0.0.184)
Origin IGP, localpref 100, weight 32768, valid, sourced, local, bestpath-from-AS Local, best
Extended Community: RT:650184:10109 ET:8 MM:3
AddPath ID: RX 0, TX 10350121
Last update: Tue Feb 14 18:40:37 2017
Displayed 1 paths for requested prefix
Examine Static MAC Addresses
You can identify static or sticky MACs in EVPN by the presence of MM:0, sticky MAC
in the Extended Community line of the output from the vtysh show bgp l2vpn evpn route vni <vni> mac <mac>
command or the net show bgp l2vpn evpn route vni <vni> mac <mac>
command.
cumulus@switch:~$ sudo vtysh
...
switch# show bgp l2vpn evpn route vni 10101 mac 00:02:00:00:00:01
BGP routing table entry for [2]:[0]:[0]:[48]:[00:02:00:00:00:01]
Paths: (1 available, best #1)
Not advertised to any peer
Route [2]:[0]:[0]:[48]:[00:02:00:00:00:01] VNI 10101
Local
172.16.130.18 from 0.0.0.0 (172.16.130.18)
Origin IGP, localpref 100, weight 32768, valid, sourced, local, bestpath-from-AS Local, best
Extended Community: ET:8 RT:60176:10101 MM:0, sticky MAC
AddPath ID: RX 0, TX 46
Last update: Tue Apr 11 21:44:02 2017
Displayed 1 paths for requested prefix
Enable FRR Debug Logs
To troubleshoot EVPN, enable FRR debug logs. The relevant debug options are:
Option | Description |
---|---|
debug zebra vxlan | Traces VNI addition and deletion (local and remote) as well as MAC and neighbor addition and deletion (local and remote). |
debug zebra kernel | Traces actual netlink messages exchanged with the kernel, which includes everything, not just EVPN. |
debug bgp updates | Traces BGP update exchanges, including all updates. The output also shows EVPN specific information. |
debug bgp zebra | Traces interactions between BGP and zebra for EVPN (and other) routes. |
ICMP echo Replies and the ping Command
When you run the ping -I
command and specify an interface, you do not receive an ICMP echo reply. However, when you run the ping
command without the -I
option, everything works as expected.
ping -I
command example:
cumulus@switch:default:~:# ping -I swp2 10.0.10.1
PING 10.0.10.1 (10.0.10.1) from 10.0.0.2 swp1.5: 56(84) bytes of data.
ping
command example:
cumulus@switch:default:~:# ping 10.0.10.1
PING 10.0.10.1 (10.0.10.1) 56(84) bytes of data.
64 bytes from 10.0.10.1: icmp_req=1 ttl=63 time=4.00 ms
64 bytes from 10.0.10.1: icmp_req=2 ttl=63 time=0.000 ms
64 bytes from 10.0.10.1: icmp_req=3 ttl=63 time=0.000 ms
64 bytes from 10.0.10.1: icmp_req=4 ttl=63 time=0.000 ms
^C
--- 10.0.10.1 ping statistics ---
4 packets transmitted, 4 received, 0% packet loss, time 3004ms
rtt min/avg/max/mdev = 0.000/1.000/4.001/1.732 ms
When you send an ICMP echo request to an IP address that is not in the same subnet using the ping -I
command, Cumulus Linux creates a failed ARP entry for the destination IP address.
For more information, refer to this article.