Re: [arch-general] Arch GNU/Linux install for beginners and new users

[Kyle Terrien via arch-general]

On Thu, 22 Sep 2016 14:22:24 -0400
Simon Gomizelj via arch-general  wrote:
> But we do have to make it clear that we won't **officially** support
> anything installed outside of the officially sanctions instructions. I
> don't see that as unreasonable.

I have to agree.  As someone who works software QA professionally, I
feel the pain.

On the QA team, we can only test a certain number of configurations
because of time.  Anything else?  Well, you are basically blazing your
own trail.  The same rule applies to support.

Yes, it is possible to debug issues on "unsupported" configurations,
but the process is almost always a time sink (i.e. low reward to cost
ratio).  This is why there is a little hostility toward the third party
tools.  They increase the amount of support work drastically.

vodik, keep up the good work!

--Kyle

-- 
The computer can't tell you the emotional story.  It can give you the
exact mathematical design, but what's missing is the eyebrows.
- Frank Zappa


pgpVlbVTz79lv.pgp
Description: OpenPGP digital signature

Re: [arch-general] What happened to the Beginner's Guide?

[Kyle Terrien via arch-general]

On Fri, 23 Sep 2016 13:00:53 +1200
Jason Ryan via arch-general  wrote:
> Then we are in agreement; the goal is to provide people with what they
> need and to encourage them to explore in more depth, or for edge
> cases, the official documentation.

Then I think I misinterpreted the snark as seriousness in your post
yesterday.  I apologize.

--Kyle


pgpgOMTmqjkcY.pgp
Description: OpenPGP digital signature

Re: [arch-general] What happened to the Beginner's Guide?

[Kyle Terrien via arch-general]

On Thu, 22 Sep 2016 18:16:27 +1200
Jason Ryan via arch-general  wrote:
> Arch has from the start been clear about its goals and intentions.
> I'll quote from a section of the wiki that people seem much less
> familiar with:
> 
> ”Whereas many GNU/Linux distributions attempt to be more
> user-friendly, Arch Linux has always been, and shall always remain
> user-centric. The distribution is intended to fill the needs of those
> contributing to it, rather than trying to appeal to as many users as
> possible. It is targeted at the proficient GNU/Linux user, or anyone
> with a do-it-yourself attitude who is willing to read the
> documentation, and solve their own problems.”
> https://wiki.archlinux.org/index.php/Arch_Linux#User_centrality
> 
> /J
> 

Yes, several people have quoted "Arch philosophy" for me.  I am
familiar with it.  I read it 3 years ago when I first installed Arch.  I
am also familiar with the fact that there are slightly different
interpretations of it.

For me, the issue is about framing the goal.  If the goal is to
"alienate new users that are unwilling to read man pages", then people
will surely find a way to alienate them.  However, if your goal is to
give people a minimal amount of information and direct them to the man
pages where they can read more, then the wiki will be more inviting.

The difference is all in the framing.  I personally prefer to see the
latter because it is more optimistic and invites those interested to
help with the distro.

--Kyle

-- 
The computer can't tell you the emotional story.  It can give you the
exact mathematical design, but what's missing is the eyebrows.
- Frank Zappa


pgpXatkR7htCq.pgp
Description: OpenPGP digital signature

Re: [arch-general] What happened to the Beginner's Guide?

[Kyle Terrien via arch-general]

On Tue, 20 Sep 2016 04:55:19 +
Sajjad Heydari via arch-general  wrote:
> It has been merged with the installation guide.
> 
> On Tue, Sep 20, 2016, 9:23 AM David C. Rankin <
> drankina...@suddenlinkmail.com> wrote:  
> 
> > Archdevs,
> >
> >   I went to review the Beginner's Guide and it was no longer listed
> > on the main
> > page. I then specifically searched for it, found it in the topics,
> > and then was
> > redirected to the bare-bones "Install" page. What happened to the
> > Beginner's Guide?
> >
> > --
> > David C. Rankin, J.D.,P.E.
> >  

On a related note, there is a stale "Beginner's Guide" link that
redirects to the new installation guide.  This link is now redundant
because there is an "Installation Guide" link right above it.

--Kyle

-- 
The computer can't tell you the emotional story.  It can give you the
exact mathematical design, but what's missing is the eyebrows.
- Frank Zappa


pgpNQuyyqeTrH.pgp
Description: OpenPGP digital signature

Re: [arch-general] What happened to the Beginner's Guide?

[Kyle Terrien via arch-general]

On Thu, 22 Sep 2016 12:58:23 +1200
Jason Ryan via arch-general <arch-general@archlinux.org> wrote:
> On 21/09/16 at 08:36pm, Dave via arch-general wrote:
> >see below
> >
> >On Wed, Sep 21, 2016 at 8:12 PM, Francis Gerund via arch-general <
> >arch-general@archlinux.org> wrote:
> >  
> >> I believe this change will only serve to alienate potential new
> >> users,  
> >
> >I strongly agree.  
> 
> This will only alienate new users that are unwilling to read man
> pages and the community provided documentation, ie., people for whom
> Arch is the wrong choice anyway…
> 
> /J
> 

So, is making people hunt for information a ritualistic hazing?

As much as I love the Arch Wiki (and I really do because it is a great
source of information), I don't understand why basic things like this
change very often.  Removing the Beginner's Guide from the home page?
This is just another Big Change made with the best of intentions.

Moving stuff around like this is like that supermarket that keeps
reorganizing its shelves.  Everything is there, but it is in a
different place, and you need to waste time looking for what you want.

Anecdote: when I installed Arch a few years ago, I followed both the
Beginner's Guide and the Installation Guide simultaneously.  The duality
was a little confusing, and I agree that if these guides can be merged
in an elegant way, they should.

But I disagree with the "hazing" attitude.  The whole point of a wiki
is to make information accessible.  Yes, while installing Arch for the
first time you will definitely hunt for a lot of information.  It is a
lot of hard work.  However, purposefully making that information harder
to find is wrong; wikis are supposed make that job easier.

I also think that expecting a new Arch user to know *exactly* what
software configuration he wants is unreasonable.  I personally know
many experienced Linux users (whom I consider far more experienced than
myself) who can't tell you which desktop environment they prefer.
However, they can blow your mind with their C and shell knowledge.

If/when these advanced users decide to learn more about distro innards
and try Arch, we should welcome them, not haze them.

--Kyle Terrien

-- 
The computer can't tell you the emotional story.  It can give you the
exact mathematical design, but what's missing is the eyebrows.
- Frank Zappa


pgpvNZ_ljeiKh.pgp
Description: OpenPGP digital signature

Re: [arch-general] efivars mounted read-write, but "operation not permitted, "

[Kyle Terrien via arch-general]

On Wed, 3 Aug 2016 13:03:41 -0700
Zachary Kline  wrote:
> Hi All,
> 
> This is admittedly more about Linux in general than Arch
> specifically, but I’m wondering if anybody has insight into why I
> can’t delete EFI variables, when efivarfs is mounted read-write. For
> anybody interested, I am wanting to remove the default boot entry
> created by systemd-boot, but receive an “Operation not permitted,”
> message when trying to do so, even as root.
> 
> Any insight would be appreciated.
> Thanks much,
> Zack.

I remember there were some kernel patches that went in a few months ago.

Brief summary of what happened:

* Someone ran 'rm -rf /' on his system to wipe it.  It was hard bricked,
  not even able to POST.  [0]  (You need an Arch BBS account to view
  that thread.)
* All Hell broke loose.  Tech blogs had a field day.  [1] A bug was
  filed in systemd [2].  For some reason beyond me, systemd requires
  that efivars be mounted read-write.  (Probably bad design)
* A kernel patch was submitted to try to protect efivars somewhat [3].
  I think you are seeing the direct consequence of this patch.

--Kyle

[0]: https://bbs.archlinux.org/viewtopic.php?id=207549
[1]: https://www.phoronix.com/scan.php?page=news_item=UEFI-rm-root-directory
[2]: https://github.com/systemd/systemd/issues/2402
[3]: https://gist.github.com/mjg59/8d9d494da56fbe6d8992

-- 
The computer can't tell you the emotional story.  It can give you the
exact mathematical design, but what's missing is the eyebrows.
- Frank Zappa


pgpag9pLw3Vxc.pgp
Description: OpenPGP digital signature

Re: [arch-general] [arch-dev-public] signoffs are dead

[Kyle Terrien via arch-general]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

On Wed, 29 Jun 2016 21:51:17 +0200
Florian Pritz via arch-dev-public  wrote:
> As pointed out on arch-general (not by me) this makes the whole
> signoff process useless. Maybe we should look into finding some
> people that want to help test stuff and give them permissions to sign
> off on packages?
> 
> Florian

Yes, this sounds like a very good idea.

Until now, I have had no reason to run a container/VM with the testing
repos enabled.  However, if signoffs are opened up to a wider community,
I think I might just try it out.

It is great to catch problems before they are pushed to production.

- --Kyle
-BEGIN PGP SIGNATURE-
Version: GnuPG v2
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=r3gL
-END PGP SIGNATURE-

Re: [arch-general] [arch-dev-public] signoffs are dead

[Kyle Terrien via arch-general]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

I am replying to arch-general because arch-dev-public is closed to most
users.

On Tue, 28 Jun 2016 08:09:41 -1000
Gaetan Bisson <bis...@archlinux.org> wrote:
> Dear all,
> 
> For a while now packages in [testing] have gotten little to no
> signoffs and I've been moving mine to [core] after a week without
> feedback. I suspect many of you have been doing this too. Here's the
> signoff reports over the last ten days:
> 
> - June 19: 0 signoffs
> - June 20: 6 from me, 4 from anthraxx
> - June 21: 0
> - June 22: 5 from me
> - June 23: 2 from demize
> - June 24: 1 from me
> - June 25: 0
> - June 26: 1 from me
> - June 27: 3 from me, 1 from eworm
> - June 28: 3 from heftig, 2 from arojas
> 
> So I've decided to shorten the wait in [testing] to 48 hours. Many
> updates to [core] packages include security fixes and they have better
> move sooner rather than later. We used to be able to gather enough
> signoffs to move these within a day or two, and that's what I intend
> to do with or without signoffs.
> 
> Any comment, and especially any other idea to fix this situation, is
> welcome.
> 
> Cheers.

First, I am an Arch user (for 3 years now) not an Arch dev, and I
realize I have no right to tell anyone how to run the distribution.
What follows is just my personal recommendation based on working
software QA professionally.

With that said, I think eliminating signoffs is a bad idea.

Signoffs ensure some form of quality control.  A signoff is an explicit
approval from someone that the package is satisfactory to his/her
standards.  A potential signee has a completely different perspective
than the packager and a different way of verifying that the packager's
package is correct.  This sort of approval process catches errors that
would otherwise escape the packager's notice.  Simply waiting a period
of time without hearing complaints is not equivalent to explicit
approval from others.

I have personally experienced several breakages in the past several
months--more than usual.  A few were big enough that simply running 'foo
- --version' should have revealed a problem (i.e.  linking problems).  A
signoff process would have very likely caught these problems.

IMHO, the correct thing to do is remind other developers of the signoff
policy.  (And the above post to arch-dev-general certainly does just
that.)  Encouraging another set of eyes to look at someone's work and
say, "This looks good to me," is a very good thing and does wonders in
terms of quality control.

If getting security fixes pushed out is a concern, then getting the
security related fixes signed off should be prioritized.  (Maybe by
putting in a flag that automatically triggers a mail to arch-dev-public)

Respectfully yours,
- --Kyle Terrien
-BEGIN PGP SIGNATURE-
Version: GnuPG v2
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=E42/
-END PGP SIGNATURE-

Re: [arch-general] Where did the touch pad tapping go?

[Kyle Terrien via arch-general]

Javier Vasquez via arch-general wrote:
> I think Xorg now uses libinput as default as opposed to prior
> versions.  I have been using libinput any ways since it was made
> available, so the transition didn't affect me, and besides I'm OK with
> the tapping offered by libinput.  Actually I removed totally the
> xf86-input-synaptics package from my Arch boxes already.  You can take
> a look at the install message:
> 
> https://git.archlinux.org/svntogit/packages.git/tree/trunk/xf86-input-synaptics.install?h=packages/xf86-input-synaptics

I saw that message while upgrading a couple of days ago.

xf86-input-synaptics driver is on maintenance mode and
xf86-input-libinput driver must be prefered over.

But I didn't understand what "must be prefered over" means (grammar
error).

You suggested that libinput is the default now.  That would make more
sense.  So is this a warning of some sort, or is it a notification to
try to encourage people to switch to libinput?

Does switching break any older desktop environments?  (e.g. MATE, Window
Maker)

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] gnome-keyring madness

[Kyle Terrien via arch-general]

Jérôme M. Berger wrote:
> On 06/14/2016 02:21 PM, Kyle Terrien via arch-general wrote:
>> Thanks, but I tried that weeks ago.  It is not the autostart feature of
>> the desktop environment that is starting gnome-keyring.  From what I can
>> figure out, it is lightdm that is starting gnome-keyring.
>>
>   Actually it's pam. Look in these files:
> 
> /etc/pam.d/lightdm
> /etc/pam.d/lightdm-autologin
> /etc/pam.d/xscreensaver
> /etc/pam.d/sddm-autologin
> 
>   Jerome

And *that* is the missing piece of information.  Thank you!

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] gnome-keyring madness

[Kyle Terrien via arch-general]

Ralf Mardorf wrote:
> Take a look at
> https://specifications.freedesktop.org/autostart-spec/autostart-spec-latest.html
>  ,
> it should be your desktop environment that is auto starting.

Thanks, but I tried that weeks ago.  It is not the autostart feature of
the desktop environment that is starting gnome-keyring.  From what I can
figure out, it is lightdm that is starting gnome-keyring.

What's doubly annoying is that gnome-keyring stays alive after I log
out, meaning that the systemd login session is never terminated.  This
can cause some bizarre issues when logging in again.

In fact, my day-to-day "desktop environment" (Window Maker) doesn't even
support xdg autostart, so I know autostart is not the culprit.  (I am
starting programs in the xprofile.)

Given the bizarre integration bugs that gnome-keyring brings, I would
rather just remove it entirely for now:

1. Seahorse can run without gnome-keyring running in the background.
Why is there a hard dependency?
2. virt-manager runs fine when gpg-agent handles SSH keys.  Why do I
need to have its dependency x11-ssh-askpass installed?

--Kyle



signature.asc
Description: OpenPGP digital signature

[arch-general] gnome-keyring madness

[Kyle Terrien via arch-general]

Seahorse 3.20.0-2 has a new hard dependency on gnome-keyring. Should
gnome-keyring instead be an optional dependency? Seahorse can cope
without gnome-keyring (although with warnings about not being able to
talk to gnome-keyring).

And for some reason, lightdm likes to launch gnome-keyring
automatically, and this breaks my setup because I use gpg-agent with
pinentry-gtk-2 to handle SSH keys. However gnome-keyring naively assumes
that it may (incorrectly) handle SSH keys. Is there any way to tell
lightdm not to launch gnome-keyring?

(And yes, the obvious answer is to uninstall gnome-keyring. However,
virt-manager has a hard dependency on seahorse.)

Update: it turns out that virt-manager has a dependency on
x11-ssh-askpass, which seahorse provides. My workaround for now is to
install x11-ssh-askpass and uninstall seahorse and gnome-keyring (even
though I will never use x11-ssh-askpass).

There is something wrong with this dependency chain. Will someone
smarter than me please look into it?

* Should gpg-agent and/or pinentry provide x11-ssh-askpass?
* Should gnome-keyring be an optional dep for seahorse, considering that
  seahorse copes without it although in a limited form?
* Should virt-manager have a hard dependency on x11-ssh-askpass?

--Kyle




signature.asc
Description: OpenPGP digital signature

Re: [arch-general] GTK2 GUIs became a PITA after upgrade

[Kyle Terrien]

Jérôme M. Berger wrote:
> On 04/12/2016 07:08 PM, Mauro Santos wrote:
>> I was using clearlooks-phenix but with gtk 3.20 it is quite broken and
>> it hasn't been updated in quite some time. I have no idea if the dev
>> will code a new version for gtk 3.20+.
>>
>   When launching a GTK3 application with the clearlooks phenix theme,
> there are several warnings that feature such-and-such has been
> deprecated and replaced by so-and-so. I've tried applying all the
> suggested replacements and that got rid of the warnings, but
> unfortunately it didn't fix the looks :( Downgrading for now and hoping
> someone who knows more about GTK3 theme will update clearlooks phenix or
> make some similar clearlooks-like theme for GTK 3.20+
> 
>   Jerome

I noticed some of the geometry is slightly different, but I just thought
it was gtk3 being libgnome.  I don't test gtk3 much (and I don't know
that much about gtk theming), so I wasn't aware of the deprecated
warnings.

But that's it!  After finals end, I am going to look into converting my
favorite gtk2 theme (Bluecurve) into gtk3 theme.  I am tired of ugly
and/or broken gtk3 themes.

Does anyone off the top of his/her head know of any good tools or guides
for converting gtk2 themes to gtk3 themes?

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] GTK2 GUIs became a PITA after upgrade

[Kyle Terrien]

Ralf Mardorf wrote:
> $ grep gnome-themes-standard /var/log/pacman.log | grep 2016-04
> [2016-04-10 19:57] [ALPM] upgraded gnome-themes-standard (3.18.0-1 -> 3.20-1)
> [2016-04-12 09:33] [PACMAN] Running 'pacman -U 
> gnome-themes-standard-3.18.0-1-x86_64.pkg.tar.xz'
> [2016-04-12 09:33] [ALPM] downgraded gnome-themes-standard (3.20-1 -> 
> 3.18.0-1)
> 
> Now the old look is back :), this not necessarily means that I also got
> rid of GTK hiccups.

:-)

>> If Adwaita really is the culprit, file a bug with GNOME.  That's their
>> domain.
> 
> OTOH, GTK2 apps are affected and the GNOME maintainers likely want to
> make GTK2 apps look consistent with GTK3 GNOME apps, so that they
> fit better to the bad design of the GNOME desktop environment. No, I
> dislike insults from upstream, when reporting an issue. I have given up
> to report to this kind of upstream, a long time ago.

Ugh!  Yeah, there probably isn't much a case to be made on their
Bugzilla.  I imagine that they will justify the change as a feature to
"make the geometry of elements match in both GTK2 and GTK3".

The best thing to do at this point is design a GTK3 theme that sucks
less (and maintain it when GNOME makes the arbitrary changes to the GTK3
theming API again).  Clearlooks-Phenix is good.  Perhaps a
Bluecurve-Phenix would be fun too.

(Someone really should adjust the attitude of the GNOME devs.  I am sick
of the crap coming out of GNOME 3 breaking stuff and making other things
ugly.)

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] GTK2 GUIs became a PITA after upgrade

[Kyle Terrien]

Ralf Mardorf wrote:
> after the last upgrade GTK2 apps running in an openbox session need more
> space, they became darker and in some windows even the fonts became
> less good readable. The reason that I dropped desktop environments was
> to get rid of chaotic designs and to keep a clear design and now even
> GTK2 apps in an openbox session became a parody.
...
> [rocketmouse@archlinux ~]$ grep -v '#' .gtkrc-2.0
> 
> gtk-theme-name = "Adwaita"
> style "user-font"
> {
>   font_name="FreeSans 12"
> }
> widget_class "*" style "user-font"
> 
> Regards,
> Ralf

My first thought was "Adwaita", and looking at the list of packages,
gnome-themes-standard is upgraded.  I suspect that Adwaita is the
culprit.

Have you tried other GTK themes?  Clearlooks-Phenix, or maybe a
gtk2-only theme such as Bluecurve?

On my side, I have been using Clearlooks-Phenix and Bluecurve over the
past few days.  (I recently switched back to Bluecurve.)  I haven't
noticed any change in GTK2 apps over the past couple days.  So, I'm apt
to blame Adwaita.

On a side note, I remember a recent GNOME propaganda video said that
GNOME had finally standardized their theming API for gtk3?  I didn't
believe them then, and I certainly don't believe them now after reading
this thread.

So, I think this might be a case of GNOME deciding that making the UI
elements even bigger is the hip thing to do.  (It's just frustrating
that GNOME is so eager to break things nowadays.)

If Adwaita really is the culprit, file a bug with GNOME.  That's their
domain.

I hope this information may be of help,
--Kyle

https://aur.archlinux.org/packages/clearlooks-phenix-gtk-theme
https://aur.archlinux.org/packages/gtk-bluecurve-engine



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Mysterious userapp .desktop files

[Kyle Terrien]

Patrick Burroughs (Celti) wrote:
> On Wed, 16 Mar 2016 22:08:46 -0700
> Kyle Terrien <kyleterr...@gmail.com> wrote:
> 
>> So, somehow my default web browser got changed from SeaMonkey back to
>> Firefox.  (Bleh!)
>>
>> [...]
>>  
>> In ~/.local/share/applications/ (where user-customized .desktop files
>> go), there were several files in the format
>> userapp--.desktop.
>>
>> userapp-Firefox-FIK3WX.desktop
>> userapp-Pale Moon-1HNY8X.desktop
>> userapp-SeaMonkey-YV09AY.desktop
>> userapp-Thunderbird-PIZCZW.desktop
> 
> If I'm not mistaken, those files and their associated mimeinfo entries
> are created every time you set Firefox/SeaMonkey/PaleMoon as you
> default browser, as the mechanism of such.
> 
> ~Celti

That's correct.  I can confirm that if I go to SeaMonkey's Edit >
Preferences and click "Set default browser", then a new userapp file is
created.  This userapp .desktop file is then inserted into
~/.config/mimeapps.list.

Good catch, and thank you!

$ ls .local/share/applications/
mimeinfo.cache  userapp-SeaMonkey-45D4DY.desktop
$ cat .local/share/applications/userapp-SeaMonkey-45D4DY.desktop 
[Desktop Entry]
Encoding=UTF-8
Version=1.0
Type=Application
NoDisplay=true
Exec=/usr/lib/seamonkey-2.40/seamonkey %u
Name=SeaMonkey
Comment=Custom definition for SeaMonkey
$ cat .local/share/applications/mimeinfo.cache 
[MIME Cache]
$ 

--Kyle



signature.asc
Description: OpenPGP digital signature

[arch-general] Mysterious userapp .desktop files

[Kyle Terrien]

So, somehow my default web browser got changed from SeaMonkey back to
Firefox.  (Bleh!)

I ended up cleaning up after xdg-open in the process of correcting my
default web browser.

In ~/.local/share/applications/ (where user-customized .desktop files
go), there were several files in the format
userapp--.desktop.

userapp-Firefox-FIK3WX.desktop
userapp-Pale Moon-1HNY8X.desktop
userapp-SeaMonkey-YV09AY.desktop
userapp-Thunderbird-PIZCZW.desktop

(Wow!  I have gone through a lot of web browsers lately.)

I remember creating a few .desktop files in this directory, but I don't
remember creating any userapp-* files.  The contents of
userapp-Firefox-FIK3WX.desktop were:

[Desktop Entry]
Encoding=UTF-8
Version=1.0
Type=Application
NoDisplay=true
Exec=/usr/lib/firefox/firefox %u
Name=Firefox
Comment=Custom definition for Firefox

This is weird.  There is only an English comment (the default has many
languages), and this one has set NoDisplay.

I deleted all my custom .desktop files.  I have no use for them anymore.

And back to the problem of default web browsers.  These userapp .desktop
files got into mimeapps.list (now located in ~/.config/).  I suppose the
short hash for SeaMonkey changed, causing mimeapps.list to fallback to
the next web browser listed in mimeapps.list.  I had to weed all the
userapp garbage out and replace them with the system-wide .desktop file
names.

What daemon creates these mysterious userapp .desktop files?  I don't
want this daemon creating this junk on my system.

--Kyle




signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Why does virtualbox-host-dkms have a hard dependency on linux-headers?

[Kyle Terrien]

Kyle Terrien wrote:
> I upgraded my desktop about 10 minutes ago, and it looks like
> virtualbox-host-dkms gained a new dependency: linux-headers.

As of virtualbox-host-dkms-5.0.16-3, linux-headers is an optional
dependency again:

$ pacman -Qi virtualbox-host-dkms 
Name: virtualbox-host-dkms
Version : 5.0.16-3
Description : VirtualBox Host kernel modules sources
Architecture: x86_64
URL : http://virtualbox.org
Licenses: GPL  custom
Groups  : None
Provides: None
Depends On  : dkms  gcc  make
Optional Deps   : linux-headers: build modules against Arch kernel
  linux-lts-headers: build modules against LTS kernel
  [installed]
  linux-zen-headers: build modules against ZEN kernel
  linux-grsec-headers: build modules against GRSEC kernel
Required By : virtualbox
Optional For: None
Conflicts With  : virtualbox-source  virtualbox-host-source
Replaces: virtualbox-source  virtualbox-host-source
  virtualbox-host-modules  virtualbox-host-modules-lts
Installed Size  : 10.23 MiB
Packager: Sébastien Luttringer <se...@seblu.net>
Build Date  : Sun 13 Mar 2016 01:24:45 PM PDT
Install Date: Mon 14 Mar 2016 08:57:17 AM PDT
Install Reason  : Explicitly installed
Install Script  : Yes
Validated By: Signature

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Why does virtualbox-host-dkms have a hard dependency on linux-headers?

[Kyle Terrien]

Eric Engestrom wrote:
> Regarding your log, it looks to me like you have both the default and
> LTS kernels installed, but only have the source for the LTS, so you're
> only building the module for LTS.

Nope.  I only have the LTS kernel installed.  linux-headers was pulled in
this morning as a dependency.

--Kyle

$ pacman -Q linux{,-lts}{,-headers}
error: package 'linux' was not found
linux-headers 4.4.5-1
linux-lts 4.1.19-1
linux-lts-headers 4.1.19-1



signature.asc
Description: OpenPGP digital signature

[arch-general] Why does virtualbox-host-dkms have a hard dependency on linux-headers?

[Kyle Terrien]

Hello,

I upgraded my desktop about 10 minutes ago, and it looks like
virtualbox-host-dkms gained a new dependency: linux-headers.

On this desktop, I have only the linux-lts kernel and its accompanying
linux-lts-headers installed, so why is linux-headers necessary?  In the
pacman -Syu output, it looks like dkms gives up when trying to build
modules for the linux (4.4) kernel.  However, it still builds the
modules for linux-lts.

Why does virtualbox-host-dkms have a hard dependency on linux-headers?

--Kyle

[2016-03-13 10:50] [ALPM] running '70-dkms-install.hook'...
[2016-03-13 10:50] [ALPM-SCRIPTLET] ==> Kernel 4.4.5-1-ARCH modules are 
missing. Nothing will be done for this kernel!
[2016-03-13 10:50] [ALPM-SCRIPTLET] ==> You have to install the matching kernel 
package to use dkms
[2016-03-13 10:50] [ALPM-SCRIPTLET] 
[2016-03-13 10:50] [ALPM-SCRIPTLET] Creating symlink 
/var/lib/dkms/vboxhost/5.0.16_OSE/source ->
[2016-03-13 10:50] [ALPM-SCRIPTLET]  
/usr/src/vboxhost-5.0.16_OSE
[2016-03-13 10:50] [ALPM-SCRIPTLET] 
[2016-03-13 10:50] [ALPM-SCRIPTLET] DKMS: add completed.
[2016-03-13 10:50] [ALPM-SCRIPTLET] 
[2016-03-13 10:50] [ALPM-SCRIPTLET] Kernel preparation unnecessary for this 
kernel.  Skipping...
[2016-03-13 10:50] [ALPM-SCRIPTLET] 
[2016-03-13 10:50] [ALPM-SCRIPTLET] Building module:
[2016-03-13 10:50] [ALPM-SCRIPTLET] cleaning build area...
[2016-03-13 10:51] [ALPM-SCRIPTLET] make KERNELRELEASE=4.1.19-1-lts -C 
/usr/lib/modules/4.1.19-1-lts/build 
M=/var/lib/dkms/vboxhost/5.0.16_OSE/build...
[2016-03-13 10:51] [ALPM-SCRIPTLET] cleaning build area...
[2016-03-13 10:51] [ALPM-SCRIPTLET] Kernel cleanup unnecessary for this kernel. 
 Skipping...
[2016-03-13 10:51] [ALPM-SCRIPTLET] 
[2016-03-13 10:51] [ALPM-SCRIPTLET] DKMS: build completed.
[2016-03-13 10:51] [ALPM-SCRIPTLET] 
[2016-03-13 10:51] [ALPM-SCRIPTLET] vboxdrv.ko:
[2016-03-13 10:51] [ALPM-SCRIPTLET] Running module version sanity check.
[2016-03-13 10:51] [ALPM-SCRIPTLET] 
[2016-03-13 10:51] [ALPM-SCRIPTLET] Good news! Module version 5.0.16_OSE for 
vboxdrv.ko
[2016-03-13 10:51] [ALPM-SCRIPTLET] exactly matches what is already found in 
kernel 4.1.19-1-lts.
[2016-03-13 10:51] [ALPM-SCRIPTLET] DKMS will not replace this module.
[2016-03-13 10:51] [ALPM-SCRIPTLET] You may override by specifying --force.
[2016-03-13 10:51] [ALPM-SCRIPTLET] 
[2016-03-13 10:51] [ALPM-SCRIPTLET] vboxnetflt.ko:
[2016-03-13 10:51] [ALPM-SCRIPTLET] Running module version sanity check.
[2016-03-13 10:51] [ALPM-SCRIPTLET] 
[2016-03-13 10:51] [ALPM-SCRIPTLET] Good news! Module version 5.0.16_OSE for 
vboxnetflt.ko
[2016-03-13 10:51] [ALPM-SCRIPTLET] exactly matches what is already found in 
kernel 4.1.19-1-lts.
[2016-03-13 10:51] [ALPM-SCRIPTLET] DKMS will not replace this module.
[2016-03-13 10:51] [ALPM-SCRIPTLET] You may override by specifying --force.
[2016-03-13 10:51] [ALPM-SCRIPTLET] 
[2016-03-13 10:51] [ALPM-SCRIPTLET] vboxnetadp.ko:
[2016-03-13 10:51] [ALPM-SCRIPTLET] Running module version sanity check.
[2016-03-13 10:51] [ALPM-SCRIPTLET] 
[2016-03-13 10:51] [ALPM-SCRIPTLET] Good news! Module version 5.0.16_OSE for 
vboxnetadp.ko
[2016-03-13 10:51] [ALPM-SCRIPTLET] exactly matches what is already found in 
kernel 4.1.19-1-lts.
[2016-03-13 10:51] [ALPM-SCRIPTLET] DKMS will not replace this module.
[2016-03-13 10:51] [ALPM-SCRIPTLET] You may override by specifying --force.
[2016-03-13 10:51] [ALPM-SCRIPTLET] 
[2016-03-13 10:51] [ALPM-SCRIPTLET] vboxpci.ko:
[2016-03-13 10:51] [ALPM-SCRIPTLET] Running module version sanity check.
[2016-03-13 10:51] [ALPM-SCRIPTLET]  - Original module
[2016-03-13 10:51] [ALPM-SCRIPTLET]- This kernel never originally had a 
module by this name
[2016-03-13 10:51] [ALPM-SCRIPTLET]  - Installation
[2016-03-13 10:51] [ALPM-SCRIPTLET]- Installing to 
/usr/lib/modules/4.1.19-1-lts/kernel/misc/
[2016-03-13 10:51] [ALPM-SCRIPTLET] 
[2016-03-13 10:51] [ALPM-SCRIPTLET] depmod...
[2016-03-13 10:51] [ALPM-SCRIPTLET] 
[2016-03-13 10:51] [ALPM-SCRIPTLET] DKMS: install completed.
[2016-03-13 10:51] [ALPM-SCRIPTLET] ==> Kernel 4.4.5-1-ARCH modules are 
missing. Nothing will be done for this kernel!
[2016-03-13 10:51] [ALPM-SCRIPTLET] ==> You have to install the matching kernel 
package to use dkms



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] thunderbird proxy env. vars for socks

[Kyle Terrien]

Javier Vasquez wrote:
> However, inside the company, I need to setup manually under TB:
> 
> SOCKS Host:  <...>  Port:  <...>
> 
> As SOCKS v5.  Where the socks host is the same as the other proxy
> hosts, but its socks port must be different than the rest.
> 
> I don't use any DE, but I have some automation through bash+xinitrc
> which depends on how to set the env. vars, so that I only need to set
> on TB/FF "use system proxy settings".

You could try tsocks, available from the extra repository.

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Link to the original efivars BBS thread

[Kyle Terrien]

Eli Schwartz wrote:
> On 02/04/2016 12:45 AM, Kyle Terrien wrote:
>> Was the thread moved?  Does anyone have the new address of the thread?
>>
>> --Kyle
>>
>> [0]: https://bbs.archlinux.org/viewtopic.php?id=207549
>>
> 
> It's still there, but it was moved to TGN so you need to be logged into
> the forums to see it.

Aha!  And I had SeaMonkey set to clear cookies on exit.

Thank you!  I didn't realize that Topics Going Nowhere is only available
to logged in users.

--Kyle



signature.asc
Description: OpenPGP digital signature

[arch-general] Link to the original efivars BBS thread

[Kyle Terrien]

Hello all,

A few days ago, there was a big stink about bricking motherboards
accidentally with an overzealous rm that descended into
/sys/firmware/efi/efivars.

It all started with one thread on the BBS titled "No POST after rm -rf
/".  After reading about the issue, I went into archivist mode and made
a small writeup for myself detailing how to remount efivars as ro.

I saved the link to the original thread [0], but now when I navigate to
it, I get the error message, "Bad request. The link you followed is
incorrect or outdated."

Was the thread moved?  Does anyone have the new address of the thread?

--Kyle

[0]: https://bbs.archlinux.org/viewtopic.php?id=207549



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Instructions to mount efivars as readonly should be linked to in Beginner's Guide

[Kyle Terrien]

Tomasz Kramkowski wrote:
> Since when does "do something dumb" and "potentially hard brick your
> motherboard" become synonymous when speaking in terms of computers?
> 
> There's doing something dumb (by accident or otherwise) and then there's
> bricking your motherboard, people make accidents all the time but since
> modern day computers are quite nice and rugged, the only losses are data
> losses.

You would think that a modern day machine is nice and rugged, but with
EFI/UEFI, it isn't.  There are way too many moving gears involved.

The preboot environment has one primary task: find a bootable medium and
boot it.  Ideally, you should be able to configure it to tell it which
medium to boot from.  In the absence of a bootable medium, it should
throw an error.  Simple!

This is how things worked before EFI.  Sure, getting an OS to load was a
magic trick in the early days ("pulling oneself up by one's
bootstraps"), but today it is a finely honed procedure.  There is
nothing broken with this procedure.  (After all, it boots!)

Enter EFI and UEFI.  From my (somewhat limited) experience with EFI, it
seems like whoever designed it attempted to solve some fringe problem
while creating 5 more problems in its place.  Why do OSes need to modify
the boot order entries?  Why do some motherboards refuse to fallback to
legacy BIOS?  To make things worse, many hardware implementations are
buggy and cannot be fixed (because there are already thousands/millions
of units in production).

So, if you want a modern day computer to be rugged:

* Use legacy BIOS.  There is nothing wrong with it.
* Mount efivars (and related stuff) as ro by default.  I read the
  systemd bug [0], but I still don't understand why so many tools need
  to write to it.  How often do you need to change motherboard
  parameters after you get an OS set up?  At that point, POST should be
  "find a device and boot it".

> I might shed a few tears over the loss of some not-backed up data, but I
> would be quite a bit more pissed off if I lost a valuable and expensive
> piece of hardware (granted, it would have to have a misconfigured and
> shitty EFI, but since when is "being misconfigured and shitty" a rare
> occurance?).

I wish I could answer the philosophical question of whether rm should be
able to brick hardware.  I suggest someone mail Brian Kernighan, Robert
Pike, or Ken Thompson.  I would be really curious to hear what they
think about this efivars thing.

--Kyle

[0]: https://github.com/systemd/systemd/issues/2402



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Instructions to mount efivars as readonly should be linked to in Beginner's Guide

[Kyle Terrien]

Maarten de Vries wrote:
> On 1 February 2016 at 23:29, Leonid Isaev 
> wrote:
> 
>>
>> Also, how can you brick a machine by simply zeroing the harddrive?
>>
>>
> You can't (well, someone can probably think of a contrived situation where
> you could, there's always someone, but generally speaking). The problem is
> with removing certain UEFI variables in buggy UEFI implementations (which
> are all too common). But in this case (with buggy UEFI implementation) a
> simple rm -rf of the wrong directory can brick your motherboard.
> 
> -- Maarten

Interesting sidenote: In Android, all the system-level stuff is
segregated to /system, which is mounted as ro by default.  This is just
another layer of security.

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Instructions to mount efivars as readonly should be linked to in Beginner's Guide

[Kyle Terrien]

Sebastiaan Lokhorst wrote:
> 2016-02-01 23:29 GMT+01:00 Leonid Isaev <leonid.is...@jila.colorado.edu>:
> 
>> On Mon, Feb 01, 2016 at 01:40:54PM -0800, Kyle Terrien wrote:
>>> Tomasz Kramkowski wrote:
>>> * Use legacy BIOS.  There is nothing wrong with it.
>> Exactly, I really don't understand this interest to UEFI (and don't mention
>> secureboot).
>>
> 
> Many new (OEM) machines do not support legacy BIOS.

And this is what is so obnoxious/stupid about the whole situation.

In their infinite wisdom, hardware manufacturers are writing broken
implementations of an over-engineered system with no way to fallback to
the tried/true system that has been around for decades.

And when problems are discovered, it is too late.  The systems are in
production, and it is much easier to blame the consumer for uninstalling
Windows than to admit there are millions of buggy boards in the wild.

I really hope some sanity will return to the world of pre-boot.

--Kyle



signature.asc
Description: OpenPGP digital signature

[arch-general] Firefox without signature checking

[Kyle Terrien]

Hello,

Are there plans to package a version of Firefox 44 that lets you disable
extension signature checking?

Background: Firefox is shipping with signature checking for addons.
Right now (in Firefox 43), there is an option to disable it if you need
to use an unsigned addon.  However, that option is being removed in
Firefox 44. [0]

Signature checking itself is actually a very good concept.  However, the
way Mozilla has implemented it in Firefox (imho) is not.  Every
extension must be signed by Mozilla, and only Mozilla, creating a walled
garden.  From my understanding, there will be no way to override this
extension check unless you recompile Firefox and build an unbranded
version.

Personally, I think this mechanism goes against the Arch Linux tenet of
user centrality [1].  As a user, you should ultimately be allowed to
decide what you want to install on your system.  Also, how are you
supposed to build your own extensions or test someone else's extension
on Firefox stable? Fedora recently discussed the possibility of
packaging an "unofficial" Firefox [2]. (Take a look at their bugtracker
for some good points.)

Is an abrowser package with the ability to disable extension signing
warranted?

I am posting to this mailing list because I have not seen much
discussion about Firefox extension signing in the Arch Linux world.
Developers, what are your thoughts?  Is it worth it packaging an
"unofficial" version of Firefox?

--Kyle Terrien

[0] https://wiki.mozilla.org/Addons/Extension_Signing
[1] https://wiki.archlinux.org/index.php/Arch_Linux#User_centrality
[2] https://fedorahosted.org/fesco/ticket/1518



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Firefox without signature checking

[Kyle Terrien]

On 01/02/2016 01:23 PM, Doug Newgard wrote:
> On Sat, 2 Jan 2016 12:17:52 -0800
> Kyle Terrien <kyleterr...@gmail.com> wrote:
> 
>> Hello,
>>
>> Are there plans to package a version of Firefox 44 that lets you disable
>> extension signature checking?
> ...
>> --Kyle Terrien
> 
> https://bugs.archlinux.org/task/45900

Thank you!  I was tempted to reopen it, but it looks like the general
consensus is that an AUR package will be submitted.

It looks like sticking to upstream trumps user-centric in this case.
(Although I guess building from AUR is a fair substitute for
user-centric.)

It's a real bummer that a bunch of users (myself included) will be
forced to compile Firefox on each release.  I really hope we can
eventually get an abrowser or firefox-nobranding (or maybe even
palemoon) into the repos.

How can we nominate a package into community or extra?

(And if you are fed-up with Mozilla's walled-garden policy like I am, I
suggest trying out Pale Moon [0].)

--Kyle Terrien

[0] https://www.palemoon.org/



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Firefox without signature checking

[Kyle Terrien]

On 01/02/2016 02:42 PM, ProgAndy wrote:
> Am 02.01.2016 um 22:52 schrieb Kyle Terrien:
>> It looks like that is only intended for release-status extensions. If I want 
>> to QA test a developer's beta build, this tells me that the developer would 
>> have to submit each build he wishes me to test to AMO. Why can't I just 
>> package the extension myself and load it? That is absurd. --Kyle Terrien 
> 
> The official stance is that developers and testers should use the "Developer 
> Version"[1] of Firefox. That release does not require extension signatures. 
> [2]
> Jan Steffens (heftig) provides us with a compiled Arch Linux package in his 
> repository. [3][4]
> 
> You can always provide or request an unofficial repository with binary 
> packages of other firefox based projects.
> 
> 1: https://www.mozilla.org/firefox/developer/
> 2: https://wiki.mozilla.org/Add-ons/Extension_Signing
> 3: https://bbs.archlinux.org/viewtopic.php?id=117157
> 4: http://pkgbuild.com/~heftig/repo/

Cool!  It's nice to know someone has my back when it comes to the
Developer Addition.

But what about regression testing?  (Testing against older "supported"
versions, or in this case the "stable" version of Firefox)  So far, the
only way it looks like is to either download the unofficial build from
Mozilla or build Firefox yourself.

--Kyle Terrien



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Firefox without signature checking

[Kyle Terrien]

On 01/02/2016 02:50 PM, Doug Newgard wrote:
> On Sat, 2 Jan 2016 15:35:01 -0700
> Leonid Isaev <leonid.is...@jila.colorado.edu> wrote:
> 
>> On Sat, Jan 02, 2016 at 02:06:05PM -0800, Kyle Terrien wrote:
>>> Thank you!  I was tempted to reopen it, but it looks like the general
>>> consensus is that an AUR package will be submitted.  
>>
>> You can only request to reopen...
> 
> And that request would be denied unless you can bring new info to the table. 
> So
> far, I haven't seen any.

The new info I have is that Mozilla is creating a walled garden.  There
is no way to override it besides rebuilding Firefox.

The Fedora bugreport I pointed at earlier [0] compares this to package
signing in RPM (or in our case pacman).  The difference with package
signing is that a user can add his own key and use that key to sign
packages.  In Firefox 44, you can do no such thing.  You are at
Mozilla's mercy.

And Mozilla's add-on checker isn't perfect either [1].

These two reasons are why I believe that Mozilla's signature policy is a
step in the wrong direction.

On the other hand, I fully understand why we would want to follow
upstream--less work for packaging and testing, as well as official
sanctioning via branding.

But I'm not affected much anyway because I'm on Pale Moon (using their
official builds).

--Kyle Terrien

[0] https://fedorahosted.org/fesco/ticket/1518
[1] 
http://danstillman.com/2015/11/23/firefox-extension-scanning-is-security-theater



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Firefox without signature checking

[Kyle Terrien]

On 01/02/2016 12:47 PM, Damjan Georgievski wrote:
>>  Aside:
>>   I don't use firefox - but curious - how would one test developer versions
>> of extensions then? Or is this no longer possible in firefox?
> 
> There will be support for that of course
> https://developer.mozilla.org/en-US/Add-ons/Distribution

It looks like that is only intended for release-status extensions.  If I
want to QA test a developer's beta build, this tells me that the
developer would have to submit each build he wishes me to test to AMO.
Why can't I just package the extension myself and load it?  That is
absurd.

--Kyle Terrien



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Firefox without signature checking

[Kyle Terrien]

On 01/02/2016 05:24 PM, Magnus Therning wrote:
> The larger, and very philosophical question is "How user un-friendly can
> upstream make it before Arch decides to *not* package as upstream
> intends?" (Answering this requires keeping in mind that Arch users are
> unlikely to fall squarely into the target group of upstream.)

This is a very good question to ask imho.  In a perfect world, we would
just fork and be done with it.  However, it is not a perfect world.
Forking requires time and effort, and it generally kills the software.
(Here I am with several installs that have both Firefox and Pale Moon
for compatibility reasons.)

Most Arch users definitely do not fall into Mozilla's target group (an
imaginary "average Joe" as I like to call it).  We made this decision
when we decided to install Arch as opposed to Ubuntu, Mint, OpenSUSE, or
any other distro that does a decent job of configuring itself
automatically.

In 2015, we have seen a surprisingly large push for user un-friendliness
in general.  Firefox proposing add-on signing and removing features such
as "complete" themes were just a couple examples.  The GNOME folks are
talking about breaking GTK themes again.  I have lost track of how many
"statically linked" QT libraries (e.g. bundled with Dropbox) that are
completely broken.  And in the world of other OSes, the Windows 10
release basically added spyware at the OS level to millions of users'
PCs.

So, the real question is where do you draw the line when something is
un-friendly?  And what do you do when the line is crossed?

--Kyle Terrien



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] [OT?] Which is most future-proof desktop environment?

[Kyle Terrien]

On 12/30/2015 04:19 AM, Maarten de Vries wrote:
> On 30 December 2015 at 10:19, Magnus Therning <mag...@therning.org> wrote:
> 
>> More generally, what do you mean by "work-flow", and how have DEs like
>> KDE and Gnome broken "the work-flow" in the past?
> 
> 
> ​For a good rant on some things that are wrong with Gnome 3 ​ (and GTK 3):
> https://igurublog.wordpress.com/2012/11/05/gnome-et-al-rotting-in-threes/

Yep, that was a good read, and it made my skin crawl.  The part that
disturbed me the most was the bug about removing the systray icon in
Transmission.  Imagine supporting all those incompatible configurations.
It's absurd!

It's amazing how the pattern of removing features and changing things
arbitrarily for the "greater good" is spreading around nowadays.  It has
invaded Firefox recently.  Mozilla is talking about deprecating XUL this
year.

(And if you are unhappy about the current state of Firefox, check out
Pale Moon [0].  We don't tolerate the nonsense that is boiling over in
GNOME 3 and Mozilla.)

> On 30 December 2015 at 12:49, Leonid Isaev <leonid.is...@jila.colorado.edu>
> wrote:
> 
>>
>> On the contrary, things like {open,flux}box and tiling WMs (i3, jwm) still
>> use
>> a design from '90s. And from olden days of Win98 we remember what it leads
>> to.
>>
>>
> In my experience, it leads to very productive and happy users that don't
> have to change the way they use their computer every time some dev or
> designer decides to "streamline the user experience" by removing useful
> features or adding extra empty space everywhere.

Amen!  I'm using Window Maker here.  It's the same GUI it has always
been for 20-something years.  Except, it is still actively maintained.
There is even an option to ignore window caption hints in GTK
applications because of the GNOME "Client Side Decoration" nonsense.

Long live no-nonsense 90s GUIs!

--Kyle Terrien

[0] https://www.palemoon.org/



signature.asc
Description: OpenPGP digital signature

[arch-general] virt-manager/spicy: no audio unless I start pulseaudio

[Kyle Terrien]

Summary: Why is there no audio in the spice viewer if PulseAudio is not
running?

I have been battling muted audio in virt-manager for some time.  I
finally figured out that it is really a spice viewer problem (spicy)
rather than a QEMU problem.

The problem is that spicy does not output any audio unless I have
PulseAudio setup and running.  If I attempt to run spicy from a terminal
and connect to a VM (running Fedora 23), I get warnings like the
following when I am not running PulseAudio:

GSpice-WARNING **: PulseAudio context failed Connection refused
GSpice-WARNING **: pa_context_connect() failed: Connection refused

And there is no sound.  If I launch PulseAudio and try again, there is
sound.

It looks like at one point the '--with-audio=gstreamer' configure
parameter was added to spice-gtk3 [0].  However, it was reverted a few
weeks later [1].  So, I have a couple questions:

- Why was the reversion necessary?  There is a forum post on that bug
where someone has the complete opposite problem I am having.
- Is there some way to make spicy output sound without PulseAudio as a
dependency?

I have filed a bug on spicy's bugtracker [2], and a developer claims
that by switching '--with-audio' to gstreamer, it is possible to make
sound output work on both PulseAudio and plain ALSA.  I have tried
recompiling with '--with-audio=gstreamer', but I was not able to get
sound output in plain ALSA.  Has anyone had success recompiling with
'--with-audio=gstreamer'?

Any insight to the root of the problem would be appreciated.

--Kyle Terrien

[0] https://bugs.archlinux.org/task/41492
[1] https://bugs.archlinux.org/task/41740
[2] https://bugs.freedesktop.org/show_bug.cgi?id=93544



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] KVM troubles

[Kyle Terrien]

On 11/27/2015 11:14 PM, Luna Moonbright wrote:
> As for it just being old Ubuntu - are the newer EOL versions of Ubuntu
> (like 9 or 10) still easy to exploit  (32 bit/no canaries/no NX) that
> are easier to get the display drivers to work for?

I can't remember when Ubuntu started supporting canaries.  (I haven't
done much Ubuntu stuff since Linux Mint 14 (based on 12.10)).

There used to be a project called Damn Vulnerable Linux, but it has
disappeared.  Even their website is gone.

A quick web search revealed some possibilities [0], although I have
never heard of them personally.  Let me know if you find any good
intentionally vulnerable distros.

You could also download old unsupported Ubuntu releases [1].  (You just
need to tweak the repository URLs after install.)

Normally, if I want/need a completely out-of-date vulnerable system to
poke at, I usually use an old distro (whatever is sitting around) and
bite the bullet to figure out what hardware it is looking for.  It's
trial and error.

> Shellshock was awesome, but my favorite exploit is the exploit in
> fingerd used by the morris worm. So simple - yet so effective. I'm
> sure us archers can appreciate that.
> 
> Thanks!

I have heard of it, but I don't know all the details.  I will definitely
look up the fingerd exploit.

--Kyle

[0] http://www.101hacker.com/2013/03/5-vulnerable-distros-for-practicing.html
[1] http://old-releases.ubuntu.com/releases/



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] KVM troubles

[Kyle Terrien]

On 11/22/2015 09:59 PM, Luna Moonbright wrote:
> Does anyone here have experience troubleshooting
> KVM installs in arch?
> 
> I am using virsh and virt-manager, the VM seems 
> to successfully boot, but the virtualization window
> is only showing a black screen with a cursor in
> the top left of the window.
> 
> (on the host) journelctl -xe only shows one related error:
> "libvirtd[298]: this function is not supported by the 
> connection driver: virConnectGetCPUModuleNames" 
> 
> If it helps: the ISO I am trying to boot from
> is the live disk that comes with "Hacking:
> The Art of Exploitation"
> 
> Thanks in advance.

I have the same book ("Hacking: The Art of Exploitation", 2nd Edition).
I just successfully booted and installed the Live CD in KVM.

The CD is a very old (EOL) custom Ubuntu Live CD (Feisty Fawn).  In
virt-manager, I selected the "Ubuntu Feisty Fawn" preset, and that
changed some of the defaults.

- The Video card is set to VMVGA (I think this is the problem.  The
  default is normally QXL, which is a really new super-fast
  paravirtualized interface iirc.)
- The hard disk is an IDE hard disk.
- Fiesty Fawn does not support the EvTouch Tablet.

For some reason, the LiveCD's bootloader took a long time to draw to the
screen.  I'm not sure why, but after I booted, everything seemed fine.
I installed and rebooted.

The repositories have been archived.  You will need to edit
/etc/apt/sources.list and change every instance of
"us.archive.ubuntu.com" with "old-releases.ubuntu.com".

After this, I was able to install openssh-server and vim-gtk on the VM.

I still need to figure out why sshfs won't work (fuse: failed to exec
fusermount: Permission denied).

I think the problem you are having is related to the QXL video driver.

Thanks for giving me an excuse to dust-off the CD.  I really should find
more time to read through that book.  It's a great book.

If you are curious, the XML dump of my VM follows.  (MAC address and
UUID are obfuscated.)

--Kyle


  hacking
  13b0cfef-b878-470f-a690-
  2097152
  2097152
  2
  
hvm

  
  



  
  
Nehalem
  
  



  
  destroy
  restart
  restart
  


  
  
/usr/sbin/qemu-system-x86_64

  
  
  
  


  
  
  
  


  


  
  


  
  


  
  



  


  


  
  
  
  


  


  


  
  




  


  


  
  






  

  





signature.asc
Description: OpenPGP digital signature

Re: [arch-general] KVM troubles

[Kyle Terrien]

On 11/27/2015 04:57 PM, Luna Moonbright wrote:
> Fantastic, I love that book and am wanting to get the rest of the way through 
> it. I didn't realize it was such an issue in KVM. I think it works in virtual 
> box, but it would be nice to get it installed on KVM so me and a few guys can 
> set up a class on exploitation. 
> 
> I'll keep working on it to.

It's an old version of Ubuntu, so you could have a lot of fun poking at
vulnerabilities, especially if you don't do an apt-get update.

Coincidentally, this VM is vulnerable to my personal favorite
exploit--Shellshock.  That's something fun you can cover in your class.

kyle@hacking:~ $ env 'x=() { :;}; echo Vulnerable' bash -c 'echo Test' 
Vulnerable
Test
kyle@hacking:~ $   

And if you set up a web server on the VM, you can demonstrate how to use
Shellshock to dump /etc/passwd by setting a malicious User-Agent. [0]

--Kyle

[0] http://blog.regehr.org/archives/1187



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Thunar sftp connection not working - access denied

[Kyle Terrien]

On 10/29/2015 06:06 PM, Friedrich Strohmaier wrote:
> Hi Kyle, *,
> 
> Am 15.10.2015 um 17:56 schrieb Kyle Terrien:
> 
>> On 10/13/2015 10:55 AM, Friedrich Strohmaier wrote:
>>> Am 12.10.2015 um 17:29 schrieb Kyle Terrien:
>>>> On 10/09/2015 05:54 PM, Friedrich Strohmaier wrote:
>>>>> Am 10.10.2015 um 02:23 schrieb Leonid Isaev:
>>>>>> On Sat, Oct 10, 2015 at 02:05:38AM +0200, Friedrich Strohmaier wrote:
> 
>>>>>>> Hi Folks,
> 
>>>>>>> since some time I cannot get thunar connect my remote servers folders 
>>>>>>> via sftp.
> 
>>> [..]
> 
>>>>>> So, does sftp work from the command line?
> 
>>>>> yes works as expected.
> 
>>>>> btw. filezilla also does..
> 
> [..]
> 
>>> digging a bit deeper..
>>> It apears to be a problem of ssh-key authentification.
> 
> Log output uploaded: 
> https://bits-fritz.de/eigene_webdateien/File/bereitstellung/messages.txt
> 
>>> Any ideas?
>>> Is this a gvfsd bug?
> 
>> Considering that CLI sftp and Filezilla work, this is probably a GVFS
>> related issue.
> 
> New facts but no solution..
> 
> After dbus update I restarted dbus by hand:
> Restarting dbus as root yields:
> =
> [root@myhost ~]# systemctl restart dbus
> PolicyKit daemon disconnected from the bus.
> We are no longer a registered authentication agent.
> =

Dbus is tied into almost everything nowadays.  Restarting it can do all
sorts of interesting things.

> After new Loggin in XFCE - tataaa sftp-connection is established without
> issues.  Cannot shutdown machine out of xfce session and have other quirks but
> this one works.
> 
> After machine restart old behaviour is back.
> 
> This tells me something's wrong with PolicyKit settings?

So, if PolicyKit is not registered as an authentication agent, gvfs SFTP
works.  I'm guessing it is falling back to some other authentication
agent.

> Forgot to mention: ssh-agent running (started by keychain)

Maybe ssh-agent is the fallback.

>> Have you tried running the ssh commands yourself?
> 
>>> Oct 13 19:36:08 my_machine gvfsd[758]: ### SFTP: spawn_ssh: ssh 
>>> -oForwardX11 no -oForwardAgent no -oPermitLocalCommand no 
>>> -oClearAllForwardings yes -oProtocol 2 -oNoHostAuthenticationForLocalhost 
>>> yes -l me -s strict.remote.host sftp
>>> Oct 13 19:36:28 my_machine gvfsd[758]: ### SFTP: spawn_ssh: ssh 
>>> -oForwardX11 no -oForwardAgent no -oPermitLocalCommand no 
>>> -oClearAllForwardings yes -oProtocol 2 -oNoHostAuthenticationForLocalhost 
>>> yes -l me -s permissive.remote.host sftp
> 
> I did (had to replace spaces with "=" between -oXX options and values).
> No result but - mmmhh - a "waiting" prompt.
> It appeared like opening a tunnel.
> 
> Kyle, many thanks for keeping up! :o))
> 

I would have to look up those ssh options.  They could very well create
a tunnel.

In your log file:

> Oct 30 00:22:22 my_machine gvfsd[1047]: ** (process:1315): WARNING **: Failed 
> to setup SSH evironment: The name org.gnome.keyring was not provided by any 
> .service files (g-dbus-error-quark, 2)

Is gnome-keyring installed?  It looks like gvfs (being a GNOME
application) is trying to start GNOME keyring.

I switched pretty much everything to gpg-agent.  In .xinitrc I have
this:

# Start the GnuPG agent and enable OpenSSH agent emulation
gnupginf="${HOME}/.gpg-agent-info"
if pgrep -x -u "${USER}" gpg-agent >/dev/null 2>&1; then
eval `cat $gnupginf`
eval `cut -d= -f1 $gnupginf | xargs echo export`
export SSH_AUTH_SOCK="${HOME}/.gnupg/S.gpg-agent.ssh"
else
eval `gpg-agent -s --enable-ssh-support --daemon --write-env-file 
"$gnupginf"`
fi

It uses some old options that I should probably clean up, but it still
starts and works.  But you will need to kill it with 'pkill gpg-agent'
before logging out.

I only use gnome-keyring-daemon on my laptop for keeping track of Wifi
passwords in NetworkManager.

Feel free to look at my configuration at

<https://github.com/KlipperKyle/dotfiles/blob/master/xautorun>

(.xautorun is sourced by .xinitrc.  Then .xinitrc calls my window
manager.)

If I get the chance, I might try to remove gnome-keyring-daemon and see
what happens if I use gvfs.

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Thunar sftp connection not working - access denied

[Kyle Terrien]

On 10/13/2015 10:55 AM, Friedrich Strohmaier wrote:
> Hi Kyle, *,
> 
> Am 12.10.2015 um 17:29 schrieb Kyle Terrien:
>> On 10/09/2015 05:54 PM, Friedrich Strohmaier wrote:
>>> Am 10.10.2015 um 02:23 schrieb Leonid Isaev:
>>>> On Sat, Oct 10, 2015 at 02:05:38AM +0200, Friedrich Strohmaier wrote:
> 
>>>>> Hi Folks,
> 
>>>>> since some time I cannot get thunar connect my remote servers folders via 
>>>>> sftp.
> 
> [..]
> 
>>>> So, does sftp work from the command line?
> 
>>> yes works as expected.
> 
>>> btw. filezilla also does..
> 
>> As a workaround, you can use sshfs.  sshfs will mount a remote
>> filesystem over ssh.
> 
>> I use sshfs quite a bit with Thunar.  Thunar lists the mount as an
>> external drive.
> 
> I know about this but well, it used to work and has stopped doing so, now.
> 
> digging a bit deeper..
> It apears to be a problem of ssh-key authentification.
> 
> While trying to connect a host with exclusvive ssh-key access, journalctl 
> tells:
> 
> Oct 13 19:36:08 my_machine gvfsd[758]: ### SFTP: spawn_ssh: ssh -oForwardX11 
> no -oForwardAgent no -oPermitLocalCommand no -oClearAllForwardings yes 
> -oProtocol 2 -oNoHostAuthenticationForLocalhost yes -l me -s 
> strict.remote.host sftp
> Oct 13 19:36:08 my_machine gvfsd[758]: ### SFTP: handle_login #1, 
> initial_connection = 1 - user: me, host: strict.remote.host, port: -1
> Oct 13 19:36:09 my_machine gvfsd[758]: ### SFTP: handle_login #1 - 
> password_save: 0
> Oct 13 19:36:09 my_machine gvfsd[758]: ### SFTP: handle_login #1 - ret_val: 1
> Oct 13 19:36:09 my_machine gvfsd[758]: ### SFTP: stderr: Permission denied 
> (publickey).
> Oct 13 19:36:09 my_machine gvfsd[758]: ** (gvfsd:758): WARNING **: 
> dbus_mount_reply: Error from org.gtk.vfs.Mountable.mount(): Zugriff verweigert
> ===
> 
> This happens connecting a host allowing password authentication:
> ===
> Oct 13 19:36:28 my_machine gvfsd[758]: ### SFTP: spawn_ssh: ssh -oForwardX11 
> no -oForwardAgent no -oPermitLocalCommand no -oClearAllForwardings yes 
> -oProtocol 2 -oNoHostAuthenticationForLocalhost yes -l me -s 
> permissive.remote.host sftp
> Oct 13 19:36:28 my_machine gvfsd[758]: ### SFTP: handle_login #1, 
> initial_connection = 1 - user: me, host: permissive.remote.host, port: -1
> Oct 13 19:36:34 my_machine gvfsd[758]: ### SFTP: handle_login #1 - prompt: 
> "m...@permissive.remote.host's password: "
> Oct 13 19:36:34 my_machine gvfsd[758]: ### SFTP: handle_login #1 - asking for 
> password...
> Oct 13 19:36:51 my_machine gvfsd[758]: ### SFTP: handle_login #1 - prompt: ""
> Oct 13 19:36:53 my_machine gvfsd[758]: ### SFTP: handle_login #1 - 
> password_save: 0
> Oct 13 19:36:53 my_machine gvfsd[758]: ### SFTP: handle_login #1 - ret_val: 1
> Oct 13 19:36:53 my_machine gvfsd[758]: ### SFTP: spawn_ssh: ssh -oForwardX11 
> no -oForwardAgent no -oPermitLocalCommand no -oClearAllForwardings yes 
> -oProtocol 2 -oNoHostAuthenticationForLocalhost yes -l me -s 
> permissive.remote.host sftp
> Oct 13 19:36:53 my_machine gvfsd[758]: ### SFTP: handle_login #2, 
> initial_connection = 0 - user: me, host: permissive.remote.host, port: -1
> Oct 13 19:36:59 my_machine gvfsd[758]: ### SFTP: handle_login #2 - prompt: 
> "m...@permissive.remote.host's password: "
> Oct 13 19:36:59 my_machine gvfsd[758]: ### SFTP: handle_login #2 - using 
> credentials from previous login attempt...
> Oct 13 19:36:59 my_machine gvfsd[758]: ### SFTP: handle_login #2 - prompt: ""
> Oct 13 19:36:59 my_machine gvfsd[758]: ### SFTP: handle_login #2 - ret_val: 1
> 
> 
> Any ideas?
> Is this a gvfsd bug?

Considering that CLI sftp and Filezilla work, this is probably a GVFS
related issue.

The part of the log output that sticks out like a sore thumb are the
lines that look like this:

> Oct 13 19:36:08 my_machine gvfsd[758]: ### SFTP: handle_login #1, 
> initial_connection = 1 - user: me, host: strict.remote.host, port: -1

"port: -1" just doesn't seem right.  At best this means "use the default
port".  But you would think that if it uses the default port, then the
log would say "port: 22".

At worst, this means literally "use port -1", which means that -1 could
overflow in unsigned integer arithmetic so it is actually a really high
port.

Have you tried running the ssh commands yourself?

> Oct 13 19:36:08 my_machine gvfsd[758]: ### SFTP: spawn_ssh: ssh -oForwardX11 
> no -oForwardAgent no -oPermitLocalCommand no -oClearAllForwardings yes 
> -oProtocol 2 -oNoHostAuthenticationForLocalhost yes -l me -s 
> strict.remote.host sftp
> Oct 13 19:36:28 my_machine gvfsd[758]: ### SFTP: spawn_ssh: ssh -oForwardX11 
> no -oForwardAgent no -oPermitLocalCommand no -oClearAllForwardings yes 
> -oProtocol 2 -oNoHostAuthenticationForLocalhost yes -l me -s 
> permissive.remote.host sftp

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Thunar sftp connection not working - access denied

[Kyle Terrien]

On 10/09/2015 05:54 PM, Friedrich Strohmaier wrote:
> Hi Leonid, *,
> 
> Am 10.10.2015 um 02:23 schrieb Leonid Isaev:
>> On Sat, Oct 10, 2015 at 02:05:38AM +0200, Friedrich Strohmaier wrote:
>>> Hi Folks,
> 
>>> since some time I cannot get thunar connect my remote servers folders via 
>>> sftp.
> 
>>> Error dialog:
>>> Failed to open "/my/remote/path on my.server"
>>> access denied
> 
>>> ssh access works as expected.
> 
>>> I feel like it stopped working after I got this arch announce:
>>> https://www.archlinux.org/news/d-bus-now-launches-user-buses/
> 
>> So, does sftp work from the command line?
> 
> yes works as expected.
> 
> btw. filezilla also does..

As a workaround, you can use sshfs.  sshfs will mount a remote
filesystem over ssh.

I use sshfs quite a bit with Thunar.  Thunar lists the mount as an
external drive.

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Cannot tether from Android using USB or bluetooth

[Kyle Terrien]

On 08/24/2015 02:13 PM, Alfredo Palhares wrote:
 I enabled the usb tethering on the phone, but ip link didn't show any new
 interface:

 snip

 The important stuff seems to be here:
 rndis_host 1-1:1.0 enp0s20u1: renamed from usb0
 usb 1-1: USB disconnect, device number 41
 rndis_host 1-1:1.0 enp0s20u1: unregister 'rndis_host' usb-:00:14.0-1, 
 RNDIS device
 usb 1-1: new low-speed USB device number 43 using xhci_hcd
 
 The interface comes up but its immediately down, for some weird reason. I 
 can't
 reproduce the problem every time.

On my phone (Nexus 4 running CM 12.1 nightlies which are based on
Android 5.1), activating USB tethering creates another Ethernet device.

The USB disconnect message is disturbing and generally indicates
hardware problems.  Furthermore, it sounds like this issue is
intermittent.  This smells of hardware problems.  I recommend cleaning
the MicroUSB port if it looks dirty.  Unfortunately, there is generally
no easy way to do this without disassembling the phone.

Personal anecdote:  My phone's MicroUSB port (same Nexus 4) once got
dirty enough to the point where the data pins no longer worked.
However, the charge pins *did* work.  I ended up taking the phone apart
to clean the MicroUSB port.  I'm still using that phone to this day.
:-)

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Subpixel antialiasing

[Kyle Terrien]

On 05/24/2015 01:52 AM, Vladimir Nikšić wrote:
 Hi guys!
 
 I've done a pacman -Syu this morning. The upgrade went fine, all the key
 components are working properly, except a slight issue I'm having with
 chromium. Chromium no longer obeys the subpixel antialiasing, but instead
 has a basic AA mode, without the subpixel hinting. This of course results
 in ugly letters across the board. What's even more weird, other
 applications seem to have it normally, like firefox, xfce4-terminal, etc...
 
 I'm using xfce4. The settings  appearance had subpixel disabled, but once
 enabled chromium still didn't obey the subpixel hinting. I don't understand
 what has changed exactly with chromium, and how do I set it to have
 subpixel once again.
 
 I went through here
 https://wiki.archlinux.org/index.php/Font_configuration#Subpixel_rendering
 but I can't really find anything pertinent to my case. Any ideas anyone? :)

In addition to Xresources, I also use a fonts.conf.  My current setup
disables antialiasing and enables full hinting (which is probably not
what you want, so you will have to modify accordingly).

Install fonts.conf to ~/.config/fontconfig/fonts.conf

https://github.com/KlipperKyle/dotfiles/blob/master/fontconfig/fonts.conf

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Connection through USB to TTL Serial Cable

[Kyle Terrien]

On 05/20/2015 11:55 AM, Csányi Pál wrote:
 Hello,
 
 I have a
 FTDI TTL-232RG-VREG3V3-WE
 USB to TTL Serial Cable.
 
 On my Arch linux I want to setup a connection through this cable.
 
 How can I achieve this goal?
 
 The output of the 'lsusb' command is:
 Bus 002 Device 005: ID 0403:6001 Future Technology Devices
 International, Ltd FT232 USB-Serial (UART) IC
 
 I'm trying to follow steps described here:
 http://linux-sunxi.org/Cubieboard/TTL
 
 but I have more difficulties.
 
 1. I can't find to install the 'cu' utility neither on Arch linux
 repository nor in AUR.
 2. There is no '/dev/ttyUSB0' on my system here.
 
 So eg.
 screen /dev/ttyUSB0 115200
 
 doesn't work here:
 Cannot exec '/dev/ttyUSB0': No such file or directory
 
 I can't find any advices on Arch Wiki.
 
 Any advices will be appreciated!

I have used a similar adapter on my system.

I used picocom (I think) to connect to the serial port.  I also had to
add myself to the uucp group because /dev/ttyUSB0 was owned by group
uucp.

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Libre Office Freezing system?

[Kyle Terrien]

On 01/26/2015 04:58 PM, Michael Rasmussen wrote:
 It does for me. 
 Have a Writer document with a embedded table, click and drag to highlight a 
 segment. 
 Boom, my user interface is locked up.  No CTRALTF# to get a terminal, 
 no toggle of the 
 caps lock indicactor LED, no CTRLALTBackspace to kill X.  No nothing. 
 I can ssh in from another system and top shows nothing unusual going on.
 
 I've disabled hardware acceleration, that did not affect the problem.
 
 Does this sound familiar?
 Suggestions for resolving?
 

I'm not sure if I can help much either, but someone could probably help
if you provide some more information about your setup.

1. Which Libreoffice are you using?  libreoffice-still or the latest
libreoffice?

2. What graphics drivers are you using?  What kernel version?

3. (Someone else asked this)  What desktop environment are you using?

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] p11-kit

[Kyle Terrien]

On 12/16/2014 04:13 AM, Ralf Mardorf wrote:
 I never add one and they don't belong to a package:
 # pacman -Qo [...].pem error: No package owns [...].pem
 
 Can I delete them?

I don't think you're supposed to.  They are created when the certificate
packages are installed.

From my understanding, in this case locally installed means anything
you installed manually.

https://bbs.archlinux.org/viewtopic.php?pid=1483835#p1483835

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] no cups web/server with new version?

[Kyle Terrien]

On 11/04/2014 06:06 PM, Javier Vasquez wrote:
 I stopped and removed the old service, reloaded daemons, started and
 enabled new services (org.cups.cupsd.service and
 cups-browsed.service).
 
 However it seems I don't have the server working, since I can't get to
 localhost:631, and:
 
 % dmesg | 'grep' cups
 [   10.006918] systemd[1]: Cannot add dependency job for unit
 cups.socket, ignoring: Unit cups.socket failed to load: No such file
 or directory.
 [   10.006965] systemd[1]: Cannot add dependency job for unit
 cups.path, ignoring: Unit cups.path failed to load: No such file or
 directory.
 [ 1652.605967] systemd[1]: Cannot add dependency job for unit
 cups.socket, ignoring: Unit cups.socket failed to load: No such file
 or directory.

cups.socket and cups.path are the old names (used before the upgrade).

Disable everything cups related by removing the symlinks.  A find
command should help you weed out stuff from before the upgrade:

find /etc/systemd/system/ -iname *cups*

Then, try reenabling and starting cups again.  (Three symlinks should be
created.)

systemctl enable org.cups.cupsd.service
systemctl start org.cups.cupsd.service

Here is my current (working) configuration:

$ find /etc/systemd/system/ -iname *cups*
/etc/systemd/system/sockets.target.wants/org.cups.cupsd.socket
/etc/systemd/system/printer.target.wants/org.cups.cupsd.service
/etc/systemd/system/multi-user.target.wants/org.cups.cupsd.path

The note about the renamed service files is on the wiki [1].

--Kyle Terrien

[1] https://wiki.archlinux.org/index.php/Cups#CUPS_daemon



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] no cups web/server with new version?

[Kyle Terrien]

On 11/04/2014 07:50 PM, Javier Vasquez wrote:
 I noticed org.cups.cupsd.service just fails:
 
 % systemctl status org.cups.cupsd.service
 ● org.cups.cupsd.service - CUPS Scheduler
Loaded: loaded (/usr/lib/systemd/system/org.cups.cupsd.service; enabled)
Active: failed (Result: exit-code) since Tue 2014-11-04 21:24:58
 CST; 8min ago
  Main PID: 7988 (code=exited, status=1/FAILURE)
 
 Nov 04 21:24:58 m1 systemd[1]: Started CUPS Scheduler.
 Nov 04 21:24:58 m1 systemd[1]: org.cups.cupsd.service: main process
 exited, code=exited, status=1/FAILURE
 Nov 04 21:24:58 m1 systemd[1]: Unit org.cups.cupsd.service entered failed 
 state.
 
 % journalctl -u org.cups.cupsd.service
 ...
 -- Reboot --
 Nov 04 21:06:59 m1 systemd[1]: Starting CUPS Scheduler...
 Nov 04 21:06:59 m1 systemd[1]: Started CUPS Scheduler.
 Nov 04 21:07:05 m1 systemd[1]: org.cups.cupsd.service: main process
 exited, code=exited, status=1/FAILURE
 Nov 04 21:07:05 m1 systemd[1]: Unit org.cups.cupsd.service entered failed 
 state.
 Nov 04 21:12:41 m1 systemd[1]: Stopped CUPS Scheduler.
 Nov 04 21:16:12 m1 systemd[1]: Starting CUPS Scheduler...
 Nov 04 21:16:12 m1 systemd[1]: Started CUPS Scheduler.
 Nov 04 21:16:12 m1 systemd[1]: org.cups.cupsd.service: main process
 exited, code=exited, status=1/FAILURE
 Nov 04 21:16:12 m1 systemd[1]: Unit org.cups.cupsd.service entered failed 
 state.
 Nov 04 21:16:42 m1 systemd[1]: Starting CUPS Scheduler...
 Nov 04 21:16:42 m1 systemd[1]: Started CUPS Scheduler.
 Nov 04 21:16:42 m1 systemd[1]: org.cups.cupsd.service: main process
 exited, code=exited, status=1/FAILURE
 Nov 04 21:16:42 m1 systemd[1]: Unit org.cups.cupsd.service entered failed 
 state.
 Nov 04 21:24:58 m1 systemd[1]: Starting CUPS Scheduler...
 Nov 04 21:24:58 m1 systemd[1]: Started CUPS Scheduler.
 Nov 04 21:24:58 m1 systemd[1]: org.cups.cupsd.service: main process
 exited, code=exited, status=1/FAILURE
 Nov 04 21:24:58 m1 systemd[1]: Unit org.cups.cupsd.service entered failed 
 state.
 
 Notice I restarted it several times after some intermediate trials...
 
 if I run cupsd directly:
 
 % sudo cupsd
 cupsd: Child exited with status 1
 
 Which is even more weird to me.  cups named packages installed:
 
 % pacman -Qs cups
 local/cups 2.0.0-2
 The CUPS Printing System - daemon package
 local/cups-filters 1.0.61-2
 OpenPrinting CUPS Filters
 local/cups-pdf 2.6.1-2
 PDF printer for cups
 local/lib32-libcups 1.7.5-1
 The CUPS Printing System - client libraries (32-bit)
 local/libcups 2.0.0-2
 The CUPS Printing System - client libraries and headers
 
 I tried re-installing cups and libcups with no sucees (just in case)...
 
 I'm still in the dark, sorry, :-(
 
 Thanks,
 

Does /var/log/cups/error_log say anything?

I suppose you could always revert to a stock configuration (using the
.pacnews) if you have to.

--Kyle Terrien



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] nm-applet and ethernet

[Kyle Terrien]

On 10/17/2014 06:39 AM, Mauro Santos wrote:
 Previously I could left-click release and the nm-applet menu would stay
 open, now I have to left-click and hold otherwise there is no menu.
 Apart from this everything seems to work correctly.

After rereading this thread, I just remembered something.  I have (in
the past) experienced a similar issue with all GTK3 programs.  If I
clicked on a submenu, instead of popping up the corresponding submenu,
it would disappear.

I never figured out why submenus disappeared when I clicked on them, but
it's what drove me to start compiling a few programs with gtk2 instead
of gtk3 (including NetworkManager's GUI).

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] nm-applet and ethernet

[Kyle Terrien]

On 10/17/2014 06:39 AM, Mauro Santos wrote:
 Previously I could left-click release and the nm-applet menu would stay
 open, now I have to left-click and hold otherwise there is no menu.
 Apart from this everything seems to work correctly.

This sounds like something with the recent GNOME/GTK3 update.

I built network-manager-applet-gtk2 from the AUR because I'm using Xfce,
which does not always play nicely with GTK3 (themes, etc.).

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] java: cannot execute - too many levels of symbolic links

[Kyle Terrien]

On 09/08/2014 05:42 PM, Thorsten Jolitz wrote:
 
 Hi List,
 
 after updating yesterday java does not work anymore for me:
 
 , | [tj@arch ~]$ LC_ALL=C java --help | /usr/bin/java: line 2:
 /usr/lib/jvm/default/bin/java: Too many levels of | symbolic links 
 | /usr/bin/java: line 2: exec: /usr/lib/jvm/default/bin/java:
 cannot | execute: Too many levels of symbolic links `
 
 with '/usr/bin/java - /usr/lib/java-common-wrapper' containing:
 
 #+BEGIN_SRC shell #!/bin/bash exec
 ${JAVA_HOME:-/usr/lib/jvm/default}/bin/${0##*/} $@ #+END_SRC
 
 I remember some Java lib was updated recently - anyone else with
 this problem?
 

I only have jre-7-openjdk installed, so I reinstalled jre7-openjdk and
jre7-openjdk-headless.  The archlinux-java fix trick works for me too.

 kyle@landru /usr/lib/jvm
 $ la
 total 148
 drwxr-xr-x   3 root root   4096 Sep  8 18:44 .
 drwxr-xr-x 233 root root 139264 Sep  8 18:35 ..
 lrwxrwxrwx   1 root root 18 Sep  8 18:44 default - java-7-openjdk/jre
 lrwxrwxrwx   1 root root 18 Sep  8 18:44 default-runtime - 
 java-7-openjdk/jre
 drwxr-xr-x   3 root root   4096 Aug 26 20:25 java-7-openjdk
 lrwxrwxrwx   1 root root  7 Sep  8 18:27 java-default-runtime - default

--Kyle

Re: [arch-general] 3.15 kernel breaking X ?

[Kyle Terrien]

On 07/07/2014 01:21 AM, Daniel Petre wrote:
 Apologies but i do not know how to reinstall the 3.14 kernel, linux is
 3.15 and linux-lts is 3.10..
 Can i get from some repo the 3.14 package?
 Thanks.
 

For the time being, I'm maintaining a custom 3.14 kernel for myself.
It's nothing fancy. It's just the default x86_64 options, except it
builds Linux 3.14.10. (Right now, it's x86_64 only because I do not run
Arch on an older 32-bit only chip.)

Feel free to build it if you want:

https://github.com/KlipperKyle/linux-custom

Kernel 3.14.11 is coming as soon as I have the chance to update the
PKGBUILD.

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Changing sizes of panes in a vimsplit resets scrolling

[Kyle Terrien]

On 06/16/2014 03:34 AM, Timothée Ravier wrote:
 On 2014-06-15 23:49, Kyle Terrien wrote:
 I have narrowed this down to gvim/vim-runtime 7.4.307-1 and later. 
 Earlier versions of the package do not have this issue.
 
 I am asking here because I am unsure if the issue is package
 related or upstream.
 
 I can reproduce this, with and without plugins. Neovim is not affected.
 
 Please file a bug upstream. If possible, you should narrow it down to
 the specific commit creating the regression.

There is no need to file upstream. I just built vim-runtime, gvim,
gvim-python3, and vim 7.4.326 (They are all one PKGBUILD.), and the
issue with adjusting vimsplits is already fixed.

Shall I flag the packages as out of date? When does Arch normally build
newer versions of Vim? (I noticed the last patchlevel change jumped from
274 to 307.)

--Kyle



signature.asc
Description: OpenPGP digital signature

[arch-general] Changing sizes of panes in a vimsplit resets scrolling

[Kyle Terrien]

Hello Archers,

Has anyone else been experiencing an issue where scrolling is reset
when changing the size of window splits in Vim?

1. Open a file in Vim that is sufficiently long enough that you need to
   scroll to view the whole buffer.
2. Scroll to somewhere in the middle of the buffer.
3. Use :split to open another file. You should have two windows open
   (2nd file on the top pane, and 1st file scrolled to somewhere in the
   middle on the bottom pane).
4. Use ^W- and/or ^W+ to adjust the relative sizes of the panes. The
   pane that is unfocused scrolls to the top of the buffer.

The quick fix is to toggle focus between the panes, which will cause the
scrolling position to reset to normal.

I have narrowed this down to gvim/vim-runtime 7.4.307-1 and later.
Earlier versions of the package do not have this issue.

I am asking here because I am unsure if the issue is package related or
upstream.

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] dnsmasq w/ad-blocking hosts file (was NTP: Possible permissions bug)

[Kyle Terrien]

Thanks everyone for all of the suggestions on how to block ads at the
DNS level. I'm currently trying out Carl Schaefer's favorite hosts file
[1], but I look forward to experimenting with dnsmasq when my finals are
over.

--Kyle

[1] http://someonewhocares.org/hosts/zero/



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] NTP: Possible permissions bug

[Kyle Terrien]

On 05/08/2014 06:34 PM, Kyle Terrien wrote:
 Hi fellow Archers,
 
 I seem to be having trouble with ntpd.service since the 4.2.7 upgrade. I
 can't get ntpd to run as the ntp user.

Well, I feel dumb.

After adding a couple more -d flags to ntpd, I got something about
unable to lookup hostnames.

I was using an old hpHosts /etc/hosts file (to block ads). I reverted it
to the stock /etc/hosts file in the package filesystem, and ntp-4.2.7
runs without thrashing.

I'm re-enabling Adblock Plus for now--until I get around to creating an
updated hosts table.

I'm still unsure why hostnames resolved for root, but not for the ntp
user. Could the hosts file have been too large for ntpd (6.4 MB)?

--Kyle



signature.asc
Description: OpenPGP digital signature

[arch-general] NTP: Possible permissions bug

[Kyle Terrien]

Hi fellow Archers,

I seem to be having trouble with ntpd.service since the 4.2.7 upgrade. I
can't get ntpd to run as the ntp user.

I'm using the /etc/ntp.conf provided by the package ntp. When starting
ntpd.service (systemctl start ntp.service), ntpd spikes the CPU for a
moment, and then spikes the CPU about every 5 minutes afterward. What's
more is ntpq -p returns No association ID's returned.

I ran ntpd -g -u ntp:ntp -p /run/ntpd.pid -d (from the Systemd unit
file). The results: http://sprunge.us/QbYL

The part I found of interest were the lines that read:

intres: EAI_SYSTEM errno 11 (Resource temporarily unavailable) means try
again, right?
select() returned -1: Interrupted system call
select() returned -1: Interrupted system call

I took out the -u ntp:ntp parameter (so ntp runs as root), and these
errors disappeared. Also, ntpq -p returns the NTP servers I'm
synchronized with. So, I'm pretty sure the issue is permissions related,
but I have no idea what it's running into. Any insight?

I currently modified /usr/lib/systemd/system/ntpd.service so that ntpd
runs as root instead of the user ntp. (Is there a cleaner way to tweak a
systemd unit file?) Everything seems to run fine for now.

On a side note, the Wiki says that timedatectl (part of systemd) can use
ntp for synchronization [1]. However, the man page for timedatectl shows
an example where chrony (another NTP server) is running. Does
ntpd.service need to be running for timedatectl to set time via NTP? Is
there a way to force timedatectl to query NTP (for testing)?

--Kyle

[1] https://wiki.archlinux.org/index.php/Ntp#systemd_services



signature.asc
Description: OpenPGP digital signature

[arch-general] system-config-printer: conflicting files

[Kyle Terrien]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hello fellow Archers,

I received the following error while upgrading system-config-printer:

 error: failed to commit transaction (conflicting files) 
 system-config-printer: /usr/share/system-config-printer/debug.pyc
 exists in filesystem Errors occurred, no packages were upgraded.

Apparently, at some point python automatically compiled debug.py into
debug.pyc to increase long term performance. However, the new package
has debug.pyc presupplied.

The solution was to delete debug.pyc, then everything installs cleanly
and works fine.

Of course, this begs the question: Should a compiled python module be
supplied in a package (debug.pyc), or should the local machine
interpret/compile/do whatever it wants with a python script (debug.py)?

- --Kyle
-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.22 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=tWUo
-END PGP SIGNATURE-

Re: [arch-general] Error in wireshark-gtk2 in show interfaces for capture

[Kyle Terrien]

On 04/10/2014 05:04 AM, Maykel Franco wrote:
 Hi, I have installed wireshark-gtk2. But when go to Capture/Interfaces
 I get this error:
 
 There are no interfaces on which a capture can be done.
 
 I follow this steps:
 
 Setting network privileges for dumpcap
 
 1. Ensure your linux kernel and filesystem supports File Capabilities
 and also you have installed necessary tools.
 
 2. setcap 'CAP_NET_RAW+eip CAP_NET_ADMIN+eip' /usr/bin/dumpcap
 
 3. Start Wireshark as non-root and ensure you see the list of
 interfaces and can do live capture.
 
 Limiting capture permission to only one group
 
 1. Create user wireshark in group wireshark.
 
 2. chgrp wireshark /usr/bin/dumpcap
 
 3. chmod 754 /usr/bin/dumpcap
 
 4. setcap 'CAP_NET_RAW+eip CAP_NET_ADMIN+eip' /usr/bin/dumpcap
 
 5. Ensure Wireshark works only from root and from a user in the
 wireshark group
 
 
 Thanks in advanced.
 

That's strange. This is the same error that appears when running
wireshark as a user not assigned to the wireshark group.

The only configuration I had to do was add myself to the wireshark group
(which wireshark-gtk2 created):

# gpasswd -a kyle wireshark

This is according to the instructions on the wiki [1].

Here is some information on how dumpcap is installed on my box:

 kyle@landru ~ $ ls -la /usr/bin/dumpcap 
 -rwxr-xr-- 1 root wireshark 85648 Apr 10 12:45 /usr/bin/dumpcap
 kyle@landru ~ $ getcap /usr/bin/dumpcap 
 /usr/bin/dumpcap = cap_net_admin,cap_net_raw+eip
 kyle@landru ~ $ stat /usr/bin/dumpcap 
   File: ‘/usr/bin/dumpcap’
   Size: 85648 Blocks: 168IO Block: 4096   regular file
 Device: 801h/2049dInode: 1069550 Links: 1
 Access: (0754/-rwxr-xr--)  Uid: (0/root)   Gid: (  150/wireshark)
 Access: 2014-04-11 10:35:22.830667985 -0700
 Modify: 2014-04-10 12:45:35.0 -0700
 Change: 2014-04-11 10:35:11.947230948 -0700
  Birth: -
 kyle@landru ~ $ lsattr /usr/bin/dumpcap
 -e-- /usr/bin/dumpcap
 kyle@landru ~ $ pacman -Qo /usr/bin/dumpcap 
 /usr/bin/dumpcap is owned by wireshark-gtk2 1.10.6-1
 kyle@landru ~ $ 

--Kyle

[1]: https://wiki.archlinux.org/index.php/Wireshark



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Bug NetworkManager when restart NetworkManager service

[Kyle Terrien]

On 04/08/2014 04:31 AM, Maykel Franco wrote:
 Hi, I change for example dns in networkmanager, when restart
 NetworkManager , systemctl restart NetworkManager the network not
 works...
 
 [maykel@maykel-arch ~]$ sudo systemctl status NetworkManager
 ● NetworkManager.service - Network Manager
Loaded: loaded (/usr/lib/systemd/system/NetworkManager.service; enabled)
Active: active (running) since mar 2014-04-08 12:24:13 CEST; 19s ago
  Main PID: 3082 (NetworkManager)
CGroup: /system.slice/NetworkManager.service
├─1223 /usr/bin/dhcpcd -B -K -L -G -c
 /usr/lib/networkmanager/nm-dhcp-client.action enp4s0f2
└─3082 /usr/bin/NetworkManager --no-daemon
 
 abr 08 12:24:22 maykel-arch NetworkManager[3082]: info (enp4s0f2):
 DHCPv4 client pid 3093 exited with status 1
 abr 08 12:24:22 maykel-arch NetworkManager[3082]: info Activation
 (enp4s0f2) Stage 4 of 5 (IPv4 Configure Timeout) scheduled...
 abr 08 12:24:22 maykel-arch NetworkManager[3082]: info Activation
 (enp4s0f2) Stage 4 of 5 (IPv4 Configure Timeout) started...
 abr 08 12:24:22 maykel-arch NetworkManager[3082]: info (enp4s0f2):
 device state change: ip-config - failed (reason
 'ip-config-unavailable') [70 120 5]
 abr 08 12:24:22 maykel-arch NetworkManager[3082]: info
 NetworkManager state is now DISCONNECTED
 abr 08 12:24:22 maykel-arch NetworkManager[3082]: info Marking
 connection 'MO2O' invalid.
 abr 08 12:24:22 maykel-arch NetworkManager[3082]: warn Activation
 (enp4s0f2) failed for connection 'MO2O'
 abr 08 12:24:22 maykel-arch NetworkManager[3082]: info Activation
 (enp4s0f2) Stage 4 of 5 (IPv4 Configure Timeout) complete.
 abr 08 12:24:22 maykel-arch NetworkManager[3082]: info (enp4s0f2):
 device state change: failed - disconnected (reason 'none') [120 30 0]
 abr 08 12:24:22 maykel-arch NetworkManager[3082]: info (enp4s0f2):
 deactivating device (reason 'none') [0]
 
 I have reboot my computer for network on again.
 
 Is a bug networkmanager?
 
 The version is:
 
 maykel-arch /home/maykel # NetworkManager --version
 0.9.8.8
 
 Thanks in advanced.
 

I did some more poking around. This is a bug in NetworkManager [1]. When
dhcpcd is running before NetworkManager (e.g. left running by a previous
instance of NetworkManager), NetworkManager does not properly restart
dhcpcd.

The workaround is to install dhclient.

[1] https://bugzilla.gnome.org/show_bug.cgi?id=723746

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Bug NetworkManager when restart NetworkManager service

[Kyle Terrien]

On 04/08/2014 04:31 AM, Maykel Franco wrote:
 Hi, I change for example dns in networkmanager, when restart
 NetworkManager , systemctl restart NetworkManager the network not
 works...
 
 ...
 
 I have reboot my computer for network on again.
 
 Is a bug networkmanager?
 
 The version is:
 
 maykel-arch /home/maykel # NetworkManager --version
 0.9.8.8
 
 Thanks in advanced.
 

Network goes down because NetworkManager leaves dhcpcd running. In fact
my network remains up after running systemctl stop
NetworkManager.service because dhcpcd is left managing enp0s25.

If NetworkManager sees dhcpcd managing an interface when NetworkManager
starts up, NetworkManager does funny things.

So, why does systemctl stop NetworkManager.service leave dhcpcd
running?

The workaround:

sudo systemctl stop NetworkManager.service
sudo killall dhcpcd
sudo systemctl start NetworkManager.service

Here is a shell session where I (effectively) do the above:

kyle@landru ~ $ sudo systemctl stop NetworkManager.service
[sudo] password for kyle:
kyle@landru ~ $ systemctl status NetworkManager.service
● NetworkManager.service - Network Manager
   Loaded: loaded (/usr/lib/systemd/system/NetworkManager.service; enabled)
   Active: inactive (dead) since Tue 2014-04-08 15:41:08 PDT; 8s ago
  Process: 2792 ExecStart=/usr/bin/NetworkManager --no-daemon (code=exited,
status=0/SUCCESS)
 Main PID: 2792 (code=exited, status=0/SUCCESS)

Apr 08 15:14:14 landru NetworkManager[2792]: info Activation (enp0s25) suc
Apr 08 15:14:21 landru dhcpcd[2822]: enp0s25: no IPv6 Routers available
Apr 08 15:14:29 landru NetworkManager[2792]: info (enp0s25): IP6 addrconf 
Apr 08 15:14:29 landru NetworkManager[2792]: info Activation (enp0s25) Sta
Apr 08 15:14:29 landru NetworkManager[2792]: info Activation (enp0s25) Sta
Apr 08 15:14:29 landru NetworkManager[2792]: info Activation (enp0s25) Sta
Apr 08 15:41:08 landru systemd[1]: Stopping Network Manager...
Apr 08 15:41:08 landru NetworkManager[2792]: info caught signal 15, shutti
Apr 08 15:41:08 landru NetworkManager[2792]: info Writing DNS information ...f
Apr 08 15:41:08 landru systemd[1]: Stopped Network Manager.
Hint: Some lines were ellipsized, use -l to show in full.
kyle@landru ~ $ systemctl status | grep -A 1 NetworkManager.service
   │   │ └─3623 grep --color=auto -A 1 NetworkManager.service
   │   └─user@2290.service
--
 ├─NetworkManager.service
 │ └─2822 /usr/bin/dhcpcd -B -K -L -G -c
/usr/lib/networkmanager/nm-dhcp-client.action enp0s25
kyle@landru ~ $ sudo kill 2822
kyle@landru ~ $ systemctl status | grep -A 1 NetworkManager.service
   │   │ └─3631 grep --color=auto -A 1 NetworkManager.service
   │   └─user@2290.service
kyle@landru ~ $ systemctl status NetworkManager.service
● NetworkManager.service - Network Manager
   Loaded: loaded (/usr/lib/systemd/system/NetworkManager.service; enabled)
   Active: inactive (dead) since Tue 2014-04-08 15:41:08 PDT; 4min 25s ago
  Process: 2792 ExecStart=/usr/bin/NetworkManager --no-daemon (code=exited,
status=0/SUCCESS)
 Main PID: 2792 (code=exited, status=0/SUCCESS)

Apr 08 15:14:29 landru NetworkManager[2792]: info (enp0s25): IP6 addrconf 
Apr 08 15:14:29 landru NetworkManager[2792]: info Activation (enp0s25) Sta
Apr 08 15:14:29 landru NetworkManager[2792]: info Activation (enp0s25) Sta
Apr 08 15:14:29 landru NetworkManager[2792]: info Activation (enp0s25) Sta
Apr 08 15:41:08 landru systemd[1]: Stopping Network Manager...
Apr 08 15:41:08 landru NetworkManager[2792]: info caught signal 15, shutti
Apr 08 15:41:08 landru NetworkManager[2792]: info Writing DNS information ...f
Apr 08 15:41:08 landru systemd[1]: Stopped Network Manager.
Apr 08 15:43:59 landru dhcpcd[2822]: received signal TERM from PID 3626, st...ng
Apr 08 15:43:59 landru dhcpcd[2822]: enp0s25: removing interface
Hint: Some lines were ellipsized, use -l to show in full.
kyle@landru ~ $ sudo systemctl start NetworkManager.service
kyle@landru ~ $

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Pulseaudio 5.1 Setup echoes Front Speakers to Rear speakers

[Kyle Terrien]

On 03/16/2014 09:59 PM, Mark Lee wrote:
 Salutations,
 
 I recently hooked up a 5.1 surround sound receiver to my Haswell setup
 via HDMI. I used the sound manager in gnome to set the system to output
 5.1 surround sound. However, while testing each individual speaker using
 the gnome sound settings, I found that front speakers were echoing
 content to rear speakers (with a small delay). I also noticed that if I
 set the fade to rear, I would lose all front speaker output. Moving the
 fade to the front reproduced the echo  again. This is noticeable in any
 5.1 audio I play as well. Has anyone experienced this and found a solution?
 
 Regards,
 Mark
 

Are you using PulseAudio? If so, try looking at the sound settings with
pavucontrol, the official PulseAudio configuration utility (in the
package extra/pavucontrol).

Correct me if I'm wrong, but if I remember correctly, the Gnome sound
manager and pavucontrol are different programs. In my experience, I have
had the best luck using pavucontrol to configure sound.

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Pulseaudio 5.1 Setup echoes Front Speakers to Rear speakers

[Kyle Terrien]

On 03/16/2014 10:21 PM, Kyle Terrien wrote:
 Are you using PulseAudio?

D'oh! I noticed the subject line said Pulseaudio right after sending
my message. Sorry for the stupid question.

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] NetworkManager 0.9.8.8-2 does not work with systemd 208

[Kyle Terrien]

On 02/23/2014 06:30 PM, Mark E. Lee wrote:
 
 Salutations,
 
 I downgraded and can confirm that fixes the situation. This was a pretty
 bad situation for me since I don't keep a package cache (I run on usb);
 I ended up rebuilding network manager 0.9.8.8-1 using an old PKGBUILD.
 Why was networkmanager pushed before systemd?
 
 Regards,
 Mark
 

Hi,

Just an FYI: Someone is maintaining an Arch Rollback Machine again. You
can find all the NetworkManager builds there (among other things).

Wiki: https://wiki.archlinux.org/index.php/Arch_Rollback_Machine
NetworkManager: http://seblu.net/a/arm/packages/n/networkmanager/

--Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Building a live USB version of ArchLinux for ARM users

[Kyle Terrien]

On 02/22/2014 04:54 PM, Hugo Rabson wrote:
 Is anyone interested in working with me to build a live USB edition and/or 
 hardened ArchLinux for ARM users (specifically, users of ARM-based 
 chromebooks)? It would be nontrivial but achievable. The question is, would 
 anyone find the product useful?
 
 -Hugo
 

I've considered putting together an Arch-based recovery disc (as an
alternative to PartedMagic now that they are commercial). It sounds like
a similar project, except for x86.

Unfortunately, I don't have an ARM-based chromebook.

Are you trying to build a full-featured Live desktop environment for
chromebooks?

- Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Building a live USB version of ArchLinux for ARM users

[Kyle Terrien]

On 02/22/2014 08:03 PM, Thiago Barroso Perrotta wrote:
 Hey Kyle,
 
 I've always thought of creating something similar too, but I didn't do
 anything about it because of its usefulness (would people use it, since
 parted magic and other similar projects already exist?) and maintainability
 issues. But it is good to know another person who also finds this a good
 idea. The Arch ISO is really great for recovery purposes without X (dd, for
 example, yesterday I used it this way!), but there are some GUI-based
 recovery tools without an easy CLI replacement.
 
 We should probably stick back to the topic, but please let me know if you
 ever do something about this.

If I ever get around to building an Arch-based recovery disc, I'll let
you know.

- Kyle



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Packages Verified with MD5

[Kyle Terrien]

On 01/12/2014 12:40 PM, Taylor Hornby wrote:
 I guess I just don't understand what happens when I type pacman -S
 firefox. Does that run the PKGBUILD on my system, or does it download
 and install pre-compiled (and signed) Firefox binaries that were
 created by one of the Arch developers using the PKGBUILD?

pacman -S firefox installs a pre-compiled binary maintained by an Arch
Dev. On the other hand, PKGBUILDs are for building packages.

And the official firefox package is cryptographically signed by the
package maintainer (not Mozilla).

Hopefully, that clears things up.

If you really want to build a firefox package yourself, you can set up
ABS. If you build a package from ABS (using makepkg), you will run the
PKGBUILD. https://wiki.archlinux.org/index.php/Abs

Kyle Terrien

PS: Great discussion on exploiting MD5.



signature.asc
Description: OpenPGP digital signature

Re: [arch-general] Packages Verified with MD5

[Kyle Terrien]

On 01/12/2014 01:13 PM, Taylor Hornby wrote:
 Thank you, that makes so much more sense!
 
 So, really, the vulnerability only exists while the Arch dev (or
 package maintainer or whatever they're called) is building the
 package. Once they do, and sign it, all Arch users will verify their
 signature to make sure they get the same file the Arch dev created.

That's correct! See these pages for more info on how pacman's signature
checking works:

https://wiki.archlinux.org/index.php/Pacman#Package_security
https://wiki.archlinux.org/index.php/Pacman-key

 That's not so bad, then, since you can't really do any better unless
 the upstream source (Mozilla) signs their files, and the package
 maintainer has their public key.

To be honest, I'm a little surprised that Mozilla doesn't sign their
Firefox source code.

Kyle



signature.asc
Description: OpenPGP digital signature