Re: [Idr] Fwd: New Version Notification for draft-spaghetti-idr-bgp-sendholdtimer-05.txt

[heasley <[e-mail]>]

Mon, Aug 15, 2022 at 11:18:43PM +0200, Robert Raszuk:
> > https://mailarchive.ietf.org/arch/msg/idr/McRvkJ6UiNwJSKvGs0GPaqDfovA/
> 
> 
> #1 - IMO it would be a pretty bad idea to apply Send_Hold_Timer to a
> booting node. If at all this timer should fire only after receiving the
> <EOR> marker on the session.

there had also been some discussion about what the appropriate timer
value is - and certainly when to apply it.  the draft can be refined
after adoption too.

but, the point is that a peer might "ignore" a KA timer as explained in
that email (or for other reasons).  It is not limited to boot-time or
initial bgp session establishment.  eg: a session might be long-
established, but the remote peer is re-establishing all of its iBGP peers.

> #2 - Authors of this draft target cases where a peer is stuck for
> "days/weeks" ... I am yet to see a BGP node taking that much to boot.

no, it targets stuck peers and does not endeavor to allow them to remain
so for long.  the "stuck for weeks" quote was a real example of the
extreme, an example of sessions that never recovered and thus had stale
NLRI.