BGP 故障排除
使用以下命令来排除 BGP 故障。
显示 BGP 配置摘要
要显示交换机上 BGP 配置的摘要，请运行 NVUE nv show router bgp 命令或 vtysh show ip bgp summary 命令。例如
cumulus@switch:~$ nv show router bgp
applied pending
------------------------------ ----------- -----------
enable on on
autonomous-system 65101 65101
router-id 10.10.10.1 10.10.10.1
policy-update-timer 5 5
graceful-shutdown off off
wait-for-install off off
graceful-restart
mode helper-only helper-only
restart-time 120 120
path-selection-deferral-time 360 360
stale-routes-time 360 360
convergence-wait
time 0 0
establish-wait-time 0 0
queue-limit
input 10000 10000
output 10000 10000
cumulus@switch:~$ sudo vtysh
...
switch# show ip bgp summary
ipv4 Unicast Summary
BGP router identifier 10.10.10.1, local AS number 65101 vrf-id 0
BGP table version 88
RIB entries 25, using 4800 bytes of memory
Peers 5, using 106 KiB of memory
Peer groups 1, using 64 bytes of memory
Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State/PfxRcd
spine01(swp51) 4 65199 31122 31194 0 0 0 1d01h44m 7
spine02(swp52) 4 65199 31060 31151 0 0 0 01:47:13 7
spine03(swp53) 4 65199 31150 31207 0 0 0 01:48:31 7
spine04(swp54) 4 65199 31042 31098 0 0 0 01:46:57 7
leaf02(peerlink.4094) 4 65101 30919 30913 0 0 0 01:47:43 12
Total number of neighbors 5
要查看 BGP 定义的路由表，请运行 vtysh show ip bgp ipv4 unicast 命令。例如
cumulus@leaf01:~$ sudo vtysh
...
leaf01# show ip bgp ipv4 unicast
BGP table version is 88, local router ID is 10.10.10.1, vrf id 0
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
* i10.0.1.1/32 peerlink.4094 0 100 0 ?
*> 0.0.0.0 0 32768 ?
*= 10.0.1.2/32 swp54 0 65199 65102 ?
*= swp52 0 65199 65102 ?
* i peerlink.4094 100 0 65199 65102 ?
*= swp53 0 65199 65102 ?
*> swp51 0 65199 65102 ?
*= 10.0.1.254/32 swp54 0 65199 65132 ?
*= swp52 0 65199 65132 ?
* i peerlink.4094 100 0 65199 65132 ?
*= swp53 0 65199 65132 ?
*> swp51 0 65199 65132 ?
*> 10.10.10.1/32 0.0.0.0 0 32768 ?
*>i10.10.10.2/32 peerlink.4094 0 100 0 ?
*= 10.10.10.3/32 swp54 0 65199 65102 ?
*= swp52 0 65199 65102 ?
* i peerlink.4094 100 0 65199 65102 ?
*= swp53 0 65199 65102 ?
*> swp51 0 65199 65102 ?
...
要显示特定邻居的更详细的细分信息，请运行 vtysh show ip bgp neighbor <neighbor> 命令或 NVUE nv show vrf <vrf> router bgp neighbor <neighbor> 命令
cumulus@switch:~$ nv show vrf default router bgp neighbor swp51
operational applied
----------------------------- ------------------------- ----------
password *
enforce-first-as off
passive-mode off
nexthop-connected-check on
description none
bfd
enable off
ttl-security
enable on off
hops 1
local-as
enable off
timers
keepalive 3 auto
hold 9 auto
connection-retry 10 auto
route-advertisement none auto
address-family
ipv4-unicast
enable on
route-reflector-client off
route-server-client off
soft-reconfiguration off
nexthop-setting auto
add-path-tx off
attribute-mod
aspath off on
med off on
nexthop off on
...
显示 BGP 对等体信息
要显示有关所有 BGP 邻居的详细信息，请运行 nv show vrf <vrf> router bgp neighbor --view=detail -o json 命令
cumulus@leaf01:mgmt:~$ nv show vrf default router bgp neighbor --view=detail -o json
{
"peerlink.4094": {
"address-family": {
"ipv4-unicast": {
"attribute-mod": {
"aspath": "off",
"med": "off",
"nexthop": "off"
},
"capabilities": {
"rx-addpath": "on",
"rx-graceful-restart": "on",
"rx-mpbgp": "on",
"rx-restart-f-bit": "off",
"tx-addpath": "off",
"tx-mpbgp": "on"
},
"community-advertise": {
"extended": "on",
"large": "on",
"regular": "on"
},
"graceful-restart": {
"rx-eof-rib": "on",
"timers": {
"stale-path": {
"timer-sec": 360
}
},
"tx-eof-rib": "on",
"tx-eof-rib-sent-after-update": "off"
},
"prefix-limits": {
"inbound": {
"warning-only": "off"
}
},
"rx-prefix": 10,
"tx-prefix": 11,
"update-group": 1
},
"l2vpn-evpn": {
"attribute-mod": {
"aspath": "off",
"med": "off",
"nexthop": "on"
},
"capabilities": {
"rx-addpath": "on",
"rx-graceful-restart": "on",
"rx-mpbgp": "on",
"rx-restart-f-bit": "off",
"tx-addpath": "off",
"tx-mpbgp": "on"
},
"graceful-restart": {
"rx-eof-rib": "on",
"timers": {
"stale-path": {
"timer-sec": 360
}
},
"tx-eof-rib": "on",
"tx-eof-rib-sent-after-update": "off"
},
"prefix-limits": {
"inbound": {
"warning-only": "off"
}
},
"rx-prefix": 50,
"tx-prefix": 70,
"update-group": 2
}
},
"bgp-version": 4,
"capabilities": {
"extended-nexthop": "on",
"rx-asn32": "off",
"rx-extended-nexthop": "off",
"rx-graceful-restart": "on",
"rx-restart-r-bit": "off",
"rx-route-refresh": "on",
"tx-asn32": "on",
"tx-graceful-restart": "on",
"tx-route-refresh": "on"
...
要查看所有 BGP 对等体的连接信息的摘要，例如状态（已建立、空闲）、正常运行时间、已接收和已发送的消息数以及连接建立的时间，请运行 nv show vrf <vrf> router bgp neighbor 命令。
cumulus@switch:~$ nv show vrf default router bgp neighbor
AS - Remote Autonomous System, PeerEstablishedTime - Peer established time in
UTC format, UpTime - Uptime in milliseconds, Afi-Safi - Address family, PfxSent
- Transmitted prefix counter, PfxRcvd - Recieved prefix counter
Neighbor AS State PeerEstablishedTime UpTime MsgRcvd MsgSent Afi-Safi PfxSent PfxRcvd
------------- ----- ----------- -------------------- -------- ------- ------- ------------ ------- -------
peerlink.4094 65102 established 2025-01-26T15:28:11Z 27073000 561127 473795 ipv4-unicast 11 10
l2vpn-evpn 70 50
swp51 65199 established 2025-01-26T15:28:16Z 27073000 548373 473791 ipv4-unicast 11 8
l2vpn-evpn 70 50
swp52 65199 established 2025-01-26T15:28:19Z 27073000 548377 473789 ipv4-unicast 11 8
l2vpn-evpn 70 50
swp53 idle 27073000 0 0 ipv4-unicast
l2vpn-evpn
swp54 idle 27073000 0 0 ipv4-unicast
l2vpn-evpn
运行 nv show vrf default router bgp neighbor -o json 命令以 json 格式显示所有 BGP 邻居的连接信息的摘要。
在 Cumulus Linux 5.11 及更早版本中，nv show vrf default router bgp neighbor -o json 命令输出显示有关 BGP 对等体的更多详细信息。要在 Cumulus Linux 5.12 及更高版本中显示更详细的 BGP 对等体信息，请运行上面显示的 nv show vrf <vrf> router bgp neighbor --view=detail -o json 命令。
检查 BGP 计时器设置
要检查 BGP 计时器，例如 BGP keepalive 间隔、保持时间和通告间隔，请运行 NVUE nv show vrf <vrf> router bgp neighbor <neighbor> timers 命令或 vtysh show ip bgp neighbor <peer> 命令。例如
cumulus@leaf01:~$ nv show vrf default router bgp neighbor swp51 timers
operational applied
------------------- ----------- -------
keepalive 3 auto
hold 9 auto
connection-retry 10 auto
route-advertisement none auto
BGP 更新组
您可以显示有关更新组事件的信息或有关特定 IPv4 或 IPv6 更新组的信息。
要显示有关更新组事件的信息，请运行 vtysh show bgp update-group 命令或运行以下 NVUE 命令
nv show vrf <vrf-id> router bgp address-family ipv4-unicast update-group用于 IPv4nv show vrf <vrf-id> router bgp address-family ipv6-unicast update-group用于 IPv6
cumulus@leaf01:~$ nv show vrf default router bgp address-family ipv4-unicast update-group
RouteMap - Outbound route map, MinAdvInterval - Minimum route advertisement
interval, CreationTime - Time when the update group was created, LocalAsChange -
LocalAs changes for inbound route, Flags - r - replace-as, x - no-prepend
UpdateGrp RouteMap MinAdvInterval CreationTime LocalAsChange Flags
--------- -------- -------------- -------------------- ------------- -----
1 0 2024-07-08T18:00:57Z
3 0 2024-07-09T20:48:11Z
要显示有关特定更新组的信息，例如对等体刷新事件数、剪枝事件数和数据包队列长度，请运行 vtysh show bgp update-group <group-id> 命令或运行以下 NVUE 命令
nv show vrf <vrf-id> router bgp address-family ipv4-unicast update-group <group-id> -o json用于 IPv4nv show vrf <vrf-id> router bgp address-family ipv6-unicast update-group <group-id> -o json用于 IPv6
cumulus@leaf01:~$ nv show vrf default router bgp address-family ipv4-unicast update-group 1 -o json
{
"create-time": "2024-10-25T14:02:24Z",
"min-route-advertisement-interval": 0,
"sub-group": {
"1": {
"adjacency-count": 13,
"coalesce-time": 1300,
"counters": {
"join-events": 5,
"merge-check-events": 0,
"merge-events": 3,
"peer-refresh-events": 0,
"prune-events": 0,
"split-events": 0,
"switch-events": 0
},
"create-time": "2024-10-25T14:02:24Z",
"needs-refresh": "off",
"neighbor": {
"peerlink.4094": {},
"swp51": {},
"swp52": {},
"swp53": {},
"swp54": {}
},
"packet-counters": {
"queue-hwm-len": 2,
"queue-len": 0,
"queue-total": 14,
"total-enqueued": 14
},
"sub-group-id": 1,
"version": 18
}
},
"update-group-id": "1"
}
显示 BGP 路由信息
您可以运行 NVUE 命令来显示 BGP 邻居的路由统计信息，例如路由数，以及有关通告和接收路由的信息。
要显示 IPv4 路由的路由表，请运行 nv show vrf <vrf> router rib ipv4 route 命令。要显示 IPv6 路由的 RIB 表，请运行 nv show vrf <vrf> router rib ipv6 route 命令。
cumulus@leaf01:mgmt:~$ nv show vrf default router rib ipv4 route
Flags - * - selected, q - queued, o - offloaded, i - installed, S - fib-
selected, x - failed
Route Protocol Distance Uptime NHGId Metric Flags
--------------- --------- -------- -------------------- ----- ------ -----
10.1.10.0/24 connected 0 2024-07-18T21:57:29Z 46 0 *Sio
10.1.20.0/24 connected 0 2024-07-18T21:57:29Z 47 0 *Sio
10.1.30.0/24 connected 0 2024-07-18T21:57:29Z 48 0 *Sio
10.1.40.0/24 bgp 20 2024-07-18T22:02:22Z 57 0 *Si
10.1.50.0/24 bgp 20 2024-07-18T22:02:22Z 57 0 *Si
10.1.60.0/24 bgp 20 2024-07-18T22:02:22Z 57 0 *Si
10.10.10.1/32 connected 0 2024-07-18T21:55:54Z 7 0 *Sio
10.10.10.2/32 bgp 20 2024-07-18T21:57:29Z 34 0 *Si
10.10.10.3/32 bgp 20 2024-07-18T22:02:22Z 57 0 *Si
10.10.10.4/32 bgp 20 2024-07-18T22:02:27Z 57 0 *Si
10.10.10.101/32 bgp 20 2024-07-18T22:01:14Z 50 0 *Si
10.10.10.102/32 bgp 20 2024-07-18T22:02:22Z 58 0 *Si
要显示本地路由表中的路由，请为 IPv4 运行 nv show vrf <vrf> router bgp address-family ipv4-unicast route 命令，为 IPv6 运行 nv show vrf <vrf> router bgp address-family ipv6-unicast route 命令。您还可以运行带有 -o json 的命令以 json 格式显示接收到的路由。
cumulus@leaf02:~$ nv show vrf default router bgp address-family ipv4-unicast route PathCount - Number of paths present for the prefix, MultipathCount - Number of
paths that are part of the ECMP, DestFlags - * - bestpath-exists, w - fib-wait-
for-install, s - fib-suppress, i - fib-installed, x - fib-install-failed
Prefix PathCount MultipathCount DestFlags
--------------- --------- -------------- ---------
10.0.1.12/32 2 1 *
10.0.1.34/32 5 4 *
10.0.1.255/32 5 4 *
10.10.10.1/32 1 1 *
10.10.10.2/32 5 1 *
10.10.10.3/32 5 4 *
10.10.10.4/32 5 4 *
10.10.10.63/32 5 4 *
10.10.10.64/32 5 4 *
10.10.10.101/32 2 1 *
10.10.10.102/32 2 1 *
10.10.10.103/32 2 1 *
10.10.10.104/32 2 1 *
要显示有关本地路由表中的特定路由的信息，请为 IPv4 运行 nv show vrf <vrf> router bgp address-family ipv4-unicast route <route> 命令，为 IPv6 运行 nv show vrf <vrf> router bgp address-family ipv6-unicast route <route> 命令。
上述 IPv4 和 IPv6 命令以简要格式显示本地路由表路由信息，以提高高规模环境的性能。您还可以运行带有 -o json 的命令以 json 格式显示接收到的路由。
cumulus@leaf01:~$ nv show vrf default router bgp address-family ipv4-unicast route 10.10.10.64/32
operational
--------------- -----------
path-count 5
multipath-count 4
path
=======
Origin - Route origin, Local - Locally originated route, Sourced - Sourced
route, Weight - Route weight, Metric - Route metric, LocalPref - Route local
preference, PathFrom - Route path origin, LastUpdate - Route last update,
NexthopCnt - Number of nexthops, Flags - = - multipath, * - bestpath, v - valid,
s - suppressed, R - removed, S - stale
Path Origin Local Sourced Weight Metric LocalPref PathFrom LastUpdate NexthopCnt Flags
---- ---------- ----- ------- ------ ------ --------- -------- -------------------- ---------- -----
1 incomplete external 2024-10-25T14:02:33Z 2 =*v
2 incomplete external 2024-10-25T14:02:42Z 2 =v
3 incomplete external 2024-10-25T14:02:36Z 2 =v
4 incomplete external 2024-10-25T14:02:36Z 2 =v
5 incomplete external 2024-10-25T14:02:33Z 2 *v
advertised-to
================
Neighbor hostname
------------- --------
peerlink.4094 leaf02
swp51 spine01
swp52 spine02
swp53 spine03
swp54 spine04
要显示路由计数，请为 IPv4 运行 nv show vrf <vrf-id> router bgp neighbor <neighbor-id> address-family ipv4-unicast route-counters 命令，为 IPv6 运行 nv show vrf <vrf-id> router bgp neighbor <neighbor-id> address-family ipv6-unicast route-counters 命令。
cumulus@leaf01:~$ nv show vrf default router bgp neighbor swp51 address-family ipv4-unicast route-counters
operational
-------------- -----------
route-count 8
adj-rib-in 0
damped 0
removed 0
history 0
stale 0
valid 8
all-rib 8
routes-counted 8
best-routes 7
usable 8
要显示所有通告的路由，请为 IPv4 运行 nv show vrf <vrf> router bgp neighbor <neighbor> address-family ipv4-unicast advertised-routes 命令，为 IPv6 运行 nv show vrf <vrf>> router bgp neighbor <neighbor> address-family ipv6-unicast advertised-routes 命令。
上述 IPv4 和 IPv6 命令以简要格式显示通告的路由，以提高高规模环境的性能。您还可以运行带有 -o json 的命令以 json 格式显示接收到的路由。
cumulus@leaf01:~$ nv show vrf default router bgp neighbor swp51 address-family ipv4-unicast advertised-routes
PathCount - Number of paths present for the prefix, MultipathCount - Number of
paths that are part of the ECMP
IPv4 Prefix PathCount MultipathCount DestFlags
--------------- --------- -------------- ---------------
10.1.10.0/24 3 1 bestpath-exists
10.1.20.0/24 3 1 bestpath-exists
10.1.30.0/24 3 1 bestpath-exists
10.1.40.0/24 3 2 bestpath-exists
10.1.50.0/24 3 2 bestpath-exists
10.1.60.0/24 3 2 bestpath-exists
10.10.10.1/32 2 1 bestpath-exists
10.10.10.2/32 3 1 bestpath-exists
10.10.10.3/32 3 2 bestpath-exists
10.10.10.4/32 3 2 bestpath-exists
10.10.10.101/32 2 1 bestpath-exists
10.10.10.102/32 2 1 bestpath-exists
要显示有关特定通告路由的信息，请为 IPv4 运行 nv show <vrf> default router bgp neighbor <neighbor> address-family ipv4-unicast advertised-routes <route> 命令，为 IPv6 运行 nv show <vrf> default router bgp neighbor <neighbor> address-family ipv6-unicast advertised-routes <route> 命令。
cumulus@leaf01:~$ nv show vrf default router bgp neighbor swp51 address-family ipv4-unicast advertised-route 10.10.10.1/32
operational
--------------- -----------
path-count 2
multipath-count 1
path
====================
Origin - Route origin, Local - Locally originated route, Sourced - Sourced
route, Weight - Route weight, Metric - Route metric, LocalPref - Route local
preference, PathFrom - Route path origin, LastUpdate - Route last update,
NexthopCnt - Number of nexthops, Flags - = - multipath, * - bestpath, v - valid,
s - suppressed, R - removed, S - stale
Path Origin Local Sourced Weight Metric LocalPref PathFrom LastUpdate NexthopCnt Flags
---- ---------- ----- ------- ------ ------ --------- -------- -------------------- ---------- -----
1 IGP on on 32768 0 2024-07-18T21:55:54Z 1 *v
2 incomplete on 32768 0 2024-07-18T21:55:54Z 1 v
...
要显示所有接收到的路由，请为 IPv4 运行 nv show vrf <vrf> router bgp neighbor <neighbor> address-family ipv4-unicast received-routes 命令，为 IPv6 运行 nv show vrf <vrf> router bgp neighbor <neighbor> address-family ipv6-unicast received-routes 命令。这些命令以简要格式显示接收到的路由，以提高高规模环境的性能。您还可以运行带有 --view=detail 的命令以查看更详细的信息，或带有 -o json 的命令以 json 格式显示接收到的路由。
要显示有关特定接收路由的信息，请为 IPv4 运行 nv show vrf <vrf> router bgp neighbor <neighbor> address-family ipv4-unicast received-routes <route> -o json 命令，为 IPv6 运行 nv show vrf <vrf> router bgp neighbor <neighbor> address-family ipv6-unicast received-routes <route> -o json 命令。
显示下一跳信息
要显示所有 BGP IPv4 或 IPv6 下一跳的摘要，请运行 nv show vrf <vrf> router bgp nexthop ipv4 或 nv show vrf <vrf> router bgp nexthop ipv6 命令。输出显示 IGP 指标、指向下一跳的路径数以及用于到达下一跳的地址或接口。
cumulus@leaf01:mgmt:~$ nv show vrf default router bgp nexthop ipv4
Nexthops
===========
PathCnt - Number of paths pointing to this Nexthop, ResolvedVia - Resolved via
address or interface, Interface - Resolved via interface
Address IGPMetric Valid PathCnt ResolvedVia Interface
----------- --------- ----- ------- ------------------------- -------------
10.0.1.34 0 on 160 fe80::4ab0:2dff:fe60:910e swp54
fe80::4ab0:2dff:fea7:7852 swp53
fe80::4ab0:2dff:fec8:8fb9 swp52
fe80::4ab0:2dff:feff:e147 swp51
10.10.10.2 0 on 15 fe80::4ab0:2dff:fe2d:495c peerlink.4094
10.10.10.3 0 on 15 fe80::4ab0:2dff:fe60:910e swp54
fe80::4ab0:2dff:fea7:7852 swp53
fe80::4ab0:2dff:fec8:8fb9 swp52
fe80::4ab0:2dff:feff:e147 swp51
10.10.10.4 0 on 15 fe80::4ab0:2dff:fe60:910e swp54
fe80::4ab0:2dff:fea7:7852 swp53
fe80::4ab0:2dff:fec8:8fb9 swp52
fe80::4ab0:2dff:feff:e147 swp51
10.10.10.63 0 on 15 fe80::4ab0:2dff:fe60:910e swp54
fe80::4ab0:2dff:fea7:7852 swp53
fe80::4ab0:2dff:fec8:8fb9 swp52
fe80::4ab0:2dff:feff:e147 swp51
10.10.10.64 0 on 15 fe80::4ab0:2dff:fe60:910e swp54
fe80::4ab0:2dff:fea7:7852 swp53
fe80::4ab0:2dff:fec8:8fb9 swp52
fe80::4ab0:2dff:feff:e147 swp51
要显示有关特定下一跳的信息，请为 IPv4 运行 vtysh NVUE nv show vrf <vrf-id> router bgp nexthop ipv4 ip-address <ip-address> 命令，为 IPv6 运行 nv show vrf <vrf-id> router bgp nexthop ipv6 ip-address <ip-address> 命令。您还可以运行 vtysh show bgp vrf default nexthop <ip-address> 命令。
cumulus@leaf01:mgmt:~$ nv show vrf default router bgp nexthop ipv4 ip-address 10.10.10.2
operational
---------------- -------------------------
valid yes
complete on
igp-metric 0
path-count 15
last-update-time 2024-10-25T14:02:32Z
[resolved-via] fe80::4ab0:2dff:fee8:57ba
要显示 BGP 通过哪个地址和接口解析特定下一跳，请为 IPv4 运行 nv show vrf <vrf-id> router bgp nexthop ipv4 ip-address <ip-address-id> resolved-via 命令，为 IPv6 运行 nv show vrf <vrf-id> router bgp nexthop ipv6 ip-address <ip-address-id> resolved-via 命令。
cumulus@leaf01:~$ nv show vrf default router bgp nexthop ipv4 ip-address 10.10.10.2 resolved-via
Nexthop interface
------------------------- ---------
fe80::4ab0:2dff:fe20:ac25 swp51
fe80::4ab0:2dff:fe93:d92d swp52
显示前缀无关收敛信息
当您启用 BGP 前缀无关收敛 (PIC) 时，您可以使用以下命令来显示有关路由源扩展社区 (SOO) 路由和路由的 SOO 对等体位索引映射的信息。
要显示有关所有路由源扩展社区 (SOO) 路由的信息，请运行 nv show <vrf> default router bgp address-family <address-family> soo-route 命令
cumulus@spine01:~$ nv show vrf default router bgp address-family ipv4-unicast soo-route
PathCnt - Number of paths for this SoO., RouteCnt - Number of routes with this
SoO, SoONhgID - Nexthop group id used by this SoO, SoORouteFlag - Indicates
Site-of-Origin route flag - I - Installed, NhgRouteCnt - Number of routes using
SoO NHG, NhgFlag - V - valid, Ip - install-pending, Dp - delete-pending
SoORouteID PathCnt RouteCnt SoONhgID SoORouteFlag NhgRouteCnt NhgFlag
---------- ------- -------- -------- ------------ ----------- -------
10.10.10.1 1 1 70328885 I 1 V
要显示有关特定 SOO 路由的信息，请运行 nv show <vrf> default router bgp address-family <address-family> soo-route 命令
cumulus@spine01:~$ nv show vrf default router bgp address-family ipv4-unicast soo-route 10.10.10.1
operational
------------------------ -------------------------
num-paths 1
nexthop-group-id 70328885
num-routes-with-soo 1
num-routes-using-soo-nhg 1
soo-route-flag I
nhg-flags-string V
nhg-flags
nhg-valid yes
nhg-install-pending no
nhg-delete-pending no
[peer-index] fe80::4ab0:2dff:fe3a:a928
bit-map
selected-path-bitmap 1
installed-path-bitmap 1
[route-with-soo] 10.0.1.12/32
要显示 SOO 路由的 SOO 对等体位索引映射，请运行 nv show <vrf> default router bgp address-family <address-family> soo-route <prefix> peer-index 命令
cumulus@spine01:~$ nv show vrf default router bgp address-family ipv4-unicast soo-route 10.10.10.1 peer-index
SooPeerBitIndex - Indicates Site-of-Origin peer bit index mapping
IPAddress SooPeerBitIndex
------------------------- ---------------
fe80::4ab0:2dff:fe3a:a928 1
要显示特定路由是否使用 PIC，请运行 nv show <vrf> default router bgp address-family <address-family> soo-route <prefix> route-with-soo 命令
cumulus@spine01:~$ nv show vrf default router bgp address-family ipv4-unicast soo-route 10.10.10.1 route-with-soo
UseSooNhg - Indicates this route-with-soo uses SooNhg, SelectedPathBitmap -
Details of Site-of-Origin selected path bitmap
Prefix UseSooNhg SelectedPathBitmap
------------ --------- ------------------
10.0.1.12/32 yes 1
要显示有关所有 SOO 路由的信息，请运行 show bgp vrf <vrf> <address-family> unicast soo route 命令
cumulus@leaf01:~$ sudo vtysh
...
leaf01# show bgp vrf default ipv4 unicast soo route
BGP: VRF default
SoO: 10.10.10.1
NHG:
NHG ID: 70328887
NHG flags: Valid, Installed
SoO route:
Number of paths: 1
Number of Routes with SoO: 1
Number of Routes with SoO using SoO NHG: 1
SoO route flags: Installed
Nexthop cache:
fe80::4ab0:2dff:feec:f8e/128 NH ifidx: 3 type: 5 flags: 5 lbw: 0
Peer BitIndex Mappings:
fe80::4ab0:2dff:feec:f8e: 1
Bitmaps:
Selected path info bitmap bits set: 1
Installed path info bitmap bits set: 1
Route with SoO:
10.0.1.12/32 uses SoO NHG Selected path info bitmap bits set: 1
SoO: 10.10.10.3
NHG:
NHG ID: 70328889
NHG flags: Valid, Installed
SoO route:
Number of paths: 1
Number of Routes with SoO: 1
Number of Routes with SoO using SoO NHG: 1
SoO route flags: Installed
Nexthop cache:
fe80::4ab0:2dff:fef2:6bc5/128 NH ifidx: 5 type: 5 flags: 5 lbw: 0
Peer BitIndex Mappings:
fe80::4ab0:2dff:fef2:6bc5: 3
Bitmaps:
Selected path info bitmap bits set: 3
Installed path info bitmap bits set: 3
Route with SoO:
10.0.1.34/32 uses SoO NHG Selected path info bitmap bits set: 3
SoO: 10.10.10.2
NHG:
NHG ID: 70328888
NHG flags: Valid, Installed
SoO route:
Number of paths: 1
Number of Routes with SoO: 1
Number of Routes with SoO using SoO NHG: 0
SoO route flags: Installed
Nexthop cache:
fe80::4ab0:2dff:fe5d:6456/128 NH ifidx: 4 type: 5 flags: 5 lbw: 0
Peer BitIndex Mappings:
fe80::4ab0:2dff:fe5d:6456: 2
Bitmaps:
Selected path info bitmap bits set: 2
Installed path info bitmap bits set: 2
Route with SoO:
10.0.1.12/32 Selected path info bitmap bits set: 2
要显示有关特定 SOO 路由的信息，请运行 show bgp vrf <vrf> <address-family> unicast soo route <prefix> 命令
cumulus@spine01:~$ sudo vtysh
...
spine01# show bgp vrf default ipv4 unicast soo route 10.10.10.3
BGP: VRF default
SoO: 10.10.10.3
NHG:
NHG ID: 70328889
NHG flags: Valid, Installed
SoO route:
Number of paths: 1
Number of Routes with SoO: 1
Number of Routes with SoO using SoO NHG: 1
SoO route flags: Installed
Nexthop cache:
fe80::4ab0:2dff:fef2:6bc5/128 NH ifidx: 5 type: 5 flags: 5 lbw: 0
Peer BitIndex Mappings:
fe80::4ab0:2dff:fef2:6bc5: 3
Bitmaps:
Selected path info bitmap bits set: 3
Installed path info bitmap bits set: 3
Route with SoO:
10.0.1.34/32 uses SoO NHG Selected path info bitmap bits set: 3
您可以以 json 格式显示上述命令。例如
cumulus@spine01:~$ sudo vtysh
...
spine01# show bgp vrf default ipv4 unicast soo route json
{
"default":[
{
"SoORoute":"10.10.10.1",
"numPaths":1,
"nexthopgroupId":70328887,
"numRoutesWithSoO":1,
"numRoutesWithSoOUsingSoONHG":1,
"nhgValid":true,
"nhgInstallPending":false,
"nhgDeletePending":false,
"SoORouteFlag":"Installed",
"nextHopCache":[
{
"prefix":"fe80::4ab0:2dff:feec:f8e/128",
"ifIndex":3,
"type":5,
"flags":5,
"lbw":0
}
],
"peerBitIndexMapping":[
{
"peerIp":"fe80::4ab0:2dff:feec:f8e",
"bitIndex":1
}
],
"bitMaps":[
{
"selectedPathBitmap":" 1",
"installedPathBitmap":" 1"
}
],
"routeWithSoO":[
{
"prefix":"10.0.1.12/32",
"usesSoONhg":true,
"selectedPathBitmap":" 1"
}
]
},
{
"SoORoute":"10.10.10.3",
"numPaths":1,
"nexthopgroupId":70328889,
"numRoutesWithSoO":1,
"numRoutesWithSoOUsingSoONHG":1,
"nhgValid":true,
"nhgInstallPending":false,
"nhgDeletePending":false,
"SoORouteFlag":"Installed",
"nextHopCache":[
{
"prefix":"fe80::4ab0:2dff:fef2:6bc5/128",
"ifIndex":5,
"type":5,
"flags":5,
"lbw":0
}
],
"peerBitIndexMapping":[
{
"peerIp":"fe80::4ab0:2dff:fef2:6bc5",
"bitIndex":3
}
],
"bitMaps":[
{
"selectedPathBitmap":" 3",
"installedPathBitmap":" 3"
}
],
"routeWithSoO":[
{
"prefix":"10.0.1.34/32",
"usesSoONhg":true,
"selectedPathBitmap":" 3"
}
]
},
{
"SoORoute":"10.10.10.2",
"numPaths":1,
"nexthopgroupId":70328888,
"numRoutesWithSoO":1,
"numRoutesWithSoOUsingSoONHG":0,
"nhgValid":true,
"nhgInstallPending":false,
"nhgDeletePending":false,
"SoORouteFlag":"Installed",
"nextHopCache":[
{
"prefix":"fe80::4ab0:2dff:fe5d:6456/128",
"ifIndex":4,
"type":5,
"flags":5,
"lbw":0
}
],
"peerBitIndexMapping":[
{
"peerIp":"fe80::4ab0:2dff:fe5d:6456",
"bitIndex":2
}
],
"bitMaps":[
{
"selectedPathBitmap":" 2",
"installedPathBitmap":" 2"
}
],
"routeWithSoO":[
{
"prefix":"10.0.1.12/32",
"usesSoONhg":false,
"selectedPathBitmap":" 2"
}
]
}
]
}
show bgp <address-family> unicast 和 show bgp <address-family> unicast <prefix> 命令也显示 SOO 信息
cumulus@spine01:~$ sudo vtysh
...
spine01# show bgp ipv4 unicast 10.10.10.3
BGP routing table entry for 10.10.10.3/32, version 29 SoO:10.10.10.3, multipath soo nhg:70328889
Paths: (2 available, best #1, table default)
Advertised to non peer-group peers:
leaf01(swp1) leaf02(swp2) leaf03(swp3) leaf04(swp4) border01(swp5) border02(swp6)
65103
fe80::4ab0:2dff:fef2:6bc5 (leaf03) from leaf03(swp3) (10.10.10.3)
(fe80::4ab0:2dff:fef2:6bc5) (used)
Origin IGP, metric 0, valid, external, bestpath-from-AS 65103, best (AS Path)
Extended Community: SoO:10.10.10.3:0
Last update: Wed Jan 22 20:12:46 2025
65104 65103
fe80::4ab0:2dff:feab:57dd (leaf04) from leaf04(swp4) (10.10.10.4)
(fe80::4ab0:2dff:feab:57dd) (used)
Origin IGP, valid, external, bestpath-from-AS 65104
Extended Community: SoO:10.10.10.3:0
Last update: Wed Jan 22 20:12:46 2025
排除 BGP Unnumbered 故障
要验证 FRR 是否通过给定接口上的 IPv6 邻居发现路由器通告学习邻居链路本地 IPv6 地址，请运行 vtysh show interface <interface> 命令。
如果您没有在接口的两端都启用 ipv6 nd suppress-ra，则 Neighbor address(s): 会显示另一端的链路本地地址（BGP 在该接口使用 BGP 时使用的地址）。
Cumulus Linux 会自动在具有 IPv6 地址的接口上启用 IPv6 路由通告 (RA)。您无需为 BGP Unnumbered 运行 no ipv6 nd suppress-ra 命令。
cumulus@switch:~$ sudo vtysh
...
switch# show interface swp51
Interface swp51 is up, line protocol is up
Link ups: 0 last: (never)
Link downs: 0 last: (never)
PTM status: disabled
vrf: default
OS Description: leaf to spine
index 8 metric 0 mtu 9216 speed 1000
flags: <UP,BROADCAST,RUNNING,MULTICAST>
Type: Ethernet
HWaddr: 10:d8:68:d4:a6:81
inet6 fe80::12d8:68ff:fed4:a681/6
Interface Type Other
protodown: off
ND advertised reachable time is 0 milliseconds
ND advertised retransmit interval is 0 milliseconds
ND advertised hop-count limit is 64 hops
ND router advertisements sent: 217 rcvd: 216
ND router advertisements are sent every 10 seconds
ND router advertisements lifetime tracks ra-interval
ND router advertisement default router preference is medium
Hosts use stateless autoconfig for addresses.
Neighbor address(s):
inet6 fe80::f208:5fff:fe12:cc8c/128
排除使用 IPv6 下一跳学习的 IPv4 前缀的故障
要显示使用 IPv6 下一跳学习的 IPv4 前缀，请运行以下命令。
以下示例显示了从 BGP 对等体通过使用 IPv6 全局地址的 IPv6 会话学习的 IPv4 前缀，但 BGP 安装的下一跳是链路本地 IPv6 地址。当会话直接在对等体之间时，并且前缀的 BGP 更新同时包含链路本地和全局 IPv6 地址时，会发生这种情况。如果全局和链路本地下一跳都存在，则 BGP 优先使用链路本地地址进行路由安装。
cumulus@spine01:mgmt:~$ sudo vtysh
...
spine01# show ip bgp ipv4 unicast summary
BGP router identifier 10.10.10.101, local AS number 65199 vrf-id 0
BGP table version 3
RIB entries 3, using 576 bytes of memory
Peers 1, using 21 KiB of memory
Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State/PfxRcd
leaf01(2001:db8:2::a00:1) 4 65101 22 22 0 0 0 00:01:00 0
Total number of neighbors 1
cumulus@spine01:mgmt:~$ sudo vtysh
...
spine01# show ip bgp ipv4 unicast
BGP table version is 3, local router ID is 10.10.10.101, vrf id 0
Default local pref 100, local AS 65199
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
10.10.10.101/32 fe80::a00:27ff:fea6:b9fe 0 0 32768 i
Displayed 1 routes and 1 total paths
cumulus@spine01:~$ sudo vtysh
...
spine01# show ip bgp ipv4 unicast 10.10.10.101/32
BGP routing table entry for 10.10.10.101/32
Paths: (1 available, best #1, table default)
Advertised to non peer-group peers:
Leaf01(2001:db8:0002::0a00:1)
3
2001:db8:0002::0a00:1 from Leaf01(2001:db8:0002::0a00:1) (10.10.10.101)
(fe80::a00:27ff:fea6:b9fe) (used)
Origin IGP, metric 0, valid, external, bestpath-from-AS 3, best (First path received)
AddPath ID: RX 0, TX 3
Last update: Mon Oct 22 08:09:22 2018
以下示例输出显示了在 FRR RIB 以及内核 FIB 中安装路由的结果。在 FRR RIB 中安装的下一跳是链路本地 IPv6 地址，Cumulus Linux 会将其转换为 IPv4 链路本地地址，这是内核 FIB 安装所必需的。
cumulus@spine01:~$ sudo vtysh
...
spine01# show ip route 10.10.10.101/32
RIB entry for 10.10.10.101/32
===========================
Routing entry for 10.10.10.101/32
Known via "bgp", distance 20, metric 0, best
Last update 2d17h05m ago
* fe80::a00:27ff:fea6:b9fe, via swp1
FIB entry for 10.10.10.101/32
===========================
10.10.10.101/32 via 10.0.1.0 dev swp1 proto bgp metric 20 onlink
如果 BGP 仅使用 IPv6 全局下一跳地址（当它通过路由反射器学习路由时）学习 IPv4 前缀，则命令输出会将 IPv6 全局地址显示为下一跳值。该命令还显示它通过路由反射器的链路本地地址递归学习。当您使用全局 IPv6 地址作为 FRR RIB 中路由安装的下一跳时，交换机仍会将其转换为 IPv4 链路本地地址以安装到内核中。
cumulus@leaf01:~$ sudo vtysh
...
leaf01# show ip bgp ipv4 unicast summary
BGP router identifier 10.10.10.1, local AS number 65101 vrf-id 0
BGP table version 1
RIB entries 1, using 152 bytes of memory
Peers 1, using 19 KiB of memory
Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State/PfxRcd
Spine01(2001:db8:0002::0a00:2) 4 1 74 68 0 0 0 00:00:45 1
Total number of neighbors 1
cumulus@leaf01:~$ sudo vtysh
...
leaf01# show ip bgp ipv4 unicast summary
BGP table version is 1, local router ID is 10.10.10.1
Status codes: s suppressed, d damped, h history, * valid, > best, = multipath,
i internal, r RIB-failure, S Stale, R Removed
Origin codes: i - IGP, e - EGP, ? - incomplete
Network Next Hop Metric LocPrf Weight Path
*>i10.1.10.0/24 2001:2:2::4 0 100 0 i
Displayed 1 routes and 1 total paths
cumulus@leaf01:~$ sudo vtysh
...
leaf01# show ip bgp ipv4 unicast 10.10.10.101/32
BGP routing table entry for 10.10.10.101/32
Paths: (1 available, best #1, table default)
Not advertised to any peer
Local
2001:2:2::4 from Spine01(2001:1:1::1) (10.10.10.104)
Origin IGP, metric 0, localpref 100, valid, internal, bestpath-from-AS Local, best (First path received)
Originator: 10.0.0.14, Cluster list: 10.10.10.111
AddPath ID: RX 0, TX 5
Last update: Mon Oct 22 14:25:30 2018
cumulus@leaf01:~$ sudo vtysh
...
leaf01# show ip route 10.10.10.1/32
RIB entry for 10.10.10.1/32
===========================
Routing entry for 10.10.10.1/32
Known via "bgp", distance 200, metric 0, best
Last update 00:01:13 ago
2001:2:2::4 (recursive)
* fe80::a00:27ff:fe5a:84ae, via swp1
FIB entry for 10.10.10.1/32
===========================
10.10.10.1/32 via 10.0.1.1 dev swp1 proto bgp metric 20 onlink
要仅使用 IPv6 全局地址进行路由安装到 FRR RIB 中，您必须在相应地址族中的邻居或对等组语句中添加额外的路由映射。当路由映射命令 set ipv6 next-hop prefer-global 应用于邻居时，如果前缀的 BGP 更新中同时存在链路本地和全局 IPv6 地址，则 BGP 使用 IPv6 全局地址进行路由安装。
通过此附加配置，直接邻居情况下的 FRR RIB 中的输出会如下所示更改
router bgp 65101
bgp router-id 10.10.10.1
neighbor 2001:db8:2::a00:1 remote-as internal
neighbor 2001:db8:2::a00:1 capability extended-nexthop
!
address-family ipv4 unicast
neighbor 2001:db8:2::a00:1 route-map GLOBAL in
exit-address-family
!
route-map GLOBAL permit 20
set ipv6 next-hop prefer-global
!
生成的 FRR RIB 输出如下
cumulus@leaf01:~$ sudo vtysh
...
leaf01# show ip route
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,
> - selected route, * - FIB route
B 0.0.0.0/0 [200/0] via 2001:2:2::4, swp2, 00:01:00
K 0.0.0.0/0 [0/0] via 10.0.2.2, eth0, 1d02h29m
C>* 10.0.0.9/32 is directly connected, lo, 5d18h32m
C>* 10.0.2.0/24 is directly connected, eth0, 03:51:31
B>* 172.16.4.0/24 [200/0] via 2001:2:2::4, swp2, 00:01:00ß
C>* 172.16.10.0/24 is directly connected, swp3, 5d18h32m
当交换机通过路由反射器学习路由时，它看起来像这样
router bgp 65101
bgp router-id 10.10.10.1
neighbor 2001:db8:2::a00:2 remote-as internal
neighbor 2001:db8:2::a00:2 capability extended-nexthop
!
address-family ipv6 unicast
neighbor 2001:db8:2::a00:2 activate
neighbor 2001:db8:2::a00:2 route-map GLOBAL in
exit-address-family
!
route-map GLOBAL permit 10
set ipv6 next-hop prefer-global
cumulus@leaf01:~$ sudo vtysh
...
leaf01# show ip route
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,
> - selected route, * - FIB route
B 0.0.0.0/0 [200/0] via 2001:2:2::4, 00:00:01
K 0.0.0.0/0 [0/0] via 10.0.2.2, eth0, 3d00h26m
C>* 10.0.0.8/32 is directly connected, lo, 3d00h26m
C>* 10.0.2.0/24 is directly connected, eth0, 03:39:18
C>* 172.16.3.0/24 is directly connected, swp2, 3d00h26m
B> 172.16.4.0/24 [200/0] via 2001:2:2::4 (recursive), 00:00:01
* via 2001:1:1::1, swp1, 00:00:01
C>* 172.16.10.0/24 is directly connected, swp3, 3d00h26m
邻居状态更改日志
Cumulus Linux 在 syslog 和 /var/log/frr/frr.log 文件中记录邻居经历的更改。例如
020-10-05T15:51:32.621773-07:00 leaf01 bgpd[10104]: %NOTIFICATION: sent to neighbor peerlink.4094 6/7 (Cease/Connection collision resolution) 0 bytes
2020-10-05T15:51:32.623023-07:00 leaf01 bgpd[10104]: %ADJCHANGE: neighbor peerlink.4094(leaf02) in vrf default Up
2020-10-05T15:51:32.623156-07:00 leaf01 bgpd[10104]: %NOTIFICATION: sent to neighbor peerlink.4094 6/7 (Cease/Connection collision resolution) 0 bytes
2020-10-05T15:51:32.623496-07:00 leaf01 bgpd[10104]: %ADJCHANGE: neighbor peerlink.4094(leaf02) in vrf default Down No AFI/SAFI activated for peer
2020-10-05T15:51:33.040332-07:00 leaf01 bgpd[10104]: [EC 33554454] swp53 [Error] bgp_read_packet error: Connection reset by peer
2020-10-05T15:51:33.279468-07:00 leaf01 bgpd[10104]: [EC 33554454] swp52 [Error] bgp_read_packet error: Connection reset by peer
2020-10-05T15:51:33.339487-07:00 leaf01 bgpd[10104]: %ADJCHANGE: neighbor swp54(spine04) in vrf default Up
2020-10-05T15:51:33.340893-07:00 leaf01 bgpd[10104]: %ADJCHANGE: neighbor swp53(spine03) in vrf default Up
2020-10-05T15:51:33.341648-07:00 leaf01 bgpd[10104]: %ADJCHANGE: neighbor swp52(spine02) in vrf default Up
2020-10-05T15:51:33.342369-07:00 leaf01 bgpd[10104]: %ADJCHANGE: neighbor swp51(spine01) in vrf default Up
2020-10-05T15:51:33.627958-07:00 leaf01 bgpd[10104]: %ADJCHANGE: neighbor peerlink.4094(leaf02) in vrf default Up
清除 BGP 路由
NVUE 提供了清除和刷新 BGP 表中路由的命令。您可以清除 BGP 表中的所有路由或 VRF 中地址族（IPv4、IPv6 或 EVPN）的所有路由。
BGP clear 命令不会清除内核或硬件中的计数器。
- 指定方向（
in或out）的 BGP clear 路由命令不会重置 BGP 邻居邻接关系。 - 当交换机配置了启用
soft-reconfiguration inbound的邻居时，执行 clear in 或 soft clear in 会清除地址族的软重配置表中的路由。这会导致根据任何应用的输入策略重新评估 BGP 表中的路由。 - 当交换机配置了未启用
soft-reconfiguration inbound选项的邻居时，执行 clear in 或 soft in 会向对等体发送路由刷新消息。 - 出站 BGP clear 命令（out 或 soft out）会将所有路由重新通告给 BGP 对等体。
- 未指定方向（
in或out）的 BGP soft clear 命令不会重置 BGP 邻居邻接关系，并且会影响入站和出站路由，如上所述，具体取决于您是否启用soft-reconfiguration inbound。
要清除和刷新所有 IPv4 入站路由
cumulus@leaf01:~$ nv action clear vrf default router bgp address-family ipv4-unicast soft in
要清除和重新发送所有 IPv6 出站路由
cumulus@leaf01:~$ nv action clear vrf default router bgp address-family ipv6-unicast soft out
要清除和刷新所有 EVPN 入站路由
cumulus@leaf01:~$ nv action clear vrf default router bgp address-family l2vpn-evpn soft in
要清除和重新发送所有出站 IPv4 路由
cumulus@leaf01:~$ nv action clear vrf default router bgp address-family ipv4-unicast soft out
要清除和重新发送到 BGP 邻居 10.10.10.101 的所有 IPv6 出站路由
cumulus@leaf01:~$ nv action clear vrf default router bgp neighbor 10.10.10.101 address-family ipv6-unicast out
要清除和重新发送 BGP 对等组 SPINES 的所有地址族（IPv4、IPv6 和 l2vpn-evpn）的出站路由
cumulus@leaf01:~$ nv action clear vrf default router bgp peer-group SPINES out
要清除和刷新所有 VRF 和地址族的所有 IPv4 入站路由
cumulus@switch:~$ nv action clear router bgp soft in
Action succeeded
要清除和刷新所有邻居、地址族和 VRF 的入站路由，并刷新出站路由过滤前缀列表
cumulus@switch:~$ nv action clear router bgp in prefix-filter
Action succeeded
要清除与所有邻居的 BGP 会话，强制邻居重新启动
cumulus@switch:~$ nv action clear router bgp
Action succeeded
要清除和刷新所有 IPv4 入站路由
cumulus@spine01:~$ sudo vtysh
...
switch# clear bgp vrf default ipv4 unicast * soft in
switch# exit
要清除和重新发送所有 IPv6 出站路由
cumulus@spine01:~$ sudo vtysh
...
switch# clear bgp vrf default ipv6 unicast * soft out
switch# exit
要清除和刷新所有 EVPN 入站路由
cumulus@spine01:~$ sudo vtysh
...
switch# clear bgp vrf default l2vpn evpn * soft in
switch# exit
要清除和重新发送所有出站 IPv4 路由
cumulus@spine01:~$ sudo vtysh
...
switch# clear bgp vrf default ipv4 unicast * soft out
switch# exit
要清除和重新发送到 BGP 邻居 10.10.10.101 的所有 IPv6 出站路由
cumulus@spine01:~$ sudo vtysh
...
switch# clear bgp vrf default ipv6 unicast 10.10.10.101 out
switch# exit
要清除和重新发送 BGP 对等组 SPINES 的所有地址族（IPv4、IPv6 和 l2vpn-evpn）的出站路由
cumulus@spine01:~$ sudo vtysh
...
switch# clear bgp vrf default peer-group SPINES out
switch# exit
要清除和刷新所有邻居、地址族和 VRF 的入站路由，并刷新出站路由过滤前缀列表
cumulus@spine01:~$ sudo vtysh
...
switch# clear bgp in prefix-filter
switch# exit
要清除与所有邻居的 BGP 会话，强制邻居重新启动
cumulus@switch:~$ sudo vtysh
...
switch# clear bgp *
switch# write memory
switch# exit