An ACX acting as a P router is seeing a forwarding issues after a link flap. The ACX has a BGP route to 10.33.24.12/30 with a backup label path:
user@acx> show route 10.33.24.12
inet.0: 973268 destinations, 3245100 routes (973268 active, 15 holddown, 27 hidden)
@ = Routing Use Only, # = Forwarding Use Only
+ = Active Route, - = Last Active, * = Both
10.33.24.12/30 *[BGP/170] 09:26:31, localpref 100
AS path: I, validation-state: unverified
> to 23.136.136.13 via et-0/0/2.0
to 23.136.136.2 via et-0/0/0.0, Push 16413
[BGP/170] 09:26:30, localpref 100, from 23.136.136.2
[BGP/170] 09:26:30, localpref 100, from 23.136.136.3
The ACX can forward packets to that destination:
user@acx> ping 10.33.24.14
Oct 24 15:24:15
PING 10.33.24.14 (10.33.24.14) 56(84) bytes of data.
64 bytes from 10.33.24.14: icmp_seq=1 ttl=64 time=0.844 ms
64 bytes from 10.33.24.14: icmp_seq=2 ttl=64 time=0.767 ms
But, routers upstream of the ACX cannot reach that destination.
From the route output, the path via et-0/0/2.0 should be chosen, with the labelled path as the backup. Pings from that upstream router fail:
user@router> ping 10.33.24.13
PING 10.33.24.13 (10.33.24.13) 56(84) bytes of data.
From 23.136.136.1 icmp_seq=1 Time to live exceeded
From 23.136.136.1 icmp_seq=2 Time to live exceeded
From 23.136.136.1 icmp_seq=3 Time to live exceeded
From 23.136.136.1 icmp_seq=4 Time to live exceeded
From 23.136.136.1 icmp_seq=5 Time to live exceeded
Traceroute shows the labelled path used on the ACX instead of the direct link:
user@router> traceroute no-resolve 10.33.24.13
traceroute to 10.33.24.13 (10.33.24.13), 30 hops max, 60 byte packets
1 * * *
2 23.136.136.1 <MPLS:L=16413,E=6,S=1,T=1> 6.199 ms 6.213 ms 6.159 ms
Even though the route has the weight set for the et-0/0/2 path:
user@acx> show route 10.33.24.13 detail
inet.0: 973473 destinations, 3245994 routes (973473 active, 15 holddown, 27 hidden)
10.33.24.12/30 (3 entries, 1 announced)
*BGP Preference: 170/-101
Next hop type: Indirect, Next hop index: 0
Address: 0x5651375b5c1c
Next-hop reference count: 12
Kernel Table Id: 0
Source: 23.136.136.13
Next hop type: Router, Next hop index: 0
Next hop: 23.136.136.13 via et-0/0/2.0 weight 0x1, selected
Label element ptr: 0x5650ddbd0948
Label parent element ptr: (nil)
Label element references: 197
Label element child references: 140
Label element lsp id: 0
Session Id: 0
Next hop: 23.136.136.2 via et-0/0/0.0 weight 0xf000
Label operation: Push 16413
Label TTL action: prop-ttl
Load balance label: Label 16413: None;
Label element ptr: 0x5650d9c0d248
Label element references: 4
Label element child references: 2
Protocol next hop: 23.136.136.13
Indirect next hop: 0x5650dc014808 49475 INH Session ID: 390
State: <Active Int Ext>
Local AS: 400146 Peer AS: 400146
Age: 9:52:16 Metric2: 10
Validation State: unverified
ORR Generation-ID: 0
Task: BGP_400146.23.136.136.13
Announcement bits (4): 0-KRT 3-Aggregate 5-BGP_RT_Background 6-Resolve tree 4
AS path: I
Communities: 65200:1
Accepted
Localpref: 100
Router ID: 23.136.136.13
Thread: junos-main
Forwarding Table perspective is also correct:
user@acx> show route forwarding-table destination 10.33.24.12 detail
Routing table: default.inet
Internet:
Destination Type RtRef Next hop Type Index NhRef Netif
10.33.24.12/30 user 0 indr 49475 1
ulst 49087 1
sftw 49473 1 et-0/0/2.0
23.136.136.13 ucst 3003 1 et-0/0/2.0
sftw Push 16413 49942 1 et-0/0/0.0
23.136.136.2 ucst 3002 1 et-0/0/0.0
Problem is tracked via PR1793786:
https://prsearch.juniper.net/PR1793786