[j-nsp] OSPF question

[Tyler Christiansen]

On Wed, Jun 4, 2014 at 7:28 PM, Andrew Miehs <andrew at 2sheds.de> wrote:

> On 5 Jun 2014, at 2:35 am, R S <dim0sal at hotmail.com> wrote:
>
> > Which can be the reason why among two MX (using irb) running OSPF there
> are around 100-150 OSPF Link State Update Packet retransmission per day ?
> > No link is interrupted, no link is fauly/CRC...
> >
> > My customer is receiving traps on its monitoring systems and I've no
> clear clue about the reason...
> > Any reccomendation/experience ?
>
> OSPF does a complete database update every 30 minutes with default
> settings.


This should not trigger the OSPF LSU Retransmit trap.  Traps are only
triggered by "error" events (which sometimes aren't really an error).

In previous experiences, I've seen devices that won't respond to packets
sent to the multicast address.  When an LSU retransmit occurs, that LSU is
sent unicast to the peer, at which point it's acknowledged.  This has the
effect that you get the trap every 30 minutes or so, coinciding with the 30
minute update timer.  The 30 minute update timer doesn't cause it, though.

In addition to the above, I would perform tcpdumps on the MX that does not
acknowledge the LSU and see if it receives the LSU when its sent via
multicast and see if it responds to it or not.

--tc


>
>
> _______________________________________________
> juniper-nsp mailing list juniper-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/juniper-nsp
>



-- 
*Tyler Christiansen | Technical Operations*
tyler <http://adap.tv/>@adap.tv <http://adap.tv/> | www.adap.tv
*m :* 864.346.4095