openSUSE Core Testing Team

[lwfinger]

As most of you are aware, openSUSE recently posted an announcement
that applications to join a Core Testing Team were being accepted, and
that they planned to choose 10 members. I'm not aware of all the
details; however, they ended up accepting 25 of the applications. So
far, I am the only member with a major interest in wifi. How the team
works, and what we do is still being planned, but it is openSUSE's
idea that the team be in charge of deciding what tests should be done,
and what priority they should receive. Our responsibilities in
conducting such tests is expected to have lower priority than the
planning. We will certainly automate tests where possible.

Given the difficulties introduced by changes in the networking code
during 11.1 upgrades, and additional failures during the 11.2 Mx
series, I will be proposing a matrix of tests to accompany every
change in the NetworkManager and other wireless components. My initial
thoughts on the combinations that need to be tested are as follows:

(a) Control with NetworkManager/traditional ifup
(b) Encryption choices: None, WEP, WPA, WPA2
(c) IP assignement by DHCP/static IP
(d) Gnome/KDE/Other(?) window manager
(e) i386/x86_64/PPC architectures
(f) drivers

Without considering alternative window managers or drivers, there are
still 96 combinations, and complete test coverage will be difficult.
If my proposal is accepted, I will do my best to test those
combinations pertinent to my hardware. I do not, however, have access
to PPC hardware. Testing there is quite important, as most other
architectures are little-endian.

Please let me know of any important categories that I have missed, and
any combinations in the matrix that you have tested, along with the
description of the distribution that you tested.

Thanks, Larry

[malcolmlewis]

Originally Posted by Larry Finger

As most of you are aware, openSUSE recently posted
an announcement that applications to join a Core Testing Team were
being accepted, and that they planned to choose 10 members. I'm not
aware of all the details; however, they ended up accepting 25 of the
applications. So far, I am the only member with a major interest in
wifi. How the team works, and what we do is still being planned, but it
is openSUSE's idea that the team be in charge of deciding what tests
should be done, and what priority they should receive. Our
responsibilities in conducting such tests is expected to have lower
priority than the planning. We will certainly automate tests where
possible.

Given the difficulties introduced by changes in the networking code
during 11.1 upgrades, and additional failures during the 11.2 Mx
series, I will be proposing a matrix of tests to accompany every
change in the NetworkManager and other wireless components. My initial
thoughts on the combinations that need to be tested are as follows:

(a) Control with NetworkManager/traditional ifup
(b) Encryption choices: None, WEP, WPA, WPA2
(c) IP assignement by DHCP/static IP
(d) Gnome/KDE/Other(?) window manager
(e) i386/x86_64/PPC architectures
(f) drivers

Without considering alternative window managers or drivers, there are
still 96 combinations, and complete test coverage will be difficult.
If my proposal is accepted, I will do my best to test those
combinations pertinent to my hardware. I do not, however, have access
to PPC hardware. Testing there is quite important, as most other
architectures are little-endian.

Please let me know of any important categories that I have missed, and
any combinations in the matrix that you have tested, along with the
description of the distribution that you tested.

Thanks, Larry

Hi Larry
Congratulations So will you be doing any IEEE 802.11i/n testing? Was
thinking of upgrading the router to an n version here now I have an
ath9k device, but wasn't sure of the status.

--
Cheers Malcolm °¿° (Linux Counter #276890)
openSUSE 11.2 Milestone 7 (i586) Kernel 2.6.31-rc9-7-desktop
up 0:10, 2 users, load average: 0.32, 0.26, 0.19
ASUS eeePC 1000HE ATOM N280 1.66GHz | GPU Mobile 945GM/GMS/GME

[lwfinger]

Malcolm wrote:
> Hi Larry
> Congratulations So will you be doing any IEEE 802.11i/n testing? Was
> thinking of upgrading the router to an n version here now I have an
> ath9k device, but wasn't sure of the status.

Thanks.

No, I won't be doing any 802.11n testing myself as I have no 802.11n
hardware. In addition, the 802.11n support in mac80211 is still
primitive; however, it is improving. As the level of support grows, we
will need more testing and testers.

[malcolmlewis]

Originally Posted by Larry Finger

Malcolm wrote:
> Hi Larry
> Congratulations So will you be doing any IEEE 802.11i/n testing?
> Was thinking of upgrading the router to an n version here now I have
> an ath9k device, but wasn't sure of the status.

Thanks.

No, I won't be doing any 802.11n testing myself as I have no 802.11n
hardware. In addition, the 802.11n support in mac80211 is still
primitive; however, it is improving. As the level of support grows, we
will need more testing and testers.

Hi
Well if you have a recommendation on a 'n' router to purchase (was
looking at linksys) that would assist in some way, please advise and am
will to help

--
Cheers Malcolm °¿° (Linux Counter #276890)
SUSE Linux Enterprise Desktop 11 (x86_64) Kernel 2.6.27.29-0.1-default
up 5 days 6:28, 2 users, load average: 0.26, 0.23, 0.17
GPU GeForce 8600 GTS Silent - Driver Version: 190.18

[Akoellh]

Advise will be difficult, as "N" is still a draft and not a standard.

As a consequence, different vendors have (slightly or even not so slightly) different implementations, meaning that sometimes "Draft N" does not work as expected with certain combinations router/client although both are "Draft N capable", very often (I don't know if it was linksys, netgear or one of the other "well known" vendors) limiting good functionality to their "own" products (client + router) with making other combinations an exciting new type of "roulette".

[malcolmlewis]

Originally Posted by Akoellh

Advise will be difficult, as "N" is still a draft and not a standard.

As a consequence, different vendors have (slightly or even not so
slightly) different implementations, meaning that sometimes "Draft N"
does not work as expected with certain combinations router/client
although both are "Draft N capable", very often (I don't know if it was
linksys, netgear or one of the other "well known" vendors) limiting good
functionality to their "own" products (client + router) with making
other combinations an exciting new type of "roulette".

Hi
Thanks for that Although maybe not for too much longer;
http://www.computerworld.com/s/artic...Wi_Fi_standard

I guess I need to read up on the atheros and their n implementation to
see if there is a good combination with a router.

--
Cheers Malcolm °¿° (Linux Counter #276890)
SUSE Linux Enterprise Desktop 11 (x86_64) Kernel 2.6.27.29-0.1-default
up 5 days 16:37, 2 users, load average: 0.99, 0.54, 0.21
GPU GeForce 8600 GTS Silent - Driver Version: 190.18

[lwfinger]

Malcolm wrote:
> Thanks for that Although maybe not for too much longer;
> http://www.computerworld.com/s/artic...Wi_Fi_standard
>
> I guess I need to read up on the atheros and their n implementation to
> see if there is a good combination with a router.

I just posted a query on the wireless mailing list where the Atheros
developers hang out asking about their implementation and the best AP
for use with ath9k. I'll post what I learn.

[Akoellh]

Originally Posted by malcolmlewis

Hi
Thanks for that Although maybe not for too much longer;
IEEE stamps 'approved' on 802.11n Wi-Fi standard

Good news in fact.

The only question is now, how long it will take for the vendors to (re)act.

[malcolmlewis]

Originally Posted by Larry Finger

Malcolm wrote:
> Thanks for that Although maybe not for too much longer;
> http://www.computerworld.com/s/artic...Wi_Fi_standard
>
> I guess I need to read up on the atheros and their n implementation to
> see if there is a good combination with a router.

I just posted a query on the wireless mailing list where the Atheros
developers hang out asking about their implementation and the best AP
for use with ath9k. I'll post what I learn.

Hi Larry
Thank you for doing that, it's much appreciated

--
Cheers Malcolm °¿° (Linux Counter #276890)
SUSE Linux Enterprise Desktop 11 (x86_64) Kernel 2.6.27.29-0.1-default
up 5 days 19:54, 2 users, load average: 0.09, 0.22, 0.19
GPU GeForce 8600 GTS Silent - Driver Version: 190.18

[malcolmlewis]

Originally Posted by Akoellh

malcolmlewis;2039587 Wrote:
> Hi
> Thanks for that Although maybe not for too much longer;
> 'IEEE stamps 'approved' on 802.11n Wi-Fi standard'
> (http://tinyurl.com/oadojt)

Good news in fact.

The only question is now, how long it will take for the vendors to
(re)act.

Windows 9 maybe?

--
Cheers Malcolm °¿° (Linux Counter #276890)
SUSE Linux Enterprise Desktop 11 (x86_64) Kernel 2.6.27.29-0.1-default
up 5 days 19:55, 2 users, load average: 0.18, 0.21, 0.18
GPU GeForce 8600 GTS Silent - Driver Version: 190.18