On Tue, Dec 15, 2020 at 10:57 AM Markku Kojo <kojo@xxxxxxxxxxxxxx>
wrote:
Hi Neal, all,
(my apologies for this message getting so long but I didn't
want to drop
any parts of it in case someone wants to get the entire
discussion in one
piece. Also, apologies for repeating some of my concerns
inline below;
Those who are clear with the problems may well opt skipping
those parts.)
Thanks Neal for the reply. I believe we are slowly
converging with many
of my comments but two of the important issues still remain
open. That is,
where can an implementer find advice for correct congestion
control
actions with RACK-TLP, when:
(1) a loss of rexmitted segment is detected
(2) an entire flight of data gets dropped (and detected),
that is, when there is no feedback available and a
timeout
is needed to detect the loss
I fully understand the intent of the RACK-TLP draft not to
include
congestion control actions in this document (or the intent
to minimise
such advise). However, when new ways to detect loss are
introduced and
published (as this document would do), IMHO it is
imperative to have such
congestion control actions readily available for an
implementer. AFAIK
we do not have in the RFC series correct actions available
for these two
cases.
I do not have a strong opinion whether the necessary
actions should be in
this document or in another document, as long as we have
them. This is
more or less a process decision, I think.
Please see more details inline.
I also investigated a bit more on RACK-TLP as specified in
the draft and
it seems that there is a problem with the pseudocode as
currently
presented (some sequence numbers seem to be off by one,
resulting in
wrong outcome in some/many? cases). Please see inline in
the end of my
reply.
On Sat, 12 Dec 2020, Neal Cardwell wrote:
> Hi Markku,
>
> Thanks for your many detailed and thoughtful comments.
Please see
> below in-line for our comments based on discussions among
RACK-TLP
> draft authors.
>
> On Tue, Dec 8, 2020 at 11:21 AM Markku Kojo
<kojo@xxxxxxxxxxxxxx> wrote:
>>
>> Hi,
>>
>> please see inline.
>>
>> On Mon, 7 Dec 2020, Yuchung Cheng wrote:
>>
>>> On Mon, Dec 7, 2020 at 8:06 AM Markku Kojo
<kojo@xxxxxxxxxxxxxx> wrote:
>>>>
>>>> Hi Yuchung,
>>>>
>>>> thanks for your reply. My major point is that IMHO
this RACK-TLP
>>>> specification should give the necessary advice w.r.t
congestion control
>>>> in cases when such advice is not available in the RFC
series or is not
>>>> that easy to interpreted correctly from the existing
RFCs.
>>>>
>>>> Even though the CC principles were available in RFC
series I believe you
>>>> agree with me that getting detailed CC actions correct
is sometimes quite
>>>> hard, especially if one needs to interpret and apply
them from another
>>>> spec than what one is implementing. Please see more
details inline below.
>>>>
>>>> In addition, we need to remember that this document is
not meant only for
>>>> TCP experts following the tcpm list and having deep
understanding of
>>>> congestion control but also those implementing TCP,
for the very first
>>>> time, for various appliances, for example. They do not
have first hand
>>>> experience in implementing TCP congestion control and
deserve clear
>>>> advice what to do.
>>>>
>>>> On Fri, 4 Dec 2020, Yuchung Cheng wrote:
>>>>
>>>>> On Fri, Dec 4, 2020 at 5:02 AM Markku Kojo
<kojo@xxxxxxxxxxxxxx> wrote:
>>>>>>
>>>>>> Hi all,
>>>>>>
>>>>>> I know this is a bit late but I didn't have time
earlier to take look at
>>>>>> this draft.
>>>>>>
>>>>>> Given that this RFC to be is standards track and
RECOMMENDED to replace
>>>>>> current DupAck-based loss detection, it is important
that the spec is
>>>>>> clear on its advice to those implementing it.
Current text seems to
>>>>>> lack important advice w.r.t congestion control, and
even though
>>>>>> the spec tries to decouple loss detection from
congestion control
>>>>>> and does not intend to modify existing standard
congestion control
>>>>>> some of the examples advice incorrect congestion
control actions.
>>>>>> Therefore, I think it is worth to correct the
mistakes and take
>>>>>> yet another look at a few implications of this
specification.
>>>>> As you noted, the intention is to decouple the two as
much as possible
>>>>>
>>>>> Unlike the 20 years ago where TCP loss detection and
congestion
>>>>> control are essentially glued in one piece, the
decoupling of the two
>>>>> (including modularizing congestion controls in
implementations) has
>>>>> helped fueled many great inventions of new congestion
controls.
>>>>> Codifying so-called-default C.C. reactions in the
loss detection is a
>>>>> step backward that the authors try their best to
avoid.
>>>>
>>>> While I fully agree with the general principle of
decoupling loss
>>>> detection from congestion control when it is possible
without leaving
>>>> open questions, I find it hard to get congestion
control right with this
>>>> spec in certain cases I raised just by following the
current standards
>>>> track CC specifications. The reason for this is that
RACK-TLP introduces
>>>> new ways to detect loss (i.e., not present in any
earlier standard track
>>>> RFC) and the current CC specifications do not provide
correct CC actions
>>>> for such cases as I try to point out below.
>>>>
>>>>> To keep the
>>>>> document less "abstract / unclear" as many WGLC
reviewers commented,
>>>>> we use examples to illustrate that includes CC
actions. But the
>>>>> details of these CC actions are likely to become
obsolete as CC
>>>>> continues to advance hopefully.
>>>>
>>>> Agreed. But I would appreciate if the CC actions in
the examples would
>>>> correctly follow what is specified in the the current
CC RFCs. And, I
>>>> would suggest explicitly citing the RFC(s) that each
of the examples is
>>>> illustriating so that there is no doubt which CC
variant the example
>>>> is valid with. Then there is no problem with the
correctness of the
>>>> example either even if the cited RFC becomes later
obsoleted.
>>>>
>>>>>>
>>>>>> Sec. 3.4 (and elsewhere when discussing recovering a
dropped
>>>>>> retransmission):
>>>>>>
>>>>>> It is very useful that RACK-TLP allows for
recovering dropped rexmits.
>>>>>> However, it seems that the spec ignores the fact
that loss of a
>>>>>> retransmission is a loss in a successive window that
requires reacting
>>>>>> to congestion twice as per RFC 5681. This advice
must be included in
>>>>>> the specification because with RACK-TLP recovery of
dropped rexmit
>>>>>> takes place during the fast recovery which is very
different
>>>>>> from the other standard algorithms and therefore
easy to miss
>>>>>> when implementing this spec.
>>>>>
>>>>> per RFC5681 sec 4.3
https://tools.ietf.org/html/rfc5681#section-4.3
>>>>> "Loss in two successive windows of data, or the loss
of a
>>>>> retransmission, should be taken as two indications
of congestion and,
>>>>> therefore, cwnd (and ssthresh) MUST be lowered
twice in this case."
>>>>>
>>>>> RACK-TLP is a loss detection algorithm. RFC5681 is
crystal clear on
>>>>> this so I am not sure what clause you suggest to add
to RACK-TLP.
>>>>
>>>> Right, this is the CC *principle* in RFC 5681 I am
refering to but I am
>>>> afraid it is not enough to help one to correctly
implement such lowering
>>>> of cwnd (and ssthresh) twice when a loss of a
retransmission is detected
>>>> during Fast Recovery. Nor do RFCs clearly advice
*when* this reduction
>>>> must take place.
>>>>
>>>> Congestion control principles tell us that congestion
must be reacted
>>>> immediately when detected. But at the same time,
standards track CC
>>>> specifications react to congestion only once during
Fast Recovery
>>>> because the losses in the current window, which Fast
Recovery repairs,
>>>> occured during the same RTT. That is, the current CC
specifications do
>>>> not handle lost rexmits during Fast Recovery but,
instead, the correct CC
>>>> reaction to a loss of a rexmit is automatically
achieved by those
>>>> specifications via RTO when cwnd is explicitly reset
upon RTO.
>>>>
>>>> Furthermore, the problem I am trying to point out is
that there is no
>>>> correct rule/formula available in the standards track
RFCs that would
>>>> give the correct way to reduce cwnd when the loss of
rexmit is detected
>>>> with RACK-TLP.
>>>>
>>>> I suggest everyone reading this message stops reading
at this point
>>>> for a while before continuing reading and figures out
themselves what
>>>> they think would be the correct equation to use in the
standards RFC
>>>> series to find the new halved value for cwnd when
RACK-TLP detects a loss
>>>> of a rexmit during Fast Recovery. I would appreciate a
comment on the
>>>> tcpm list from those who think they found the correct
answer immediately
>>>> and easily, and what was the formula to use.
>>>>
>>>> ...
>>>>
>>>> I think the best advice one may find by looking at RFC
6675 (and RFC
>>>> 5681) is to set
>>>>
>>>> ssthresh = cwnd = (FlightSize / 2) (RFC 6675, Sec 5,
algorithm step 4.2)
>>>>
>>>> Now, let's modify the example in Sec 3.4 of the draft:
>>>>
>>>> 1. Send P0, P1, P2, ..., P20
>>>> [Assume P1, ..., P20 dropped by network]
>>>>
>>>> 2. Receive P0, ACK P0
>>>> 3a. 2RTTs after (2), TLP timer fires
>>>> 3b. TLP: retransmits P20
>>>> ...
>>>> 5a. Receive SACK for P20
>>>> 5b. RACK: marks P1, ..., P20 lost
>>>> set cwnd=ssthresh=FlightSize/2=10
>>>> 5c. Retransmit P1, P2 (and some more depending on how
CC implemented)
>>>> [P1 retransmission dropped by network]
>>>>
>>>> Receive SACK P2 & P3
>>>> 7a. RACK: marks P1 retransmission lost
>>>> As per RFC 6675: set
cwnd=ssthresh=FlightSize/2=20/2=10
>>>> 7b. Retransmit P1
>>>> ...
>>>>
>>>> So, is the new value of the cwnd (10MSS) correct and
halved twice? If not,
>>>> where is the correct formula to do it?
>>>>
>>>> Before RFC 5681 was published we had a discussion on
FlighSize and that
>>>> during Fast Recovery it does not reflect the amount of
segments in
>>>> flight correctly but may also have a too large value.
It was decided not
>>>> to try to correct it because it only has an impact
when RTO fires during
>>>> Fast Recovery and in such a case cwnd is reset to 1
MSS. Having too large
>>>> ssthresh for RTO recovery in some cases was not
considered that bad
>>>> because a TCP sender anyway takes to most conservative
CC action with the
>>>> cwnd and would slow start from cwnd = 1 MSS. But now
when RACK-TLP enables
>>>> detecting loss of a rexmit during Fast Recovery we
have an unresolved
>>>> problem.
>>>
>>>> 1. Send P0, P1, P2, ..., P20
>>>> [Assume P1, ..., P20 dropped by network]
>>>>
>>>> 2. Receive P0, ACK P0
>>>> 3a. 2RTTs after (2), TLP timer fires
>>>> 3b. TLP: retransmits P20
>>>> ...
>>>> 5a. Receive SACK for P20
>>>> 5b. RACK: marks P1, ..., P20 lost
>>>> set cwnd=ssthresh=FlightSize/2=10
>>>> 5c. Retransmit P1, P2 (and some more depending on how
CC implemented)
>>>> [P1 retransmission dropped by network]
>>>>
>>>> Receive SACK P2 & P3
>>>> 7a. RACK: marks P1 retransmission lost
>>>> As per RFC 6675: set
cwnd=ssthresh=FlightSize/2=20/2=10
>>>> 7b. Retransmit P1
>>>
>>> To account for your points, IMO clearly stating the
existing CC RFC
>>> interactions w/o mandating any C.C. actions are the
best way to move
>>> forward. Here are the text diff I proposed:
>>>
>>> "Figure 1, above, illustrates ...
>>> Notice a subtle interaction with existing congestion
control actions
>>> on event 7a. It essentially starts another new episode
of congestion
>>> due to the detection of lost retransmission. Per
RFC5681 (section 4.3)
>>> that loss in two successive windows of data, or the
loss of a
>>> retransmission, should be taken as two indications of
congestion as a
>>> principle. But RFC6675 that introduces the pipe concept
does not
>>> specify such a second window reduction. This document
reminds RACK-TLP
>>
>> This is not quite correct characterization of RFC 6675.
>> RFC 6675 does not repeat all guidelines of RFC 5681. RFC
6675 DOES specify
>> the two indications of congestion implicitly by clearly
stating in the
>> intro that it follows the guidelines set in RFC 5681.
>> And RFC 5681 articulates a set of MUSTs which ALL
alternative loss
>> recovery algorithms MUST follow in order to become RFCs.
>>
>> (*) RFC 5681, Sec 4.3 (pp. 12-13):
>>
>> "While this document does not standardize any of the
>> specific algorithms that may improve fast
retransmit/fast recovery,
>> ...
>> That is, ...
>> Loss in two successive windows of data, or the loss
of a
>> retransmission, should be taken as two indications of
congestion and,
>> therefore, cwnd (and ssthresh) MUST be lowered twice
in this case."
>>
>>
>>> implementation to carefully consider the new multiple
congestion
>>> episode cases in the corresponding congestion control."
>>
>> I am sorry to say that this is very fuzzy and leaves an
implementer
>> all alone what to do.
>>
>> More inportantly, AFAIK we have no discussion nor
consensus in the IETF
>> on what is the correct CC action when lost rexmit is
detected.
>> Should one reset cwnd like current CCs do, or would
"halving again" be
>> fine, or something else? IMHO this requires
experimentation to decide.
>>
>> I assumed there is an implementation of RACK-TLP with
corresponding CC
>> actions and experimental results to devise what are the
implications of
>> the selected CC action(s) with various levels of
congestion. But it seems
>> we do not have such implementation nor experimental
evidence? Am I wrong?
>
> [RACK-TLP-team:]
>
> We do have such implementations of RACK-TLP with
corresponding CC
> actions, and experimental results were reported in
RACK-TLP
> presentations and earlier revisions of the RACK-TLP
draft. With
Yes, that was my understanding what is implemented. Given
there is this
implementation that reacts appropriately to congestion when
RACK-TLP
detects loss of a retransmitted segment, could you please
illuminate the
congestion control actions taken in the implementation with
which there
is experimental experience.
If I understood it correctly, the RACK-TLP in question is
implemented
with PRR. Let's assume another implementer wants to add
RACK-TLP with PRR
in the stack she/he is working on. I believe that the
implementer would
think that detecting a loss of a rexmit during fast
recovery would start
a new congestion episode in PRR (and that's what Yuchung
also indicated
in the new proposed text in his previeous reply shown
above). With
that advice and reading RFC 6937, an implementer would
reinitialize PRR
state for the new congestion episode. So how should the
implementer
calculate new values for ssthresh and RecoverFS? Should
she/he use what
is currently described in RFC 6937, Sec 3 (to keep
discussion simple
assume the implementer wants to have an IETF Standards
Track comformant
implementation, so use CongCtrlAlg() = FlightSize/2):
ssthresh = CongCtrlAlg() // Target cwnd after recovery
prr_delivered = 0 // Total bytes delivered during
recovery
prr_out = 0 // Total bytes sent during
recovery
RecoverFS = snd.nxt-snd.una // FlightSize at the start of
recovery
Or should one use something else? AFAIK depending on the
loss pattern
neither FlightSize nor "snd.nxt-snd.una" would give correct
values for
the intended behavior. Depending on the loss pattern, at
the time
of detecting a loss of rexmit FlighSize may be ~ 50% larger
than in the
beginning of fast recovery. This would result in increasing
ssthresh
(send rate), instead of decreasing it. It seems that
something else in
PRR algo would also need to be modified to get sndcnt
correct but I
cannot figure it out right now.
If one wants to implement RACK-TLP without PRR, that is,
with RFC 6675,
we have the same probelm with FlightSize in determining a
new value for
cwnd and ssthresh.
My point is that using what we currently have out there in
the RFC
series gives incorrect CC advice. And, having correct CC
actions in this
case is far from trivial so IMHO high quality standards
specifications
should not leave an implementer of RACK-TLP alone to figure
it out.
In addition, if a new episode of fast recovery is started
for CC purposes
how would that play together with RACK-TLP loss detection?
In Sec 7.1 the
text says initialize (reset) TLP.end_seq and TLP.is_retrans
when
initiating fast recovery. But detecting a lost rexmit
seemingly should
not reinitialize TLP because PTO is disabled during fast
recovery as per
Sec 7.2. If CC actions are in another document and loss
detection in this
document, it would be of high importance to clarify when
each of the
fast recovery episodies (CC vs. loss detection) are
supposed to end: is
the end condition the same or different?
> respect to implementations, there is the Linux TCP stack,
which has
> been using RACK-TLP as the default loss recovery
algorithm since Linux
> v4.18 in August 2018. The exact commit is:
>
> b38a51fec1c1 tcp: disable RFC6675 loss detection
>
> With Linux, the default sender behavior since that commit
has been
> governed by CUBIC, PRR, and RACK-TLP.
>
> RACK-TLP is used by all Google TCP traffic, including
internal RPC
> traffic and external YouTube and google.com traffic over
the public
> Internet. At Google we have years of high-traffic-volume
experience
> with RACK-TLP with three different CC schemes: CUBIC+PRR,
BBRv1, and
> BBRv2.
>
> My understanding is that Netflix FreeBSD TCP traffic uses
RACK-TLP,
> and Microsoft Windows TCP traffic has also been using
RACK-TLP.
And I wonder what would be the CC actions taken in each of
these
implementations when a loss of rexmited segment is
detected?
>> I sincerely apologize that this got raised this late in
the prosess and
>> I know how irritating it may be. I like the idea of
RACK-TLP and by no
>> means my intention is not to hold up the process but the
lack of evidence
>> makes me quite concerned. In particular, when this
document says that
>> the current loss detection SHOULD be replaced in all
implementations with
>> RACK-TLP.
>>
>>> and to emphasize in section 9.3 Interaction with
congestion control
>>>
>>> "9.3. Interaction with congestion control
>>>
>>> RACK-TLP intentionally decouples loss detection ...
this appropriate.
>>> As mentioned in Figure 1 caption, RFC5681 mandates a
principle that
>>> Loss in two successive windows of data, or the loss of
a
>>> retransmission, should be taken as two indications of
congestion, and
>>> therefore reacted separately. However implementation of
RFC6675 pipe
>>> algorithm may not directly account for this newly
detected congestion
>>> events properly. Therefore the documents reminds
RACK-TLP
>>> implementation to carefully consider these implications
in its
>>> corresponding congestion control.
>>>
>>> ..."
>>>
>>>
>>>
>>>>
>>>>
>>>>>> Sec 9.3:
>>>>>>
>>>>>> In Section 9.3 it is stated that the only
modification to the existing
>>>>>> congestion control algorithms is that one
outstanding loss probe
>>>>>> can be sent even if the congestion window is fully
used. This is
>>>>>> fine, but the spec lacks the advice that if a new
data segment is sent
>>>>>> this extra segment MUST NOT be included when
calculating the new value
>>>>>> of ssthresh as per the equation (4) of RFC 5681.
Such segment is an
>>>>>> extra segment not allowed by cwnd, so it must be
excluded from
>>>>>> FlightSize, if the TLP probe detects loss or if
there is no ack
>>>>>> and RTO is needed to trigger loss recovery.
>>>>>
>>>>> Why exclude TLP (or any data) from FlightSize? The
congestion control
>>>>> needs precise accounting of the flight size to react
to congestion
>>>>> properly.
>>>>
>>>> Because FlightSize does not always reflect the correct
amount of data
>>>> allowed by cwnd. When a TCP sender is not already in
loss recovery and
>>>> it detects loss, this loss indicates the congestion
point for the TCP
>>>> sender, i.e., how much data it can have outstanding.
It is this amount of
>>>> data that it must use in calculating the new value of
cwnd (and
>>>> ssthresh), so it must not include any data sent beyond
the congestion
>>>> point. When TLP sends a new data segment, it is beyond
the congestion
>>>> point and must not be included. Same holds for the
segments sent via
>>>> Limited Transmit: they are allowed to be send out by
the packet
>>>> conservation rule (DupAck indicates a pkt has left the
network, but does
>>>> not allow increasing cwnd), i.e., the actual amount of
data in flight
>>>> remains the same.
>>>>
>>>>>> In these cases the temporary over-commit is not
accounted for as DupAck
>>>>>> does not decrease FlightSize and in case of an RTO
the next ACK comes too
>>>>>> late. This is similar to the rule in RFC 5681 and
RFC 6675 that prohibits
>>>>>> including the segments transmitted via Limitid
Transmit in the
>>>>>> calculation of ssthresh.
>>>>>>
>>>>>> In Section 9.3 a few example scenarios are used to
illustriate the
>>>>>> intended operation of RACK-TLP.
>>>>>>
>>>>>> In the first example a sender has a congestion
window (cwnd) of 20
>>>>>> segments on a SACK-enabled connection. It sends
10 data segments
>>>>>> and all of them are lost.
>>>>>>
>>>>>> The text claims that without RACK-TLP the ending
cwnd would be 4 segments
>>>>>> due to congestion window validation. This is
incorrect.
>>>>>> As per RFC 7661 the sender MUST exit the
non-validated phase upon an
>>>>>> RTO. Therefore the ending cwnd would be 5 segments
(or 5 1/2 segments if
>>>>>> the TCP sender uses the equation (4) of RFC 5681).
>>>>>>
>>>>>> The operation with RACK-TLP would inevitably result
in congestion
>>>>>> collapse if RACK-TLP behaves as described in the
example because
>>>>>> it restores the previous cwnd of 10 segments after
the fast recovery
>>>>>> and would not react to congestion at all! I think
this is not the
>>>>>> intended behavior by this spec but a mistake in the
example.
>>>>>> The ssthresh calculated in the beginning of loss
recovery should
>>>>>> be 5 segments as per RFC 6675 (and RFC 5681).
>>>>> To clarify, would this text look more clear?
>>>>>
>>>>> 'an ending cwnd set to the slow start threshold of 5
segments (half of
>>>>> the original congestion window of 10 segments)'
>>>>
>>>> This is correct, but please replace:
>>>>
>>>> (half of the original congestion window of 10
segments)
>>>> -->
>>>> (half of> the original FlightSize of 10 segments)
>>>
>>> sure will do
>>>
>>>>
>>>> cwnd in the example was 20 segments.
>>>>
>>>> Please also correct the ending cwnd for "without RACK"
scenario.
>>>> I poined out wrong equation number IN RFC 5681 and
INCORRECT cwnd value,
>>>> my apologies. It MAENT equation (3) AND that results
in ending cwnd of 5
>>>> and 2/5 MSS (not 5 and 1/2 MSS).
>>>> NB: and if a TCP sender implements entering CA when
cwnd > ssthresh, then
>>>> ending cwnd would be 6 and 1/6 MSS).
>>>>
>>>>>>
>>>>>> Furthermore, it seems that this example with
RACK-TLP refers to using
>>>>>> PRR_SSRB which effectively implements regular slow
start in this
>>>>>> case(?). From congestion control point of view this
is correct because
>>>>>> the entire flight of data as well as ack clock was
lost.
>>>>>>
>>>>>> However, as correctly discussed in Sec 2, congestion
window must be reset
>>>>>> to 1 MSS when an entire flight of data is and Ack
clock is lost. But how
>>>>>> can an implementor know what to do if she/he is not
implementing the
>>>>>> experimental PRR algrorithm? This spec articulates
specifying an
>>>>>> alternative for DupAck counting, indicating that TLP
is used to trigger
>>>>>> Fast Retransmit & Fast Recovery only, not a loss
recovery in slow start.
>>>>>> This means that without an additional advise an
implementation of this
>>>>>> spec would just halve the cwnd and ssthresh and send
a potentially very
>>>>>> large burst of segments in the beginning of the Fast
Recovery because
>>>>>> there is no ack clock. So, this spec begs for an
advise (MUST) when to
>>>>>> slow start and reset cwnd and when not, or at least
a discussion of
>>>>>> this problem and some sort of advise what to do and
what to avoid.
>>>>>> And, maybe a recommendation to implement it with
PRR?
>>>>>
>>>>> It's wise to decouple loss detection (RACK-TLP) vs
congestion/burst
>>>>> control (when to slow-start). The use of PRR is just
an example to
>>>>> illustrate and not meant for a recommendation.
>>>>
>>>> I understand the use of PRR was just an example, but
my point is that if
>>>> one wants to implement RACK-TLP but does not intend to
implement PRR but
>>>> RFC 6675 then we do not have a rule in RFC 6675 to
correctly implement
>>>> CC for the case when an entire flight is lost and loss
is detected with
>>>> TLP. Congestion control principle for this is clear
and also stated in
>>>> this draft but IMHO it is not enough to ensure correct
implementation.
>>>>
>>>> To my understanding we only have implementation
experience for RACK-TLP
>>>> only togerher with PRR, which has the necessary rule
to handle this kind
>>>> of scenario correctly.
>>>>
>>>> So, my question is how can one implement CC correctly
without PRR such
>>>> a scenario where entire inflight is lost?
>>>> Which rule and where in the RFC series gives the
necessary guidance to
>>>> reset cwnd and slow start when TCP detets loss of an
entire flight?
>>>
>>> I think we're going in loops. To move forward it'd help
if you suggest
>>> some text you like to change.
>>
>> Unfortunately I do not have an immediate solution. I
assumed there is an
>> implementation and you could enlighten what was the
solution there and
>> experimental results showing how it seems to work. If I
have understood
>> it correctly there is a solution/implementation and
experience of RACK-TLP
>> that works together with PRR but no
solution/implementation nor experience
>> how it works without PRR. Am I correct?
>
> [RACK-TLP-team:]
>
> The implementation solutions and experiments were
mentioned above. Our
> team's experience is with RACK-TLP in combination with
either
> CUBIC+PRR, BBRv1, and BBRv2. We are not sure if there are
other TCP
> implementations that implement RACK-TLP without PRR.
So, in addition to the lack of experimental evidence using
RACK-TLP
without PRR we have a specification problem for which we
have no answer:
The RACK-TLP text suggests that if PTO fires and TLP
detects a loss, the
TCP sender enters fast recovery. But
PTO = timeout that expires if there is no feedback (ACKs)
from
the network, i.e., entire flight of data is lost.
TLP = a single segment that is sent when timer expires.
What is
different compared to RTO is that PTO may expire
faster
and a different segment is (re)transmitted, allowing
to
detect more than one lost segment with a single Ack
w/ SACK
triggered by TLP (with regular RTO recovery only the
next
unacknowledged segment can reliably be considered
lost)
>From congestion control point of view this means that PTO
is no different
from RTO. The congestion control principle as stated in Sec
2.2 of this
document tells us that a loss of entire flight deserves a
cautious
response, i.e., reset cwnd and slow start.
When RACK-TLP is implemented with PRR we have appropriate
CC actions
specified in RFC 6937 that effectively reset cwnd and force
the TCP sender
to slow start, even though it tecnically enters fast
recovery. This is
safe.
However, if an implementer wants to implement RACK-TLP
without PRR,
she/he has to consult RFC 6675 which gives an incorrect CC
response for
this case, resulting in an unsafe implementation. The
problem is not in
the fast recovery algorithm in RFC 6675 because it was not
intended to be
initiated with TLP-kind of loss detection.
So the question is how to ensure an implementer has the
necessary advice
to implement correct congestion control actions for
RACK-TLP once this
document is published. The lack of correct advice becomes
even more
prominent as this document requires (all SACK capable) TCPs
to implement
RACK-TLP.
>> Given my understanding of PRR, the problem of an entire
flight being
>> dropped is quite nicely solved with it. However,
implementing RACK-TLP
>> without PRR begs for a solution. Here is my current
understanding:
>>
>> (1) RACK part is what can be called to "replace" DupAck
counting
>> (2) TLP part is effectively a new retransmission timeout
to detect
>> a loss of an entire flight of data (i.e., it is
only invoked
>> when a tail of the current flight becomes lost
which equals to
>> the case of losing an entire flight since all
segments before
>> the tail loss will get cumulatively Acked and hence
these segments
>> are not anymore a part of the current flight at the
time the loss
>> is detected. And we an assume application limited
TCP sender.
>> (3) Current CC principles require resetting cwnd in such
a case
>> and entering slow start (and so effectively does
PRR though it is
>> not explicitly stated in RFC 6937). Slow start
avoids a big burst
>> if the lost flight is big.
>> (4) RACK-TLP would possibly like to allow that cwnd is
set to a half
>> of the lost flight and not to slow start. This
means that the bigger
>> the lost flight is, the bigger is the burst that
gets transmitted at
>> the beginning of the recovery which is bad. So,
this approach would
>> need at least a rule/advice for burst avoidance
(slow start, pacing,
>> ...).
>> (5) When the lost flight is small (<= 4 segments) there
is no difference
>> in recovery efficiency between (3) and (4). If the
lost flight is
>> > 5 segments, then (4) takes less RTTs to complete
the recovery
>> but generates a burst. Note, even if pacing over
one SRTT would be
>> used, it is still a burst.
>>
>> Now, it would be useful to have experimental data to
know how the size of
>> the lost flight is distributed. Is it typically just a
few segments as
>> illustriated in the examples of the draft or is it often
larger?
>>
>> My advice would be to add a rule that cwnd MUST be reset
to 1 MSS
>> and the sender MUST enter slow start if TLP detects loss
of an entire
>> flight. This would be safe. Otherwise, without
experimental evidence from
>> a wide range of different network conditions and
workloads it feels
>> unsafe to allow more aggressive approach.
>
> [RACK-TLP-team:]
>
> Our team's position is that the RACK-TLP draft documents
a loss
> detection algorithm, and normative declarations about
congestion
> control behavior (like your "cwnd MUST be reset to 1
MSS", above)
> should be in a separate document focused on congestion
control.
I do not disagree with you what becomes to the role of
congestion control
in the RACK-TLP draft. It may well be in another document
but IMHO such
document must be available at the time this docment gets
published.
How can we otherwise ensure correct implementations?
So this is more of a process question than critique towards
the draft.
>>>>> Section 3 has a lengthy section to elaborate the key
point of RACK-TLP
>>>>> is to maximize the chance of fast recovery. How C.C.
governs the
>>>>> transmission dynamics after losses are detected are
out of scope of
>>>>> this document in our authors' opinions.
>>>>>
>>>>>
>>>>>>
>>>>>> Another question relates to the use of TLP and
adjusting timer(s) upon
>>>>>> timeout. In the same example discussed above, it is
clear that PTO
>>>>>> that fires TLP is just a more aggressive retransmit
timer with
>>>>>> an alternative data segment to (re)transmit.
>>>>>>
>>>>>> Therefore, as per RFC 2914 (BCP 41), Sec 9.1, when
PTO expires, it is in
>>>>>> effect a retransmission timout and the timer(s) must
be backed-off.
>>>>>> This is not adviced in this specification. Whether
it is the TCP RTO
>>>>>> or PTO that should be backed-off is an open
question. Otherwise,
>>>>>> if the congestion is persistent and further
transmission are also lost,
>>>>>> RACK-TLP would not react to congestion properly but
would keep
>>>>>> retransmitting with "constant" timer value because
new RTT estimate
>>>>>> cannot be obtained.
>>>>>> On a buffer bloated and heavily congested bottleneck
this would easily
>>>>>> result in sending at least one unnecessary
retransmission per one
>>>>>> delivered segment which is not advisable (e.g., when
there are a huge
>>>>>> number of applications sharing a constrained
bottleneck and these
>>>>>> applications are sending only one (or a few)
segments and then
>>>>>> waiting for an reply from the peer before sending
another request).
>>>>>
>>>>> Thanks for pointing to the RFC. After TLP, RTO
timers will
>>>>> exp-backoff (as usual) for stability reasons
mentioned in sec 9.3
>>>>> (didn't find 9.1 relevant).
>>>>
>>>> My apologies for refering to the wrong section of RFC
2914, Yes, I meant
>>>> Sec 9.3.
>>>>
>>>>> In your scenario, you presuppose the
>>>>> retransmission is unnecessary so obviously TLP is not
good. Consider
>>>>> what happens without TLP where all the senders fire
RTO spuriously and
>>>>> blow up the network. It is equally unfortunate
behavior. "bdp
>>>>> insufficient of many flows" is a congestion control
problem
>>>>
>>>> If (without TLP) RTO is spurious, it may result in
unnecessary
>>>> retransmissions. But we have F-RTO (RFC 5682) and
Eifel (RFC 3522) to
>>>> detect and resolve it without TLP, so I don't find it
as a problem.
>>>>
>>>> To clarify more, what I am concerned about. Think
about a scenario where a
>>>> (narrow) bottleneck becomes heavily congested by a
huge number of
>>>> competing senders such that the available capacity per
sender is less
>>>> than 1 segment (or << 1 MSS).
>>>> This is a situation that network first enters before
congestion collapse
>>>> gets realized. So, it is extremely important that all
CC and timer
>>>> mechanisms handle it properly. Regular TCP handles it
via RFC 6298 by
>>>> backing off RTO expotentially and keeping this
backed-off RTO until an
>>>> new ACK is received for new data. This saves RACK-TLP
from full congestion
>>>> collapse. But consider what happens: even though RTO
is backed off, each
>>>> time a TCP sender manages to get one segment through
(with cwnd = 1 MSS)
>>>> it always first arms PTO with more or less constant
value of 2*SRTT. If
>>>> the bottleneck is buffer bloated the actual RTT easily
exceeds 2*SRTT and
>>>> TLP becomes spurious. After a spurious TLP, RTO
expires (maybe more than
>>>> once before exponential back-off of RTO results in
large enough value)
>>>> and a new RTT sample is not received. So, SRTT remains
unchanged and even
>>>> if sometimes a new sample is received, SRTT gets very
slowly adjusted. As
>>>> a result, each TCP sender would keep on sending a
spurious TLP for each
>>>> new segment resulting in at least 50% of the packets
being unnecessary
>>>> rexmitted and the utilization of the bottleneck is <
50%. This would not
>>>> be a full congestion collapse but has unwanted
symptoms towards
>>>> congestion collapse (Note: there is no clear line for
the level of
>>>> reduction in delivery of useful data is considered as
congestion
>>>> collapse).
>>>
>>> AFAIK you are saying: under extreme congestion shared
by many short
>>> flows, RACK-TLP can cause more packet losses because of
the more
>>> aggressive PTO timer. I agree and can add this to the
"section 9.3".
>>
>> No, that was not what I meant. This happens simply when
enough flows are
>> competing on the same bottleneck. It may be long flows
or maybe
>> applications with more or less continuous request-reply
exhanges.
>> E.g., if the bottleneck bit rate is 1 Mbits/s, RTT is
500msecs and PMTU
>> is 1500B, then a bit over 40 simultaneous flows would
mean that the
>> bottleneck becomes fully utilized with the equal share
of roughly 1 MSS
>> per flow. With a quite typical bottleneck buffer size
roughly equaling to
>> BDP, about 80+ flows would fill up the buffer as well
and increase
>> RTT >= 1 secs. A buffer bloated bottleneck buffer would
mean even larger
>> RTT and allow more flows sharing the bottleneck without
loss.
>>
>> If the number of TCP flows is > 90 (or >> 90) RACK-TLP
would ends up
>> (almost) always unnecessarily retransmitting each new
segment once (PTO
>> being 1 sec or around 1 sec). RTO back-off after a few
rounds saves each
>> TCP sender from additional unnecessary rexmits but still
~ 50% of the
>> deivered packets are not making any useful progress.
>>
>> If the bottleneck buffer is small, it would result in
more losses as you
>> suggest but would not be that much of problem because
majority of the
>> unnecessary rexmits would not get delivered over the
bottleneck link.
>> Instead, they wuold get dropped at the bottleneck.
Unnecessary rexmits
>> would just create extra load on the poth before the
bottleneck (which of
>> course is not a non-problem either).
>
> [RACK-TLP-team:]
>
> If we are correctly understanding the scenario you are
outlining, the
> PTO would likely not be 1 sec in this case. In this
scenario, the RTT
> is around 1sec, so the PTO would be around 2*srtt, which
would be
> around 2 secs. With a 1sec RTT, and even considering
delayed ACKs, it
> seems very unlikely that a 2 sec PTO would cause a
significant number
> of unnecessary TLP probes.
Well. Roughly the first initiated 40 flows will have SRTT ~
0.5 sec. If
next ~ 40 flows arrive within an RTT or so they will get
RTT samples
roughly between 0.5 - 1.x secs. The rest 80 - N hundreds
flows will not
get an RTT sample and will use initial/default PTO = 1 sec
(unless using
timestamps). The larger the buffer bloat the worse things
get.
Once ~200 flows are competing over the bottleneck, RTT will
be well
beyond 2 secs and grows when more flows join, guaranteeing
spurious TLP
for each new segment (unless timestamps are in use).
> Perhaps your postulated PTO of 1 sec is coming from the
pseudocode
> path in TLP_calc_PTO() where, if SRTT is not available,
the PTO is 1
> second? However, this case where there is no SRTT only
happens if (a)
> the connection is *not* using TCP timestamps (enabled for
most
> connections on the public Internet, due to
default-enabled support in
> Linux/Android and iOS/MacOS), and (b) the connection has
*never* had a
> single RTT sample from a non-retransmitted packet. To
meet both of
> those conditions is quite rare in practice. Is this the
path that is
> causing your concerns about a PTO of 1 sec?
Sure. Having timestamps helps a lot in case the bottleneck
buffer is
buffer bloated (though RTT sample will not include delayed
Ack effect for
rexmits that are immediately acked). But RACK-TLP does not
require
timestamps and there are a lot of TCP stacks out there in
addition to
Linux/Android and iOS/MacOS. And more are coming with IoT
expansion,
for example. In addition, it is quite suprising how old
(mobile) devices
many people are using, some of which are old enough and
have not been
eligible for an OS upgrade for many years.
But my quick example scenario was just to illustriate the
problem, maybe
typical in some areas where the only Internet access is
over a
geostationary satellite connection and people quite likely
not using
mobile/cellular devices. And, I selected a buffer-bloated
scenario
because that generates unnecessary rexmits which are a
prerequisite for
classical congestion collapse.
It may well be in some environments that these hundreds of
competing
flows are application limited (request-reply continuously)
and get
started over a longer period of time, each getting RTT
sample(s) over a
lightly loaded bottleneck and then becoming idle or sending
data only
sporadically. Then later they all become active more or
less at the same
time (maybe by a common trigger). Base RTT may then be much
lower than
500 ms as long as there is enough buffer bloat.
Furthermore, timestamps are not that helpful in case of a
heavily
congested small buffer bottleneck. Having approximately
accurate SRTT
does not alone help in reducing the sending rate below 1
segment/RTT in
heavily congested scenarios. Exponential backoff of timer
is a must.
I believe that either of us know what's really out there
employing
Internet for communication purposes. It does not matter
whether it is a
notable percentage or not as long there is a group of users
who are
encountering the envisioned problem.
It may very well be a small minority, but if this kind of a
problem
escalates for this group very often (maybe almost always),
it means we
are disabling use of Internet technology for them.
> Furthermore, in cases like this where the number of flows
is roughly
> equal to or greater than the BDP of the path,
Reno/[RFC5681] and CUBIC
> will spend the majority of their time in loss recovery
(the minimum
> ssthresh from [RFC5681] being 2*SMSS).
Sure there is surprisingly lot to improve with TCP when the
window needs
to be kept lower than 2 MSS or when the available share
would be even
less than 1 MSS/RTT. RFC 5681 would exponentially back-off
on every RTO,
leaving opportunity for other flows to make progress during
the
backed-off interval. Once the sender is lucky and gets an
Ack for a new
segment, min ssthresh of 2*MSS unfortunately allows it to
accelerate to
2*MSS/RTT and maybe more until RTO expires again. We cannot
avoid some
drops or unnecessary rexmits. Actually, in such a heavily
congested
scenario the combination of Karn's exponential back-off and
slow start
results in Multiplicative Increase Multiplicative Decrease
(MIMD)
behaviorwhich does not converge that well to fairness. But
it avoids
congestion collapse. It is kind a tradeoff between safe and
stable
behavior during heavy congestion and being able to ramp up
fast when the
congestion fades away. Even though the current behavior is
not ideal
for fairness, we should not make it even worse.
Currently only way to avoid spending most of the time in
loss
recovery seems to be ECN that would make a TCP sender to
become rate
controlled once cwnd hits the minimum cwnd of 1 MSS(*). But
there is a
lot to optimise also with ECN in this kind of scenario ...
(*) to avoid anyone repeating the misconception that
minimum cwnd is
2 MSS per equation (4) in RFC 5681: equation (4) sets
ssthresh,
not cwnd. RFC 3168 clearly states that cwnd must be
halved on
an arrival of ECE; this holds until cwnd = 1 MSS.
After this,
the sending rate of new segments is (RTO) timer
controlled and
the timer avlue exponentially backed-off if more ECEs
arrive.
At least two major OSs implement this incorrectly
unless now
already fixed.
> And the RACK-TLP ID specifies
> that TLP probes are not sent while in loss recovery.
Maybe I missed it but I couldn't find an end condition of
RTO recovery
for RACK-TLP purposes in the document. Regular RTO does not
have such end
condition because it is not needed. Congestion control and
loss recovery
are quite decoupled during RTO recovery; slow start
follows its own
rules and ends when cwnd > ssthresh and loss recovery has
its own
rules to select segments to rexmit and just moves on
sending new data
once all lost segments have been retransmitted. Some
implementations may
have such variable/condition but it didn't help me.
Without timestamps, if TLP is not backed-off, a TLP is
likely to be sent
for each new segment (assuming certain conditions). Each
not backed-off
rexmit is potentially stealing an opportunity from another
flow to get
its segment being delivered, thereby contributing to unfair
use of
bottleneck resources.
>>> What authors disagree is that RTO must be back-off on
the first
>>> instance if TLP is not acked. While your suggestion
helps the
>>> congestion case, it may also hurt the recovery in other
cases when the
>>> TLP is dropped due to light/burst/transient congestion.
Arguing which
>>> scenarios matter more subjectively is not productive.
>>
>> When RACK-TLP is required to be implemented by all TCP
stacks, it
>> is extremely important that it always works in a safe
way and congestion
>> is always the primary concern to get properly handled.
Without arguing
>> more, I just want to point out that TCP must work
reasonably for all
>> Internet users. That means it must not generate a
situation where even a
>> small minority of the Internet users often or almost
always encounter
>> severe problems with their connectivity.
>>
>> For example, also users in developing countries where
possibly an entire
>> village shares just one mobile (cellular, maybe 3G
possibly only GPRS)
>> connection for their Internet access and pays per amount
of data should
>> get reasonable TCP behavior. In such a case, I cannot
agree with
>> engineering that results in almost always getting only
half of the
>> already scarce bandwidth whle paying double price fot
the useful data.
>>
>> But thinking about a way forward. Karn's algorithm would
require backing
>> off PTO even if you get Ack of PTO. Relaxing this does
not sound me that
>> bad at all, because there is often a pause before TLP
and a TCP sender
>> gets feedback, so apparently conditions are not that bad
and loss
>> recovery gets triggered (hopefully in slow start).
>> If TLP is not acked, RTO is needed and recovery is
completed in using RTO
>> recovery in slow start. Now, if the RTO recovery is
successful (no
>> losses during RTO recovery), it should be quite likely
that the TCP
>> sender is able to successfully send also one new
segment, because it
>> enters CA when it is sending at the half of the previous
rate. So, once
>> you get Ack for the new segment, the TLP back off can be
removed and it
>> is unlikely that the TLP back off slowed down next loss
detection.
>> On the other hand, if the RTO recovery after an
unsuccessful TLP is not
>> successful (more losses are detected), it is quite
likely that congestion
>> has not been resolved. So, it is important to be
conservative and have a
>> backed-off PTO (or even turn it off) to avoid (further)
unnecessary
>> rexmits. If PTO is not backed of, I'd envision PTO
mainly failing to get
>> an Ack in such a case and thereby it not being that
useful.
>>
>> This certainly would require experimental data in a
heavily congested
>> setting to really figure out the actual impact of
different alternatives.
>>
>>> So the question we need to look at is if RACK-TLP
2RTT-PTO + regular
>>> RTO-backoff is going to cause major stability issues in
extreme
>>> congestion. My understanding based on my officemate Van
Jacobson's
>>> explanation is, as long as there's eventual exponential
backoff, we'll
>>> avoid the repeated shelling the network.
>>
>> Right. But the problem is that with TLP as specified we
do not have full
>> exponential back off. It lacks Karn's clamped retransmit
backoff (as it
>> is called in Van Jacobson's seminal paper) which
requires keeping
>> backed-off timer of a retransmitted segment for the next
(new) segment and
>> ensures that there is eventual exponential backoff.
>> Backing off just RTO is not enough, because "fixed" PTO
for each new
>> segment breaks this "back-off chain", that is, the
exponential backoff
>> is not continueed until there is evidence that
congestion has been
>> resolved (a cumulative Ack arrives for new data). But as
I said, backing
>> off RTO saves RACK-TLP from a full congestion collapse.
Still, wasting
>> 50% of the available network capacity in certain usage
scenarios does not
>> sound acceptable for me.
>
> [RACK-TLP-team:]
>
> To address concerns related to Karn's algorithm, perhaps
it would be
> sufficient to specify that after a RACK-TLP sender sends
a TLP probe
> that is a retransmitted segment, it temporarily disables
further TLP
> retransmissions until the sender obtains a valid RTT
sample (i.e. from
> a non-retransmitted segment, or a segment with [RFC7323]
TCP
> timestamps)?
As a basic rule: disabling TLP probes for the period of
loss recovery as
currently specified sounds like a correct approach provided
that the end
condition
"until the sender obtains a valid RTT sample"
is added (this would be without timestamps, see some
considerations with
timestamps later below).
Then some justification and maybe some exeptions can also
be specified
(all these without timestamps):
If fast recovery is not entered via detecting loss using
TLP, the
sender does not need to wait for a valid RTT sample. It
should be enough
that RecoveryPoint is reached, i.e., use normal exit
condition for fast
recovery. However, I am not sure whether detecting a loss
of a rexmit
during fast recovery might change this?
I think it does not matter whether a TLP probe is a
retransmitted segment
or not:
(a) if there is no ack after TLP probe and RTO expires it
should be clear
that TLP must be disabled until an ack for a new previously
not rexmitted
segment is received, regardless of the probe segment sent
by TLP.
This is Karn's algorithm.
(b) if TLP triggers an Ack that is used to detect a loss of
an entire
flight of data, this may later during loss recovery result
in an RTO
which definitely requires disabling TLP as per the basic
rule. There
might be some exeptions when the basic rule is not
necessary but it is
hard for me to figure one right now, except
(c) if TLP rexmit detects and repairs a single loss, then
CC is anyway
invoked and Ack is received. If the arriving Ack is
triggered by the TLP
probe segment, it indicates that the conditions are likely
not that bad
(RTO not needed). So it might be ok, to rearm PTO for the
next time
without any backoff (without disbling it). However,
(d) if TLP was spurious rexmit, then PTO value obviously
was too low
(effective RTT may have increased suddenly and Acks got
just delayed).
This is possibly detected only one RTT later when the
DupAck of the TLP
rexmit arrives. TLP may have got rearmed by that time again
with too
low value, so it should be rearmed with doubled value to
avoid another
spurious TLP PTO (or disabled).
Then with timestamps. I think the simple rule for disabling
TLP "until
the sender obtains a valid RTT sample" is not necessarily
giving the
desired outcome in all cases. For example, if a TLP probe
is send and
then a late ack arrives for a segment sent before the TLP
probe and gives
a valid RTT sample with help of timestamps. This would
allow a new TLP
even though the sender has not got an Ack for a new segment
sent after
the TLP probe?
So, also with timestamps the requirement must be that the
sender obtains
a valid RTT sample for a segment sent after the TLP probe.
And, at least,
if RTO recovery is entered, then the Ack must be for a new
segment.
Cannot figure out now all potential exceptions when a valid
RTT sample
with timestamps would allow re-enabling TLP earlier. Too
complex
algorithm with too many possible scenarios to investigate
right now ;)
A single, simple rule would be advisable here. Otherwise,
this may end up
being too complex to get it right.
>>> As a matter of fact some
>>> major TCP (!= Linux) implementation has implemented
linear backoff of
>>> first N RTOs before exp-backoff.
>>
>> But that's quite ok as long as there is eventual
exponential backoff,
>> including Karn's clamped retransmit backoff. Linear
back-off in the
>> beginning just makes resolving (heavy) congestion a bit
slower.
>>
>>>>
>>>>>>
>>>>>> Additional notes:
>>>>>>
>>>>>> Sec 2.2:
>>>>>>
>>>>>> Example 2:
>>>>>> "Lost retransmissions cause a resort to RTO
recovery, since
>>>>>> DUPACK-counting does not detect the loss of the
retransmissions.
>>>>>> Then the slow start after RTO recovery could cause
burst losses
>>>>>> again that severely degrades performance
[POLICER16]."
>>>>>>
>>>>>> RTO reovery is done in slow start. The last sentence
is confusing as
>>>>>> there is no (new) slow-start after RTO recovery (or
more precisely
>>>>>> slow start continues until cwnd > ssthresh). Do you
mean: if/when slow
>>>>>> start still continues after RTO Recovery has
repaired lost segments,
>>>>>> it may cause burst losses again?
>>>>> I mean the slow start after (the start of) RTO
recovery. HTH
>>>>
>>>> Tnx. I'd appreciate if the text could be clarified to
reflect this more
>>>> accurately. Maybe something along the lines(?):
>>>>
>>>> "Then the RTO recovery in slow start could cause
burst
>>>> losses again that severely degrades performance
[POLICER16]."
>>>>
>>>>>>
>>>>>> Example 3:
>>>>>> "If the reordering degree is beyond DupThresh, the
DUPACK-
>>>>>> counting can cause a spurious fast recovery and
unnecessary
>>>>>> congestion window reduction. To mitigate the
issue, [RFC4653]
>>>>>> adjusts DupThresh to half of the inflight size to
tolerate the
>>>>>> higher degree of reordering. However if more
than half of the
>>>>>> inflight is lost, then the sender has to resort
to RTO recovery."
>>>>>>
>>>>>> This seems to be somewhat incorrect description of
TCP-NCR specified in
>>>>>> RFC 4653. TCP-NCR uses Extended Limited Transmit
that keeps on sending
>>>>>> new data segments on DupAcks that makes it likely to
avoid an RTO in
>>>>>> the given example scenario, if not too many of the
the new data
>>>>>> segments triggered by Extended Limited Transmit are
lost.
>>>>> sorry I don't see how the text is wrong describing
RFC4653,
>>>>> specifically the algorithm in adjusting ssthresh
>>>>
>>>> To my understanding RFC4653 initializes DupThresh to
half of the inflight
>>>> size in the beginning of the Extended Limited
Transmit. Then on each
>>>> DupAck it adjusts (recalculates) DupThresh again such
that ideally a cwnd
>>>> worth of DupAcks are received before packet loss is
declared (or
>>>> reordering detected). So, if I am not incorrect, loss
of a half of the
>>>> inflight does not necessarily result in RTO recovery
with TCP-NCR.
>>> Could you suggest the text you'd like on NCR
description.
>>
>> I'm not an expert nor closely acquainted with NCR. There
might be many
>> different packet loss patterns that may affect the
behavior. So, my
>> advice is to simply drop the last sentence starting with
"However ...",
>> because it seems incorrect and replace the second but
last sentence:
>>
>> To mitigate the issue, [RFC4653]
>> adjusts DupThresh to half of the inflight size to
tolerate the
>> higher degree of reordering.
>>
>> -->
>>
>> To mitigate the issue, TCP-NCR [RFC4653]
>> increases the DupThresh from the current fixed value
of three duplicate
>> ACKs [RFC5681] to approximately a congestion window
of data having left
>> the network.
>
> [RACK-TLP-team:]
>
> Sure, we will revise accordingly.
Tnx.
>
>>>>>>
>>>>>> Sec. 3.5:
>>>>>>
>>>>>> "For example, consider a simple case where one
>>>>>> segment was sent with an RTO of 1 second, and then
the application
>>>>>> writes more data, causing a second and third
segment to be sent right
>>>>>> before the RTO of the first segment expires.
Suppose only the first
>>>>>> segment is lost. Without RACK, upon RTO
expiration the sender marks
>>>>>> all three segments as lost and retransmits the
first segment. When
>>>>>> the sender receives the ACK that selectively
acknowledges the second
>>>>>> segment, the sender spuriously retransmits the
third segment."
>>>>>>
>>>>>> This seems incorrect. When the sender receives the
ACK that selectively
>>>>>> acknowledges the second segment, it is a DupAck as
per RFC 6675 and does
>>>>>> not increase cwnd and cwnd remains as 1 MSS and pipe
is 1 MSS. So, the
>>>>>> rexmit of the third segment is not allowad until the
cumulative ACK of
>>>>>> the first segment arrives.
>>>>> I don't see where RFC6675 forbids growing cwnd. Even
if it does, I
>>>>> don't think it's a good thing (in RTO-slow-start) as
DUPACK clearly
>>>>> indicates a delivery has been made.
>>>>
>>>> SACKed sequences with DUpAcks indicate that those
sequences were
>>>> delivered but it does not tell when they were sent.
The basic principle
>>>> of slow start is to reliably determine the available
network capacity
>>>> during slow start. Therefore, slow start must ensure
it uses only
>>>> segments sent during the slow start to increase cwnd.
Otherwise, a TCP
>>>> sender may encounter exactly the problem of
unnecessary retransmission
>>>> envisioned in this example of RACK-TCP draft (and
increase cwnd on not
>>>> valid Acks).
>>>>
>>>> RFC 6675 does re-specify DupAck with SACK option but
it does not include
>>>> the rule for slow start. Slow start is specified in
RFC 5681. It is
>>>> crystal clear in allowing increase in cwnd only on
cumulative
>>>> Acks, i.e., forbidding to increase cwnd on DupAcks
(RFC 5681, Sec 3,
>>>> page 6:
>>>>
>>>> During slow start, a TCP increments cwnd by at most
SMSS bytes for
>>>> each ACK received that cumulatively acknowledges
new data.
>>>>
>>>> Maybe this example in the RACK-TLP draft was inspired
by a incorrect
>>>> implementation of SACK-based loss recovery?
>>>>
>>>> FYI: when we were finalizing RFC 6675 I suggested
including also an
>>>> algorithm for RTO recovery with SACK in RFC 6675. The
reason was exactly
>>>> that it might be not easy to gather info from multiple
documents and
>>>> hence help the implementor to have all necessary
advice in a single
>>>> document. This unfortunately did not get realized
though.
>>>
>>> I am honestly lost quibbling these RFC6675
implementations and feeling
>>> pedantic to these standards largely serve as guiding
principles
>>> instead of line-by-line code. Is cwnd one packet bigger
or smaller in
>>> this example making any different in advancing the
Internet's
>>> capability to do better loss detection. I do not think
so.
>>
>> Sorry, I don't understand what was quibbling here. I
believe I did not
>> argue anything about cwnd size with this example at hand
in Sec. 3.5.
>> My point is that it describes incorrect slow start
behavior. Correctly
>> implemented slow start does not have the problem
illustriated in the
>> example.
>>
>>> At this point please suggest a text you like to change.
>>
>> That is simple. Please remove the description of the
behavior without
>> RACK. Or correct it: the behavior would be exactly the
same as with RACK,
>> but the reason for not unnecessarily retransmitting
third segment can be
>> described to be different.
>> And, please correct also the description with RACK. With
RACK, third
>> segment does not get unnecessarly rexmitted for the
reason already
>> indicated and also because cwnd=1 MSS. And, no new
segments are allowed
>> either by cwnd=1.
>>
>
> [RACK-TLP-team:]
>
> Thanks. Perhaps it makes sense to consider a slightly
different RTO
> scenario to illustrate the point about how RACK RTO
behavior can avoid
> spurious retransmissions.
>
> Here is the proposed new text:
>
> "3.5. An Example of RACK-TLP in Action: RTO
>
> In addition to enhancing fast recovery, RACK improves
the accuracy of
> RTO recovery by reducing spurious retransmissions.
>
> Without RACK, upon RTO timer expiration the sender
marks all the
> unacknowledged segments lost. This approach can lead
to spurious
> retransmissions. For example, consider a simple case
where one
> segment was sent with an RTO of 1 second, and then the
application
> writes more data, causing a second and third segment to
be sent right
> before the RTO of the first segment expires. Suppose
none of the
> segments were lost. Without RACK, if there is a
spurious RTO then
> the sender marks all three segments as lost and
retransmits the
> first segment. If the ACK for the original copy of the
first segment
> arrives right after the spurious RTO retransmission,
then the
> sender continues slow start and spuriously retransmits
the second
> and third segments, since it (erroneously) presumed
they are lost.
That seems fine. Would it be useful to add an additional
note saying that
if F-RTO is in use, then further unnecessary
retransmissions can be
avoided?
> With RACK, upon RTO timer expiration the only segment
automatically
> marked lost is the first segment (since it was sent an
RTO ago); for
> all the other segments RACK only marks the segment lost
if at least
> one round trip has elapsed since the segment was
transmitted.
> Consider the previous example scenario, this time with
RACK. With
> RACK, when the RTO expires the sender only marks the
first segment as
> lost, and retransmits that segment. The other two very
recently sent
> segments are not marked lost, because they were sent
less than one
> round trip ago and there were no ACKs providing
evidence that they
> were lost. Upon receiving the ACK for the RTO
retransmission the
> RACK sender would not yet retransmit the second or
third segment. but
> rather would rearm the RTO timer and wait for a new RTO
interval to elapse
> before marking the second or third segments as lost.
> "
This behavior with RACK-TLP does not quite match with my
reading of
the draft in all parts, so I have some questions.
First, wouldn't the PTO timer be first running for the
first segment
and when the sender sends the 2nd and 3rd segment it would
rearm PTO for
the 3rd segment? My understanding based on Sec 8 is that
RTO cannot be
armed until PTO fires? And when the PTO fires, the sender
would
retransmit the 3rd segment? Maybe I am missing something?
Anyways, an RTO timer may expire in this scenario (e.g., if
there is no
Ack for TLP).
Let's assume the RTO timer expires as described. When the
first Ack after
the rexmit of the 1st segment arrives, isn't that the late
ACK for the
original copy of the first segment, not the ACK for the RTO
retransmission? Anyways, when this first Ack arrives after
the RTO, the
second and third segment are not marked as lost as
described, which
nicely prevents unnecessary retransmissions of those
segments.
However, if the 2nd and 3rd segments were lost, wouldn't
rearming RTO
with exponentially backed-off timer value (2*RTO) make the
recovery of
the 2nd and 3rd segment very slow? After 2*RTO the timer
expires and
the 2nd segment is rexmitted and only after one additional
RTT when the
ack for the 2nd segment arrives, the sender rexmits the 3rd
segment?
Wouldn't it be much more efficient to disable RACK loss
detection during
an RTO recovery and employ F-RTO? Maybe I am missing
something again?
Then to the problem with the pseudocode I mentioned in the
beginning of
this message. If I am not incorrect it seems to me that
setting a value
to TLP.end_seq in TLP_send_probe() is off by one byte?
Maybe something
similar is wrong somewhere else too, didn't check. Please
apologise me if
I missed something.
Let's use an example with RACK-TLP recovery:
Send P1, ..., P9, P10
[P1,.., P9 delayed, P10 dropped]
/* I believe P1-9 delayed is not important here */
After 2*SRTT PTO expires
P10 rexmitted by TLP
[ TLP_send_probe(): set TLP.end_seq = SND.NXT (= SEG.P11)
]
[after TLP, the application possibly provides more data
to send]
Ack for original P1 arrives
(P11 sent, if more data by application)
Ack for original P2 arrives
(P12 sent, if more data by application)
...
Ack for original P9 arrives
(P19 sent, if more data by application)
Ack for P10 triggered by TLP rexmit of P10 arrives
[TLP_process_ack(ACK):
...
If ACK's ack. number > TLP.end_seq
returns False
since ACK's ack. number = SEQ.ACK = TLP.end_seq =
SEG.P11]
Loss of P10 is undetected and no reaction to congestion]
My understanding of draft text is that a loss of P10 is not
concealed.
Please advise.
Best regards,
/Markku
> best regards,
>
> neal
>
>> Please remove also the last paragraph of the Sec. 3.5,
it being also
>> incorrect description of behavior.
>>
>> BR,
>>
>> /Markku
>>
>>>>
>>>> BR,
>>>>
>>>> /Markku
>>>>
>>>>>>
>>>>>> Best regards,
>>>>>>
>>>>>> /Markku
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Mon, 16 Nov 2020, The IESG wrote:
>>>>>>
>>>>>>>
>>>>>>> The IESG has received a request from the TCP
Maintenance and Minor Extensions
>>>>>>> WG (tcpm) to consider the following document: -
'The RACK-TLP loss detection
>>>>>>> algorithm for TCP'
>>>>>>> <draft-ietf-tcpm-rack-13.txt> as Proposed Standard
>>>>>>>
>>>>>>> The IESG plans to make a decision in the next few
weeks, and solicits final
>>>>>>> comments on this action. Please send substantive
comments to the
>>>>>>> last-call@xxxxxxxx mailing lists by 2020-11-30.
Exceptionally, comments may
>>>>>>> be sent to iesg@xxxxxxxx instead. In either case,
please retain the beginning
>>>>>>> of the Subject line to allow automated sorting.
>>>>>>>
>>>>>>> Abstract
>>>>>>>
>>>>>>>
>>>>>>> This document presents the RACK-TLP loss
detection algorithm for TCP.
>>>>>>> RACK-TLP uses per-segment transmit timestamps and
selective
>>>>>>> acknowledgements (SACK) and has two parts: RACK
("Recent
>>>>>>> ACKnowledgment") starts fast recovery quickly
using time-based
>>>>>>> inferences derived from ACK feedback. TLP ("Tail
Loss Probe")
>>>>>>> leverages RACK and sends a probe packet to
trigger ACK feedback to
>>>>>>> avoid retransmission timeout (RTO) events.
Compared to the widely
>>>>>>> used DUPACK threshold approach, RACK-TLP detects
losses more
>>>>>>> efficiently when there are application-limited
flights of data, lost
>>>>>>> retransmissions, or data packet reordering
events. It is intended to
>>>>>>> be an alternative to the DUPACK threshold
approach.
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> The file can be obtained via
>>>>>>>
https://datatracker.ietf.org/doc/draft-ietf-tcpm-rack/
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> No IPR declarations have been submitted directly on
this I-D.
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> tcpm mailing list
>>>>>>> tcpm@xxxxxxxx
>>>>>>> https://www.ietf.org/mailman/listinfo/tcpm
>>>>>>>
>>>>>
>>>
>
