vMX

I am trying to get BFD to work on vMX. It's configured exactly how it would be in our production network, but for some reason, there are no BFD packets leaving the local vMX. I have a vMX interface attached to a vSRX and I can see the vSRX sending BFD, but the vMX is not sending anything back. I've tried it on several vMX installs and it's all the same behavior. Is there something I am missing in getting this working?

I'm on 18.1 in the vMX, running a PREMIUM license. There are no firewall filters applied to the RE.

group as65222 {
    local-address 192.168.50.33;
    peer-as 65222;
    bfd-liveness-detection {
        minimum-interval 1000;
        multiplier 3;
        session-mode single-hop;
    }

Output from show bfd session:

admin@vmx1> show bfd session extensive 
                                                  Detect   Transmit
Address                  State     Interface      Time     Interval  Multiplier
192.168.50.34             Init      xe-0/0/3.0     6.000     2.000        3   
 Client BGP, TX interval 1.000, RX interval 1.000
 Local diagnostic None, remote diagnostic None
 Remote state Down, version 1
 Session type: Single hop BFD
 Min async interval 1.000, min slow interval 2.000
 Adaptive async TX interval 2.000, RX interval 2.000
 Local min TX interval 2.000, minimum RX interval 1.000, multiplier 3
 Remote min TX interval 2.000, min RX interval 2.000, multiplier 3
 Local discriminator 17, remote discriminator 16
 Echo mode disabled/inactive, no-absorb, no-refresh
  Session ID: 0x140

1 sessions, 1 clients
Cumulative transmit rate 0.5 pps, cumulative receive rate 0.5 pps

Output from monitor traffic:

admin@vmx1> monitor traffic interface xe-0/0/3.0 no-resolve size 1500 matching "udp port 3784" 
verbose output suppressed, use <detail> or <extensive> for full protocol decode
Address resolution is OFF.
Listening on xe-0/0/3.0, capture size 1500 bytes

10:46:53.858927 In IP 192.168.50.34.49152 > 192.168.50.33.3784: BFDv1, One-hop Control, State Down, Flags: [none], length: 24
10:47:06.736092 In IP 192.168.50.34.49152 > 192.168.50.33.3784: BFDv1, One-hop Control, State Down, Flags: [none], length: 24
10:47:18.930322 In IP 192.168.50.34.49152 > 192.168.50.33.3784: BFDv1, One-hop Control, State Down, Flags: [none], length: 24

bfd traceoptions output:

May 27 10:51:54 [THROTTLE]bfdd_rate_limit_program_timer_expiry: Session 192.168.50.34 is moved state = 2 Time Elapsed since last packet = 02:30:01
May 27 10:51:54 [THROTTLE]bfdd_rate_limit_program_timer_expiry: Session 192.168.50.34 is removed form program therad
May 27 10:51:54 bfdd_rate_limit_program_timer_expiry: Slow start revert
May 27 10:51:54 [RE_PPS]bfdd_slow_start_revert_poll_batch: RE PPS was 0 max_pps 5000 orginal_max_pps 5000 pend_pps 0 outstnading_poll 0
 
May 27 10:51:54 [SLOW_START]bfdd_slow_start_revert_poll_batch: processed 0 sessions
May 27 10:51:54 [SLOW_START]bfdd_slow_start_timer_expiry: Slow start thread not empty, calling slow start reversal
May 27 10:51:57 Received Downstream RcvPkt (19) len 102:
May 27 10:51:57    IfIndex (3) len 4: 332
May 27 10:51:57    Protocol (1) len 1: BFD
May 27 10:51:57    SrcAddr (5) len 8: 192.168.50.34
May 27 10:51:57    Data (9) len 24: (hex) 20 40 03 18 00 00 00 10 00 00 00 00 00 1e 84 80 00 1e 84
May 27 10:51:57    PktError (26) len 4: 0
May 27 10:51:57    RtblIdx (24) len 4: 0
May 27 10:51:57    MultiHop (64) len 1: (hex) 00
May 27 10:51:57    Seamless (245) len 1: 0
May 27 10:51:57    Unknown (213) len 1: (hex) 00
May 27 10:51:57    Unknown (168) len 1: (hex) 00
May 27 10:51:57    Authenticated (121) len 1: (hex) 00
May 27 10:51:57 BFD packet from 192.168.50.34 (IFL 332), len 24
May 27 10:51:57    Ver 1, diag 0, mult 3, len 24
May 27 10:51:57    Flags: ,  State: Down
May 27 10:51:57    My discr 0x00000010, your discr 0x00000000
May 27 10:51:57    Tx ivl 2000000, rx ivl 2000000, echo rx ivl 0
May 27 10:51:57 [THROTTLE]bfdd_rate_limit_can_accept_pkt: session 192.168.50.34 pkt rcvd for a session which might be deleted or down so adding that session to program/pending thread
May 27 10:51:57 [SLOW_START]bfdd_slow_start_start: 192.168.50.34 is already in slow start thread
May 27 10:51:57 [THROTTLE] bfdd_rate_limit_start_program_timer: Starting program timer
May 27 10:51:57 [THROTTLE]bfdd_rate_limit_add_ppm_thread:Session 192.168.50.34 added to program thread
May 27 10:51:57 [THROTTLE]bfdd_rate_limit_can_accept_pkt: session 192.168.50.34 moved from pending to program
May 27 10:51:57 (bfdd_update_tx_intervals:510) Session 192.168.50.34 (IFL 332): cur tx ivl 2000000, new_invl 0(2,0,2000000,2000000)
 
May 27 10:51:57 (bfdd_update_tx_intervals:537) Session 192.168.50.34 (IFL 332): cur tx ivl 2000000, new_invl 2000000
 
May 27 10:51:57 (bfdd_update_tx_intervals:555) Session 192.168.50.34 (IFL 332): cur tx ivl 2000000, new_invl 2000000

How are you running vMX? vmware? kvm (and what OS)?

What physical NICs are you using? X520? X710? etc.

I'm running on VMware 5.5 and I am not sure of what the physical NICs are. Does the physical NIC matter? If so, is there at least some mention of this in the documentation?

I should also mention that it doesn't work in a KVM environment, either.

It can matter, it shouldn't, but it can due to bugs. We've seen no end of issues with Intel X710 NICs (I'd assume XL710 are the same, given the same i40e driver).  The most recent issue we've seen as with OSPF - One side would send and receive Hello packets, the other side the host node would receive Hello, but vMX never received the Hello, only sent them out; so one side was stuck at init because it could see the incoming Hello, the other didn't have a clue.

Hello,

Have You tried on VMX:

set routing-options ppm no-delegate-processing

HTH

Thx

Alex

Thanks, Alex! That worked. I'll check out documentation to get more details about that command. Much appreciated.