Sent from my iPhone

> On May 2, 2016, at 8:49 PM, Carlos Pignataro (cpignata) <[email protected]> 
> wrote:
> 
> Hi, Kathleen,
> 
>> On May 2, 2016, at 8:34 PM, [email protected] wrote:
>> 
>> Hi,
>> 
>> Thanks for the quick responses.  See inline 
>> 
>> Sent from my iPhone
>> 
>>> On May 2, 2016, at 8:04 PM, Carlos Pignataro (cpignata) 
>>> <[email protected]> wrote:
>>> 
>>> Kathleen,
>>> 
>>>>> On May 2, 2016, at 7:48 PM, Manav Bhatia <[email protected]> wrote:
>>>>> 
>>>>> Hi Kathleen,
>>>>> 
>>>>> DISCUSS:
>>>>> ----------------------------------------------------------------------
>>>>> 
>>>>> This should be pretty easy to address.  In the security consideration
>>>>> section, the following recommendation appears:
>>>>> 
>>>>>  o  SBFDReflector MUST NOT look at the crypto sequence number before
>>>>>       accepting the packet.
>>>>> 
>>>>> Could you please add text to say what happens (what attacks are possible)
>>>>> if this is looked at?  There is nothing to stop the crypt sequence number
>>>>> from being looked at, right?  Is there a way to actually prevent that?
>>>> 
>>>> SBFD is state-less. The SBFDReflector is NOT maintaining any BFD peer 
>>>> state, and is thus incapable of doing the crypto-sequence checks. It has 
>>>> no idea of last sequence number that it had seen from a BFD peer, and 
>>>> hence CANNOT compare the new sequence number. Its for this reason that we 
>>>> mandate that the reflectors MUST NOT look at the sequence numbers.
>>> 
>>> Further to this, the SBFDReflector can be receiving S-BFD packets from 
>>> multiple SBFDInitiators. Hence, the authentications can be used from BFD 
>>> but not the sequence numbers.
>>> 
>>>> We cant prevent a peer from looking at the sequence number -- thats an 
>>>> implementation specific issue. The implementation is violating the 
>>>> standard. Not sure what we can do to prevent that.
>>>> 
>>>> Does this help?
>>> 
>>> We could also explain the rational behind this requirement a bit better. 
>>> Would that help?
>> Yes, that would be helpful.  I'm glad to see that this is not an issue.
> 
> Indeed — thanks. I added the following, ready in our working copy:
> 
> @@ -916,7 +916,10 @@
>        configured.
>  
>     o  SBFDReflector MUST NOT look at the crypto sequence number before
> -      accepting the packet.
> +      accepting the packet.  This is because the SBFDReflector does not
> +      maintain S-BFD peer state, and because the SBFDReflector can
> +      receive S-BFD packets from multiple SBFDInitiators.  Consequently,
> +      BFD authentication can be used but not the sequence number.
>  
>     o  SBFDReflector MAY look at the Auth Key ID in the incoming packet
>        and verify the authentication data.

Excellent, thank you.  I'll move it to a comment in the morning for the 
shepherd & AD to track when approving the draft.

Best regards,
Kathleen 

> 
> 
> Thanks,
> 
> — Carlos.
> 
>> Thanks,
>> Kathleen 
>> 
>>> Thanks,
>>> 
>>> — Carlos.
>>> 
>>>> Cheers, Manav
> 

Reply via email to