Re: [dnssd] Genart last call review of draft-ietf-dnssd-push-20

[Tom Pusateri <[e-mail]>]

If we detect a PUSH protocol error and don’t want the client to immediately reconnect and repeat the same error, we can send a retry delay TLV with a long delay and then close gracefully with a TLS close_notify in lieu of a TCP RST. 

Tom

> On Jul 8, 2019, at 7:05 PM, David Schinazi <[e-mail]> wrote:
> 
> In general the "TLS Alerts" error codes are specific to the operation of TLS itself, not the application running over TLS.
> 
> If you want to send a graceful close, the tool of choice is close_notify.
> If you detect an unrecoverable error and want to abort the connection, I see two options:
> (1) forcibly terminate the connection at the DNS layer by sending a DNS error message followed by a TLS close_notify
> (2) forcibly terminate the connection at the TCP layer by sending a RST
> 
> As a client sending, I don't see much value in (1) since all the server can do in either case is free the resources associated with this connection.
> As a server sending, I suspect (1) is best unless you were unable to parse anything in which case (2) makes sense.
> 
> David
> 
>> On Mon, Jul 8, 2019 at 3:53 PM Stuart Cheshire <[e-mail]> wrote:
>> On 2 Jul 2019, at 12:36, David Schinazi <[e-mail]> wrote:
>> 
>> > Hi Tom,
>> > 
>> > If the protocol is restricted to TLS over TCP, it should send a TLS close_notify, not a TCP RST.
>> > TLS close_notify is cryptographically guaranteed to originate from the peer,
>> > whereas TCP RST can be injected by an on-path entity to cause truncation attacks.
>> 
>> In TCP we use FIN for a graceful close, and RST for an abortive close. The former is normal operation; the latter means your code has a bug you need to fix.
>> 
>> Is there an appropriate equivalent in TLS? It would be good to differentiate normal operation from a fatal protocol error that causes a forcible termination.
>> 
>> I see in the TLS 1.3 spec, RFC 8446, Section 6.2. “Error Alerts” says:
>> 
>>    Whenever an implementation encounters a fatal error condition, it
>>    SHOULD send an appropriate fatal alert and MUST close the connection
>>    without sending or receiving any additional data.
>> 
>> <https://tools.ietf.org/html/rfc8446#section-6.2>
>> 
>> Are any of these error alerts appropriate to perform this abortive disconnect, like perhaps the decode_error code?
>> 
>>    decode_error:  A message could not be decoded because some field was
>>       out of the specified range or the length of the message was
>>       incorrect.  This alert is used for errors where the message does
>>       not conform to the formal protocol syntax.  This alert should
>>       never be observed in communication between proper implementations,
>>       except when messages were corrupted in the network.
>> 
>> Or are these TLS error alerts reserved for TLS-layer error conditions?
>> 
>> Stuart Cheshire
>>