[OASIS Issue Tracker] (MQTT-306) Clarify Handling of Malformed DISCONNECT command in WD04

[OASIS Issues Tracker <workgroup_mailer@lists.oasis-open.org>]

    [ https://issues.oasis-open.org/browse/MQTT-306?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=63551#comment-63551 ] 

Ken Borgendale commented on MQTT-306:
-------------------------------------

I do not understand the requirement for this.  After sending a DISCONNECT the sender is required to close the connection.  We certainly could in any case have DISCONNECT packets are flowing in both directions, but this does not cause any problems.

I suggest we reject this JIRA

> Clarify Handling of Malformed DISCONNECT command in WD04
> --------------------------------------------------------
>
>                 Key: MQTT-306
>                 URL: https://issues.oasis-open.org/browse/MQTT-306
>             Project: OASIS Message Queuing Telemetry Transport (MQTT) TC
>          Issue Type: New Feature
>          Components: core
>    Affects Versions: wd04
>            Reporter: Ed Briggs
>              Labels: Proposed
>
> Proposed textual addition to WD04 following review of MQTT-270 (Server Initiated Disconnect) which has been closed.
> WD04 Section 3.14.1 describes the DISCONNECT command packet fixed header.   Line 1798 says
> "The Client or Server MUST validate that the reserved bits are set to zero and close the Network Connection if they are not zero."
> As a precaution I propose we should add
> "A Client or Server MUST NOT send a DISCONNECT in response to a malformed DISCONNECT."
> Rationale:  The DISCONNECT error table in table 3.13 includes values for  MALFORMED CONTROL PACKET and PROTOCOL ERROR.  Sending a DISCONNECT in response to a malformed DISCONNECT introduces the possibility of an endless cycle. (i.e. the two session partners disagree on the format of the disconnect and reject each others disconnects.) 



--
This message was sent by Atlassian JIRA
(v6.2.2#6258)