BGP Peer Hold Timer Error

RouterOS Forwarding Protocols
1

I am receiving the following BGP notifications and errors and the remote user said the problem is on my side:

Connection terminated
RemoteAddress=10.251.253.129
Received notification
Hold timer expired, subcode=0

Peer config:
name=“Provider1” instance=default remote-address=10.251.253.129 remote-as=####### tcp-md5-key=“” nexthop-choice=default multihop=no route-reflect=no hold-time=3m ttl=default in-filter=Microsoft-in out-filter=Microsoft-out address-families=ip default-originate=never remove-private-as=no as-override=no passive=no use-bfd=no

We aren’t seeing packet loss via the cross connect. The BGP connection last some nights for 16 hours and once it start timing out, it continues and doesn’t last more then 5min

2

We’ve had a similar problem on our backbone. It seems that was generated by a flapping link.
It seems that in the middle of the exchange some data get lost, generating that error. The ugliest thing is the error is spread to all other peers.

3

Had same problem other side somehow fixed it but after upgrading ros 7 i had to disable them on all peers.