Message header error / Connection Not Synchronized

[petra83]

Hi,

I'm having trouble getting the BGP connection to work between my router and server running Ubuntu. I'm running exabgp v3.4.3.

debugs logs say that the server is attempting the connection and then times out after 1 min.

Mon, 24 Nov 2014 12:27:28 | DEBUG | 4309 | wire | Attempting connection to y.y.y.y
Mon, 24 Nov 2014 12:27:30 | DEBUG | 4309 | wire | session 1 outgoing x.x.x.x / y.y.y.y SENDING FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF 00AF 0104 FFDD 0E10 80E6 430E 9202 3040 2E84 B000 0102 8000 0286 8000 0280 8000 0104 8000 0185 8000 0186 8000 0201 8000 0285 8000 0180 8000 0101 8000 1941 8002 0601 0400 0100 0102 0601 0400 0100 0202 0601 0400 0100 0402 0601 0400 0100 8002 0601 0400 0100 8502 0601 0400 0100 8602 0601 0400 0200 0102 0601 0400 0200 8002 0601 0400 0200 8502 0601 0400 0200 8602 0601 0400 1900 4102 0641 0400 00FF DD

Mon, 24 Nov 2014 12:27:30 | INFO | 4309 | message | Peer y.y.y.y ASN 65500 >> OPEN version=4 asn=65501 hold_time=3600 router_id=x.x.x.x capabilities=[Multiprotocol(ipv4 unicast,ipv4 multicast,ipv4 nlri-mpls,ipv4 mpls-vpn,ipv4 flow,ipv4 flow-vpn,ipv6 unicast,ipv6 mpls-vpn,ipv6 flow,ipv6 flow-vpn,l2vpn vpls), Graceful Restart Flags 0x8 Time 1200 ipv4/multicast=0x80 ipv6/flow-vpn=0x80 ipv6/mpls-vpn=0x80 ipv4/nlri-mpls=0x80 ipv4/flow=0x80 ipv4/flow-vpn=0x80 ipv6/unicast=0x80 ipv6/flow=0x80 ipv4/mpls-vpn=0x80 ipv4/unicast=0x80 l2vpn/vpls=0x80, ASN4(65501)]

Mon, 24 Nov 2014 12:27:30 | DEBUG | 4309 | timers | peer y.y.y.y ASN 65500 Receive Timer 59 second(s) left
Mon, 24 Nov 2014 12:27:31 | DEBUG | 4309 | timers | peer y.y.y.y ASN 65500 Receive Timer 59 second(s) left
Mon, 24 Nov 2014 12:27:32 | DEBUG | 4309 | timers | peer y.y.y.y ASN 65500 Receive Timer 57 second(s) left


Mon, 24 Nov 2014 12:28:29 | DEBUG | 7253 | wire | session 1 outgoing x.x.x.x / y.y.y.y SENDING FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF 004D 0301 0177 6169 7465 6420 666F 7220 6F70 656E 2074 6F6F 206C 6F6E 672C 2077 6520 646F 206E 6F74 206C 696B 6520 7374 7563 6B20 696E 2061 6374 6976 65
Mon, 24 Nov 2014 12:28:29 | INFO | 7253 | message | Peer y.y.y.y ASN 65500 >> NOTIFICATION (1,1,"waited for open too long, we do not like stuck in active")
Mon, 24 Nov 2014 12:28:29 | INFO | 7253 | network | Peer y.y.y.y ASN 65500 out loop, peer reset, message [notification sent (1,1)] error[Message header error / Connection Not Synchronized / waited for open too long, we do not like stuck in active]

*****************************

exbgp config on the server:

group edgerouters {
peer-as 65500;
local-as 65501;
hold-time 3600;
router-id x.x.x.x;
local-address x.x.x.x;
graceful-restart 1200;

static {
}

process blocker-dynamic {
run /home/blocker/blocker_env/bin/blocker-exabgp; ## to manage null routes
}

neighbor y.y.y.y {
description "edge-1";
}

}

**************************************
BGP config on the router:

router bgp 65500
bgp log-neighbor-changes
neighbor blackhole-group peer-group
neighbor blackhole-group remote-as 65501
neighbor blackhole-group ebgp-multihop 3
neighbor blackhole-group update-source Loopback1
neighbor blackhole-group timers 10 3600
neighbor x.x.x.x peer-group blackhole-group
neighbor z.z.z.z peer-group blackhole-group
!
address-family ipv4
neighbor blackhole-group activate
neighbor blackhole-group soft-reconfiguration inbound
neighbor blackhole-group route-map deny-all out
neighbor x.x.x.x peer-group blackhole-group
neighbor z.z.z.z peer-group blackhole-group
no auto-summary
no synchronization
exit-address-family

******************************

Server specs:

lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 12.04.5 LTS
Release: 12.04
Codename: precise


uname -a
Linux track10 3.2.0-70-generic #105-Ubuntu SMP Wed Sep 24 19:49:16 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux


I have another server that belongs to the remote AS peer group running exabgp 3.1.12 and using the similar exabgp config - and that seems to have no trouble connecting with the router.

Any thoughts as to what may be causing this?


Thanks!

[Thomas Mangin]

Hello,

> I'm having trouble getting the BGP connection to work between my router and server running Ubuntu. I'm running exabgp v3.4.3.

Sorry to hear that.

> debugs logs say that the server is attempting the connection and then times out after 1 min.
>
> Mon, 24 Nov 2014 12:27:28 | DEBUG | 4309 | wire | Attempting connection to y.y.y.y
> Mon, 24 Nov 2014 12:27:30 | DEBUG | 4309 | wire | session 1 outgoing x.x.x.x / y.y.y.y SENDING FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF 00AF 0104 FFDD 0E10 80E6 430E 9202 3040 2E84 B000 0102 8000 0286 8000 0280 8000 0104 8000 0185 8000 0186 8000 0201 8000 0285 8000 0180 8000 0101 8000 1941 8002 0601 0400 0100 0102 0601 0400 0100 0202 0601 0400 0100 0402 0601 0400 0100 8002 0601 0400 0100 8502 0601 0400 0100 8602 0601 0400 0200 0102 0601 0400 0200 8002 0601 0400 0200 8502 0601 0400 0200 8602 0601 0400 1900 4102 0641 0400 00FF DD
>
> Mon, 24 Nov 2014 12:27:30 | INFO | 4309 | message | Peer y.y.y.y ASN 65500 >> OPEN version=4 asn=65501 hold_time=3600 router_id=x.x.x.x capabilities=[Multiprotocol(ipv4 unicast,ipv4 multicast,ipv4 nlri-mpls,ipv4 mpls-vpn,ipv4 flow,ipv4 flow-vpn,ipv6 unicast,ipv6 mpls-vpn,ipv6 flow,ipv6 flow-vpn,l2vpn vpls), Graceful Restart Flags 0x8 Time 1200 ipv4/multicast=0x80 ipv6/flow-vpn=0x80 ipv6/mpls-vpn=0x80 ipv4/nlri-mpls=0x80 ipv4/flow=0x80 ipv4/flow-vpn=0x80 ipv6/unicast=0x80 ipv6/flow=0x80 ipv4/mpls-vpn=0x80 ipv4/unicast=0x80 l2vpn/vpls=0x80, ASN4(65501)]
>
> Mon, 24 Nov 2014 12:27:30 | DEBUG | 4309 | timers | peer y.y.y.y ASN 65500 Receive Timer 59 second(s) left
> Mon, 24 Nov 2014 12:27:31 | DEBUG | 4309 | timers | peer y.y.y.y ASN 65500 Receive Timer 59 second(s) left
> Mon, 24 Nov 2014 12:27:32 | DEBUG | 4309 | timers | peer y.y.y.y ASN 65500 Receive Timer 57 second(s) left
>
>
> Mon, 24 Nov 2014 12:28:29 | DEBUG | 7253 | wire | session 1 outgoing x.x.x.x / y.y.y.y SENDING FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF 004D 0301 0177 6169 7465 6420 666F 7220 6F70 656E 2074 6F6F 206C 6F6E 672C 2077 6520 646F 206E 6F74 206C 696B 6520 7374 7563 6B20 696E 2061 6374 6976 65
> Mon, 24 Nov 2014 12:28:29 | INFO | 7253 | message | Peer y.y.y.y ASN 65500 >> NOTIFICATION (1,1,"waited for open too long, we do not like stuck in active")
> Mon, 24 Nov 2014 12:28:29 | INFO | 7253 | network | Peer y.y.y.y ASN 65500 out loop, peer reset, message [notification sent (1,1)] error[Message header error / Connection Not Synchronized / waited for open too long, we do not like stuck in active]

Ok, so ExaBGP sent the OPEN message and is waiting for the peer to reply for 60 seconds with an OPEN but nothing comes back.

> *****************************
>
> exbgp config on the server:
>
> group edgerouters {
> peer-as 65500;
> local-as 65501;
> hold-time 3600;
> router-id x.x.x.x;
> local-address x.x.x.x;
> graceful-restart 1200;
>
> static {
> }
>
> process blocker-dynamic {
> run /home/blocker/blocker_env/bin/blocker-exabgp; ## to manage null routes
> }
>
> neighbor y.y.y.y {
> description "edge-1";
> }
>
> }

This looks right.

> **************************************
> BGP config on the router:
>
> router bgp 65500
> bgp log-neighbor-changes
> neighbor blackhole-group peer-group
> neighbor blackhole-group remote-as 65501
> neighbor blackhole-group ebgp-multihop 3
> neighbor blackhole-group update-source Loopback1
> neighbor blackhole-group timers 10 3600
> neighbor x.x.x.x peer-group blackhole-group
> neighbor z.z.z.z peer-group blackhole-group
> !
> address-family ipv4
> neighbor blackhole-group activate
> neighbor blackhole-group soft-reconfiguration inbound
> neighbor blackhole-group route-map deny-all out
> neighbor x.x.x.x peer-group blackhole-group
> neighbor z.z.z.z peer-group blackhole-group
> no auto-summary
> no synchronization
> exit-address-family
>
> ******************************

A long time since I configured a Cisco router but it looks right too.
Is it that the multi-hop need to be increased from the new location ?

You could restrict the family negotiated (can not harm):

family {
ipv4 unicast;
ipv6 unicast;

}


> I have another server that belongs to the remote AS peer group running exabgp 3.1.12 and using the similar exabgp config - and that seems to have no trouble connecting with the router.

If not multihop, could you please re-run exabgp with "-d" to gather more information as AFAICS it should just work.

Sincerely,

Thomas

[petra83]

You were right... The problem was with the multihop value - the second server was 4 hops away; increasing the ebgp-multihop value fixed it... Thanks a lot!!