Grazie di nuovo, Giuseppe.

So that "every 1800 seconds" thingy is the "paranoid update," eh? But only the LSA owner (you mean originator?) sends it? 'Cause I had thought the paranoid update was when routers synchronize their _entire_ topology table with their neighbors.

I wonder what the purpose is of storing a Type 1 LSA for any length of time when the link is known to be down.

Hello Stuey,

yes the originator is the LSA owner and the 1800 seconds.

Note: each LSA has its own age so they don't become too old all together.

Anyway the bandwidth usage is very low in comparison to RIP or IGRP.

What you have seen has been reported by other collegues.

My guess is that OSPF is able to detect that the router LSA is not connected anymore to the topology, so instead of having some node to take the role to purge it, it is simply left there to expire in silence.

Hope to help

Giuseppe

you can try under ospf config the next.

router osp 1

timers pacing lsa-group xxx

where xxx is seconds to age-out the LSA..

The-great-l0k1

mx

Anybody can generate a maxAge LSA if the age of an LSA in its database reaches the max age of 3600. Nobody can prematurely age somebody else's LSA, though; a router may only prematurely age its own self-originated LSAs.