Re: [gentoo-user] USB device (ATtiny86 w. mcronucleus bootloader) not recognized ?

[R0b0t1]

Hello again, I believe I have cause to bring this thread back to life.

On Sun, Aug 6, 2017 at 11:50 AM,   wrote:
> When I plug in such a little board into my PC, demesg
> reports:
> [ 1429.834140] usb 7-4: new low-speed USB device number 15 using ohci-pci
> [ 1429.965142] usb 7-4: device descriptor read/64, error -62
> [ 1430.203151] usb 7-4: device descriptor read/64, error -62
> [ 1430.438161] usb 7-4: new low-speed USB device number 16 using ohci-pci
> [ 1430.569151] usb 7-4: device descriptor read/64, error -62
> [ 1430.803174] usb 7-4: device descriptor read/64, error -62
> [ 1431.038184] usb 7-4: new low-speed USB device number 17 using ohci-pci
> [ 1431.456157] usb 7-4: device not accepting address 17, error -62
> [ 1431.582204] usb 7-4: new low-speed USB device number 18 using ohci-pci
> [ 1432.000209] usb 7-4: device not accepting address 18, error -62
> [ 1432.000244] usb usb7-port4: unable to enumerate USB device
>

This is with a NUCLEO-L432KC, an STM32 development board:

[1817074.986238] usb usb3-port14: unable to enumerate USB device
[1817075.429167] usb 3-14: new low-speed USB device number 50 using xhci_hcd
[1817075.586133] usb 3-14: device descriptor read/64, error -71
[1817075.847136] usb 3-14: device descriptor read/64, error -71
[1817076.110100] usb 3-14: new low-speed USB device number 51 using xhci_hcd
[1817076.267102] usb 3-14: device descriptor read/64, error -71
[1817076.527036] usb 3-14: device descriptor read/64, error -71
[1817076.790003] usb 3-14: new low-speed USB device number 52 using xhci_hcd
[1817076.790470] usb 3-14: Device not responding to setup address.
[1817076.994569] usb 3-14: Device not responding to setup address.
[1817077.201986] usb 3-14: device not accepting address 52, error -71
[1817077.357969] usb 3-14: new low-speed USB device number 53 using xhci_hcd
[1817077.358525] usb 3-14: Device not responding to setup address.
[1817077.562503] usb 3-14: Device not responding to setup address.
[1817077.769930] usb 3-14: device not accepting address 53, error -71
[1817077.769959] usb usb3-port14: unable to enumerate USB device
[1817080.134131] usb 3-5: USB disconnect, device number 29

Admittedly, I cut apart a USB cable and am using socketed pin headers
to attach it to the development board (I'd like to keep it looking
nice). However, this same board properly identifies as a USB CDC
device to Windows.

There may be an issue with the kernel driver, Meino. I'm going to see
how I should follow up on this. Can you test your ATtiny85 with
Windows?

R0b0t1.

Re: [gentoo-user] Online hosting recommendation - VMs?

[Stroller]

> On 26 Mar 2017, at 03:57, Stroller  wrote:
> 
> In the next few weeks I need to move my email server (a very old Gentoo 
> installation) from the closet in my home, into the cloud so that I can go 
> travelling and access my mail from anywhere.

A few months ago I asked for hosting recommendations, and was surprised not to 
receive any mention of Amazon's cloud services.

I thought reason might be that Amazon's cloud servers are different from a 
regular VM, but today saw someone on the Postfix list state that they're 
running it on an AWS instance.

Has anyone tried running Gentoo on AWS or did this go unmentioned because it's 
impossible?

Stroller.

Re: [gentoo-user] No beep.

[Alan Mackenzie]

Hello, Corbin.

On Mon, Aug 14, 2017 at 21:06:56 -0500, Corbin Bird wrote:
> On 08/14/2017 01:22 PM, Alan Mackenzie wrote:
> > Hello, Gentoo.

> > I've almost got my new(ish) machine up and running.  In particular, I've
> > got an email server (s/qmail) running on it, the most difficult part of
> > bringing up a new box.

> > However, during the building, I discovered to my disgust that there was
> > no loudspeaker in my new case.  So, none of the helpful booting beeps
> > that the motherboard uses to tell us that something's wrong.  Still, I
> > got past that stage.

> > But just typing into a virtual tty, sometimes the system (should) beep.
> > It doesn't.  It should.

> > It should, because I compiled the kernel with CONFIG_SND_HDA_INPUT_BEEP
> > and CONFIG_SND_HDA_INPUT_BEEP_MODE set to 1.

> > My loudspeakers are working, as shown by running speaker-test.  I'm
> > using the kernel I think I am, verified by checking those two config
> > settings in /proc/config.gz.

> > Would somebody please suggest to me where I am going wrong, here?

> > Thanks!


> Apologies, if this is just a stupid question ...

> Did you purchase / install the old BIOS / Case speaker?
> ... yes, that 80's piece of speaker tech :)

Not as such.  I expected the speaker to be part of the case, as it always
has been in the past.

> Newer motherboards and cases don't come with one.
> Had to spend ... $3 for a new one.
> (U)EFI won't go BEEP! without one.

You can buy PC case loudspeakers?  I'll need to search one out.  Ideal
would be one which screwed into a disk drive mounting (of which I have
rather a lot spare).

> Corbin

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: [gentoo-user] No beep.

[Alan Mackenzie]

Hello, Joost.

On Tue, Aug 15, 2017 at 06:38:31 +0200, J. Roeleveld wrote:
> On 14 August 2017 20:22:54 GMT+02:00, Alan Mackenzie  wrote:
> >Hello, Gentoo.

> >I've almost got my new(ish) machine up and running.  In particular,
> >I've
> >got an email server (s/qmail) running on it, the most difficult part of
> >bringing up a new box.

> >However, during the building, I discovered to my disgust that there was
> >no loudspeaker in my new case.  So, none of the helpful booting beeps
> >that the motherboard uses to tell us that something's wrong.  Still, I
> >got past that stage.

> >But just typing into a virtual tty, sometimes the system (should) beep.
> >It doesn't.  It should.

> >It should, because I compiled the kernel with CONFIG_SND_HDA_INPUT_BEEP
> >and CONFIG_SND_HDA_INPUT_BEEP_MODE set to 1.

> >My loudspeakers are working, as shown by running speaker-test.  I'm
> >using the kernel I think I am, verified by checking those two config
> >settings in /proc/config.gz.

> >Would somebody please suggest to me where I am going wrong, here?

> >Thanks!

> Silly question, but did you check the volume settings?

Not as such, but with speaker-test, the pink noise came through perfectly
audibly.  I don't think there's anything specifically to adjust the beep
volume.

> --
> Joost
> -- 
> Sent from my Android device with K-9 Mail. Please excuse my brevity.

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: [gentoo-user] Re: No beep.

[Alan Mackenzie]

Hello, Nikos.

On Tue, Aug 15, 2017 at 17:02:41 +0300, Nikos Chantziaras wrote:
> On 14/08/17 21:22, Alan Mackenzie wrote:
> > However, during the building, I discovered to my disgust that there was
> > no loudspeaker in my new case.

> You sure? These days, it's not shaped like a speaker anymore, but more 
> like... a clip-on microphone? Not sure how to describe it. It's really 
> small and easy to overlook in the mainboard's package along the other 
> cables and connectors. It's this one:

Yes, I'm sure.  I think.  There's no mention in the motherboard's manual
about a speaker being included, and there was no mention in the case's
manual about hooking up the loudspeaker.

> https://www.newegg.com/Product/Product.aspx?Item=9SIA2RP0YW4962

I'll have a look at that, sometime.  I've just migrated "everything" to
my new PC, and (apart from the beeplessness) am happy about it.
Libreoffice emerged in just ~25 minutes (plus ~11 minutes for the other
42 packages which went with it).

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: [gentoo-user] cleaning /var/lib/portage/world

[Neil Bothwick]

On Wed, 16 Aug 2017 13:23:28 -0300, Francisco Ares wrote:

> 2017-08-16 10:40 GMT-03:00 Arve Barsnes :
> 
> > On 16 August 2017 at 13:32, Francisco Ares  wrote:
> >  
> >> But, after backing up the original "world" file and replacing with
> >> the one built by the script, things don't work as expected, as a lot
> >> of packages were orphaned, by checking with "depclean".
> >>
> >> Anyone could tell me what did I miss?
> >>  
> >
> > You're probably getting false positives from equery, as you haven't
> > taken USE flags into consideration.
> >
> > An example:
> > # equery d unrar
> > app-arch/rar-5.5.0_p20170811 (all_sfx ? app-arch/unrar)
> > app-emulation/winetricks-20170614 (rar ? app-arch/unrar)
> >
> > I don't have the USE="rar" set on winetricks, so although it is
> > correct that rar would keep unrar available, since I have
> > USE="all_sfx" set, your script will make the wrong choice if I don't
> > have rar installed.

> Good point!  Back to work.

emerge -cpv cat/pkg will tell you whether the package can be removed or
if it is needed by something else. It will take a while to run it for
every atom in @world but it will give reliable results, as determined by
portage.


-- 
Neil Bothwick

Nostalgia isn't what it used to be.


pgpY_5dk45BkU.pgp
Description: OpenPGP digital signature

Re: [gentoo-user] cleaning /var/lib/portage/world

[Francisco Ares]

2017-08-16 10:40 GMT-03:00 Arve Barsnes :

> On 16 August 2017 at 13:32, Francisco Ares  wrote:
>
>> But, after backing up the original "world" file and replacing with the
>> one built by the script, things don't work as expected, as a lot of
>> packages were orphaned, by checking with "depclean".
>>
>> Anyone could tell me what did I miss?
>>
>
> You're probably getting false positives from equery, as you haven't taken
> USE flags into consideration.
>
> An example:
> # equery d unrar
> app-arch/rar-5.5.0_p20170811 (all_sfx ? app-arch/unrar)
> app-emulation/winetricks-20170614 (rar ? app-arch/unrar)
>
> I don't have the USE="rar" set on winetricks, so although it is correct
> that rar would keep unrar available, since I have USE="all_sfx" set, your
> script will make the wrong choice if I don't have rar installed.
>
> Cheers,
> Arve
>

Good point!  Back to work.

Thanks,
Francisco

Re: [gentoo-user] cleaning /var/lib/portage/world

[Arve Barsnes]

On 16 August 2017 at 13:32, Francisco Ares  wrote:

> But, after backing up the original "world" file and replacing with the one
> built by the script, things don't work as expected, as a lot of packages
> were orphaned, by checking with "depclean".
>
> Anyone could tell me what did I miss?
>

You're probably getting false positives from equery, as you haven't taken
USE flags into consideration.

An example:
# equery d unrar
app-arch/rar-5.5.0_p20170811 (all_sfx ? app-arch/unrar)
app-emulation/winetricks-20170614 (rar ? app-arch/unrar)

I don't have the USE="rar" set on winetricks, so although it is correct
that rar would keep unrar available, since I have USE="all_sfx" set, your
script will make the wrong choice if I don't have rar installed.

Cheers,
Arve

Re: [gentoo-user] Codeblocks Fails to Exit Cleanly in Awesome WM

[Dutch Ingraham]

On Wed, Aug 16, 2017 at 01:14:13AM -0400, P Levine wrote:
> On Tue, Aug 15, 2017 at 10:28 AM, Dutch Ingraham  wrote:
> 
> > Specifically, when exiting in any manner, the GUI will disappear,
> > but the Codeblocks process continues to run.
> >
> 
> ​Maybe related to https://github.com/awesomeWM/awesome/issues/1193?

Thanks, that's an interesting bug with some apparent similarities.  However,
the process state after closing Codeblocks in Awesome is not a zombie, but
Ssl, which indicates some event handling issue.  I was looking at this bug:

https://github.com/awesomeWM/awesome/issues/1975

as the issue seems related to some event handler or IPC issue, but I haven't
progressed too far with that.

Re: [gentoo-user] Codeblocks Fails to Exit Cleanly in Awesome WM

[Dutch Ingraham]

On Tue, Aug 15, 2017 at 11:21:48AM -0500, R0b0t1 wrote:
> On Tue, Aug 15, 2017 at 9:28 AM, Dutch Ingraham  wrote:
> > Hi all:
> >
> > I'm having a problem with Codeblocks not exiting cleanly when using the
> > Awesome WM.  I've filed a bug report[1] but the wrangler closed it almost
> > immediately without any testing or attempt to confirm, so I'm asking for
> > your help here.
> >
> 
> I don't necessarily agree with the closure of the bug report. I
> suspect such closures happen because the first responder does not know
> how to troubleshoot the issue. However, the Gentoo developers care a
> lot more than, say, the Mozilla developers. Every bug report I have
> sent to Mozilla trackers has been ignored.

Thanks for your response.

I've had issues with this particular wrangler before.  See:
https://bugs.gentoo.org/show_bug.cgi?id=618148

> 
> It seems like a tossup: Awesome might be launching Code::Blocks in a
> way that leaves the process unable to close itself, but seeing as I
> don't actually know how that would happen, you might have better luck
> asking the Code::Blocks developers as they will know how to
> troubleshoot their project. They then might be able to figure out that
> it is something else's fault.
> 

Given how bug reports are handled on non-critical issues, I think a likely 
response from CodeBlocks would be something like "You stated Codeblocks shuts 
down cleanly in Arch, Fedora, and Debian when using Awesome.  This is a Gentoo 
issue."  A likely response from Awesome would be "You stated this only occurs 
with Codeblocks on Gentoo; this is a Codeblocks or Gentoo problem" And around 
and around we go...


> > I am using Codeblocks-16.01 (the only unmasked version, which is keyworded
> > unstable) on both a full-unstable installation and on a mostly-stable
> > installation, both with only the 'contrib' use flag set.  On both 
> > installations,
> > Codeblocks, when started from a menu in the Awesome WM, will not exit
> > cleanly.  Specifically, when exiting in any manner, the GUI will disappear,
> > but the Codeblocks process continues to run.
> >
> > However, when started from a terminal, Codeblocks does not show this 
> > behavior,
> > i.e., it does shut down completely and cleanly.
> >
> > Additionally, this behavior does not present when using either the Plasma
> > desktop or the Fluxbox WM, i.e., Codeblocks will shutdown cleanly whether
> > started from a menu or from a terminal.
> >
> > This is likely not strictly an upstream issue. This is almost certainly a 
> > Gentoo
> > issue, as issue I presented does not appear in Arch, Debian, or Fedora.
> > All three have the exact same version of Codeblocks as Gentoo; all three 
> > have
> > similar versions of Awesome (all in the 4.x branch) and Arch is using the
> > exact same version of Awesome (4.2) that I am using on both Gentoo
> > installations noted above.
> >
> > Any help or other insights in debugging this issue are appreciated.
> >
> 
> When I encounter issues like this I try to find the newest version of
> the project. In this case, it looks like 16.01 is the latest release
> of Code::Blocks. You might try building from their repository. It will
> take time to become acquainted with the different build systems
> (automake, CMake, ...) but you can generally specify a custom
> installation directory -- ~/.local is a good place. Add the binary
> paths inside of that directory to your PATH.
> 
> You can also run a command to kill Code::Blocks after you exit,

As noted, Codeblocks will exit cleanly if started from a terminal emulator, so
that seems like the cleanest path to run the program for now.

> or use
> another IDE (like Geany). Some example commands:
> 
> ps | awk '/codeblocks/ { print $1; }' | xargs kill
> kill `pgrep codeblocks`
> pkill codeblocks
> 
> Be careful that you don't kill processes that contain the name you
> give but aren't the one you mean. The regex in the first command will
> be easier to customize. Pkill and pgrep are nonstandard.
> 
> R0b0t1.
> 

[gentoo-user] cleaning /var/lib/portage/world

[Francisco Ares]

Hello,

During the process of learning how to manage packages in our beloved distro
- and even after knowing how to, but due to being in a hurry - I have my
"world" package set a bit cumbered with package names that would not need
to be there, as they are dependencies of other packages, as some libraries,
for instance.

So I thought on finding out which of those packages could be removed from
"world", and I have come up with the following script, which results in 3
files in the "/tmp" directory:

- world_remove_candidates
- world_remove_candidates_reasons
- world

- the first is a simple list of packages found in "world" that _probably_
could be removed from "world";
- the second includes which packages depends on each of the above mentioned
packages;
- the third, the cleaned up "world", without overwriting the original.

But, after backing up the original "world" file and replacing with the one
built by the script, things don't work as expected, as a lot of packages
were orphaned, by checking with "depclean".

Anyone could tell me what did I miss?

Here is the script:

#---
#! /bin/bash
cd /tmp
if [ -e world_remove_candidate ]
then
rm -f world_remove_candidate
fi
if [ -e world_remove_candidate_reasons ]
then
rm -f world_remove_candidate_reasons
fi
for i in `cat /var/lib/portage/world`
do
A=`equery d $i`
if [ "a$A" != "a" ]
then
echo $i >> world_remove_candidate
echo "* * * * * * * * * * * * * "$i >>
world_remove_candidate_reasons
for j in $A
do
echo $j >> world_remove_candidate_reasons
done
fi
done

cp -a /var/lib/portage/world .
for i in `cat world_remove_candidate`
do
S=`echo $i | sed s+/+/+`  # this is to format the string
accorgingly to 'sed' (ugly, isn't it?)
cat world | sed -e /"$S"/d > world.x
rm -f world
mv world.x world
done

# the following loop is for announcing it has finished.
A=20
while [ "$A" != "0" ]
do
A=$(( $A - 1 ))
sleep 1
echo -n -e \\a
done

ls -halF world*
#---


Thanks!
Francisco

Re: [gentoo-user] Install on ZFS

[J. Roeleveld]

On Wednesday, August 16, 2017 6:12:03 AM CEST Grant wrote:
> >>I'm trying to get started on my first Gentoo install on ZFS.  My
> >>remote server is booted to a Gentoo LiveCD and I've SSH'ed in but it
> >>looks like the CD doesn't have ZFS tools:
> >>
> >>livecd ~ # zpool status
> >>-bash: zpool: command not found
> >>
> >>Can I pull this off from here or do I need to mail a different CD to my
> >>host?
> >>
> >>This server has 6 SSDs and I'm planning to set up a 2-way stripe and
> >>3-way mirror.  Any tips that might help me pull this off?
> >>
> >>- Grant
> >>
> > I did this not too long ago, using the boot disc and howto from
> > FearedBliss: https://wiki.gentoo.org/wiki/User:Fearedbliss
> 
> This looks like a great guide.  Are there instructions for extracting
> a portage archive over my running LiveCD OS so I can then emerge zfs
> and execute this guide without mailing another CD-R to my host?
> 
> - Grant

On this, I don't know.
There are guides on how to build a sysresccd yourself with additional software 
added. I was thinking of having to do that myself.
But when I found this one, I stopped investigating that.

--
Joost

Re: [gentoo-user] Install on ZFS

[C .]

"J. Roeleveld"  writes:

> On 15 August 2017 16:13:37 GMT+02:00, "C."  wrote:
>>Rich Freeman  writes:
>>> Is this your first Gentoo install, or your first Gentoo+ZFS install?
>>> If the former, you're definitely not doing this the easy way...
>>
>>Not easy, no. But definitely possible. I managed to do so.
>>
>>One thing to keep in mind, though, is that it might be best to _not_
>>put
>>"/boot" on ZFS. I did and can't upgrade my zpool just yet, because - as
>>far as I know (please corret me if I am wrong)- Grub ain't able to boot
>>from a zpool that has been upgraded to latest ZoL (> 0.7). It works
>>just
>>fine with 0.6.5.11 though.
>>
>>Kind regards
>>Christian.
>
> That's why the one I linked to actually tells you to create 2
> zpools. 1 for /boot and 1 for the rest.

Right, sorry, must've overlooked your link. Just as I have forgotten
about this during installation. Ah well.

-- 
[ Insert favourite quote here. ]