介面

擷取自本次部署其中一台 CE，時間為 2026-07-28。sitecli/capture-manifest.json 記錄了擷取的節點，而 scripts/capture-sitecli.sh --check 會針對線上 CE 重新驗證指令介面。

這些指令回報的是核心的觀點。Argo 資料平面則保有自己的觀點，可透過 vif 查看，兩者可能不一致。

ip

接受一個 ip 子指令。addr 是最常用的；route、link 與 neigh 也 可以使用。

{"command":["ip","addr"]}

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
2: a-i-eth0: <BROADCAST,MULTICAST,ALLMULTI,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 60:45:bd:eb:3f:f2 brd ff:ff:ff:ff:ff:ff
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 60:45:bd:eb:38:ea brd ff:ff:ff:ff:ff:ff
    inet 10.0.2.4/26 brd 10.0.2.63 scope global noprefixroute eth1
       valid_lft forever preferred_lft forever
4: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 60:45:bd:eb:3d:f3 brd ff:ff:ff:ff:ff:ff
    inet 10.0.3.4/26 brd 10.0.3.63 scope global noprefixroute eth2
       valid_lft forever preferred_lft forever
    inet6 fe80::55af:2cd4:23a1:7e06/64 scope link noprefixroute
       valid_lft forever preferred_lft forever
5: docker0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default
    link/ether 62:61:2b:33:cd:42 brd ff:ff:ff:ff:ff:ff
    inet 100.64.255.1/29 brd 100.64.255.7 scope global docker0
       valid_lft forever preferred_lft forever
10: tap_eth0: <BROADCAST,MULTICAST,ALLMULTI,UP,LOWER_UP> mtu 1500 qdisc multiq state UNKNOWN group default qlen 1000
    link/ether 60:45:bd:eb:3f:f2 brd ff:ff:ff:ff:ff:ff
11: vhost0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UNKNOWN group default qlen 1000
    link/ether 60:45:bd:eb:3f:f2 brd ff:ff:ff:ff:ff:ff
    inet 10.0.1.4/26 brd 10.0.1.63 scope global noprefixroute vhost0
       valid_lft forever preferred_lft forever
13: vethfrr0@if12: <BROADCAST,MULTICAST,PROMISC,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 02:01:02:03:04:05 brd ff:ff:ff:ff:ff:ff link-netns frrns
15: argo-envoy-0@if14: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1520 qdisc noqueue state UP group default qlen 1000
    link/ether 00:00:5e:00:01:00 brd ff:ff:ff:ff:ff:ff link-netns envoy-ns
17: host-envoy-0@if16: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether ee:95:ea:b5:8c:2a brd ff:ff:ff:ff:ff:ff link-netns envoy-ns
    inet 169.254.100.1/30 scope global host-envoy-0
       valid_lft forever preferred_lft forever
18: xc-br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether aa:db:2f:e1:15:2c brd ff:ff:ff:ff:ff:ff
19: argo-end@xc-br0-end: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 76:e0:23:f6:48:8b brd ff:ff:ff:ff:ff:ff
20: xc-br0-end@argo-end: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master xc-br0 state UP group default qlen 1000
    link/ether aa:db:2f:e1:15:2c brd ff:ff:ff:ff:ff:ff
22: linkquality0@if21: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 590 qdisc noqueue state UP group default
    link/ether 00:00:5e:00:01:00 brd ff:ff:ff:ff:ff:ff link-netns ns-link-quality
24: vethbfd0@if23: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 00:00:5e:00:01:00 brd ff:ff:ff:ff:ff:ff link-netns bfdns
25: keepalived-tap: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc fq_codel state DOWN group default qlen 1000
    link/ether 32:1d:e4:ae:46:0e brd ff:ff:ff:ff:ff:ff
    inet 169.254.10.13/30 brd 169.254.10.15 scope global keepalived-tap
       valid_lft forever preferred_lft forever
27: vethdnsmasq0@if26: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 02:01:02:03:04:07 brd ff:ff:ff:ff:ff:ff link-netnsid 4
    inet 169.254.254.103/32 brd 169.254.255.255 scope global vethdnsmasq0
       valid_lft forever preferred_lft forever

CE 依設計具有三張 Azure 網卡 —— 管理/SLO、外部、內部/SLI —— 但 客體作業系統並非依照該順序所暗示的方式呈現它們，而這正是誤讀 輸出會浪費時間的地方。

在客體系統中	承載
a-i-eth0	完全沒有位址 —— 這是被重新命名的實體 eth0，由 Argo 資料平面所擁有
vhost0	管理/SLO 位址，也就是 BGP 對等連線的來源位址，以及站點物件所回報的位址
eth1 與 eth2	外部與內部/SLI 位址 —— 但順序並不固定

eth1 與 eth2 之中哪一個是內部介面，各節點並不相同

從本次部署的三台 CE 上即時讀取的結果：在 f5-xc-ce-vm-01 與 f5-xc-ce-vm-03 上，內部/SLI 位址位於 eth2；而在 f5-xc-ce-vm-02 上則位於 eth1。設定中並沒有任何項目決定這一點。

因此，絕不要從介面名稱推斷子網路，也絕不要把某個節點的對應關係套用到 另一個節點。請改以位址進行比對 —— CE 位址所屬的子網路是穩定的，即使 承載它的介面並不穩定。

a-i-eth0 沒有承載任何位址，也正是為什麼無法透過管理位址連上 sshd 的完整原因：那裡根本沒有監聽程序，而不是連接埠被封鎖。請參閱 SSH。

ip-link-show

僅顯示連結層資訊：狀態、MTU、佇列規則、MAC。

{"command":["ip-link-show"]}

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: a-i-eth0: <BROADCAST,MULTICAST,ALLMULTI,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000
    link/ether 60:45:bd:eb:3f:f2 brd ff:ff:ff:ff:ff:ff
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000
    link/ether 60:45:bd:eb:38:ea brd ff:ff:ff:ff:ff:ff
4: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000
    link/ether 60:45:bd:eb:3d:f3 brd ff:ff:ff:ff:ff:ff
5: docker0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN mode DEFAULT group default
    link/ether 62:61:2b:33:cd:42 brd ff:ff:ff:ff:ff:ff
10: tap_eth0: <BROADCAST,MULTICAST,ALLMULTI,UP,LOWER_UP> mtu 1500 qdisc multiq state UNKNOWN mode DEFAULT group default qlen 1000
    link/ether 60:45:bd:eb:3f:f2 brd ff:ff:ff:ff:ff:ff
11: vhost0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UNKNOWN mode DEFAULT group default qlen 1000
    link/ether 60:45:bd:eb:3f:f2 brd ff:ff:ff:ff:ff:ff
13: vethfrr0@if12: <BROADCAST,MULTICAST,PROMISC,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether 02:01:02:03:04:05 brd ff:ff:ff:ff:ff:ff link-netns frrns
15: argo-envoy-0@if14: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1520 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether 00:00:5e:00:01:00 brd ff:ff:ff:ff:ff:ff link-netns envoy-ns
17: host-envoy-0@if16: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether ee:95:ea:b5:8c:2a brd ff:ff:ff:ff:ff:ff link-netns envoy-ns
18: xc-br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether aa:db:2f:e1:15:2c brd ff:ff:ff:ff:ff:ff
19: argo-end@xc-br0-end: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether 76:e0:23:f6:48:8b brd ff:ff:ff:ff:ff:ff
20: xc-br0-end@argo-end: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master xc-br0 state UP mode DEFAULT group default qlen 1000
    link/ether aa:db:2f:e1:15:2c brd ff:ff:ff:ff:ff:ff
22: linkquality0@if21: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 590 qdisc noqueue state UP mode DEFAULT group default
    link/ether 00:00:5e:00:01:00 brd ff:ff:ff:ff:ff:ff link-netns ns-link-quality
24: vethbfd0@if23: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether 00:00:5e:00:01:00 brd ff:ff:ff:ff:ff:ff link-netns bfdns
25: keepalived-tap: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc fq_codel state DOWN mode DEFAULT group default qlen 1000
    link/ether 32:1d:e4:ae:46:0e brd ff:ff:ff:ff:ff:ff
27: vethdnsmasq0@if26: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether 02:01:02:03:04:07 brd ff:ff:ff:ff:ff:ff link-netnsid 4

當問題是「連結是否已啟用，以及它的 MTU 是多少」時，這比 ip addr 更快。 MTU 不一致是隧道能夠建立但無法傳遞流量的常見原因。

ip-link-set Exec 層級 —— 具變更性

啟用或停用某個介面或群組。

ip-link-set (<device>||<group>) (up||down)

此處未執行，且無法以通常方式存取

這是此版本中唯一屬於 Exec 層級的指令。它會送往 exec 端點 而非 exec-user，並且會變更節點：在遠端 CE 上停用 eth0 會移除你用來管理它的路徑，包括除錯 API 本身。

本頁刻意沒有擷取任何輸出。擷取工具會拒絕整個 Exec 層級， 並且是從線上目錄讀取層級資訊，而非使用硬編碼清單，因此即使 未來版本將其他指令提升至該層級，這一點依然成立。