From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756030Ab3AWOW2 (ORCPT ); Wed, 23 Jan 2013 09:22:28 -0500 Received: from mail-qa0-f43.google.com ([209.85.216.43]:47666 "EHLO mail-qa0-f43.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755924Ab3AWOWX (ORCPT ); Wed, 23 Jan 2013 09:22:23 -0500 Message-ID: <50FFF218.1010204@gmail.com> Date: Wed, 23 Jan 2013 09:22:16 -0500 From: Vlad Yasevich User-Agent: Mozilla/5.0 (X11; Linux i686; rv:17.0) Gecko/17.0 Thunderbird/17.0 MIME-Version: 1.0 To: Neil Horman CC: xufengzhang.main@gmail.com, sri@us.ibm.com, davem@davemloft.net, linux-sctp@vger.kernel.org, netdev@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] sctp: set association state to established in dupcook_a handler References: <1358926720-25557-1-git-send-email-xufengzhang.main@gmail.com> <20130123134650.GA3512@hmsreliant.think-freely.org> In-Reply-To: <20130123134650.GA3512@hmsreliant.think-freely.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 01/23/2013 08:46 AM, Neil Horman wrote: > On Wed, Jan 23, 2013 at 03:38:40PM +0800, xufengzhang.main@gmail.com wrote: >> From: Xufeng Zhang >> >> While sctp handling a duplicate COOKIE-ECHO and the action is >> 'Association restart', sctp_sf_do_dupcook_a() will processing >> the unexpected COOKIE-ECHO for peer restart, but it does not set >> the association state to SCTP_STATE_ESTABLISHED, so the association >> could stuck in SCTP_STATE_SHUTDOWN_PENDING state forever. >> This violates the sctp specification: >> RFC 4960 5.2.4. Handle a COOKIE ECHO when a TCB Exists >> Action >> A) In this case, the peer may have restarted. ..... >> After this, the endpoint shall enter the ESTABLISHED state. >> >> Fix this problem by adding a SCTP_CMD_NEW_STATE cmd to the command >> list so as to set the restart association to SCTP_STATE_ESTABLISHED >> state properly. >> >> Signed-off-by: Xufeng Zhang >> --- >> net/sctp/sm_statefuns.c | 2 ++ >> 1 files changed, 2 insertions(+), 0 deletions(-) >> >> diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c >> index 618ec7e..528f1c8 100644 >> --- a/net/sctp/sm_statefuns.c >> +++ b/net/sctp/sm_statefuns.c >> @@ -1779,6 +1779,8 @@ static sctp_disposition_t sctp_sf_do_dupcook_a(struct net *net, >> >> /* Update the content of current association. */ >> sctp_add_cmd_sf(commands, SCTP_CMD_UPDATE_ASSOC, SCTP_ASOC(new_asoc)); >> + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, >> + SCTP_STATE(SCTP_STATE_ESTABLISHED)); >> sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl)); >> sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, SCTP_ULPEVENT(ev)); >> return SCTP_DISPOSITION_CONSUME; >> -- >> 1.7.0.2 >> >> > > Looks reasonable to me, thanks > > nit: The RFC indicate the association should enter the ESTABLISHED state after > preforming all other actions, so it seems that the state change should occur > after the ULP event is sent > I have a slight concern here that if we change state last, then any data that may have been bundled with COOKIE-ACK as part of CMD_REPLY will get the SACK_IMMEDIATE flag set since we are still in the SHUTDOWN_PENDING state. I would be more correct (and would match sctp_sf_do_5_1D_ce) to do it in this order: UPDATE_ASSOC - resets all the congestion/association variables EVENT_UP - send RESTART to USER NEW_STATE - set ESTABLISHED state (as per spec) REPLY - Send cookie-ack along with any pending data. -vlad > Neil >