[Cerowrt-devel] bugs using the 1200ac as a target

[Cerowrt-devel] bugs using the 1200ac as a target

[Dave Taht]

it has something deeply wrong in it at a gbit. In this test I'd failed
over to a faster link at a gigE to show WTF it does - one flow totally
wipes out the others.

http://blog.cerowrt.org/flent/failing_over_faster/linksys_1200_majorly_headblocking.svg

I've seen this behavior from this device since the very first one I
got 6 months back.

In other news:

http://blog.cerowrt.org/post/failing_over_faster/
http://blog.cerowrt.org/post/some_hackerboards_for_wifi/

-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Cerowrt-devel] bugs using the 1200ac as a target

[Richard Smith]

On 04/25/2016 11:56 PM, Dave Taht wrote:
> it has something deeply wrong in it at a gbit. In this test I'd failed
> over to a faster link at a gigE to show WTF it does - one flow totally
> wipes out the others.
>
> http://blog.cerowrt.org/flent/failing_over_faster/linksys_1200_majorly_headblocking.svg

What's your test conditions?  I'll duplicate the test on my 1900acs and 
see if it's there as well.


-- 
Richard A. Smith

Re: [Cerowrt-devel] bugs using the 1200ac as a target

[Dave Taht]

run netserver on the router. It doesn't seem to matter what qdisc is
on the router.

hit it with the flent rrul_be or rrul tests. (or just multiple netperfs)

You'll see one flow completely take over the router in this case and
the others starve.

On Wed, Apr 27, 2016 at 3:04 PM, Richard Smith <smithbone@gmail.com> wrote:
> On 04/25/2016 11:56 PM, Dave Taht wrote:
>>
>> it has something deeply wrong in it at a gbit. In this test I'd failed
>> over to a faster link at a gigE to show WTF it does - one flow totally
>> wipes out the others.
>>
>>
>> http://blog.cerowrt.org/flent/failing_over_faster/linksys_1200_majorly_headblocking.svg
>
>
> What's your test conditions?  I'll duplicate the test on my 1900acs and see
> if it's there as well.
>
>
> --
> Richard A. Smith
> _______________________________________________
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Cerowrt-devel] bugs using the 1200ac as a target

[Aaron Wood]

[-- Attachment #1: Type: text/plain, Size: 1539 bytes --]

That's behavior I've seen multiple times in the past, but I never had a
good feel for what the cause was.

-Aaron

On Wed, Apr 27, 2016 at 3:08 PM, Dave Taht <dave.taht@gmail.com> wrote:

> run netserver on the router. It doesn't seem to matter what qdisc is
> on the router.
>
> hit it with the flent rrul_be or rrul tests. (or just multiple netperfs)
>
> You'll see one flow completely take over the router in this case and
> the others starve.
>
> On Wed, Apr 27, 2016 at 3:04 PM, Richard Smith <smithbone@gmail.com>
> wrote:
> > On 04/25/2016 11:56 PM, Dave Taht wrote:
> >>
> >> it has something deeply wrong in it at a gbit. In this test I'd failed
> >> over to a faster link at a gigE to show WTF it does - one flow totally
> >> wipes out the others.
> >>
> >>
> >>
> http://blog.cerowrt.org/flent/failing_over_faster/linksys_1200_majorly_headblocking.svg
> >
> >
> > What's your test conditions?  I'll duplicate the test on my 1900acs and
> see
> > if it's there as well.
> >
> >
> > --
> > Richard A. Smith
> > _______________________________________________
> > Cerowrt-devel mailing list
> > Cerowrt-devel@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/cerowrt-devel
>
>
>
> --
> Dave Täht
> Let's go make home routers and wifi faster! With better software!
> http://blog.cerowrt.org
> _______________________________________________
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel
>

[-- Attachment #2: Type: text/html, Size: 2747 bytes --]

Re: [Cerowrt-devel] bugs using the 1200ac as a target

[Richard Smith]

On 04/27/2016 06:08 PM, Dave Taht wrote:
> run netserver on the router. It doesn't seem to matter what qdisc is
> on the router.
>
> hit it with the flent rrul_be or rrul tests. (or just multiple netperfs)


> You'll see one flow completely take over the router in this case and
> the others starve.

The behavior of my 1900acs is different but still not very good.  Seems 
to be a mix of total fail on the Tx side (ie no data reported) to what 
looks ok but much slower than 1Gbit.

https://drive.google.com/folderview?id=0B-P0wCbNmKvATHNKb3RjbVoxXzA&usp=sharing

-- 
Richard A. Smith