If you were trying to ping both of your BGP peers (and failing), I'm
guessing both are on directly connected subnets as per eBGP rules.
As they are directly connected, routing protocols shouldn't have been
an issue so it was probably the firewall.
If I remember correctly, if you block outbound pings you get a 'no
route to host'. So it sounds like it could have been a PF resource
problem.
Do you run PF and how many states do you have usually?
On Sat 12 Apr 2014 18:54:11 BST, Randhir Prakash wrote:
Hi,
I have been using Openbgpd on Openbsd from last 3 years without any issue.
We are having two upstream providers with full route. Suddenly last week, i
observed that our network became unreachable from internet and sometimes
fluctuating in terms of reachability. I checked the bgp session using
bgpctl show summary which was showing bgp session alive and
PrfRcvd(Received prefix) from our peers was fine.
When i tried to ping to both peers WAN IP or any inetrnet ip, it was
telling "no route to host".
When i connected peers wan ip to standalone system and tried pinging, it
was pinging.
I was not able to figure out the issue and after 45-60 minutes,
automatically everything got rectified and became normal.
I am not having any idea about this issue and want to explore to avoid such
situations in production environment. Kindly assist.
