RE: [gentoo-user] IPv6 not ready here; Hmmm

[Pandu Poluan]

-original message-
Subject: Re: [gentoo-user] IPv6 not ready here; Hmmm
From: Paul Hartman 
Date: 2011-06-10 03:52

>And another bonus is that there are plenty of
funny things we can spell in hexadecimal. ;)

While I'm sure I'll tag the C001:D00D address for my workstation, I'm not sure 
upper management will appreciate me naming some servers DEAD:BEEF or BAD:D06 or 
A55:401E ... :-P


Rgds,
--
FdS Pandu E Poluan
~ IT Optimizer ~

Sent from Nokia E72-1

[gentoo-user] RE: Kernel Modules

[Pandu Poluan]

-original message-
Subject: Re: [gentoo-user] Re: Kernel Modules
From: Alan McKinnon 
Date: 2011-06-10 02:06

>Apparently, though unproven, at 20:52 on  Thursday 09 June 2011, Bill Longman 
>did opine thusly:
>
>> On 06/09/2011 11:18 AM, Alan McKinnon wrote:
>> > eselect OTOH, is something I always have to run bit by bit to recall the
>> > invocation. That's just way too much effort for this here old git
>> 
>> Wasn't on this list that I saw the correct procedure for eselect?
>> 
>> eselect
>> eselect kernel
>> eselect kernel list
>> eselect kernel set 6
>> 
>>  It's so true
>
>In my case there's usually a lot of wondering which one it is after step 1. 
>Try this, poke around, nope. Try that, nope not that one. Sometime after the 
>third try I find it.
>
>-- 
>alan dot mckinnon at gmail dot com 

eselect bashcomp enable gentoo
eselect bashcomp enable eselect

I always do those when doing a new install. Of course, don't forget to do 
USE=bash-completion and the subsequent emerge --update --newuse --deep @world ;)

Rgds,
--
FdS Pandu E Poluan
~ IT Optimizer ~

Sent from Nokia E72-1

Re: [gentoo-user] Re: Error while `emerge grub`

[Dale]

walt wrote:

On 06/08/2011 10:54 PM, Pandu Poluan wrote:
   

Okay, what's going on here...

While `emerge grub` I got "configure: error: unsupported CPU type"

Here's "/var/tmp/portage/sys-boot/grub-0.97-r10/work/grub-0.97/config.log" :

This file contains any messages produced by compilers while
running configure, to aid debugging if configure makes a mistake.

It was created by GRUB configure 0.97, which was
generated by GNU Autoconf 2.65.  Invocation command line was

   $ ./configure --prefix=/usr --build=x86-pc-linux-gnu
 

Hm.  That should be --build=i686-pc-linux-gnu (for 32-bit machines)

My first thought would be to grep through /etc/* for x86.  At this
minute I don't have a second thought :(

   


I did a bit of googlin and the only thing I can think of is to make sure 
gcc is pointing correctly.  That would be gcc-config -l and that is a 
little L or short for list.  I mention in case you are a bit new.


Hope that helps.

Dale

:-)  :-)

[gentoo-user] Re: Error while `emerge grub`

[walt]

On 06/08/2011 10:54 PM, Pandu Poluan wrote:
> Okay, what's going on here...
> 
> While `emerge grub` I got "configure: error: unsupported CPU type"
> 
> Here's "/var/tmp/portage/sys-boot/grub-0.97-r10/work/grub-0.97/config.log" :
> 
> This file contains any messages produced by compilers while
> running configure, to aid debugging if configure makes a mistake.
> 
> It was created by GRUB configure 0.97, which was
> generated by GNU Autoconf 2.65.  Invocation command line was
> 
>   $ ./configure --prefix=/usr --build=x86-pc-linux-gnu

Hm.  That should be --build=i686-pc-linux-gnu (for 32-bit machines)

My first thought would be to grep through /etc/* for x86.  At this
minute I don't have a second thought :(

[gentoo-user] Yet another Flash security problem

[walt]

This is quoted from today's sans.org security update:

Title: Adobe Flash Player Cross-Site Scripting
Description: Adobe Flash Player is a multimedia application available
for multiple platforms. The application is exposed to an unspecified
cross-site scripting issue. Adobe Flash Player 10.3.181.16 and prior
versions for Windows, Macintosh, Linux and Solaris operating systems
and Adobe Flash Player 10.3.185.22 and prior versions for Android are
affected.
Ref: http://www.adobe.com/support/security/bulletins/apsb11-13.html

Re: [gentoo-user] Support sent you a message: Unique Offer

[Dale]

YouTube Support wrote:
YouTube  	help center 
 | e-mail options 
 | report spam 



Support  has sent you a 
message:


Unique Offer
To:gentoo-user

Unique Offer from YouTube partners:
http://www.youtube.com/t/partnerships_showcase 



You can reply to this message by visiting your inbox 
.


© 2011 YouTube, LLC
901 Cherry Ave, San Bruno, CA 94066



OK.  This is getting old.  Let me introduce you to my little friend.

http://www.youtube.com/watch?v=Tza2L6kfl8E&feature=youtu.be

Dale

:-)  :-)

Re: [gentoo-user] Re: umm, google.... WAS: Kernel Modules

[Dale]

Alan McKinnon wrote:

How exceptionally interesting, as I didn't mail youtube...

   



Shhh.  It's going to pop.  Don't mess with a can of SPAM !!!  It's 
worse than a can of worms.


lol

Dale

:-)  :-)

Re: [gentoo-user] IPv6 not ready here; Hmmm

[Paul Hartman]

On Thu, Jun 9, 2011 at 3:27 PM, Mick  wrote:
> I take it that loading this module would cut both ways.  If I were to allow
> connections to my server only for *my* IP address, then that would be quite
> difficult to achieve if my IP address changed every few minutes.

Since you can have multiple addresses on an interface, you could
theoretically create a separate route just for your connections to
your server, using a consistent address rather than the randomized
one.

And of course in a world where we all get a large pool of address
space to do with what we please, you could skip the whole
auto-configuration thing entirely and just manually assign whatever
addresses you want. And another bonus is that there are plenty of
funny things we can spell in hexadecimal. ;)

Re: [gentoo-user] Can I delete "/usr/lib64/perl5/*_perl/5.12.2/"?

[Mick]

On Thursday 09 Jun 2011 20:35:18 Alan McKinnon wrote:
> Apparently, though unproven, at 21:21 on Thursday 09 June 2011, Grant did
> 
> opine thusly:
> > A while back I installed a bundle of perl modules outside of portage.
> > I never got around to using a cruft script to get rid of them and now
> > that I've moved from perl-5.12.2 to 5.12.3, I'm thinking it might be
> > as easy as:
> > 
> > # rm /usr/lib64/perl5/*_perl/5.12.2
> > 
> > Is that safe?  Should it eliminate all traces of the stale perl modules?
> 
> Yes, that's the way it is done. Also look in /usr/share/{doc|man} and the
> same dir structures in /usr/local/
> 
> If you run "equery files" on a random bunch of perl ebuilds, you'll see
> that most only put files in those perl libs and /usr/share/doc/, so you've
> probably got most stuff covered

and don't forget perl-cleaner.
-- 
Regards,
Mick


signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-user] Re: Can't find b43 in kernel config

[Mick]

On Thursday 09 Jun 2011 19:14:38 Grant wrote:
> >> > I've satisfied these requirements:
> >> > 
> >> > 
> >> > 
> >> > Depends on: NETDEVICES [=y] && WLAN [=y] && SSB_POSSIBLE [=y] &&
> >> > 
> >> > MAC80211 [=n] && HAS_DMA [=y]
> >> > 
> >> > 
> >> > 
> >> > but I don't see the option here:
> >> > 
> >> > 
> >> > 
> >> > Location:
> >> > 
> >> > -> Device Drivers
> >> > 
> >> > -> Network device support (NETDEVICES [=y])
> >> > 
> >> > -> Wireless LAN (WLAN [=y])
> >> > 
> >> > 
> >> > 
> >> > Does anyone know how to make it appear?
> >> > 
> >> > 
> >> > 
> >> > - Grant
> >> 
> >> What is 'b43'?
> >> 
> >> 
> >> 
> >> When I have that problem I typically start with
> >> 
> >> 
> >> 
> >> cat /usr/src/.config | grep b43
> >> 
> >> 
> >> 
> >> which returns nothing implying to me that b43 doesn't exist at all.
> >> 
> >> 
> >> 
> >> - Mark
> > 
> > Mark this isn't a good test. It only says that your config doesn't
> > contain references to b43.
> > 
> > If you do "make menuconfig" and then search for b43 (type slash to reach
> > the search functionality) many entries will show up: I count 23 hits in
> > 2.6.39-gentoo-r1 kernel.
> > 
> > Grant: you say that the requirements are satisfied, yet I see "MAC80211
> > [=n]", could be it? Same way as above to see where the option you lack
> > can be activated
> > 
> > HTH
> > 
> > Francesco
> 
> Ah, thank you Francesco.  I was reading that to mean that it was
> required for MAC80211 to *not* be selected.  Now I've got it.
> 
> - Grant

You'll invariably also need the appropriate firmware too.  You'll need to 
emerge net-wireless/b43-fwcutter and have a look at:

http://linuxwireless.org/en/users/Drivers/b43

PS. Ignore the notes for b43legacy and bcm43xx drivers.
-- 
Regards,
Mick


signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-user] Re: Kernel Modules

[Mick]

On Thursday 09 Jun 2011 21:06:12 Paul Hartman wrote:
> On Thu, Jun 9, 2011 at 2:37 PM, Alan McKinnon  
wrote:
> > Apparently, though unproven, at 21:03 on Thursday 09 June 2011, Dale did
> > opine
> > 
> > thusly:
> >> Alan McKinnon wrote:
> >> > Sure, but I can use ln in my sleep.
> >> > 
> >> > eselect OTOH, is something I always have to run bit by bit to recall
> >> > the invocation. That's just way too much effort for this here old
> >> > git
> >> 
> >> I'm still not used to eselect and its options.  They are sensible but I
> >> just haven't got the hang of it.  I been practicing tho.  I do eselect
> >> modules list then go from there.  You are right tho Alan, in the end, it
> >> takes longer.  I bet it was supposed to save time too. lol
> > 
> > On the whole, eselect is a good idea, just some of it's modules are
> > extreme.
> > 
> > eselect opengl and eselect python are just two examples of things I have
> > no idea how to accomplish manually
> 
> I'm okay with most of them, but whenever I need to use "eselect news"
> my brain comes up empty.

Aha!  I had to memorise that because it kept popping up every time I would run 
emerge (and couldn't be bothered to run eselect at the time).  So it is:

  eselect news read new

it sort of rhymes.  ;-)
-- 
Regards,
Mick


signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-user] IPv6 not ready here; Hmmm

[Mick]

On Thursday 09 Jun 2011 16:51:29 Paul Hartman wrote:
> On Thu, Jun 9, 2011 at 12:46 AM, Mick  wrote:
> >> BTW, Windows Vista and 7 generate randomized host IDs for public IPv6
> >> addresses, it's generally advised to disable that. You can do that by
> >> running this at administrator cmd prompt:
> >> netsh interface ipv6 set global randomizeidentifiers=disabled
> > 
> > I was looking at the same in the Linux kernel scratching my head if I
> > should enable this or not ...
> > 
> > What does it do - not sure I understand what such temporary addresses are
> > used for:
> > 
> >  IPv6: Privacy Extensions (RFC 3041) support
> 
> > CONFIG_IPV6_PRIVACY:
> Sorry, I described the problem poorly. More specifically I should have
> said that it should be disabled because Windows does it /wrong/. :)
> 
> In IPv6, link-local address is required (begins with fe80::) even when
> an internet-routable address exists. It is derived from your network
> prefix and your MAC address. Normally, the public IPv6 address also
> contains your MAC address. Every IPv6 interface is going to have at
> least 2 different addresses.
> 
> Imagine a world where IPv6 is everywhere. You take your laptop home,
> to the cafe, to work, to a hotel on a business trip. Despite using
> different networks in each place, your MAC address will tie them all
> together. The governments and corporations are tracking this and now
> know even more about you. At least, that's what people worry about.
> 
> In Linux, enabling the privacy extensions adds an additional,
> temporary IPv6 address to the interface, with a randomized "MAC" part,
> and it changes regularly (every hour or two? something like that). The
> link-local address still contains the MAC-based IPv6 address, and the
> standard routable IPv6 address is also available but not used by
> default for outgoing connections. So, inside your network, things are
> predictable and unchanging, which makes management of clients, routing
> of traffic, firewall rules, etc. easier to deal with. To the outside
> world, your IP address is constantly changing and can't be used to
> track you as easily as it would be if the MAC portion of the address
> were consistent.
> 
> In Windows, however, when that option is enabled, they wrongly
> randomize ALL of the addresses, even the local, rather than just
> creating a temp random public address. Which means every time that
> machine reboots it's going to look like a new client on the local
> network, and any local network setup you have pertaining to a certain
> IP are going to be a pain to maintain. Depending on your usage, maybe
> that doesn't matter, but in general, on Windows machines, it's
> considered a buggy implementation and is undesired.
> 
> In Linux, it should be absolutely fine to use. In your
> /etc/sysctl.conf you can add these lines to enable it on every
> interface by default, assuming you enabled in your kernel config:
> 
> net.ipv6.conf.all.use_tempaddr = 2
> net.ipv6.conf.default.use_tempaddr = 2

Excellent explanation!  Thank you.  :-)

Now was it that difficult to add a couple of meaningful lines in the kernel 
documentation, so that any other than the kernel hacker who wrote that module 
would learn that its there to anonymise your ipv6 address for privacy 
purposes?

I take it that loading this module would cut both ways.  If I were to allow 
connections to my server only for *my* IP address, then that would be quite 
difficult to achieve if my IP address changed every few minutes.
-- 
Regards,
Mick


signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-user] Re: umm, google.... WAS: Kernel Modules

[Alan McKinnon]

How exceptionally interesting, as I didn't mail youtube...

Headers of what I did send:

From: Alan McKinnon 
X-KMail-Identity: 1726407999
X-KMail-Transport: Gmail
X-KMail-Fcc: .Community.directory/gentoo-user
To: gentoo-user@lists.gentoo.org
Subject: Re: [gentoo-user] Re: Kernel Modules
Date: Thu, 9 Jun 2011 21:37:30 +0200
User-Agent: KMail/1.13.7 (Linux/2.6.39-ck; KDE/4.6.3; x86_64; ; )
References: <4df0a5fa.1020...@gmail.com> 
<201106092018.37311.alan.mckin...@gmail.com> <4df118fc.2030...@gmail.com>
In-Reply-To: <4df118fc.2030...@gmail.com>
X-KMail-Link-Message: 96095
X-KMail-Link-Type: reply
X-KMail-QuotePrefix: > 
MIME-Version: 1.0
Content-Type: Text/Plain;
  charset="utf-8"
Content-Transfer-Encoding: 7bit
Message-Id: <201106092137.30415.alan.mckin...@gmail.com>
Status: RO
X-Status: RS
X-KMail-EncryptionState: N
X-KMail-SignatureState: N
X-KMail-MDN-Sent:  

I wonder if this one will do the same.



Apparently, though unproven, at 21:44 on Thursday 09 June 2011, YouTube 
Support did opine thusly:

> You are receiving this automated reply because you have sent mail to
> an invalid email address. If you are trying to contact YouTube,
> please visit the YouTube Help Center at:
> http://www.google.com/support/youtube
> 
> If you're unable to find the answer to your question in the Help
> Center, you may use the contact forms located there to send us your
> question.
> 
> 
> 
> Original Message Follows:
> 
> From: Alan McKinnon 
> Subject: [gentoo-user] Re: Kernel Modules
> Date: Thu, 9 Jun 2011 21:37:30 +0200
> 
> Apparently, though unproven, at 21:03 on Thursday 09 June 2011, Dale did
> opine
> 
> thusly:
> > Alan McKinnon wrote:
> > > Sure, but I can use ln in my sleep.
> > > 
> > > eselect OTOH, is something I always have to run bit by bit to recall
> 
> the
> 
> > > invocation. That's just way too much effort for this here old git
> > 
> > I'm still not used to eselect and its options.  They are sensible but I
> > just haven't got the hang of it.  I been practicing tho.  I do eselect
> > modules list then go from there.  You are right tho Alan, in the end, it
> > takes longer.  I bet it was supposed to save time too. lol
> 
> On the whole, eselect is a good idea, just some of it's modules are
> extreme.
> 
> eselect opengl and eselect python are just two examples of things I have
> no
> idea how to accomplish manually

-- 
alan dot mckinnon at gmail dot com

Re: [gentoo-user] Re: Kernel Modules

[Paul Hartman]

On Thu, Jun 9, 2011 at 2:37 PM, Alan McKinnon  wrote:
> Apparently, though unproven, at 21:03 on Thursday 09 June 2011, Dale did opine
> thusly:
>
>> Alan McKinnon wrote:
>> > Sure, but I can use ln in my sleep.
>> >
>> > eselect OTOH, is something I always have to run bit by bit to recall the
>> > invocation. That's just way too much effort for this here old git
>>
>> I'm still not used to eselect and its options.  They are sensible but I
>> just haven't got the hang of it.  I been practicing tho.  I do eselect
>> modules list then go from there.  You are right tho Alan, in the end, it
>> takes longer.  I bet it was supposed to save time too. lol
>
> On the whole, eselect is a good idea, just some of it's modules are extreme.
>
> eselect opengl and eselect python are just two examples of things I have no
> idea how to accomplish manually

I'm okay with most of them, but whenever I need to use "eselect news"
my brain comes up empty.

Re: [gentoo-user] Re: Kernel Modules

[YouTube Support]

You are receiving this automated reply because you have sent mail to
an invalid email address. If you are trying to contact YouTube,
please visit the YouTube Help Center at:
http://www.google.com/support/youtube

If you're unable to find the answer to your question in the Help
Center, you may use the contact forms located there to send us your
question.



Original Message Follows:

From: Alan McKinnon 
Subject: [gentoo-user] Re: Kernel Modules
Date: Thu, 9 Jun 2011 21:37:30 +0200

Apparently, though unproven, at 21:03 on Thursday 09 June 2011, Dale did
opine 
thusly:

> Alan McKinnon wrote:
> > Sure, but I can use ln in my sleep.
> > 
> > eselect OTOH, is something I always have to run bit by bit to recall
the
> > invocation. That's just way too much effort for this here old git
> 
> I'm still not used to eselect and its options.  They are sensible but I
> just haven't got the hang of it.  I been practicing tho.  I do eselect
> modules list then go from there.  You are right tho Alan, in the end, it
> takes longer.  I bet it was supposed to save time too. lol

On the whole, eselect is a good idea, just some of it's modules are
extreme.

eselect opengl and eselect python are just two examples of things I have
no 
idea how to accomplish manually


-- 
alan dot mckinnon at gmail dot com

[gentoo-user] Support sent you a message: Unique Offer

[YouTube Support]

Support has sent you a message:
  Unique Offer
  To:gentoo-user

Unique Offer from YouTube partners:
http://www.youtube.com/t/partnerships_showcase

You can reply to this message by visiting your inbox:
  http://www.youtube.com/inbox

Re: [gentoo-user] Re: Kernel Modules

[Alan McKinnon]

Apparently, though unproven, at 21:03 on Thursday 09 June 2011, Dale did opine 
thusly:

> Alan McKinnon wrote:
> > Sure, but I can use ln in my sleep.
> > 
> > eselect OTOH, is something I always have to run bit by bit to recall the
> > invocation. That's just way too much effort for this here old git
> 
> I'm still not used to eselect and its options.  They are sensible but I
> just haven't got the hang of it.  I been practicing tho.  I do eselect
> modules list then go from there.  You are right tho Alan, in the end, it
> takes longer.  I bet it was supposed to save time too. lol

On the whole, eselect is a good idea, just some of it's modules are extreme.

eselect opengl and eselect python are just two examples of things I have no 
idea how to accomplish manually


-- 
alan dot mckinnon at gmail dot com

Re: [gentoo-user] Can I delete "/usr/lib64/perl5/*_perl/5.12.2/"?

[Alan McKinnon]

Apparently, though unproven, at 21:21 on Thursday 09 June 2011, Grant did 
opine thusly:

> A while back I installed a bundle of perl modules outside of portage.
> I never got around to using a cruft script to get rid of them and now
> that I've moved from perl-5.12.2 to 5.12.3, I'm thinking it might be
> as easy as:
> 
> # rm /usr/lib64/perl5/*_perl/5.12.2
> 
> Is that safe?  Should it eliminate all traces of the stale perl modules?

Yes, that's the way it is done. Also look in /usr/share/{doc|man} and the same 
dir structures in /usr/local/

If you run "equery files" on a random bunch of perl ebuilds, you'll see that 
most only put files in those perl libs and /usr/share/doc/, so you've probably 
got most stuff covered

-- 
alan dot mckinnon at gmail dot com

[gentoo-user] Can I delete "/usr/lib64/perl5/*_perl/5.12.2/"?

[Grant]

A while back I installed a bundle of perl modules outside of portage.
I never got around to using a cruft script to get rid of them and now
that I've moved from perl-5.12.2 to 5.12.3, I'm thinking it might be
as easy as:

# rm /usr/lib64/perl5/*_perl/5.12.2

Is that safe?  Should it eliminate all traces of the stale perl modules?

- Grant

Re: [gentoo-user] Re: Kernel Modules

[Alan McKinnon]

Apparently, though unproven, at 20:52 on Thursday 09 June 2011, Bill Longman 
did opine thusly:

> On 06/09/2011 11:18 AM, Alan McKinnon wrote:
> > eselect OTOH, is something I always have to run bit by bit to recall the
> > invocation. That's just way too much effort for this here old git
> 
> Wasn't on this list that I saw the correct procedure for eselect?
> 
> eselect
> eselect kernel
> eselect kernel list
> eselect kernel set 6
> 
>  It's so true

In my case there's usually a lot of wondering which one it is after step 1. 
Try this, poke around, nope. Try that, nope not that one. Sometime after the 
third try I find it.


-- 
alan dot mckinnon at gmail dot com 

Re: [gentoo-user] Re: Kernel Modules

[Dale]

Alan McKinnon wrote:


Sure, but I can use ln in my sleep.

eselect OTOH, is something I always have to run bit by bit to recall the
invocation. That's just way too much effort for this here old git

   


I'm still not used to eselect and its options.  They are sensible but I 
just haven't got the hang of it.  I been practicing tho.  I do eselect 
modules list then go from there.  You are right tho Alan, in the end, it 
takes longer.  I bet it was supposed to save time too. lol


Dale

:-)  :-)

Re: [gentoo-user] Re: Kernel Modules

[Bill Longman]

On 06/09/2011 11:18 AM, Alan McKinnon wrote:
> eselect OTOH, is something I always have to run bit by bit to recall the 
> invocation. That's just way too much effort for this here old git

Wasn't on this list that I saw the correct procedure for eselect?

eselect
eselect kernel
eselect kernel list
eselect kernel set 6

 It's so true

Re: [gentoo-user] Can't find b43 in kernel config

[Grant]

>> I've satisfied these requirements:
>>
>> Depends on: NETDEVICES [=y] && WLAN [=y] && SSB_POSSIBLE [=y] &&
>> MAC80211 [=n] && HAS_DMA [=y]
>
> It looks like you did not satisfy MAC80211.

Yeah, I misunderstood the depends line.

> Where did copy that Depends from? When I use 'h' in menuconfig mine shows up
> as [=m]

I didn't know about that 'h' command.  I got the Depends from
2.6.38-hardened-r6 by using '/'.

>> but I don't see the option here:
>>
>> Location:
>> -> Device Drivers
>> -> Network device support (NETDEVICES [=y])
>> -> Wireless LAN (WLAN [=y])
>
> It's very close to the middle item in the list (so you know where it goes once
> it's visible)

Got it now, thank you.

- Grant


>> Does anyone know how to make it appear?
>>
>> - Grant

Re: [gentoo-user] Re: Kernel Modules

[Alan McKinnon]

Apparently, though unproven, at 19:32 on Thursday 09 June 2011, Nikos 
Chantziaras did opine thusly:

> On 06/09/2011 03:12 PM, Ignas Anikevicius wrote:
> > On 09/06/11 12:06, Alan McKinnon wrote:
> >> why you making so much work for yourself?
> >> 
> >> set the /usr/src/linux symlink to each set of installed sources in turn,
> >> 
> >>   run emerge @module-rebuild
> >>   or run module-rebuild rebuild
> >> 
> >> you could even script it
> >> 
> >> cd /usr/src
> >> for I in linux-*
> >> do
> >> 
> >>ln -sfn $I linux
> >>module-rebuild rebuils
> >> 
> >> done
> >> 
> >> Fixing my bash syntax errors is left as an exercise for the interested
> >> reader
> > 
> > Thanks very much!
> 
> You actually don't need to the symlinks yourself.  Try:
> 
>eselect kernel list
> 
> Then choose one with something like:
> 
>eselect kernel set 2

Sure, but I can use ln in my sleep.

eselect OTOH, is something I always have to run bit by bit to recall the 
invocation. That's just way too much effort for this here old git


-- 
alan dot mckinnon at gmail dot com

Re: [gentoo-user] debugfs

[Alan McKinnon]

Apparently, though unproven, at 18:40 on Thursday 09 June 2011, Stéphane 
Guedon did opine thusly:

> 16:05 root@luciole /boot # mount
> rootfs on / type rootfs (rw)
> /dev/root on / type ext3 (rw,commit=0)
> proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
> rc-svcdir on /lib64/rc/init.d type tmpfs
> (rw,nosuid,nodev,noexec,relatime,size=1024k,mode=755)
> sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,relatime)
> debugfs on /sys/kernel/debug type debugfs (rw,nosuid,nodev,noexec,relatime)
> udev on /dev type tmpfs (rw,nosuid,relatime,size=10240k,mode=755)
> devpts on /dev/pts type devpts (rw,nosuid,noexec,relatime,gid=5,mode=620)
> shm on /dev/shm type tmpfs (rw,nosuid,nodev,noexec,relatime)
> /dev/sda6 on /home type ext4 (rw,commit=0)
> 
> Is debugfs usefull ? If not, is there a way not to mount it ?

deslect it in kernel config

It's not something you mount like a regular disk-based fs


-- 
alan dot mckinnon at gmail dot com

Re: [gentoo-user] debugfs

[Paul Hartman]

On Thu, Jun 9, 2011 at 12:16 PM, Stéphane Guedon  wrote:
> But it seems every debugfs is disabled in my kernel. But it always mount !

Hmmm, I think the main setting is in "Kernel Hacking" section. In your
.config it is CONFIG_DEBUG_FS=n

Re: [gentoo-user] Can't find b43 in kernel config

[Alan McKinnon]

Apparently, though unproven, at 19:10 on Thursday 09 June 2011, Grant did 
opine thusly:

> I've satisfied these requirements:
> 
> Depends on: NETDEVICES [=y] && WLAN [=y] && SSB_POSSIBLE [=y] &&
> MAC80211 [=n] && HAS_DMA [=y] 

It looks like you did not satisfy MAC80211.

Where did copy that Depends from? When I use 'h' in menuconfig mine shows up 
as [=m]

> 
> but I don't see the option here:
> 
> Location:
> -> Device Drivers
> -> Network device support (NETDEVICES [=y])
> -> Wireless LAN (WLAN [=y])

It's very close to the middle item in the list (so you know where it goes once 
it's visible)


> 
> Does anyone know how to make it appear?
> 
> - Grant

-- 
alan dot mckinnon at gmail dot com

Re: [gentoo-user] Re: Can't find b43 in kernel config

[Grant]

>> > I've satisfied these requirements:
>
>> >
>
>> > Depends on: NETDEVICES [=y] && WLAN [=y] && SSB_POSSIBLE [=y] &&
>
>> > MAC80211 [=n] && HAS_DMA [=y]
>
>> >
>
>> > but I don't see the option here:
>
>> >
>
>> > Location:
>
>> > -> Device Drivers
>
>> > -> Network device support (NETDEVICES [=y])
>
>> > -> Wireless LAN (WLAN [=y])
>
>> >
>
>> > Does anyone know how to make it appear?
>
>> >
>
>> > - Grant
>
>>
>
>> What is 'b43'?
>
>>
>
>> When I have that problem I typically start with
>
>>
>
>> cat /usr/src/.config | grep b43
>
>>
>
>> which returns nothing implying to me that b43 doesn't exist at all.
>
>>
>
>> - Mark
>
> Mark this isn't a good test. It only says that your config doesn't contain
> references to b43.
>
> If you do "make menuconfig" and then search for b43 (type slash to reach the
> search functionality) many entries will show up: I count 23 hits in
> 2.6.39-gentoo-r1 kernel.
>
> Grant: you say that the requirements are satisfied, yet I see "MAC80211
> [=n]", could be it? Same way as above to see where the option you lack can
> be activated
>
> HTH
>
> Francesco

Ah, thank you Francesco.  I was reading that to mean that it was
required for MAC80211 to *not* be selected.  Now I've got it.

- Grant

[gentoo-user] Re: Can't find b43 in kernel config

[Mark Knecht]

On Thu, Jun 9, 2011 at 11:07 AM, Francesco Talamona
 wrote:

>> What is 'b43'?
>
>>
>
>> When I have that problem I typically start with
>
>>
>
>> cat /usr/src/.config | grep b43
>
>>
>
>> which returns nothing implying to me that b43 doesn't exist at all.
>
>>
>
>> - Mark
>
> Mark this isn't a good test. It only says that your config doesn't contain
> references to b43.
>
> If you do "make menuconfig" and then search for b43 (type slash to reach the
> search functionality) many entries will show up: I count 23 hits in
> 2.6.39-gentoo-r1 kernel.
>
> Grant: you say that the requirements are satisfied, yet I see "MAC80211
> [=n]", could be it? Same way as above to see where the option you lack can
> be activated
>
> HTH
>
> Francesco
>

Thanks Francesco. I didn't know that the config would eliminate
options. My experience in the past had been that it would be in the
config but commented out and saying 'not set'.

Good to know this if I have a problem like this again in the future.

Cheers,
Mark

[gentoo-user] Re: Can't find b43 in kernel config

[Francesco Talamona]

On Thursday 09 June 2011, Mark Knecht wrote:
> On Thu, Jun 9, 2011 at 10:10 AM, Grant  wrote:
> > I've satisfied these requirements:
> > 
> > Depends on: NETDEVICES [=y] && WLAN [=y] && SSB_POSSIBLE [=y] &&
> > MAC80211 [=n] && HAS_DMA [=y]
> > 
> > but I don't see the option here:
> > 
> > Location:
> > -> Device Drivers
> > -> Network device support (NETDEVICES [=y])
> > -> Wireless LAN (WLAN [=y])
> > 
> > Does anyone know how to make it appear?
> > 
> > - Grant
> 
> What is 'b43'?
> 
> When I have that problem I typically start with
> 
> cat /usr/src/.config | grep b43
> 
> which returns nothing implying to me that b43 doesn't exist at all.
> 
> - Mark

Mark this isn't a good test. It only says that your config doesn't 
contain references to b43.

If you do "make menuconfig" and then search for b43 (type slash to reach 
the search functionality) many entries will show up: I count 23 hits in 
2.6.39-gentoo-r1 kernel.

Grant: you say that the requirements are satisfied, yet I see "MAC80211 
[=n]", could be it? Same way as above to see where the option you lack 
can be activated

HTH
Francesco

-- 
Linux Version 2.6.39-gentoo, Compiled #1 SMP PREEMPT Thu May 26 20:13:26 
CEST 2011
Two 2GHz AMD Athlon 64 X2 Processors, 4GB RAM, 8041.06 Bogomips Total
aemaeth

Re: [gentoo-user] Can't find b43 in kernel config

[Grant]

>> I've satisfied these requirements:
>>
>> Depends on: NETDEVICES [=y] && WLAN [=y] && SSB_POSSIBLE [=y] &&
>> MAC80211 [=n] && HAS_DMA [=y]
>>
>> but I don't see the option here:
>>
>> Location:
>> -> Device Drivers
>> -> Network device support (NETDEVICES [=y])
>> -> Wireless LAN (WLAN [=y])
>>
>> Does anyone know how to make it appear?
>>
>> - Grant
>
> What is 'b43'?
>
> When I have that problem I typically start with
>
> cat /usr/src/.config | grep b43
>
> which returns nothing implying to me that b43 doesn't exist at all.
>
> - Mark

If you do a make menuconfig and search for b43, lots of stuff comes up
including "B43".

- Grant

[gentoo-user] Re: Kernel Modules

[Nikos Chantziaras]

On 06/09/2011 03:12 PM, Ignas Anikevicius wrote:

On 09/06/11 12:06, Alan McKinnon wrote:

why you making so much work for yourself?

set the /usr/src/linux symlink to each set of installed sources in turn,
  run emerge @module-rebuild
  or run module-rebuild rebuild

you could even script it

cd /usr/src
for I in linux-*
do
   ln -sfn $I linux
   module-rebuild rebuils
done

Fixing my bash syntax errors is left as an exercise for the interested reader

Thanks very much!


You actually don't need to the symlinks yourself.  Try:

  eselect kernel list

Then choose one with something like:

  eselect kernel set 2

Re: [gentoo-user] Can't find b43 in kernel config

[Mark Knecht]

On Thu, Jun 9, 2011 at 10:10 AM, Grant  wrote:
> I've satisfied these requirements:
>
> Depends on: NETDEVICES [=y] && WLAN [=y] && SSB_POSSIBLE [=y] &&
> MAC80211 [=n] && HAS_DMA [=y]
>
> but I don't see the option here:
>
> Location:
> -> Device Drivers
> -> Network device support (NETDEVICES [=y])
> -> Wireless LAN (WLAN [=y])
>
> Does anyone know how to make it appear?
>
> - Grant

What is 'b43'?

When I have that problem I typically start with

cat /usr/src/.config | grep b43

which returns nothing implying to me that b43 doesn't exist at all.

- Mark

Re: [gentoo-user] debugfs

[Stéphane Guedon]

On Thursday 09 June 2011 18:52:29 Paul Hartman wrote:
> On Thu, Jun 9, 2011 at 11:40 AM, Stéphane Guedon  
wrote:
> > 16:05 root@luciole /boot # mount
> > rootfs on / type rootfs (rw)
> > /dev/root on / type ext3 (rw,commit=0)
> > proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
> > rc-svcdir on /lib64/rc/init.d type tmpfs
> > (rw,nosuid,nodev,noexec,relatime,size=1024k,mode=755)
> > sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,relatime)
> > debugfs on /sys/kernel/debug type debugfs
> > (rw,nosuid,nodev,noexec,relatime) udev on /dev type tmpfs
> > (rw,nosuid,relatime,size=10240k,mode=755) devpts on /dev/pts type devpts
> > (rw,nosuid,noexec,relatime,gid=5,mode=620) shm on /dev/shm type tmpfs
> > (rw,nosuid,nodev,noexec,relatime)
> > /dev/sda6 on /home type ext4 (rw,commit=0)
> > 
> > Is debugfs usefull ? If not, is there a way not to mount it ?
> 
> If you don't use it, it's not useful. :) Disable it in your kernel and
> it will go away.
> 
> debugfs is not a real filesystem, it's a virtual way to access
> debugging info from various modules/programs. Kind of like /proc is a
> virtual filesystem which shows info about processes. If you don't use
> any of that debugging info, then it's useless to you.

of course I won't use it
But it seems every debugfs is disabled in my kernel. But it always mount !
-- 
Stéphane Guedon
page web : http://www.22decembre.eu/
carte de visite : http://www.22decembre.eu/downloads/Stephane-Guedon.vcf
clé publique gpg : http://www.22decembre.eu/downloads/Stephane-Guedon.asc


signature.asc
Description: This is a digitally signed message part.

[gentoo-user] Can't find b43 in kernel config

[Grant]

I've satisfied these requirements:

Depends on: NETDEVICES [=y] && WLAN [=y] && SSB_POSSIBLE [=y] &&
MAC80211 [=n] && HAS_DMA [=y]

but I don't see the option here:

Location:
-> Device Drivers
-> Network device support (NETDEVICES [=y])
-> Wireless LAN (WLAN [=y])

Does anyone know how to make it appear?

- Grant

Re: [gentoo-user] debugfs

[Paul Hartman]

On Thu, Jun 9, 2011 at 11:40 AM, Stéphane Guedon  wrote:
> 16:05 root@luciole /boot # mount
> rootfs on / type rootfs (rw)
> /dev/root on / type ext3 (rw,commit=0)
> proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
> rc-svcdir on /lib64/rc/init.d type tmpfs
> (rw,nosuid,nodev,noexec,relatime,size=1024k,mode=755)
> sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,relatime)
> debugfs on /sys/kernel/debug type debugfs (rw,nosuid,nodev,noexec,relatime)
> udev on /dev type tmpfs (rw,nosuid,relatime,size=10240k,mode=755)
> devpts on /dev/pts type devpts (rw,nosuid,noexec,relatime,gid=5,mode=620)
> shm on /dev/shm type tmpfs (rw,nosuid,nodev,noexec,relatime)
> /dev/sda6 on /home type ext4 (rw,commit=0)
>
> Is debugfs usefull ? If not, is there a way not to mount it ?

If you don't use it, it's not useful. :) Disable it in your kernel and
it will go away.

debugfs is not a real filesystem, it's a virtual way to access
debugging info from various modules/programs. Kind of like /proc is a
virtual filesystem which shows info about processes. If you don't use
any of that debugging info, then it's useless to you.

[gentoo-user] debugfs

[Stéphane Guedon]

16:05 root@luciole /boot # mount
rootfs on / type rootfs (rw)
/dev/root on / type ext3 (rw,commit=0)
proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
rc-svcdir on /lib64/rc/init.d type tmpfs 
(rw,nosuid,nodev,noexec,relatime,size=1024k,mode=755)
sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,relatime)
debugfs on /sys/kernel/debug type debugfs (rw,nosuid,nodev,noexec,relatime)
udev on /dev type tmpfs (rw,nosuid,relatime,size=10240k,mode=755)
devpts on /dev/pts type devpts (rw,nosuid,noexec,relatime,gid=5,mode=620)
shm on /dev/shm type tmpfs (rw,nosuid,nodev,noexec,relatime)
/dev/sda6 on /home type ext4 (rw,commit=0)

Is debugfs usefull ? If not, is there a way not to mount it ?
-- 
Stéphane Guedon
page web : http://www.22decembre.eu/
carte de visite : http://www.22decembre.eu/downloads/Stephane-Guedon.vcf
clé publique gpg : http://www.22decembre.eu/downloads/Stephane-Guedon.asc


signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-user] IPv6 not ready here; Hmmm

[Paul Hartman]

On Thu, Jun 9, 2011 at 6:16 AM, Tanstaafl  wrote:
> On 2011-06-08 9:25 PM, Paul Hartman wrote:
>> After that, machines on my local network (including wifi) can get both
>> IPv4 and IPv6 addresses from the router and can talk to the outside
>> world on either network.
>
> I'm getting a headache...
>
> Is there a decent guide that explains IPV6 for noobs who don't speak IP?
> Meaning, in plain english, how to set it up and make it work, without
> having to understand all of the granular technical aspects?

Short version - if your ISP and your networking hardware
(gateway/router/firewall/whatever) already support IPv6, you simply
need to enable all the IPv6 stuff in your kernel, enable "ipv6" USE
flag in your /etc/make.conf and rebuild affected packages. If you use
DHCP/autoconfig it should just work automatically.

Otherwise, you need to jump through the hoops we're talking about to
establish tunnels or other means of getting IPv6 over a network that
is not IPv6-capable. You can decide if you care enough about that kind
of thing to shed your noob-ness and get into it more. :)

The only real benefit of being on IPv6 at the moment is that every
device has an internet-accissble address. That means no more NAT
forwarding from your router to ports on certain devices. Otherwise,
there's basically no perceivable benefits from using IPv6, other than
the geek cred you earn by saying you're on IPv6. :)

Another benefit, a side-effect of the fact that that most places are
NOT ready for IPv6 yet, means many internet filters and loggers ignore
IPv6 packages entirely. I've read that using IPv6 is one way to get
around the great firewalls of oppressive regimes like China, Iran and
universities. :) I don't expect that to last very long once more
people start using it.

For an unscientific example of how many people are using IPv6, the
Mainline DHT network shows several million clients connected on IPv4
but only 78 clients on IPv6...

> Also - how long is it going to be before there are parts of the internet
> that you can't get to without speaking IPV6?

There are some v6-only sites now, but they are basically sites that
also exist on ipv4 internet and are used for testing/proof-of-concept.
Given the billions of non-v6-capable devices on the internet, it would
be commercial suicide for a company to leave the IPv4 Internet any
time soon. I would guess you should be fine for another 10 years using
IPv4-only.

Re: [gentoo-user] sys-fs/udev-171-r1 dependency conflict

[Einux]

it gets normal after I re-synced the portage tree.

Thanks for your detailed analysis ^_^

On Thu, Jun 9, 2011 at 9:06 PM, Alan McKinnon wrote:

> Apparently, though unproven, at 12:38 on Thursday 09 June 2011, Einux did
> opine thusly:
>
> > I just updated the portage tree. But there seems to be a dependency
> > conflict when I try to update the system. This is the first time I
> > encounter this kind of problem. Could you guys help me out? Thanks in
> > advance^_^
> >
> > ---
> >
> > $ sudo emerge -avuDN world
> >
> > These are the packages that would be merged, in order:
> >
> > Calculating dependencies... done!
> > [ebuild  NS   ~] sys-kernel/gentoo-sources-2.6.39-r1 [2.6.39] USE="-build
> > -deblob -symlink" 72 kB
> > [ebuild U  ] dev-libs/libdbusmenu-qt-0.8.2 [0.6.2] USE="-debug -doc%
> > -test" 36 kB
> > [ebuild U ~] sys-fs/udev-171-r1 [151-r4] USE="acl%* gudev%* hwdb%*
> > keymap%* rule_generator%* -action_modeswitch% -debug% -edd% -floppy%
> > (-introspection) (-selinux) -test (-devfs-compat%)
> > (-extras%*) (-old-hd-rules%)" 595 kB
> > [ebuild U  ] net-ftp/lftp-4.2.3 [4.1.3] USE="nls ssl -gnutls -socks5"
> > 1,296 kB
> > [ebuild U ~] media-sound/pulseaudio-0.9.22-r2 [0.9.22] USE="X alsa
> > asyncns bluetooth caps dbus glib ipv6 tcpd udev -avahi -doc -gnome -jack
> > -libsamplerate -lirc (-oss) -realtime
> > (-system-wide) -test" 0 kB
> > [ebuild U ~] sys-fs/udisks-1.0.2-r4 [1.0.2-r1] USE="nls
> > -bash-completion -debug -doc -remote-access" 0 kB
> >
> > Total: 6 packages (5 upgrades, 1 in new slot), Size of downloads: 1,997
> kB
> >
> > !!! Multiple package instances within a single package slot have been
> > pulled !!! into the dependency graph, resulting in a slot conflict:
> >
> > sys-fs/udev:0
> >
> >   (sys-fs/udev-171-r1::gentoo, ebuild scheduled for merge) pulled in by
> >
> > >=sys-fs/udev-171[hwdb] required by
> >
> > (media-sound/pulseaudio-0.9.22-r2::gentoo, ebuild scheduled for merge)
> >
> >   (sys-fs/udev-151-r4::gentoo, installed) pulled in by
> >
> > >=sys-fs/udev-147[extras] required by
> (sys-fs/udisks-1.0.2-r4::gentoo,
> >
> > ebuild scheduled for merge)
> > (and 3 more with the same problem)
>
> There's no way to fix this right now.
>
> You need udev version 171 or greater with the hwdb USE flag set as required
> by
> pulseaudio (which is fine, your system will do this OK)
>
> You also need udev version 147 or greater with the extras USE flag set as
> required by udisks. udev-171 does not meet these requirements as it does
> not
> have the extras USE flag (check the output of -p above, it is not listed).
>
> So portage will look for earlier and earlier versions of udev to find one
> that
> does support extras, but the first one it finds conflict with what
> pulseaudio
> needs.
>
> You have some options:
>
> - it's a bug, it needs to be reported. But you will not need to do this as
> every unstable user has it also and there will be 100s of bugs already
> (udev
> is needed by almost everyone)
>
> - mask/downgrade pulseaudio to a version that can use an earlier udev
>
> - wait a day, resync, try again.
>
> The third option is usually the one that works best.
>
>
> >
> >
> > It may be possible to solve this problem by using package.mask to
> > prevent one of those packages from being selected. However, it is also
> > possible that conflicting dependencies exist such that they are
> > impossible to satisfy simultaneously.  If such a conflict exists in
> > the dependencies of two different packages, then those packages can
> > not be installed simultaneously. You may want to try a larger value of
> > the --backtrack option, such as --backtrack=30, in order to see if
> > that will solve this conflict automatically.
> >
> > For more information, see MASKED PACKAGES section in the emerge man
> > page or refer to the Gentoo Handbook.
>
> --
> alan dot mckinnon at gmail dot com
>
>


-- 
Best Regards,
Einux

Re: [gentoo-user] IPv6 not ready here; Hmmm

[Paul Hartman]

On Thu, Jun 9, 2011 at 12:46 AM, Mick  wrote:
>> BTW, Windows Vista and 7 generate randomized host IDs for public IPv6
>> addresses, it's generally advised to disable that. You can do that by
>> running this at administrator cmd prompt:
>> netsh interface ipv6 set global randomizeidentifiers=disabled
>
> I was looking at the same in the Linux kernel scratching my head if I should
> enable this or not ...
>
> What does it do - not sure I understand what such temporary addresses are used
> for:
> 
>  IPv6: Privacy Extensions (RFC 3041) support
>
> CONFIG_IPV6_PRIVACY:

Sorry, I described the problem poorly. More specifically I should have
said that it should be disabled because Windows does it /wrong/. :)

In IPv6, link-local address is required (begins with fe80::) even when
an internet-routable address exists. It is derived from your network
prefix and your MAC address. Normally, the public IPv6 address also
contains your MAC address. Every IPv6 interface is going to have at
least 2 different addresses.

Imagine a world where IPv6 is everywhere. You take your laptop home,
to the cafe, to work, to a hotel on a business trip. Despite using
different networks in each place, your MAC address will tie them all
together. The governments and corporations are tracking this and now
know even more about you. At least, that's what people worry about.

In Linux, enabling the privacy extensions adds an additional,
temporary IPv6 address to the interface, with a randomized "MAC" part,
and it changes regularly (every hour or two? something like that). The
link-local address still contains the MAC-based IPv6 address, and the
standard routable IPv6 address is also available but not used by
default for outgoing connections. So, inside your network, things are
predictable and unchanging, which makes management of clients, routing
of traffic, firewall rules, etc. easier to deal with. To the outside
world, your IP address is constantly changing and can't be used to
track you as easily as it would be if the MAC portion of the address
were consistent.

In Windows, however, when that option is enabled, they wrongly
randomize ALL of the addresses, even the local, rather than just
creating a temp random public address. Which means every time that
machine reboots it's going to look like a new client on the local
network, and any local network setup you have pertaining to a certain
IP are going to be a pain to maintain. Depending on your usage, maybe
that doesn't matter, but in general, on Windows machines, it's
considered a buggy implementation and is undesired.

In Linux, it should be absolutely fine to use. In your
/etc/sysctl.conf you can add these lines to enable it on every
interface by default, assuming you enabled in your kernel config:

net.ipv6.conf.all.use_tempaddr = 2
net.ipv6.conf.default.use_tempaddr = 2

Re: [gentoo-user] IPv6 not ready here; Hmmm

[Paul Hartman]

On Thu, Jun 9, 2011 at 12:46 AM, Mick  wrote:
>> On my wife's Windows 7 laptop, it just worked perfectly after I
>> enabled it on my router and her wifi reconnected. All tests on
>> test-ipv6.com pass except for the last DNS test. She can go to sites
>> like http://www.v6.facebook.com no problems.
>
> That's because Windows7 use toredo servers/relays to resolve and connect to
> ipv6 addresses.

I disagree. :) It gets an IPv6 address from my ISP and no traffic
flows through Microsoft's relay servers. It works the same as my
gentoo boxes and my cell phone. The only tunnel is the one between my
router and my ISP's 6RD server. Everything on my side of the network
just uses native regular IPv6 as far as the clients are concerned.

AFAIK, Win vista/7 only uses teredo when no actual IPv6 internet
connectivity exists. And when it's not part of a corporate domain. And
when the host you're connecting to does not have an IPv4 address.
Again, that's AFAIK. :)

Re: [gentoo-user] IPv6 not ready here; Hmmm

[Mick]

On 9 June 2011 12:16, Tanstaafl  wrote:
> On 2011-06-08 9:25 PM, Paul Hartman wrote:
>> After that, machines on my local network (including wifi) can get both
>> IPv4 and IPv6 addresses from the router and can talk to the outside
>> world on either network.
>
> I'm getting a headache...
>
> Is there a decent guide that explains IPV6 for noobs who don't speak IP?
> Meaning, in plain english, how to set it up and make it work, without
> having to understand all of the granular technical aspects?
>
> Also - how long is it going to be before there are parts of the internet
> that you can't get to without speaking IPV6?

There's a number of howto's in google, but I'm not sure how they work
if you are running a laptop and connect through different networks
with/without ipv6 provision.

This is the Gentoo guide:  http://www.gentoo.org/doc/en/ipv6.xml
-- 
Regards,
Mick

Re: [gentoo-user] sys-fs/udev-171-r1 dependency conflict

[Alan McKinnon]

Apparently, though unproven, at 12:38 on Thursday 09 June 2011, Einux did 
opine thusly:

> I just updated the portage tree. But there seems to be a dependency
> conflict when I try to update the system. This is the first time I
> encounter this kind of problem. Could you guys help me out? Thanks in
> advance^_^
> 
> ---
> 
> $ sudo emerge -avuDN world
> 
> These are the packages that would be merged, in order:
> 
> Calculating dependencies... done!
> [ebuild  NS   ~] sys-kernel/gentoo-sources-2.6.39-r1 [2.6.39] USE="-build
> -deblob -symlink" 72 kB
> [ebuild U  ] dev-libs/libdbusmenu-qt-0.8.2 [0.6.2] USE="-debug -doc%
> -test" 36 kB
> [ebuild U ~] sys-fs/udev-171-r1 [151-r4] USE="acl%* gudev%* hwdb%*
> keymap%* rule_generator%* -action_modeswitch% -debug% -edd% -floppy%
> (-introspection) (-selinux) -test (-devfs-compat%)
> (-extras%*) (-old-hd-rules%)" 595 kB
> [ebuild U  ] net-ftp/lftp-4.2.3 [4.1.3] USE="nls ssl -gnutls -socks5"
> 1,296 kB
> [ebuild U ~] media-sound/pulseaudio-0.9.22-r2 [0.9.22] USE="X alsa
> asyncns bluetooth caps dbus glib ipv6 tcpd udev -avahi -doc -gnome -jack
> -libsamplerate -lirc (-oss) -realtime
> (-system-wide) -test" 0 kB
> [ebuild U ~] sys-fs/udisks-1.0.2-r4 [1.0.2-r1] USE="nls
> -bash-completion -debug -doc -remote-access" 0 kB
> 
> Total: 6 packages (5 upgrades, 1 in new slot), Size of downloads: 1,997 kB
> 
> !!! Multiple package instances within a single package slot have been
> pulled !!! into the dependency graph, resulting in a slot conflict:
> 
> sys-fs/udev:0
> 
>   (sys-fs/udev-171-r1::gentoo, ebuild scheduled for merge) pulled in by
> 
> >=sys-fs/udev-171[hwdb] required by
> 
> (media-sound/pulseaudio-0.9.22-r2::gentoo, ebuild scheduled for merge)
> 
>   (sys-fs/udev-151-r4::gentoo, installed) pulled in by
> 
> >=sys-fs/udev-147[extras] required by (sys-fs/udisks-1.0.2-r4::gentoo,
> 
> ebuild scheduled for merge)
> (and 3 more with the same problem)

There's no way to fix this right now.

You need udev version 171 or greater with the hwdb USE flag set as required by 
pulseaudio (which is fine, your system will do this OK)

You also need udev version 147 or greater with the extras USE flag set as 
required by udisks. udev-171 does not meet these requirements as it does not 
have the extras USE flag (check the output of -p above, it is not listed).

So portage will look for earlier and earlier versions of udev to find one that 
does support extras, but the first one it finds conflict with what pulseaudio 
needs.

You have some options:

- it's a bug, it needs to be reported. But you will not need to do this as 
every unstable user has it also and there will be 100s of bugs already (udev 
is needed by almost everyone)

- mask/downgrade pulseaudio to a version that can use an earlier udev

- wait a day, resync, try again.

The third option is usually the one that works best.


> 
> 
> It may be possible to solve this problem by using package.mask to
> prevent one of those packages from being selected. However, it is also
> possible that conflicting dependencies exist such that they are
> impossible to satisfy simultaneously.  If such a conflict exists in
> the dependencies of two different packages, then those packages can
> not be installed simultaneously. You may want to try a larger value of
> the --backtrack option, such as --backtrack=30, in order to see if
> that will solve this conflict automatically.
> 
> For more information, see MASKED PACKAGES section in the emerge man
> page or refer to the Gentoo Handbook.

-- 
alan dot mckinnon at gmail dot com

Re: [gentoo-user] Kernel Modules

[Ignas Anikevicius]

On 09/06/11 12:06, Alan McKinnon wrote:
> why you making so much work for yourself?
>
> set the /usr/src/linux symlink to each set of installed sources in turn,
>  run emerge @module-rebuild
>  or run module-rebuild rebuild
>
> you could even script it
>
> cd /usr/src
> for I in linux-*
> do
>   ln -sfn $I linux
>   module-rebuild rebuils
> done
>
> Fixing my bash syntax errors is left as an exercise for the interested reader
Thanks very much!

I am a former Arch user, so I was used for a lot of inconvenience while
doing such things, but gentoo seems to make my life easier and easier.

Ignas

Re: [gentoo-user] Kernel Modules

[Volker Armin Hemmann]

On Thursday 09 June 2011 11:52:42 Ignas Anikevicius wrote:
> Hello list,
> 
> I was wondering if it is possible to have a tool with which it would be
> possible to have external modules installed for _all_ kernel versions in
> my computer. Now I am using 2.6.38 kernel, but would like to try 2.6.39
> and the thing is that I would like to have tp_smapi and phc-intel
> modules in both kernels. Is it possible to have it without any serious
> hacking?
> 
> I have only 3 ideas how I could achieve that:
> * Making a custom ebuild, which would build the modules, but install
> itself as a different package depending on the kernel version (eg
> tp_smapi-2.6.39-gentoo)?
> * Making a custom ebuild, which would build the modules for all kernel
> versions in one go... (is this possible?)
> * patching the gentoo-sources each time.
> 
> Is any of these solutions sensible?
> 
> Cheers,
> Ignas

why not emerging them several times with linux pointing to the different 
kernels?

I that does not work:
ebuild  unpack
ebuild  compile
ebuild  install
cp .ko from image directory to modules directory
depmod -ae


-- 
#163933

Re: [gentoo-user] IPv6 not ready here; Hmmm

[Tanstaafl]

On 2011-06-08 9:25 PM, Paul Hartman wrote:
> After that, machines on my local network (including wifi) can get both
> IPv4 and IPv6 addresses from the router and can talk to the outside
> world on either network.

I'm getting a headache...

Is there a decent guide that explains IPV6 for noobs who don't speak IP?
Meaning, in plain english, how to set it up and make it work, without
having to understand all of the granular technical aspects?

Also - how long is it going to be before there are parts of the internet
that you can't get to without speaking IPV6?

Re: [gentoo-user] Re: sshd no longer starting when it should.

[Tanstaafl]

On 2011-06-08 5:09 PM, Alan McKinnon wrote:
> Yes, that stuff can get confusing and it's easy to get it mixed up. 
> Te way it's done is the only really sane way - consider how it would
> play out if the setting was a value or a list of possibilities - you
> couldn't put a commented example in there that is the opposite of the
> default

Ummm... the sane way would simply be to say what was the default in the
comments.

Re: [gentoo-user] Kernel Modules

[Alan McKinnon]

Apparently, though unproven, at 12:52 on Thursday 09 June 2011, Ignas 
Anikevicius did opine thusly:

> Hello list,
> 
> I was wondering if it is possible to have a tool with which it would be
> possible to have external modules installed for _all_ kernel versions in
> my computer. Now I am using 2.6.38 kernel, but would like to try 2.6.39
> and the thing is that I would like to have tp_smapi and phc-intel
> modules in both kernels. Is it possible to have it without any serious
> hacking?
> 
> I have only 3 ideas how I could achieve that:
> * Making a custom ebuild, which would build the modules, but install
> itself as a different package depending on the kernel version (eg
> tp_smapi-2.6.39-gentoo)?
> * Making a custom ebuild, which would build the modules for all kernel
> versions in one go... (is this possible?)
> * patching the gentoo-sources each time.

why you making so much work for yourself?

set the /usr/src/linux symlink to each set of installed sources in turn,
 run emerge @module-rebuild
 or run module-rebuild rebuild

you could even script it

cd /usr/src
for I in linux-*
do
  ln -sfn $I linux
  module-rebuild rebuils
done

Fixing my bash syntax errors is left as an exercise for the interested reader


-- 
alan dot mckinnon at gmail dot com

Re: [gentoo-user] Kernel Modules

[YoYo Siska]

On Thu, Jun 09, 2011 at 11:52:42AM +0100, Ignas Anikevicius wrote:
> Hello list,
> 
> I was wondering if it is possible to have a tool with which it would be
> possible to have external modules installed for _all_ kernel versions in
> my computer. Now I am using 2.6.38 kernel, but would like to try 2.6.39
> and the thing is that I would like to have tp_smapi and phc-intel
> modules in both kernels. Is it possible to have it without any serious
> hacking?
> 
> I have only 3 ideas how I could achieve that:
> * Making a custom ebuild, which would build the modules, but install
> itself as a different package depending on the kernel version (eg
> tp_smapi-2.6.39-gentoo)?
> * Making a custom ebuild, which would build the modules for all kernel
> versions in one go... (is this possible?)
> * patching the gentoo-sources each time.


kernel modules are CONFIG_PROTECTED, so they are not automatically
removed when you uninstall / remerge the package (you have to remove
them manually), so you just have to remerge the package after you change
the /usr/src/linux symlink
there is also the module-rebuild utility, that automatically remerges
packages that installed a kernel module

i.e.

ln -sfn linux-VERSION1 /usr/src/linux
module-rebuild -X rebuild
ln -sfn linux-VERSION2 /usr/src/linux
module-rebuild -X rebuild
...

you might have to do 
module-rebuild populate
before the first time...


yoyo

Re: [gentoo-user] Recommendations for N WLAN adapter, PCI or PCIe

[JDM]

Tp-Link TL-WN951N
Works well for me on both linux and windows. Linux drivers (atheros ath9k) are 
available in kernel sources. Found this very reliable and quite cheap on amazon
JDM

-Original Message-
From: Adam Carter 
Date: Thu, 9 Jun 2011 17:07:20 
To: 
Reply-to: gentoo-user@lists.gentoo.org
Subject: [gentoo-user] Recommendations for N WLAN adapter, PCI or PCIe

It will be used to make an AP out of a Gentoo box. Let me know what you
like.

[gentoo-user] Kernel Modules

[Ignas Anikevicius]

Hello list,

I was wondering if it is possible to have a tool with which it would be
possible to have external modules installed for _all_ kernel versions in
my computer. Now I am using 2.6.38 kernel, but would like to try 2.6.39
and the thing is that I would like to have tp_smapi and phc-intel
modules in both kernels. Is it possible to have it without any serious
hacking?

I have only 3 ideas how I could achieve that:
* Making a custom ebuild, which would build the modules, but install
itself as a different package depending on the kernel version (eg
tp_smapi-2.6.39-gentoo)?
* Making a custom ebuild, which would build the modules for all kernel
versions in one go... (is this possible?)
* patching the gentoo-sources each time.

Is any of these solutions sensible?

Cheers,
Ignas

[gentoo-user] sys-fs/udev-171-r1 dependency conflict

[Einux]

I just updated the portage tree. But there seems to be a dependency conflict
when I try to update the system. This is the first time I encounter this
kind of problem. Could you guys help me out? Thanks in advance^_^

---

$ sudo emerge -avuDN world

These are the packages that would be merged, in order:

Calculating dependencies... done!
[ebuild  NS   ~] sys-kernel/gentoo-sources-2.6.39-r1 [2.6.39] USE="-build
-deblob -symlink" 72 kB
[ebuild U  ] dev-libs/libdbusmenu-qt-0.8.2 [0.6.2] USE="-debug -doc%
-test" 36 kB
[ebuild U ~] sys-fs/udev-171-r1 [151-r4] USE="acl%* gudev%* hwdb%*
keymap%* rule_generator%* -action_modeswitch% -debug% -edd% -floppy%
(-introspection) (-selinux) -test (-devfs-compat%)
(-extras%*) (-old-hd-rules%)" 595 kB
[ebuild U  ] net-ftp/lftp-4.2.3 [4.1.3] USE="nls ssl -gnutls -socks5"
1,296 kB
[ebuild U ~] media-sound/pulseaudio-0.9.22-r2 [0.9.22] USE="X alsa
asyncns bluetooth caps dbus glib ipv6 tcpd udev -avahi -doc -gnome -jack
-libsamplerate -lirc (-oss) -realtime
(-system-wide) -test" 0 kB
[ebuild U ~] sys-fs/udisks-1.0.2-r4 [1.0.2-r1] USE="nls -bash-completion
-debug -doc -remote-access" 0 kB

Total: 6 packages (5 upgrades, 1 in new slot), Size of downloads: 1,997 kB

!!! Multiple package instances within a single package slot have been pulled
!!! into the dependency graph, resulting in a slot conflict:

sys-fs/udev:0

  (sys-fs/udev-171-r1::gentoo, ebuild scheduled for merge) pulled in by
>=sys-fs/udev-171[hwdb] required by
(media-sound/pulseaudio-0.9.22-r2::gentoo, ebuild scheduled for merge)

  (sys-fs/udev-151-r4::gentoo, installed) pulled in by
>=sys-fs/udev-147[extras] required by (sys-fs/udisks-1.0.2-r4::gentoo,
ebuild scheduled for merge)
(and 3 more with the same problem)


It may be possible to solve this problem by using package.mask to
prevent one of those packages from being selected. However, it is also
possible that conflicting dependencies exist such that they are
impossible to satisfy simultaneously.  If such a conflict exists in
the dependencies of two different packages, then those packages can
not be installed simultaneously. You may want to try a larger value of
the --backtrack option, such as --backtrack=30, in order to see if
that will solve this conflict automatically.

For more information, see MASKED PACKAGES section in the emerge man
page or refer to the Gentoo Handbook.


-- 
Best Regards,
Einux

Re: [gentoo-user] IPv6 not ready here; Hmmm

[Dale]

Alan McKinnon wrote:

Apparently, though unproven, at 08:52 on Thursday 09 June 2011, Dale did opine
thusly:

   

Paul Hartman wrote:
 

On Wed, Jun 8, 2011 at 11:50 PM, Dale   wrote:
   

I'm just hoping html5 will improve some things.  May not but doesn't
hurt to hope.  I just hope it will eliminate some of the things that
are such power or memory hogs now.  It may not at first but eventually
 

Some stuff is already looking good, outside of streaming video, like
http://chrome.angrybirds.com/ which probably would not be possible
without Flash or Java just a couple years ago.
   

So things like youtube won't change any?
 

Why do you think it will change at all?

IPv4 is not going away. There is still lots of space available and it has a
very long life ahead of it still.
   


I was just hoping thing like youtube might change for the better.  What 
little I know about html5, it sounds like it is going to be better.  If 
something as big as youtube changes to better its site then others will 
too.


We sort of switched subject from IPv6 to HTML5.  I'm sure IPv4 will be 
around a long time tho.


Now back to finding the leak in my car A/C.  It won't hold a vacuum at 
the moment.  Weird.


Dale

:-)  :-)

Re: [gentoo-user] IPv6 not ready here; Hmmm

[Alan McKinnon]

Apparently, though unproven, at 08:52 on Thursday 09 June 2011, Dale did opine 
thusly:

> Paul Hartman wrote:
> > On Wed, Jun 8, 2011 at 11:50 PM, Dale  wrote:
> >> I'm just hoping html5 will improve some things.  May not but doesn't
> >> hurt to hope.  I just hope it will eliminate some of the things that
> >> are such power or memory hogs now.  It may not at first but eventually
> > 
> > Some stuff is already looking good, outside of streaming video, like
> > http://chrome.angrybirds.com/ which probably would not be possible
> > without Flash or Java just a couple years ago.
> 
> So things like youtube won't change any?

Why do you think it will change at all?

IPv4 is not going away. There is still lots of space available and it has a 
very long life ahead of it still.


-- 
alan dot mckinnon at gmail dot com

[gentoo-user] Recommendations for N WLAN adapter, PCI or PCIe

[Adam Carter]

It will be used to make an AP out of a Gentoo box. Let me know what you
like.

Re: [gentoo-user] IPv6 not ready here; Hmmm

[Volker Armin Hemmann]

On Wednesday 08 June 2011 23:56:32 Dale wrote:
> Volker Armin Hemmann wrote:
> > Read the FAQ and Info posted.
> > 
> > > From the website:
> > If this test fails: it means that the DNS resolver you are using,
> > requires IPv4 to reach the DNS authoritative servers of your favoriate
> > web sites. In the near future, every web site of consequence will
> > remain accessible in this form, so there is no immediate danger.

There is no 'falling behind'. If there is a need to migrate them to ipv6 they 
will do that. As end user you have nothing to worry about.

-- 
#163933