BGP disappearing problems (still in 2.9.29)

RouterOS General
1

Hi, 2.9.29 has fixed routing test memory leak, but i still have the annoying “loosing settings” problem.
At completely (seems) random intervals all the peers and instances settings disappear, and bgp sessions of course stop working.
The RB is still responsive and i can telnet to it from the neighbour, and a simple /routing bgp peer print solves the situation.
First time /routing bgp peer print shows nothing , a second /routing bgp peer print show all the peers again and the sessions come back up.



  MMM      MMM       KKK                          TTTTTTTTTTT      KKK
  MMMM    MMMM       KKK                          TTTTTTTTTTT      KKK
  MMM MMMM MMM  III  KKK  KKK  RRRRRR     OOOOOO      TTT     III  KKK  KKK
  MMM  MM  MMM  III  KKKKK     RRR  RRR  OOO  OOO     TTT     III  KKKKK
  MMM      MMM  III  KKK KKK   RRRRRR    OOO  OOO     TTT     III  KKK KKK
  MMM      MMM  III  KKK  KKK  RRR  RRR   OOOOOO      TTT     III  KKK  KKK

  MikroTik RouterOS 2.9.29 (c) 1999-2006       http://www.mikrotik.com/












Terminal vt102 detected, using multiline input mode
[admin@aim] > /routing bgp peer prin
Flags: X - disabled 
 

admin@aim] > /routing bgp peer prin
Flags: X - disabled 
 0   name="peer1" instance=default remote-address=80.79.50.10 remote-as=65001 
     tcp-md5-key="" nexthop-choice=force-self multihop=no route-reflect=no 
     hold-time=3m ttl=1 in-filter="" out-filter="" 

 1   name="peer2" instance=aim-local remote-address=80.79.50.1 remote-as=34695 
     tcp-md5-key="" nexthop-choice=force-self multihop=no route-reflect=no 
     hold-time=3m ttl=1 in-filter=infracom out-filter="" 

 2   name="aim2" instance=default remote-address=80.79.50.22 remote-as=65051 
     tcp-md5-key="" nexthop-choice=force-self multihop=no route-reflect=no 
     hold-time=3m ttl=1 in-filter="" out-filter="" 
[admin@aim] >

A quick hack is to run a scheduled job that runs a /routing bgp peer print every some minutes.
The problem shows up with 2.9.27 , 2.9.28 and 2.9.29

What can i do to help you solve the problem?
Ideally the scheduled scripts should avoid keeping the sessions stuck, and the problem seems not to show on lab.

I can send either configuration files, or a supout.
I can also turn bgp debugging on.
Bye,
Ricky

2

it just happened again..
I started generating the supot file, and as soon as i gave /system sup-out and it started generating the file the bgp sessions came back up…
I’m sending it..
Bye,
Ricky

3

I do not have the answer to your question but had one for you.

Are you using a pc or routerboard? How much ram? The reason I ask is, I have done some bgp stuff but always use bgpd. Have played with mikrotik but ran into issues with not enough ram. I havent put it on a pc with mikrotik yet but I guess I should soon to see how it will do.

4

Hi :smiley:

It’s on a routerboard, they’re used for the wireless network.

32 megs, but the routing tables is around 700 entries only.
Free memory is about 17/19 megs, so i guess it’s not an issue.

It depends what you need to do.. surely you cannot put a full table on a RB, but i can’t tell you how MT bgp performs with more memory on a pc, in that situation i always used Quagga.
Bye,
Ricky