From: Kobby Carmona <kobby.Carmona@qlogic.com>
To: "tcpm@ietf.org" <tcpm@ietf.org>
Thread-Topic: Possible deadlock scenario with retransmission on both sides at
 the same time
Thread-Index: AdHth2zqZmjYio0mRrKVRRoD8V9Fcg==
Date: Thu, 4 Aug 2016 09:08:58 +0000
Message-ID: <MWHPR11MB1374A50BC599B093EA09668984070@MWHPR11MB1374.namprd11.prod.outlook.com>
Accept-Language: en-US
Content-Language: en-US
received-spf: None (protection.outlook.com: qlogic.com does not designate
 permitted sender hosts)
spamdiagnosticoutput: 1:99
spamdiagnosticmetadata: NSPM
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-originalarrivaltime: 04 Aug 2016 09:08:58.4051 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 0d68a1f9-1490-4d0e-8767-a87dab3ef2ba
X-MS-Exchange-Transport-CrossTenantHeadersStamped: MWHPR11MB1376
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/Y-64mSrPhdWYusLKMZLnoZFiUWM>
X-Mailman-Approved-At: Thu, 04 Aug 2016 10:43:38 -0700
Subject: [tcpm] Possible deadlock scenario with retransmission on both sides
 at the same time
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>,
 <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>,
 <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 04 Aug 2016 09:10:16 -0000

Hi all,
While running a bidirectional scenario with random drops in a network simul=
ator of our (QLogic's NIC) TCP stack we found a case where it seems there i=
s deadlock in the TCP protocol (the connection will keep sending pure acks =
from both sides until RTO will expire multiple times and a RST will sent to=
 close the connection).
The scenario is as follows (there is an example with numbers for each stage=
 assuming the MSS and each packet is 1000B):
1. Both sides are transmitting data and a single packet is dropped on eithe=
r side and the next two packets are received properly
	Side A - SND.MAX=3D3000, SND.NXT=3D3000, SND.UNA=3D1000, RCV.NXT=3D11000, =
out-of-order block 12000-13000
	Side B - SND.MAX =3D13000, SND.NXT =3D13000, SND.UNA=3D11000, RCV.NXT=3D10=
00, out-of-order block 2000-3000
2. RTO timer expires on both sides
	Side A - SND.MAX=3D3000, SND.NXT=3D1000, SND.UNA=3D1000, RCV.NXT=3D11000, =
out-of-order block 12000-13000
	Side B - SND.MAX =3D13000, SND.NXT=3D11000, SND.UNA=3D11000, RCV.NXT=3D100=
0, out-of-order block 2000-3000
3. Both sides transmit a single packet to the peer:
	A->B - pkt.seq=3D1000, pkt.ack=3D11000, len=3D1000
	B->A - pkt.seq=3D11000, pkt.ack=3D1000, len=3D1000
3. Both sides receive the packets and update the receive context:
	Side A - SND.MAX=3D3000, SND.NXT=3D2000, SND.UNA=3D1000, RCV.NXT=3D13000
	Side B - SND.MAX=3D13000, SND.NXT=3D12000, SND.UNA=3D11000, RCV.NXT=3D3000
4. Both sides send another segment:
	A->B - pkt.seq=3D2000, pkt.ack=3D13000, len=3D1000
	B->A - pkt.seq=3D12000, pkt.ack=3D3000, len=3D1000
5. Both sides don't accept the packet (and don't update SND.UNA) since the =
sequence on the packet is less than RCV.NXT (sequence number check in page =
69 of RFC793) and send a pure ACK instead
	A->B - pkt.seq=3D2000, pkt.ack=3D13000, len=3D0 (pure ACK)
	B->A - pkt.seq=3D12000, pkt.ack=3D3000, len=3D0 (pure ACK)
6. This will continue forever (until the connection will be terminated by R=
ST) since every packet that ends before RCV.NXT (even a retransmit from SND=
.UNA) will be dropped.

Did anyone encountered this issue before? Is the anything we missed on this=
 sequence?
If this is indeed a real deadlock, there might be several solutions to this=
 which will require a modification in receive processing of RFC793. But I w=
ould like to know if you think this is a real issue before dealing with sol=
utions.
Thanks,

	Kobby