Re: Summary of responses so far and proposal moving forward[WasRe: [tcpm] Is this a problem?]

[Joe Touch <[e-mail]>]

Mark Allman wrote:
> Joe-
> 
> You and I are just going to disagree.
> 
>> That OS is required to reserve per-connection resources when
>> connections are created. It can halt new connections.
> 
> If this is the general reading of the spec then TCP has a built-in DoS
> vulnerability that we need to fix.

There's nothing in TCP that says that new connections will ALWAYS succeed.

Or can you find that in the requirements?

...
> It is absurd to me that a TCP (or any protocol) would allow a peer to
> indefinitely tie up a local resource without being subject to local
> policies on that resource.

It's very specifically in 1122. Local policies can kill applications,
but can't go around gleaning connections it *thinks* aren't useful.

If you want a new RFC that allows connection revocation, then it's
*definitely* a standards change.

I would favor a BCP that says that:

---
an OS under attack MAY terminate applications to recover their
resources, and the decision of which applications to terminate is a
matter of local policy. when such applications are terminated, their TCP
sessions would, as usual, be aborted
---

Note also that DOS attacks would likely not keep TCP connections around
with zero windows AND continue to ACK - they'd stop ACKing, the
connection would drop for *that* reason, and be recovered.

I would object to an RFC that allows TCP connection revocation to
compensate for not wanting to kill the apps.

Joe

_______________________________________________
tcpm mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/tcpm