Re: [RFC PATCH 28/28] tcp: AccECN sysctl documentation

From: Yuchung Cheng <ycheng@google.com>
To: "Ilpo Järvinen" <ilpo.jarvinen@cs.helsinki.fi>
Cc: Dave Taht <dave.taht@gmail.com>,
	Linux Kernel Network Developers <netdev@vger.kernel.org>,
	Neal Cardwell <ncardwell@google.com>,
	Eric Dumazet <eric.dumazet@gmail.com>,
	Olivier Tilmans <olivier.tilmans@nokia-bell-labs.com>
Subject: Re: [RFC PATCH 28/28] tcp: AccECN sysctl documentation
Date: Tue, 24 Mar 2020 10:05:16 -0700	[thread overview]
Message-ID: <CAK6E8=dTKvUbNQeA5wnSNCmW3ARAawxcM7Q4oLBbWMbNnOm-8Q@mail.gmail.com> (raw)
In-Reply-To: <alpine.DEB.2.20.2003241449310.14439@whs-18.cs.helsinki.fi>

On Tue, Mar 24, 2020 at 5:50 AM Ilpo Järvinen
<ilpo.jarvinen@cs.helsinki.fi> wrote:
>
> On Mon, 23 Mar 2020, Yuchung Cheng wrote:
>
> > On Mon, Mar 23, 2020 at 6:34 AM Ilpo Järvinen
> > <ilpo.jarvinen@cs.helsinki.fi> wrote:
> > >
> > > On Fri, 20 Mar 2020, Yuchung Cheng wrote:
> > >
> > > > On Fri, Mar 20, 2020 at 3:40 PM Ilpo Järvinen
> > > > <ilpo.jarvinen@cs.helsinki.fi> wrote:
> > > > >
> > > > > On Thu, 19 Mar 2020, Dave Taht wrote:
> > > > >
> > > > > > On Wed, Mar 18, 2020 at 2:44 AM Ilpo Järvinen <ilpo.jarvinen@helsinki.fi> wrote:
> > > > > > >
> > > > > > > From: Ilpo Järvinen <ilpo.jarvinen@cs.helsinki.fi>
> > > > > > >
> > > > > > > Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@cs.helsinki.fi>
> > > > > > > ---
> > > > > > >  Documentation/networking/ip-sysctl.txt | 12 +++++++++---
> > > > > > >  1 file changed, 9 insertions(+), 3 deletions(-)
> > > > > > >
> > > > > > > diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
> > > > > > > index 5f53faff4e25..ecca6e1d6bea 100644
> > > > > > > --- a/Documentation/networking/ip-sysctl.txt
> > > > > > > +++ b/Documentation/networking/ip-sysctl.txt
> > > > > > > @@ -301,15 +301,21 @@ tcp_ecn - INTEGER
> > > > > > >                 0 Disable ECN.  Neither initiate nor accept ECN.
> > > > > > >                 1 Enable ECN when requested by incoming connections and
> > > > > > >                   also request ECN on outgoing connection attempts.
> > > > > > > -               2 Enable ECN when requested by incoming connections
> > > > > > > +               2 Enable ECN or AccECN when requested by incoming connections
> > > > > > >                   but do not request ECN on outgoing connections.
> > > > > >
> > > > > > Changing existing user-behavior for this default seems to be overly
> > > > > > optimistic. Useful for testing, but...
> > > > >
> > > > > I disagree.
> > > > >
> > > > > The kernel default on ECN is/has been "do nothing" like forever. Yet,
> > > > > passively allowing ECN on servers is a low risk operation because nothing
> > > > > will change before client actively asks for it. However, it was obvious
> > > > > that the servers didn't do that. The servers could have set tcp_ecn to 1
> > > > > (before 2 was there) which is low risk for _servers_ (unlike for clients)
> > > > > but only very very few did. I don't believe servers would now
> > > > > intentionally pick 2 when they clearly didn't pick 1 earlier either.
> > > > >
> > > > > Adding 2 is/was an attempt to side-step the need for both ends to make
> > > > > conscious decision by setting the sysctl (which servers didn't want to
> > > > > do). That is, 2 gives decision on what to do into the hands of the client
> > > > > side which was the true intent of 2 (in case you don't know, I made that
> > > > > change).
> > > > What can a server configure to process only RFC3168 ECN if it prefers to?
> > >
> > > That's why I suggested the flag-based approach?
> >
> > That's assuming an admin that has control of sysctls can also change
> > individual applications (easily). In reality it often is not the case.
> > The default sysctl choices in this patch seem risky to me.
> >
> > > > > If "full control" is the way to go, I think it should be made using flags
> > > > > instead, along these lines:
> > > > >
> > > > > 1: Enable RFC 3168 ECN in+out
> > > > > 2: Enable RFC 3168 ECN in (default on)
> > > > > 4: Enable Accurate ECN in (default on)
> > > > > 8: Enable Accurate ECN in+out
> > > > >
> > > > > Note that I intentionally reversed the in and in/out order for 4&8
> > > > > (something that couldn't be done with 1&2 to preserve meaning of 1).
> > >
> > > It should address any except "out" but no "in" (the meaning of 1 cannot
> > > be changed I think). But out w/o in doesn't sound very useful.
>
> So you mean you'd want to have control that is finer-grained than what the
> sysctls offer?
I recommend having separate sysctl values for AccECN so that servers
configured to use existing values do not behave differently (on ECN)
after kernel upgrade, similar to what Dave Taht suggested.

>
>
> --
>  i.