OSPF neighborship stays up eventhough int is unreachable

RouterOS Forwarding Protocols
1

Hello,
I witnessed a strange behaviour on Mikrotik 450G v5.2
Two mikrotiks, directly connected interfaces via wireless, in same subnet, say one of them is 192.168.1.1 and the other one is 192.168.1.2
They can ping&telnet each other, OSPF neighborship is up, interfaces configured as point to point in OSPF settings. Everything is great.
But the wireless goes down, and 192.168.1.1 cannot ping 192.168.1.2, they cannot telnet eachother, no connectivity, BUT OSPF neighborship is still up!

Any ideas?

2

OSPf neighbor is up until dead timer expires. By default it is 40 seconds.

3

Thanks mrz, I know, I should ve mentioned that I waited for the dead timer.
Unreachable for 40+ seconds but the adjacency is still up.

4

It can’t stay up if it is not reachable. Perhaps the same neighbor is reachable via other interface.

5

Thanks mrz,
All OSPF interfaces are in their respective different subnets.
It is happening right in front of my eyes, absolutely no sense.
I see that 1 ping out of 100 hardly ever successes with a very high latency. May it be sending the hello packet right at that time and keeping the adjacency?

6

So that means that link is not completely down, and at least one hello packet can reach neighbor within 40 seconds.

7

06:47:49 route,ospf,debug RECV: Hello <- x.x.x.41 on eth2 xxxxx (x.x.x.42)
06:47:49 route,ospf,debug received options: E
06:47:50 route,ospf,debug SEND: Hello x.x.x.42 -> 224.0.0.5 on eth2 xxxxx

06:47:59 route,ospf,debug RECV: Hello <- x.x.x.41 on eth2 xxxxx (x.x.x.42)
06:47:59 route,ospf,debug received options: E
06:48:00 route,ospf,debug SEND: Hello x.x.x.42 -> 224.0.0.5 on eth2 xxxxx



[zzzzzz] > ping x.x.x.41
HOST SIZE TTL TIME STATUS
X.X.X.41 timeout
X.X.X.41 timeout
X.X.X.41 timeout
X.X.X.41 timeout
X.X.X.41 timeout
X.X.X.41 timeout
X.X.X.41 timeout
sent=7 received=0 packet-loss=100%

[zzzzzz] /system> telnet x.x.x.41
Trying x.x.x.41...
telnet: Unable to connect to remote host: No route to host

[zzzzzz] > /ip route print
Flags: X - disabled, A - active, D - dynamic, C - connect, S - static, r - rip, b - bgp, o - ospf, m - mme,
B - blackhole, U - unreachable, P - prohibit

38 ADC x.x.x.40/29 x.x.x.42 eth2 xxxxx 0

[zzzzzz] > ping 224.0.0.5
HOST SIZE TTL TIME STATUS
224.0.0.5 timeout
224.0.0.5 timeout
224.0.0.5 timeout
224.0.0.5 timeout
224.0.0.5 timeout
224.0.0.5 timeout
224.0.0.5 timeout
"

[zzzzzz] /ip address> print
Flags: X - disabled, I - invalid, D - dynamic

ADDRESS NETWORK INTERFACE



1 ;;; xxxxxx
x.x.x.42/29 x.x.x.40 eth2 xxxxxx

Also the MPLS neighborship is up

Do you know any type of mechanism to place the interface into some type of listening state just like in STP, while the ping stats are below certain threshold? Can I do this with BFD?

8

Tell me please, your mikrotik ospf-router answer to ping 224.0.0.5 when link is up ?

9

No - annoyingly… :wink:

Are you using area type NSSA? I have seen that do very strange things on ROS 5.

10

Thanks for answer!

I use just “default” type.

I need reply from mikrotik-ospf-router (by 224.0.0.5) very much for diagnostik my real network.
I wrote trouble ticket in support about it.