Re: [Idr] draft-spaghetti-idr-bgp-sendholdtimer - Feedback requested

[Jeffrey Haas <[e-mail]>]

Jakob,

On Sun, Apr 25, 2021 at 11:28:43PM +0000, Jakob Heitz (jheitz) wrote:
> My understanding of TCP zero window is different.
> It is the peer sending 0 in the "window" field of the TCP header
> in the 16 bit word, at offset 14 bytes into the header.
> That means it will not accept any TCP data from you,
> but it is perfectly capable of sending you data and
> accepting your acks for it.

You're quite correct and I was mistaken here.

> If it were not so, then it would not be able to send you keepalives
> and your hold timer would expire.

Eventually, which was my incorrect thinking.

> It does not mean that the data plane is broken.
> If that were the case, you would get either no keepalives
> from the peer or no more acks to your own keepalives.

For my part, I don't believe that "broken data plane" is the main issue
here.  It's simply the fact that since the peer is blocked from sending
updates, we have a flavor of the dreaded "stuck route" problem.

> If forwarding is broken one way from you to your peer, then
> you will no longer get ACKs back from keepalives or other
> BGP packets you send to your peer.

Forwarding for directly attached peers (same subnet) often will behave well
when forwarding for remotely attached peers may fail.  A quirk of various
silicon.  And certainly not a guaranteed behavior.

>  This is not a zero window
> condition. This may be another condition we wish to monitor.
> Note, there are other ways to monitor this condition, e.g. BFD.

I might know a little about that.

-- Jeff