[Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[Michael Richardson]

On Monday I had VDSL2 installed at my home office.

50Mb/s down, 10Mb/s up from storm.ca. Native IPv6, but as far as I can tell,
they aren't speaking DHCPv6; anyway I kept my /56 from my previous
connection, but agreed to swap my IPv4/25 for an IPv4/28.
I moved all my IPv4 to traditional NAT'ed, IPv6 all around, and /32
routed the IPv4 to the various machines that actually need it.

The ISP (www.storm.ca. Great local ISP.) provided me a SmartRG router in
bridged mode.  I'd like to get into it; but they don't answer questions from
"end-users"....

With a laptop hooked up, I saw the full bandwidth.
With the Netgear 3800 running 3.10.44, I saw a max download of 36Mb/s.
Initially, I was seeing 7Mb/s, 640Kb/s up, as well.. that's what the QoS
parameters were set to from my bridge-DSL setup :-)
(So, the good news is that the scripts definitely *do* something...)

I had problems last week with getting bridges over wireless to work,
and later on problems with getting all the wireless devices to come up.

My PPPoE interface doesn't come up on it's own.  What is the "@ge00"
part about?  I'm doing it in /etc/rc.local, which has all sorts of
problems, including failing to include the pppoe-* interface into the
iptables, etc.

At this point my guess is that netifd has some kind of limit on the number of
interfaces it will bring up.  I have 27 interfaces in my the ifconfig,
including the 8 "ifbX" ones, the tun,tap, and my VLANs ("se00.XX") and the
like.  Maybe the number is around 16... I have been looking at netifd source,
and I don't see any obvious struct interfaces[16] or something like that.

Is there a way to enable debugging on netifd?  The -d option to it?
The /etc/rc.common stuff is... a bit impenetrable to me...

--
]               Never tell me the odds!                 | ipv6 mesh networks [
]   Michael Richardson, Sandelman Software Works        | network architect  [
]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on rails    [

Re: [Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[Dave Taht]

On Tue, Jun 24, 2014 at 3:48 PM, Michael Richardson <mcr@sandelman.ca> wrote:
>
> On Monday I had VDSL2 installed at my home office.
>
> 50Mb/s down, 10Mb/s up from storm.ca. Native IPv6, but as far as I can tell,
> they aren't speaking DHCPv6; anyway I kept my /56 from my previous

Cool. I note that the wndr3800 runs out of horsepower at about 50mbit/10.

The edgerouter (which has fq_codel in their 1.5 firmware release), doesn't.

(and it took 10 minutes to port the sqm-scripts over to it)
(and I haven't got around to booting an openwrt version on it, their firmware
 is 3.4 based)

That said I am seeing stuff that indicates inbound htb is increasingly
inaccurate on both products starting at about 20 mbit. (I had long
assumed before now that it was a cpu limitation as I don't see this on
x86) The 36mbit number you got matches mine, try increasing the rate
limit to 64mbits and see what happens.

Also, please measure the bloat you get on the openrg without sqm.... I
would hope the downstream bloat is less horrible than cable modems.

> connection, but agreed to swap my IPv4/25 for an IPv4/28.
> I moved all my IPv4 to traditional NAT'ed, IPv6 all around, and /32
> routed the IPv4 to the various machines that actually need it.

nifty. When I get a /29 from comcast I don't have that ability, I think.
did you do that via proxy arp, or did they give you an ipv4 gateway
and routed the /28 ?

> The ISP (www.storm.ca. Great local ISP.) provided me a SmartRG router in
> bridged mode.  I'd like to get into it; but they don't answer questions from
> "end-users"....
>
> With a laptop hooked up, I saw the full bandwidth.
> With the Netgear 3800 running 3.10.44, I saw a max download of 36Mb/s.
> Initially, I was seeing 7Mb/s, 640Kb/s up, as well.. that's what the QoS
> parameters were set to from my bridge-DSL setup :-)
> (So, the good news is that the scripts definitely *do* something...)
>
> I had problems last week with getting bridges over wireless to work,
> and later on problems with getting all the wireless devices to come up.
>
> My PPPoE interface doesn't come up on it's own.  What is the "@ge00"
> part about?  I'm doing it in /etc/rc.local, which has all sorts of
> problems, including failing to include the pppoe-* interface into the
> iptables, etc.

regrettably how pppoe works is a mystery to me. I'd nuke the @ge00 and
try creating the pppoe interface from the gui for a start.

> At this point my guess is that netifd has some kind of limit on the number of
> interfaces it will bring up.  I have 27 interfaces in my the ifconfig,

Well there are other hard limits - for example ifb's number is created by
a line in /etc/modules.d/ (ifb numifbs=0)

> including the 8 "ifbX" ones, the tun,tap, and my VLANs ("se00.XX") and the
> like.  Maybe the number is around 16... I have been looking at netifd source,
> and I don't see any obvious struct interfaces[16] or something like that.

You definately aren't a typical user.

> Is there a way to enable debugging on netifd?  The -d option to it?
> The /etc/rc.common stuff is... a bit impenetrable to me...

Hit #openwrt-devel for questions - netifd is a rather complex state machine.

>
> --
> ]               Never tell me the odds!                 | ipv6 mesh networks [
> ]   Michael Richardson, Sandelman Software Works        | network architect  [
> ]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on rails    [
>
>
>
> _______________________________________________
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel



-- 
Dave Täht

NSFW: https://w2.eff.org/Censorship/Internet_censorship_bills/russell_0296_indecent.article

Re: [Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[Dave Taht]

On Fri, Jun 27, 2014 at 12:37 PM, Dave Taht <dave.taht@gmail.com> wrote:
> On Tue, Jun 24, 2014 at 3:48 PM, Michael Richardson <mcr@sandelman.ca> wrote:
>>
>> On Monday I had VDSL2 installed at my home office.
>>
>> 50Mb/s down, 10Mb/s up from storm.ca. Native IPv6, but as far as I can tell,
>> they aren't speaking DHCPv6; anyway I kept my /56 from my previous
>
> Cool. I note that the wndr3800 runs out of horsepower at about 50mbit/10.

And when I say that, I mean it. HTB eats all the cpu, leaving nothing
for wifi , wifi crypto, the web interface, etc, at these rates. I have
long been looking for something more efficient than htb without much
luck.

So assuming your downstream bloat isn't "bad", and you stick with cero
on this interface, I'd disable the downstream rate limiter entirely
(set it to 0), and just use sqm on the upstream. It is increasingly
hard to hit the higher bandwidth limits with normal traffic.

(but I'd love a rrul measurement all the same... )


> The edgerouter (which has fq_codel in their 1.5 firmware release), doesn't.
>
> (and it took 10 minutes to port the sqm-scripts over to it)
> (and I haven't got around to booting an openwrt version on it, their firmware
>  is 3.4 based)
>
> That said I am seeing stuff that indicates inbound htb is increasingly
> inaccurate on both products starting at about 20 mbit. (I had long
> assumed before now that it was a cpu limitation as I don't see this on
> x86) The 36mbit number you got matches mine, try increasing the rate
> limit to 64mbits and see what happens.
>
> Also, please measure the bloat you get on the openrg without sqm.... I
> would hope the downstream bloat is less horrible than cable modems.
>
>> connection, but agreed to swap my IPv4/25 for an IPv4/28.
>> I moved all my IPv4 to traditional NAT'ed, IPv6 all around, and /32
>> routed the IPv4 to the various machines that actually need it.
>
> nifty. When I get a /29 from comcast I don't have that ability, I think.
> did you do that via proxy arp, or did they give you an ipv4 gateway
> and routed the /28 ?
>
>> The ISP (www.storm.ca. Great local ISP.) provided me a SmartRG router in
>> bridged mode.  I'd like to get into it; but they don't answer questions from
>> "end-users"....
>>
>> With a laptop hooked up, I saw the full bandwidth.
>> With the Netgear 3800 running 3.10.44, I saw a max download of 36Mb/s.
>> Initially, I was seeing 7Mb/s, 640Kb/s up, as well.. that's what the QoS
>> parameters were set to from my bridge-DSL setup :-)
>> (So, the good news is that the scripts definitely *do* something...)
>>
>> I had problems last week with getting bridges over wireless to work,
>> and later on problems with getting all the wireless devices to come up.
>>
>> My PPPoE interface doesn't come up on it's own.  What is the "@ge00"
>> part about?  I'm doing it in /etc/rc.local, which has all sorts of
>> problems, including failing to include the pppoe-* interface into the
>> iptables, etc.
>
> regrettably how pppoe works is a mystery to me. I'd nuke the @ge00 and
> try creating the pppoe interface from the gui for a start.
>
>> At this point my guess is that netifd has some kind of limit on the number of
>> interfaces it will bring up.  I have 27 interfaces in my the ifconfig,
>
> Well there are other hard limits - for example ifb's number is created by
> a line in /etc/modules.d/ (ifb numifbs=0)
>
>> including the 8 "ifbX" ones, the tun,tap, and my VLANs ("se00.XX") and the
>> like.  Maybe the number is around 16... I have been looking at netifd source,
>> and I don't see any obvious struct interfaces[16] or something like that.
>
> You definately aren't a typical user.
>
>> Is there a way to enable debugging on netifd?  The -d option to it?
>> The /etc/rc.common stuff is... a bit impenetrable to me...
>
> Hit #openwrt-devel for questions - netifd is a rather complex state machine.
>
>>
>> --
>> ]               Never tell me the odds!                 | ipv6 mesh networks [
>> ]   Michael Richardson, Sandelman Software Works        | network architect  [
>> ]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on rails    [
>>
>>
>>
>> _______________________________________________
>> Cerowrt-devel mailing list
>> Cerowrt-devel@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/cerowrt-devel
>
>
>
> --
> Dave Täht
>
> NSFW: https://w2.eff.org/Censorship/Internet_censorship_bills/russell_0296_indecent.article



-- 
Dave Täht

NSFW: https://w2.eff.org/Censorship/Internet_censorship_bills/russell_0296_indecent.article

Re: [Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[Baptiste Jonglez]

[-- Attachment #1: Type: text/plain, Size: 700 bytes --]

On Fri, Jun 27, 2014 at 12:37:51PM -0700, Dave Taht wrote:
> On Tue, Jun 24, 2014 at 3:48 PM, Michael Richardson <mcr@sandelman.ca> wrote:
> >
> > On Monday I had VDSL2 installed at my home office.
> >
> > 50Mb/s down, 10Mb/s up from storm.ca. Native IPv6, but as far as I can tell,
> > they aren't speaking DHCPv6; anyway I kept my /56 from my previous
> 
> Cool. I note that the wndr3800 runs out of horsepower at about 50mbit/10.

My wndr3800 handles 300M/50M without issues.  I'm using the default
pfifo_fast queue, though, running OpenWRT.

Note that my ISP provides IPv6 through a L2TP tunnel, and the WNDR3800
saturates around 100M with IPv6, since it must decapsulate packets.

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

Re: [Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[dpreed]

[-- Attachment #1: Type: text/plain, Size: 1189 bytes --]


You get 300M/50M in Lyons?  May I ask who your provider is, and what that costs?
 
Only a few years ago, one had to use dialup in most areas outside Paris, as I recall.
 


On Sunday, June 29, 2014 4:33am, "Baptiste Jonglez" <baptiste.jonglez@ens-lyon.fr> said:



> _______________________________________________
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel
> On Fri, Jun 27, 2014 at 12:37:51PM -0700, Dave Taht wrote:
> > On Tue, Jun 24, 2014 at 3:48 PM, Michael Richardson <mcr@sandelman.ca>
> wrote:
> > >
> > > On Monday I had VDSL2 installed at my home office.
> > >
> > > 50Mb/s down, 10Mb/s up from storm.ca. Native IPv6, but as far as I can
> tell,
> > > they aren't speaking DHCPv6; anyway I kept my /56 from my previous
> >
> > Cool. I note that the wndr3800 runs out of horsepower at about 50mbit/10.
> 
> My wndr3800 handles 300M/50M without issues. I'm using the default
> pfifo_fast queue, though, running OpenWRT.
> 
> Note that my ISP provides IPv6 through a L2TP tunnel, and the WNDR3800
> saturates around 100M with IPv6, since it must decapsulate packets.
> 

[-- Attachment #2: Type: text/html, Size: 1985 bytes --]

Re: [Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[Michael Richardson]

Dave Taht <dave.taht@gmail.com> wrote:
    > The edgerouter (which has fq_codel in their 1.5 firmware release),
    > doesn't.

Yes, I have one, and I've been thinking it might be a better choice for me.

    > (and I haven't got around to booting an openwrt version on it, their firmware
    > is 3.4 based)

good to know.

    > That said I am seeing stuff that indicates inbound htb is increasingly
    > inaccurate on both products starting at about 20 mbit. (I had long
    > assumed before now that it was a cpu limitation as I don't see this on
    > x86) The 36mbit number you got matches mine, try increasing the rate
    > limit to 64mbits and see what happens.

I'll try that today.

    > Also, please measure the bloat you get on the openrg without sqm.... I
    > would hope the downstream bloat is less horrible than cable modems.

Bell Canada and nortel ATM equipment is involved.

    >> connection, but agreed to swap my IPv4/25 for an IPv4/28.
    >> I moved all my IPv4 to traditional NAT'ed, IPv6 all around, and /32
    >> routed the IPv4 to the various machines that actually need it.

    > nifty. When I get a /29 from comcast I don't have that ability, I think.
    > did you do that via proxy arp, or did they give you an ipv4 gateway
    > and routed the /28 ?

v4 gateway and routed /28.

    >> My PPPoE interface doesn't come up on it's own.  What is the "@ge00"
    >> part about?  I'm doing it in /etc/rc.local, which has all sorts of
    >> problems, including failing to include the pppoe-* interface into the
    >> iptables, etc.

    > regrettably how pppoe works is a mystery to me. I'd nuke the @ge00 and
    > try creating the pppoe interface from the gui for a start.

I went through that... it's all there. Netifd just ignores it.

    >> At this point my guess is that netifd has some kind of limit on the number of
    >> interfaces it will bring up.  I have 27 interfaces in my the ifconfig,

    > Well there are other hard limits - for example ifb's number is created by
    > a line in /etc/modules.d/ (ifb numifbs=0)

okay, I'll try making that zero and see if that makes netifd happier.

    >> including the 8 "ifbX" ones, the tun,tap, and my VLANs ("se00.XX") and the
    >> like.  Maybe the number is around 16... I have been looking at netifd source,
    >> and I don't see any obvious struct interfaces[16] or something like that.

    > You definately aren't a typical user.

    >> Is there a way to enable debugging on netifd?  The -d option to it?
    >> The /etc/rc.common stuff is... a bit impenetrable to me...

    > Hit #openwrt-devel for questions - netifd is a rather complex state machine.

I'll try there.
There used to be images/nightlies of barrier breaker, but I can't find them
anymore.  I have two authenticate 54GLs that I want to upgrade...

--
]               Never tell me the odds!                 | ipv6 mesh networks [
]   Michael Richardson, Sandelman Software Works        | network architect  [
]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on rails    [

Re: [Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[Sebastian Moeller]

Hi Michael,

On Jun 30, 2014, at 15:24 , Michael Richardson <mcr@sandelman.ca> wrote:

> 
> Dave Taht <dave.taht@gmail.com> wrote:
>> The edgerouter (which has fq_codel in their 1.5 firmware release),
>> doesn't.
> 
> Yes, I have one, and I've been thinking it might be a better choice for me.
> 
>> (and I haven't got around to booting an openwrt version on it, their firmware
>> is 3.4 based)
> 
> good to know.
> 
>> That said I am seeing stuff that indicates inbound htb is increasingly
>> inaccurate on both products starting at about 20 mbit. (I had long
>> assumed before now that it was a cpu limitation as I don't see this on
>> x86) The 36mbit number you got matches mine, try increasing the rate
>> limit to 64mbits and see what happens.
> 
> I'll try that today.
> 
>> Also, please measure the bloat you get on the openrg without sqm.... I
>> would hope the downstream bloat is less horrible than cable modems.
> 
> Bell Canada and nortel ATM equipment is involved.

	Oh, that would be quite interesting vdsl2 and atm at the same time. Could you by chance post the status page of your vdsl modem (the more detailed the statistics the better)? (Typically vdsl2 should use packet transfer mode (PTM) instead of ATM between modem and DSLAM, though the DSLAM might still connected to an ATM fabric. It would be interesting to measure whether you still "see" the effects of ATM encapsulation along your link.
	At least in europe the ATM fabric seems on the way out and gets replaced by ethernet links to the FTTC cabinets. 
Note I think that your link will not require link layer adatations for ATM (unless modem and slam talk atm that is).

Best Regards
	sebastian

> 
>>> connection, but agreed to swap my IPv4/25 for an IPv4/28.
>>> I moved all my IPv4 to traditional NAT'ed, IPv6 all around, and /32
>>> routed the IPv4 to the various machines that actually need it.
> 
>> nifty. When I get a /29 from comcast I don't have that ability, I think.
>> did you do that via proxy arp, or did they give you an ipv4 gateway
>> and routed the /28 ?
> 
> v4 gateway and routed /28.
> 
>>> My PPPoE interface doesn't come up on it's own.  What is the "@ge00"
>>> part about?  I'm doing it in /etc/rc.local, which has all sorts of
>>> problems, including failing to include the pppoe-* interface into the
>>> iptables, etc.
> 
>> regrettably how pppoe works is a mystery to me. I'd nuke the @ge00 and
>> try creating the pppoe interface from the gui for a start.
> 
> I went through that... it's all there. Netifd just ignores it.
> 
>>> At this point my guess is that netifd has some kind of limit on the number of
>>> interfaces it will bring up.  I have 27 interfaces in my the ifconfig,
> 
>> Well there are other hard limits - for example ifb's number is created by
>> a line in /etc/modules.d/ (ifb numifbs=0)
> 
> okay, I'll try making that zero and see if that makes netifd happier.
> 
>>> including the 8 "ifbX" ones, the tun,tap, and my VLANs ("se00.XX") and the
>>> like.  Maybe the number is around 16... I have been looking at netifd source,
>>> and I don't see any obvious struct interfaces[16] or something like that.
> 
>> You definately aren't a typical user.
> 
>>> Is there a way to enable debugging on netifd?  The -d option to it?
>>> The /etc/rc.common stuff is... a bit impenetrable to me...
> 
>> Hit #openwrt-devel for questions - netifd is a rather complex state machine.
> 
> I'll try there.
> There used to be images/nightlies of barrier breaker, but I can't find them
> anymore.  I have two authenticate 54GLs that I want to upgrade...
> 
> --
> ]               Never tell me the odds!                 | ipv6 mesh networks [
> ]   Michael Richardson, Sandelman Software Works        | network architect  [
> ]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on rails    [
> 
> _______________________________________________
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel

Re: [Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[Michael Richardson]

Dave Taht <dave.taht@gmail.com> wrote:
    > On Fri, Jun 27, 2014 at 12:37 PM, Dave Taht <dave.taht@gmail.com> wrote:
    >> On Tue, Jun 24, 2014 at 3:48 PM, Michael Richardson <mcr@sandelman.ca> wrote:
    >>>
    >>> On Monday I had VDSL2 installed at my home office.
    >>>
    >>> 50Mb/s down, 10Mb/s up from storm.ca. Native IPv6, but as far as I can tell,
    >>> they aren't speaking DHCPv6; anyway I kept my /56 from my previous
    >>
    >> Cool. I note that the wndr3800 runs out of horsepower at about 50mbit/10.

    > And when I say that, I mean it. HTB eats all the cpu, leaving nothing
    > for wifi , wifi crypto, the web interface, etc, at these rates. I have
    > long been looking for something more efficient than htb without much
    > luck.

Right now, with it still set at 50000, I see 49.86Mb/s from my ISPs' speedtest.storm.ca.
I'll take your advice and turn off download policing.

    > So assuming your downstream bloat isn't "bad", and you stick with cero
    > on this interface, I'd disable the downstream rate limiter entirely
    > (set it to 0), and just use sqm on the upstream. It is increasingly
    > hard to hit the higher bandwidth limits with normal traffic.

    > (but I'd love a rrul measurement all the same... )

I'll try to get it setup.... anecdotally, my ssh is really fast all the time.
I spend a lot of time ssh'ed into my SOHO desktop from my office, running
emacsclient in "-nw" mode.  In the pre-CODEL days, this resulted regularly in
me getting 20-40seconds ahead of the screen updates.  Post cerowrt/CODEL,
there were occasional periods of time when codel couldn't protected my typing
SSH from the ssh that was running sshfs/rsync.
(sshfs/rsync needs to invoke ssh with some option to make it apply DSCP to
the packets)

I'm think that with 10Mb/s up, I just won't care :-)

--
]               Never tell me the odds!                 | ipv6 mesh networks [
]   Michael Richardson, Sandelman Software Works        | network architect  [
]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on rails    [

Re: [Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[Michael Richardson]

Sebastian Moeller <moeller0@gmx.de> wrote:
    >> Bell Canada and nortel ATM equipment is involved.

    > Oh, that would be quite interesting vdsl2 and atm at the same
    > time. Could you by chance post the status page of your vdsl modem (the
    > more detailed the statistics the better)? (Typically vdsl2 should use
    > packet transfer mode (PTM) instead of ATM between modem and DSLAM,
    > though the DSLAM might still connected to an ATM fabric. It would be
    > interesting to measure whether you still "see" the effects of ATM
    > encapsulation along your link.

I can't say what the VDSL2 interface is, but the backhaul network with Bell
has been historically all nortel passport ATM (with way over buffered LANE).

--
]               Never tell me the odds!                 | ipv6 mesh networks [
]   Michael Richardson, Sandelman Software Works        | network architect  [
]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on rails    [

Re: [Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[Dave Taht]

On Mon, Jun 30, 2014 at 8:35 AM, Michael Richardson <mcr@sandelman.ca> wrote:
>
> Dave Taht <dave.taht@gmail.com> wrote:
>     > On Fri, Jun 27, 2014 at 12:37 PM, Dave Taht <dave.taht@gmail.com> wrote:
>     >> On Tue, Jun 24, 2014 at 3:48 PM, Michael Richardson <mcr@sandelman.ca> wrote:
>     >>>
>     >>> On Monday I had VDSL2 installed at my home office.
>     >>>
>     >>> 50Mb/s down, 10Mb/s up from storm.ca. Native IPv6, but as far as I can tell,
>     >>> they aren't speaking DHCPv6; anyway I kept my /56 from my previous
>     >>
>     >> Cool. I note that the wndr3800 runs out of horsepower at about 50mbit/10.
>
>     > And when I say that, I mean it. HTB eats all the cpu, leaving nothing
>     > for wifi , wifi crypto, the web interface, etc, at these rates. I have
>     > long been looking for something more efficient than htb without much
>     > luck.
>
> Right now, with it still set at 50000, I see 49.86Mb/s from my ISPs' speedtest.storm.ca.
> I'll take your advice and turn off download policing.
>
>     > So assuming your downstream bloat isn't "bad", and you stick with cero
>     > on this interface, I'd disable the downstream rate limiter entirely
>     > (set it to 0), and just use sqm on the upstream. It is increasingly
>     > hard to hit the higher bandwidth limits with normal traffic.
>
>     > (but I'd love a rrul measurement all the same... )
>
> I'll try to get it setup.... anecdotally, my ssh is really fast all the time.
> I spend a lot of time ssh'ed into my SOHO desktop from my office, running
> emacsclient in "-nw" mode.  In the pre-CODEL days, this resulted regularly in
> me getting 20-40seconds ahead of the screen updates.  Post cerowrt/CODEL,
> there were occasional periods of time when codel couldn't protected my typing
> SSH from the ssh that was running sshfs/rsync.
> (sshfs/rsync needs to invoke ssh with some option to make it apply DSCP to
> the packets)

I switched to mosh for terminal sessions 2+ years ago and I'm never
going back. Also, I'd supplied a patch to mark those packets as AF42,
as well, which moves it up in priority in the SQM system. This was in
mainline mosh for a while, until it was found to mess up two systems
(out of 10s of thousands, 2 would stop forwarding AF42 marked packets
after 10 seconds), and rather than switch it to tos immediate, they
merely reverted the patch.

This probably won't re-apply but would be simple to add back in.

http://snapon.lab.bufferbloat.net/~d/0001-Set-AF42-on-mosh-by-default.patch


Secondly, I made patches available for rsync's native protocol to use
another tcp algorithm and mark the packets as CS1 (background), also
two years ago. (I don't know how to do the right thing to ssh, but
should probably get around to it one day)

These two patches should (in addition to fq_codel) reduce your
interruptions of interactive service from pretty darn good with
fq_codel, to nearly never, with sqm.

https://git.samba.org/cifs-utils.git/?p=rsync-patches.git;a=history;f=congestion.diff;h=5fb8a79edd399b6e8cdefb251dce0e53c0951c65;hb=7578744b8595da0d9ee785c96c74fff810e721ea

I don't think this landed in rsync head, either.

>
> I'm think that with 10Mb/s up, I just won't care :-)

Measure...
>
> --
> ]               Never tell me the odds!                 | ipv6 mesh networks [
> ]   Michael Richardson, Sandelman Software Works        | network architect  [
> ]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on rails    [
>



-- 
Dave Täht

NSFW: https://w2.eff.org/Censorship/Internet_censorship_bills/russell_0296_indecent.article

Re: [Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[Sebastian Moeller]

Hi Michael,

On Jun 30, 2014, at 17:37 , Michael Richardson <mcr@sandelman.ca> wrote:

> 
> Sebastian Moeller <moeller0@gmx.de> wrote:
>>> Bell Canada and nortel ATM equipment is involved.
> 
>> Oh, that would be quite interesting vdsl2 and atm at the same
>> time. Could you by chance post the status page of your vdsl modem (the
>> more detailed the statistics the better)? (Typically vdsl2 should use
>> packet transfer mode (PTM) instead of ATM between modem and DSLAM,
>> though the DSLAM might still connected to an ATM fabric. It would be
>> interesting to measure whether you still "see" the effects of ATM
>> encapsulation along your link.
> 
> I can't say what the VDSL2 interface is, but the backhaul network with Bell
> has been historically all nortel passport ATM (with way over buffered LANE).

	I would be amazed if they would manage to extend the reach of their ATM network while most other telco's seem to be moving away from ATM… Since bell advertises this as a fiber to the cabinet product, my bet is on ethernet over fiber from the slam/msan to the core network, but I digress. Not that it matters, only for the modem slam connection one needs to take ATM encapsulation into account for bandwidth shaping purposes. So what is your direct uplink to the slam using PTM or ATM?

Best Regards
	Sebastian

> 
> --
> ]               Never tell me the odds!                 | ipv6 mesh networks [
> ]   Michael Richardson, Sandelman Software Works        | network architect  [
> ]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on rails    [
> 

Re: [Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[Michael Richardson]

Sebastian Moeller <moeller0@gmx.de> wrote:
    > Hi Michael,

    > On Jun 30, 2014, at 17:37 , Michael Richardson <mcr@sandelman.ca> wrote:

    >>
    >> Sebastian Moeller <moeller0@gmx.de> wrote:
    >>>> Bell Canada and nortel ATM equipment is involved.
    >>
    >>> Oh, that would be quite interesting vdsl2 and atm at the same
    >>> time. Could you by chance post the status page of your vdsl modem (the
    >>> more detailed the statistics the better)? (Typically vdsl2 should use
    >>> packet transfer mode (PTM) instead of ATM between modem and DSLAM,
    >>> though the DSLAM might still connected to an ATM fabric. It would be
    >>> interesting to measure whether you still "see" the effects of ATM
    >>> encapsulation along your link.
    >>
    >> I can't say what the VDSL2 interface is, but the backhaul network with Bell
    >> has been historically all nortel passport ATM (with way over buffered LANE).

    > I would be amazed if they would manage to extend the reach of their ATM
    > network while most other telco's seem to be moving away from ATM… Since
    > bell advertises this as a fiber to the cabinet product, my bet is on

OC192 still counts as fiber :-)
Bell Canada is still very much a Nortel captured company.

    > ethernet over fiber from the slam/msan to the core network, but I
    > digress. Not that it matters, only for the modem slam connection one
    > needs to take ATM encapsulation into account for bandwidth shaping
    > purposes. So what is your direct uplink to the slam using PTM or ATM?

I have no idea and no real way to find out. I'll ask my ISP.

--
]               Never tell me the odds!                 | ipv6 mesh networks [
]   Michael Richardson, Sandelman Software Works        | network architect  [
]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on rails    [

Re: [Cerowrt-devel] CeroWRT and "FTTN" 50/10 VDSL2 (aka "FIBE")

[Sebastian Moeller]

Hi Michael,


On Jul 2, 2014, at 22:45 , Michael Richardson <mcr@sandelman.ca> wrote:

> 
> Sebastian Moeller <moeller0@gmx.de> wrote:
>> Hi Michael,
> 
>> On Jun 30, 2014, at 17:37 , Michael Richardson <mcr@sandelman.ca> wrote:
> 
>>> 
>>> Sebastian Moeller <moeller0@gmx.de> wrote:
>>>>> Bell Canada and nortel ATM equipment is involved.
>>> 
>>>> Oh, that would be quite interesting vdsl2 and atm at the same
>>>> time. Could you by chance post the status page of your vdsl modem (the
>>>> more detailed the statistics the better)? (Typically vdsl2 should use
>>>> packet transfer mode (PTM) instead of ATM between modem and DSLAM,
>>>> though the DSLAM might still connected to an ATM fabric. It would be
>>>> interesting to measure whether you still "see" the effects of ATM
>>>> encapsulation along your link.
>>> 
>>> I can't say what the VDSL2 interface is, but the backhaul network with Bell
>>> has been historically all nortel passport ATM (with way over buffered LANE).
> 
>> I would be amazed if they would manage to extend the reach of their ATM
>> network while most other telco's seem to be moving away from ATM… Since
>> bell advertises this as a fiber to the cabinet product, my bet is on
> 
> OC192 still counts as fiber :-)

	Interesting pointer, it seems that an oc-192 link could carry atm or ethernet equally well (or even at the same time ;))

> Bell Canada is still very much a Nortel captured company.

	Looking at http://www.wholesale.bell.ca/wavelength-service/ I would say the backbone seems protocol agnostic, so no clear indicator how the dslams are connected upstream.

> 
>> ethernet over fiber from the slam/msan to the core network, but I
>> digress. Not that it matters, only for the modem slam connection one
>> needs to take ATM encapsulation into account for bandwidth shaping
>> purposes. So what is your direct uplink to the slam using PTM or ATM?
> 
> I have no idea and no real way to find out. I'll ask my ISP.

	Some modems have a detailed statistics page that reveals relevant information (then again some modems hide this information (behind obscure links) or do not have them in the first place). Your ISP will know for sure, at least the technicians (not necessarily the first level help-desk personal though). Good luck in finding out.
	Oh theoretically you could try to find out empirically whether you are on an ATM link, but at 50/10 that measurements might take too long to be feasible (that is might require too many data points).

Best Regards
	sebastian


> 
> --
> ]               Never tell me the odds!                 | ipv6 mesh networks [
> ]   Michael Richardson, Sandelman Software Works        | network architect  [
> ]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on rails    [
> 
> 
>