[gentoo-user] Re: (G)vim 9.0.0099 won't start

[nunojsilva]

On 2022-08-21, Philip Webb wrote:

> Today, I updated to the latest stable Vim 9.0.0099 + GVim same
> & it refused to open.  No problem back with 8.2.4586.
>
> Has anyone else encountered this ?  Does anyone have a suggestion ?

No *vim here, but: what does happen exactly? That is, what do you mean
with "refused to open"? Is there any error message? (If it hasn't been
started from a terminal or terminal emulator, try that way and see if
there is any relevant output.)

-- 
Nuno Silva

[gentoo-user] Re: Layman adding git over ssh only repository

[nunojsilva]

On 2022-08-15, Alexander Puchmayr wrote:

> Hi there,
>
> I tried to add a repository which is only reachable via git over ssh; no 
> access via git port (9418 according to /etc/services) or http(s). Only ssh 
> connections with public key are accepted.
>
> I tried 
> Layman -o ssh://git@myserver/path/to/repo.git -f -a myrepo, but failed with 
> "invalid url".
> Then I tried the same with "git://git@..." , also failed.
>
> It seems like it does not accept the url scheme and refusing to connect at 
> all. Next thing I tried is to create a repository.xml and put it to the local 
> file system:
>
> 
> 
> 
>   
> my-repo
> My overlay
> 
>   alexander.puchmayr#my-email.com
>   Alexander Puchmayr>
> 
>   
> git://myserver/path/to/repo.git
> 
>
> Layman -o repository.xml -f -a myrepo now tries to fetch it via the git port 
> (9418), failing with timeout.
>
> Replacing source type="git" with "ssh", "ssh+git", "git+ssh" also did not 
> succeed. 
>
> I'm pretty sure not being the first one trying this, but I cannot find any 
> useful information how to do this. So how can I achieve this?

This is not something I have tried myself, but a quick glance at the
online manual page for layman shows a git+ssh example (under "Overlay
lists format") - the source type is still "git", "git+ssh" is only in
the protocol/scheme part of the URL, so perhaps it needs to be something
like:

git+ssh://myserver/path/to/repo.git

-- 
(Apologies if this is not helpful,)
Nuno Silva

[gentoo-user] Re: KDE update puzzle

[nunojsilva]

On 2021-09-18, Philip Webb wrote:

> It mb that I have to remove the whole of the present versions
> in order to (re-)install the new ones ; it usually happens with Qt.
> I've never understood why Portage can't handle that itself.

Portage might be able to handle this automatically if you upgrade @world
directly, but if you only ask to upgrade specific packages (the case
here?), you may be leaving out some package which needs to be upgraded
at the same time.

If that is the problem, solving it is probably a matter of identifying
packages which are not being upgraded and that depend on currently
installed versions of packages that will be upgraded. The output you
have included in the message I'm replying to is, unfortunately, not
enough - unless I'm misreading, I think it only says that:

- kcrash needs to be upgraded to allow the kcoreaddons upgrade, and

- kglobalaccel needs to be upgraded to allow kconfig, kwindowsystem,
  kdbusaddon and kcrash upgrades

But these two are already part of your emerge command, so there is
possibly something else that is not in this list. Could you please share
the output with --verbose-conflicts?

-- 
Nuno Silva

[gentoo-user] Re: KDE update puzzle

[nunojsilva]

On 2021-09-18, Philip Webb wrote:

> There is an update available for KDE Frameworks.
> When I try to perform it, I get a conflict statement which begins
>
>   !!! Multiple package instances within a single package slot have been pulled
>   !!! into the dependency graph, resulting in a slot conflict:
>   kde-frameworks/kcoreaddons:5
>
>   (kde-frameworks/kcoreaddons-5.85.0:5/5.85::gentoo, ebuild scheduled for 
> merge) USE="-debug -doc (-fam) -nls -test" ABI_X86="(64)" pulled in by
> =kde-frameworks/kcoreaddons-5.85*:5 required by 
> (kde-frameworks/kglobalaccel-5.85.0-r1:5/5.85::gentoo, ebuild scheduled for 
> merge) USE="-debug -doc -nls -test" ABI_X86="(64)"
> ^   ^^^
>
> The  ^  signs emphasise  =kde-frameworks/kcoreaddons-5.85*:5
>  ^   ^^^
>
> I can't make sense of this : surely, '5.85*' includes '5.85.0',
> so the requirement sb fulfilled.  I don't remember seeing this before.

Is this the complete message about the conflicts? If there is nothing
more, please try again with the --verbose-conflicts emerge parameter.

-- 
Nuno Silva

[gentoo-user] Re: Motif compile fails

[nunojsilva]

On 2021-08-28, Philip Webb wrote:

> As part of updating to the latest stable Xscreensaver,
> I tried to emerge its requirement Motif, which failed with this message :
[...]
> /usr/lib/gcc/x86_64-pc-linux-gnu/10.2.0/../../../../x86_64-pc-linux-gnu/bin/ld:
> /usr/lib/gcc/x86_64-pc-linux-gnu/10.2.0/../../../../lib64/libfl.a(libmain.o):
> relocation R_X86_64_PC32 against symbol `exit@@GLIBC_2.2.5' can not be
> used when making a PIE object; recompile with -fPIE
> /usr/lib/gcc/x86_64-pc-linux-gnu/10.2.0/../../../../x86_64-pc-linux-gnu/bin/ld:
> final link failed: bad value
> collect2: error: ld returned 1 exit status
> make: *** [Makefile:502: wmluiltok] Error 1
> make: Leaving directory
> '/var/tmp/portage/portage/x11-libs/motif-2.3.8-r2/work/motif-2.3.8-abi_x86_64.amd64/tools/wml'
>  * ERROR: x11-libs/motif-2.3.8-r2::gentoo failed (compile phase):
>  *   emake failed
>
> -- end of Portage output --
>
> This looks as if the problem is with Gcc flags, which is rather technical.
>
> Has anyone else run into this ?  Does anyone have suggestions ?

As the error involves libfl.a, from sys-devel/flex, I'd try rebuilding
that package. See bug 583842,

https://bugs.gentoo.org/show_bug.cgi?id=583842

-- 
Nuno Silva

[gentoo-user] Re: --depclean wants to remove openrc. Yikes!

[nunojsilva]

On 2021-08-04, Philip Webb wrote:

> 210804 Dr Rainer Woitok wrote:
>> On Tue, 03 Aug 2021 07:45:27 -0400, Alec Ten Harmsel wrote:
>>> $ emerge --ask --depclean | less
>> I tested this using "most" from within a "urxvt" terminal
>> and at least with default options it did not work for me.
>
> Why not write the output to a file ?  -- eg
> 'emerge --ask --depclean > '.
> Then you can look at the output at leisure, even on another machine.

Other possibilities probably include using --pretend instead of --ask
(was this mentioned already? if so, my apologies) or using the "tee"
utility.

-- 
Nuno Silva

[gentoo-user] Re: dev-lang/perl:0

[nunojsilva]

On 2021-07-30, nunojsi...@ist.utl.pt wrote:

> On 2021-07-30, the...@sys-concept.com wrote:
>
>> I'm trying to run an upgrade but I got stuck on perl:
> [...]
>>   (dev-lang/perl-5.34.0:0/5.34::gentoo, ebuild scheduled for merge)
>> USE="gdbm -berkdb -debug -doc -ithreads -minimal" ABI_X86="(64)"
>> pulled in by
> [...]
>> Running:  "perl-cleaner --all" gives me:
> [...]
>>   (dev-lang/perl-5.32.1:0/5.32::gentoo, installed) USE="gdbm -berkdb -debug 
>> -doc -ithreads -minimal" ABI_X86="(64)" pulled in by
>> dev-lang/perl:0/5.32= required by (app-text/po4a-0.57:0/0::gentoo, 
>> installed) USE="-test" ABI_X86="(64)"
>>   
>
> Are you upgrading @world or just dev-lang/perl directly? If it's the
> latter, try to upgrade po4a at the same time.

(In this case, it will probably be an upgrade (as 0.57-r1 is stable for
amd64), but I should have written "upgrade or rebuild".)

-- 
Nuno Silva

[gentoo-user] Re: dev-lang/perl:0

[nunojsilva]

On 2021-07-30, the...@sys-concept.com wrote:

> I'm trying to run an upgrade but I got stuck on perl:
[...]
>   (dev-lang/perl-5.34.0:0/5.34::gentoo, ebuild scheduled for merge) USE="gdbm 
> -berkdb -debug -doc -ithreads -minimal" ABI_X86="(64)" pulled in by
[...]
> Running:  "perl-cleaner --all" gives me:
[...]
>   (dev-lang/perl-5.32.1:0/5.32::gentoo, installed) USE="gdbm -berkdb -debug 
> -doc -ithreads -minimal" ABI_X86="(64)" pulled in by
> dev-lang/perl:0/5.32= required by (app-text/po4a-0.57:0/0::gentoo, 
> installed) USE="-test" ABI_X86="(64)"
>   

Are you upgrading @world or just dev-lang/perl directly? If it's the
latter, try to upgrade po4a at the same time.

>From a quick glance at the po4a ebuild and the perl-module eclass, the
default seems to be a run-time dependency on perl with the same slot and
subslot as the perl present when the package was installed, which would
explain why po4a needs perl:0/5.32.

-- 
Nuno Silva

[gentoo-user] Re: world update gone astray portage will not let me continue

[nunojsilva]

On 2021-07-20, John Covici wrote:

> On Tue, 20 Jul 2021 02:53:30 -0400,
> Neil Bothwick wrote:
>> 
>> [1  ]
>> On Tue, 20 Jul 2021 07:36:27 +0100, (Nuno Silva) wrote:
>> 
>> > As this is a problem with generating documentation, have you tried the
>> > following as a workaround?:
>> > 
>> > USE=-doc emerge -1a media-libs/harfbuzz
>> 
>> It is recommended that you do not enable the doc USE flag globally, but
>> only for those packages where you need the extra documentation.
>> 
>> % grep \^doc /var/portage/profiles/use.desc
>> doc - Add extra documentation (API, Javadoc, etc). It is recommended
>> to enable per package instead of globally
> Well, I finally did manage to get harfbuzz installed which brings up
> another question.  First, I tried to find the package for the
> /usr/bin/gtkdoc packages and found none.  So, on a hunch, I emerged
> dev-utils/gtk-doc and discovered I did not have the package in my
> install.  So, I did it anyway and after that harfbuzz installed.  So,
> the question is, should I put that in my world file -- which seems
> wrong to me, or is there a missing dependency somewhere or what?

If I'm reading the ebuild correctly, currently having USE=-doc does not
disable building the documentation.  Setting USE=doc would probably have
worked as a workaround, because then emerge would pull gtk-doc as a
dependency.

I've mentioned this problem on IRC, but I haven't filled anything at
bugzilla.


I wonder if replacing

$(meson_native_use_feature doc)

with

$(meson_native_use_feature doc docs)

is enough? (I haven't tested it, though)

-- 
Nuno Silva

[gentoo-user] Re: world update gone astray portage will not let me continue

[nunojsilva]

On 2021-07-19, John Covici wrote:

> Hi.  I am having major problems with my world update started a couple
> of days ago.  There is a long history which I won't go into now, but
> what I have immediately is that media-libs/harfbuzz will not install
> and even though I have --keep-going, portage won't let me continue.
>
> Here is the last part of the build log:
>
> /var/tmp/portage/media-libs/harfbuzz-2.8.2-r1/work/harfbuzz-2.8.2/docs/../src/hb-unicode.cc:180:
> warning: unknown annotation "Xconstructor" in documentation for
> hb_unicode_funcs_create.^M
> See harfbuzz-undeclared.txt for the list of undeclared symbols.^M
> ./harfbuzz-unused.txt:1: warning: 46 unused declarations.They should
> be added to harfbuzz-sections.txt in the appropriate place.^M
> ^M
> ERROR: Error in gtkdoc helper script:^M
> ^M
> ERROR: ['/usr/bin/gtkdoc-mkhtml',
> '--path=/var/tmp/portage/media-libs/harfbuzz-2.8.2-r1/work/harfbuzz-2.8.2/docs:/var/tmp/portage/media-libs/harfbuzz-2.8.2-r1/work/harfbuzz-2.8.2-abi_x86_64.amd64/docs',
> 'harfbuzz', '../harfbuzz-docs.xml'] \failed with status 6^M
> ^M
> --- stderr ---^M
> ^M


As this is a problem with generating documentation, have you tried the
following as a workaround?:

USE=-doc emerge -1a media-libs/harfbuzz

Gentoo bug 704550[1], makes me wonder if it's worth a try trying to
update gtk-doc before harfbuzz, or is it already updated?

[1] https://bugs.gentoo.org/show_bug.cgi?id=704550

-- 
Nuno Silva

[gentoo-user] Re: [OT] tar exclude syntax tip

[nunojsilva]

On 2021-05-05, tastytea wrote:

> On 2021-05-05 09:33-0400 "Walter Dnes"  wrote:
>
>>   tar version
>> 
>> 
>> tar (GNU tar) 1.34
>> Copyright (C) 2021 Free Software Foundation, Inc.
>> License GPLv3+: GNU GPL version 3 or later
>> . This is free software: you are
>> free to change and redistribute it. There is NO WARRANTY, to the
>> extent permitted by law.
>> 
>> Written by John Gilmore and Jay Fenlason.
>> 
>> 
>>   I'm passing on this solution to help others avoid my frustration and
>> wasted time.  If you've done "RTFM" on tar, you'll find out that "TFM"
>> is broken or out-of-date or whatever, re: "--exclude=PATTERN".  I'm
>> fighting the urge to turn this into a rant.  Here's my situation...
>> 
>>   I either log in as root or "su -" and then "cd /home".  I want to
>> tar up /home/waltdnes, and transfer it to another machine.  While I'm
>> at it, I want to exlude directory /home/waltdnes/.cache/ and all *.xz
>> files in directory /home/waltdnes/pm/  The "--exclude=" never worked.
>>  After much hair pulling, I was ready to give up on the exclude, and
>> simply transfer all the unnecessary garbage.
>> 
>>   Then "I asked Mr. Google".  It seems that I wasn't the only person
>> running into problems.  After some searching, I finally found a syntax
>> that works...
>> 
>> 
>> #!/bin/bash
>> export GZIP=-9
>> tar cvzf wd.tgz --exclude ".cache/*" --exclude "pm/*.xz" waltdnes
>> 
>> 
>> Notes...
>> 
>> 1) This is obviously not in line with the man page.  Specifically,
>> "--exclude" is followed by one space, not an equals sign.
>> 
>> 2) ***THERE MUST BE EXACTLY ONE SPACE BETWEEN EACH WORD***
>> 
>> 3) All directories and/or files to exclude must be listed as relative
>> paths to the directory being tarred, i.e. last parameter on the
>> command line.
>> 
>> 4) I don't know the maximum line-length, which would limit the number
>> of --exclude entries.  In those cases, I wonder if
>> "--exclude-from=FILE" works as "--exclude-from FILE".
>> 
>
> This works fine here with “tar (GNU tar) 1.34”:
>
> $ mkdir -p a/b
> $ touch a/file a/b/file
> $ touch a/file.xz a/b/file.xz
> $ tree a
> a
> ├── b
> │   ├── file
> │   └── file.xz
> ├── file
> └── file.xz
>
> 1 directory, 4 files
> $ tar -cvzf test.tar.gz --exclude="a/file" --exclude="a/b/*.xz" a
> a/
> a/file.xz
> a/b/
> a/b/file
> $ tar -tf test.tar.gz
> a/
> a/file.xz
> a/b/
> a/b/file
>
> You can find out the maximum length of the command-line with 
> `getconf ARG_MAX`.


But does it work with a space instead of = as well? According to the
online manual page, it should work both ways.

-- 
Nuno Silva

[gentoo-user] Re: [OT] Saving an image as black and white

[nunojsilva]

On 2021-03-01, Wols Lists wrote:

> On 01/03/21 12:11, (Nuno Silva) wrote:
>> On 2021-03-01, Wols Lists wrote:
>> 
>>> I've got a bunch of scans, let's assume they're text documents. And
>>> they're rather big ... I want to email them.
>>>
>>> How on earth do I convert them to TRUE b documents? At the moment they
>>> are jpegs that weigh in at 3MB, and I guess they're using about 5 bytes
>>> to store all the colour, luminance, whatever, per pixel. But actually,
>>> there's only ONE BIT of information there - whether that pixel is black
>>> or white.
>>>
>>> I'm using imagemagick, but so far all my attempts to strip out the
>>> surplus information have resulted in INcreasing the file size ???
>>>
>>> So basically, how do I save an image as "one bit per pixel" like you'd
>>> think you'd send to a B printer?
>>>
>>> Even at 300dpi, I make that 300*300/8 ~= 10KB/in^2 or 800KB of
>>> uncompressed info for a page of A4, not 3MB.
>>>
>>> Cheers,
>>> Wol
>> 
>> Somebody else might have a better suggestion, or perhaps a better
>> understanding of the JPEG format and of what needs to be tuned, but, for
>> example:
>> 
>> convert origin.jpg -threshold 70% -monochrome result.jpg
>> 
>> (And adjust the "-threshold percent" if needed. It might be that you
>> don't need thresholding at all, but if you do, it apparently must go
>> before "-monochrome".)
>> 
>> (Depending on the receiving end, you could also explore other
>> formats. Here, if the scanned document can be stored in monochrome, I
>> usually use djvu.)
>> 
> Thanks but no, I've already tried that. It makes matters worse!
>
> I've messed about with the scanner, so it is now creating 800KB images,
> but I don't want to rescan everything I've done.
>
> The problem is that it is clearly saving the images as greyscale, not as
> black And when I search for help, what I want is swamped by all
> the false positives for greyscale.
>
> Oh - and for Nuno - sorry tesseract is no use, they are NOT text. That's
> why I used the word "assume" - to make it clear that I want a
> 1-bit/pixel palette, not a 5-byte/pixel greyscale.
>
> Cheers,
> Wol

Sorry, my bad - I was checking the file sizes, but I didn't notice the
larger one was the new, "monochrome" version. More coffee needed, it
seems.

-- 
Nuno Silva

[gentoo-user] Re: [OT] Saving an image as black and white

[nunojsilva]

On 2021-03-01, Wols Lists wrote:

> I've got a bunch of scans, let's assume they're text documents. And
> they're rather big ... I want to email them.
>
> How on earth do I convert them to TRUE b documents? At the moment they
> are jpegs that weigh in at 3MB, and I guess they're using about 5 bytes
> to store all the colour, luminance, whatever, per pixel. But actually,
> there's only ONE BIT of information there - whether that pixel is black
> or white.
>
> I'm using imagemagick, but so far all my attempts to strip out the
> surplus information have resulted in INcreasing the file size ???
>
> So basically, how do I save an image as "one bit per pixel" like you'd
> think you'd send to a B printer?
>
> Even at 300dpi, I make that 300*300/8 ~= 10KB/in^2 or 800KB of
> uncompressed info for a page of A4, not 3MB.
>
> Cheers,
> Wol

Somebody else might have a better suggestion, or perhaps a better
understanding of the JPEG format and of what needs to be tuned, but, for
example:

convert origin.jpg -threshold 70% -monochrome result.jpg

(And adjust the "-threshold percent" if needed. It might be that you
don't need thresholding at all, but if you do, it apparently must go
before "-monochrome".)

(Depending on the receiving end, you could also explore other
formats. Here, if the scanned document can be stored in monochrome, I
usually use djvu.)

-- 
Nuno Silva

[gentoo-user] Re: No logging output to tty12

[nunojsilva]

On 2020-12-12, Walter Dnes wrote:

>   I used to get stuff going to tty12, e.g.when attaching a USB drive,
> etc.  My recent fresh install doesn't have this output.  What am I doing
> wrong?

Have you installed something for syslog? In my installs, I think it is
syslog-ng that does this. At least in /etc/syslog-ng/syslog-ng.conf, I
see the following:

destination console_all { file("/dev/tty12"); };

-- 
Nuno Silva

[gentoo-user] Re: error trying to mount Samsung: Galaxy android models (MTP)

[nunojsilva]

On 2020-11-22, edward m wrote:

> hi, i receive an error when trying to mount an android phone under
> gentoo. im  wondering if a fix, workaround exists or more likely its a bug.
> thanks in advance.

I think with MTP there can be issues owing to implementation details -
some phones will have trouble with some MTP tools.

If you can, please try sys-fs/jmtpfs.


>
>
> error:
> mtpfs AndroidDevice/
> Listing raw device(s)
> Device 0 (VID=04e8 and PID=6860) is a Samsung Galaxy models (MTP).
>Found 1 device(s):
>Samsung: Galaxy models (MTP) (04e8:6860) @ bus 2, dev 12
> Attempting to connect device
> Error 1: Get Storage information failed.
> Error 2: PTP Layer error 02fe: get_all_metadata_fast(): could not get
> proplist of all objects.
> Error 2: Error 02fe: PTP Data Expected
> Error 2: PTP Layer error 02fe: get_handles_recursively(): could not get
> object handles.
> Error 2: Error 02fe: PTP Data Expected
> Listing File Information on Device with name: Galaxy A10e
> LIBMTP_Get_Storage() failed:-1

-- 
Nuno Silva

[gentoo-user] Re: dhcpd versus fixed IP addresses

[nunojsilva]

On 2020-10-05, Walter Dnes wrote:

> On Sun, Oct 04, 2020 at 06:31:44PM +0100, Neil Bothwick wrote
>
>> However, I would wait until the new modem arrives as most do allow
>> you to turn off DHCP for the LAN and the information you have may
>> only apply to the WAN connection.
>
>   On their support board, the ISP's tech support told me...
>
> a) My computer or router needs to be configured to get the IP
>automatically by DHCP to get the connection up and running.

So the address your computer or router will get via the cable modem is a
WAN address?

-- 
Nuno Silva

[gentoo-user] Re: gnokii emerge failed

[nunojsilva]

On 2020-09-05, n952162 wrote:

> gnokii wanted to emerge 6 packages and failed on the last:
>
> /txm1 /var/tmp/portage/app-mobilephone/gnokii-0.6.31-r1 # less
> /var/tmp/portage/app-mobilephone/gnokii-0.6.31-r1/temp/autoconf.out//
> //* autoconf *//
> //* PWD:
> /var/tmp/portage/app-mobilephone/gnokii-0.6.31-r1/work/gnokii-0.6.31//
> //* autoconf --force//
> //
> //configure.ac:74: error: possibly undefined macro: AM_LANGINFO_CODESET//
> //  If this token and others are legitimate, please use
> m4_pattern_allow.//
> //  See the Autoconf documentation./
>
> It looks like this could be an ebuild error.  Any hints, how to proceed?
>
>

Given the macro name, it looks like this could be bug 685832. If it is
that one, it is worth a try to emerge version 0.6.31-r3.

-- 
Nuno Silva

[gentoo-user] Re: problems with slack and zoom: Was: ALSA wizard...

[nunojsilva]

On 2020-04-26, Jorge Almeida wrote:

> On Sun, Apr 26, 2020 at 8:28 PM Jack  wrote:
>>
>> On 2020.04.26 15:08, Jorge Almeida wrote:
>> [snip]
>
>> I run a mostly stable amd64 system with selected ~amd64 packages.  I
>> use both slack and zoom reasonably often, and have not had any install
>> related problems.  (I don't necessarily TRUST either of them, but my
>> use of them is for a project with nothing particularly confidential of
>> private.)  I'd say if you care enough to bother - start a separate
>> thread about why those two fail emerge for you.  I seem to have
>> recurring problems with audio and/or video, but I'm pretty sure that is
>> hardware and/or system related - not due directly to either slack or
>> zoom.
>>
> I'll see about pulseaudio first, and if discord doesn't work even so
> I'll try that. Meanwhile, the ebuilds may change, or even go stable.

Any chance apulse works?

I don't use slack, discord or zoom, but last time I used software that
required pulseaudio (skype), apulse was enough.

-- 
Nuno Silva

[gentoo-user] Re: sci-libs/spqr-1.2.3-r1 failed to emerge

[nunojsilva]

On 2020-04-09, gevisz wrote:

> ср, 8 апр. 2020 г. в 19:31, gevisz :
>>
>> ср, 8 апр. 2020 г. в 18:13, Michael Orlitzky :
>> >
>> > The other thing you should be aware of is that dev-lang/julia is, from a
>> > software engineering point of view, an absolute trainwreck (upstream,
>> > not the Gentoo maintainer's fault). Your experience so far is going to
>> > be typical, I fear.
>>
>> After I added all these files to the
>> /etc/portage/package.accept_keywords/julia file,
>> julia finally emerged. Thank you. However, I am affraid what will be
>> when I try to update my Gentoo system in a month or so.
>
> A small addition: I've just noticed that I've managed to emerge only
> julia-1.3.0;
> julia-1.4.0-r1 still fails to be emerged (at a compile phase).

What errors?

There is bug 716434[1] where julia fails because it tries to download
more tarballs (yes, during compilation). An ebuild change was made to
address this, so it might be worth syncing the tree and trying to emerge
again.

[1] https://bugs.gentoo.org/show_bug.cgi?id=716434

-- 
Nuno Silva

[gentoo-user] Re: sci-libs/spqr-1.2.3-r1 failed to emerge

[nunojsilva]

On 2020-04-08, gevisz wrote:

> I've tried to install julia-1.4.0 but one of its dependencies, namely,
> sci-libs/spqr-1.2.3-r1 failed to emerge with the following error message:
>
> * Failed Running automake !
>  *
>  * Include in your bugreport the contents of:
>  *
>  *   /var/tmp/portage/sci-libs/spqr-1.2.3-r1/temp/automake.out
[...]
> I tried to look into the automake.out file mentioned above but it even
> does not exists (or empty).



Could this be bug 586582? (But that one has an automake.out?)

https://bugs.gentoo.org/show_bug.cgi?id=586582

And see also: https://bugs.gentoo.org/show_bug.cgi?id=627094

>From a quick read, it looks like there is no fix for 1.2.3-r1, but that
the plan is to drop 1.2.3-r1 if 2.0.9 works.


-- 
Nuno Silva

[gentoo-user] Re: zoom?

[nunojsilva]

On 2020-04-01, William Kenworthy wrote:

> On 1/4/20 4:55 pm, Neil Bothwick wrote:
>> On Wed, 1 Apr 2020 09:12:46 +0800, William Kenworthy wrote:
>>
>>> [blocks B  ] media-libs/mesa[-libglvnd(-)] 
>>> ("media-libs/mesa[-libglvnd(-)]" is blocking media-libs/libglvnd-1.3.1)
>> It looks like you need to emerge mesa with USE="libglvnd".
>
> no, there is something else at play - I have tried adding it to
> package.accept_keywords, package.provided etc. but they fail.  If I use
> --nodeps it wants to overwrite a number of mesa files so I am thinking
> its not actually needed.

Any chance it is the libglvnd USE flag that is masked?

-- 
Nuno Silva

[gentoo-user] Re: swaps mounted randomly [not out of the woods yet]

[nunojsilva]

On 2020-03-19, n952162 wrote:

> On 2020-03-19 09:36, n952162 wrote:
>> On 2020-03-19 09:33, Neil Bothwick wrote:
>>> On Thu, 19 Mar 2020 08:17:58 +0100, n952162 wrote:
>>>
 A couple of years back, I bought these drives to install RAID on them,
 but gave up on that.  Now, I've decided to do "manual" RAID, but I'm
 wondering if the fact that the two drives have the same UUID is causing
 whoever it is who sets up /dev/disk (I'm still trying to find that
 culprit) is croaking on two different devices with the same UUID.
>>> udev creates /dev
>>>
 Where is the UUID determined?  I'd presumed that it was derived from
 some characteristics of the drive, determined by the device controller,
 but now I'm wondering if my initial RAID configuration set some
 drive-internal variable to be identical?

 And, how does one /*reset*/ it?
>>> tune2fs -U [UUID] /dev/sdX
>>>
>>> UUID can be either a string in the standard format or the word random.
>>>
>>>
>>
>> Cool!  I missed that about the "random" keyword.
>>
>>
> I changed the UUID of all the partitions of the second drive and now all
> my devices are linked to in /dev/disk/by-uuid. I still have
> no/dev/disk/by-label, though.  Also, my swap file on a mounted drive
> wasn't mounted, which was my original problem  ;-(

Any chance the "swap" service is what you need here?

/etc/conf.d/swap has examples for different setups - it seems you'd need
this service and the line rc_need="localmount"

-- 
Nuno Silva

[gentoo-user] Re: Loading Issue

[nunojsilva]

On 2020-02-19, Dr Rainer Woitok wrote:

> Roger,
>
> On Wednesday, 2020-02-19 09:51:20 +1300, you wrote:
>
>> ...
>> rc_interactive means I can press "I" or "i" during boot and 
>> INTERRUPT
>> the boot process; *otherwise* it will boot as per normal.
>> 
>> Is that not the behaviour you are seeking?
>
> Speaking not for the original poster but only for me:
>
> In a mathematical sense it surely is,  because the interactive interface
> you are confronted with allows you not to start any services you're dis-
> liking.  In a practical sense it surely requires more than patience, be-
> cause for each and every service available  the user will be prompted to
> enter a number:
>
>1. Start this sevice.
>2. Skip this service.
>3. Start all remaining services.
>4. Start a privileged shell.
>
> So instead of pressing the "1" key umpteen times (plus one time too oft-
> en, ARGH :-) it's probably really easier to boot from a live CD.
>
> And the privileged shell offered here  is also restricted  in that it is
> only usable for the  minority of 330 million people  using a US keyboard
> layout,  while the vast  majority of  7.4 billion people  NOT using a US
> keyboard layout will most probably not even manage  to enter their pass-
> word :-(

If the US layout is causing too much trouble, it might be worth it to
run "loadkeys" manually.

-- 
Nuno Silva

[gentoo-user] Re: Web browsers crash when trying to print.

[nunojsilva]

On 2020-02-15, Mick wrote:

> On Saturday, 15 February 2020 04:54:21 GMT Dale wrote:
>> Howdy,
>> 
>> I first discovered this with Seamonkey.  I then tested this with Firefox
>> and got the same results.  When I go to File and select Print, the print
>> dialog window pops up for just a second and then the web browser
>> crashes.  Both seem to use the same print software.  I tried a different
>> version of Seamonkey but it does the same.  I also tested a fresh
>> profile of Seamonkey as well, Firefox also.  I don't print from Firefox
>> much.  Also, if I select Print Preview from the menu, it opens normally
>> but as soon as I click print, crash.
>> 
>> I then tried a newer unstable version of cups just in case it would
>> help.  After that, I opened Kwrite and tried to print.  Its print dialog
>> opened and waited for me to hit print.  LOo did the same.  However, both
>> of those use a different software or at least they look very different
>> to print with. 
>> 
>> Usually going back a version or up a version fixes things like this. 
>> Given that this didn't work in this case, I'm not sure where to go.  Two
>> versions of Seamonkey and Firefox both crash.  A newer version of cups
>> and it still crashes.  I did a search on BGO and didn't find anything
>> except for a fixed version of Chrome which I don't have on here.  I
>> suspect it uses different software to print anyway.  So no help there. 
>> Forums had a thread that was from 2010.  It mentioned a USE flag which
>> cups doesn't even have anymore.  No solution on the forums.
>> 
>> One other thing that may or may not be related.  I did a emerge -e world
>> a week or so ago.  Before that, I could print fine.  The reason I did
>> that was because I switched to a new gcc and I just wanted to be sure
>> everything was stable.  I went from gcc-8 to gcc-9.  It may not have
>> been needed but it was cold here and I didn't mind the extra heat. 
>> Plus, it sometimes fixes other quirks I may not even see.  Here is the
>> info for Seamonkey, Firefox and cups.
[...]
>> 
>> I'm not sure if the printing is done within Seamonkey itself or if
>> Seamonkey and Firefox use some common external print software.  I'd
>> think the later since both behave the same way.  I'm just not sure.
>> 
>> Any ideas or thoughts??

Try getting a backtrace with gdb, and build involved packages with debug
symbols as needed to get a more detailed backtrace.

With the backtrace, you will at least have an idea of where it is
crashing.

If you need documentation on this, the following page is probably a good
place to start: https://wiki.gentoo.org/wiki/Debugging_with_GDB

-- 
Nuno Silva

[gentoo-user] Re: Bouncing messages from gentoo-user@lists.gentoo.org

[nunojsilva]

On 2020-01-13, james wrote:

> On 1/13/20 11:32 AM, gentoo-user+ow...@lists.gentoo.org wrote:
>>
>> Some messages to you could not be delivered. If you're seeing this
>> message it means things are back to normal, and it's merely for your
>> information.
>>
>> Here is the list of the bounced messages:
>> - 189231

How does one get to the message from that number? Is it possible to get
an URL to the archived copy at http://archives.gentoo.org/ using that
number?

> Anyone else getting these?

It might just be the DMARC policy thing again. 

Here's a thread from last October/November about a similar message:
https://marc.info/?t=15725373431

-- 
Nuno Silva

Re: [gentoo-user] Re: trying to upgrade some old, never upgraded image for an embedded system …

[nunojsilva]

On 2019-12-19, Thomas Schweikle wrote:

>>> > On 2019-12-18,  (Nuno Silva) < 
>>> > nunojsi...@ist.utl.pt> wrote:
>>> >
>>> > > The EAPI problem is in a package that is pulled as a dependency of
>>> > > portage.
>>> > >
>>> > > Unless there's a simple hack to solve this, you will need to use older
>>> > > ebuilds or split the update in several steps, using older versions of
>>> > > the portage tree. The following notes show a way of achieving this:
>>> > >
>>> > > https://wiki.gentoo.org/wiki/User:NeddySeagoon/HOWTO_Update_Old_Gentoo
[...]
> So I've tried now to upgrade in various ways:
> 1. the one given in https://anongit.gentoo.org/git/repo/sync/gentoo.git
>   But this fails as soon as I try to emerge git. python-exec is at version
> 2.4.6 now. Without any 2.0.1 packed, legal versions left. Same for all
> other dependencies. Not really a way to go ...

Looking at this again, the installed version of portage says (in the
output quoted in your initial post) that it supports EAPI 6, which is
also used by the python-exec-2.4.6 ebuild (not in the tree anymore, the
one using EAPI 7 is 2.4.6-r1). So you could give the 2.4.6 ebuild a try:

https://gitweb.gentoo.org/repo/gentoo.git/plain/dev-lang/python-exec/python-exec-2.4.6.ebuild?id=20664dd65ec565233f460b94efc0337249b84550

Hopefully this will allow you to upgrade portage, unless there are more
dependencies of portage in similar situations. If this ebuild is not
enough, any chance you have another machine where you could do the
date-based checkout and then copy the entire portage tree?

-- 
Nuno Silva

[gentoo-user] Re: trying to upgrade some old, never upgraded image for an embedded system …

[nunojsilva]

On 2019-12-18, Ilya Trukhanov wrote:

> On Sun, Dec 15, 2019 at 07:01:15PM +0100, Thomas Schweikle wrote:
>> The current version of portage supports EAPI '6'. You must upgrade to a
>> newer version of portage before EAPI masked packages can be installed.
>
> The message is pretty clear. You need to upgrade portage first. Try
> `emerge -1 portage`. This should take care of the EAPI message.

The EAPI problem is in a package that is pulled as a dependency of
portage.

Unless there's a simple hack to solve this, you will need to use older
ebuilds or split the update in several steps, using older versions of
the portage tree. The following notes show a way of achieving this:

https://wiki.gentoo.org/wiki/User:NeddySeagoon/HOWTO_Update_Old_Gentoo

-- 
Nuno Silva

[gentoo-user] Re: Network config problem

[nunojsilva]

On 2019-11-30, Peter Humphrey wrote:

> I want to start playing with IPv6 (thanks, Ralph S.) but first I need
> to clean up something wrong with my existing IPv4.
>
> # /etc/init.d/net.eth0 restart
[...]
>  * Bringing down interface eth0
> RTNETLINK answers: No such file or directory
> Error talking to the kernel
>  * Bringing up interface eth0
[...]

Which version of netifrc do you have? Could this be bug 642774[1]? It
depends on the request to stabilize netifrc-0.6.1, so if you have an
older version, you could try that one.

>From the commit itself[2], it looks like netifrc is checking for L2TP in
a way which produces this output when L2TP isn't supported.

[1] https://bugs.gentoo.org/show_bug.cgi?id=642774
[2] 
https://gitweb.gentoo.org/proj/netifrc.git/commit/?id=1e14262524d65918ed6d1d13f2abd87b2f11425b

-- 
Nuno Silva

[gentoo-user] Re: Bounced messages

[nunojsilva]

On 2019-10-31, Mick wrote:

> On Thursday, 31 October 2019 16:06:15 GMT Dale wrote:
>> Mick wrote:
>
>> > I'm getting the same.  On the same message(s).  On user@ only.
>> 
>> I sent for and got the help thingy and tried to get it to send me the
>> missed messages.  So far, I've yet to get anything including a error
>> that I did it wrong.  Have you tried to retrieve the missed messages and
>> if so, did it work?  Unless it takes a while to process, it seems it
>> isn't working. 
>> 
>> Do we need to alert someone to this?  I saw in the message source that
>> there is a owner email addy or we could file a bug??
>
> This is not a problem for me.  I can read the original bounced message either 
> via responses provided to it, or if not (fully) quoted in the response, at 
> the 
> Mailing List archives - e.g. for message 188380:
>
> https://archives.gentoo.org/gentoo-user/message/
> 805899603d74dd747515c1d5e63f5ebe
>
> Incidentally, I think most frequently the bounced messages were sent from 
> Mr.Grimes.

I suppose it might be Yahoo's DMARC policy, see, for example:

https://wiki.list.org/DEV/DMARC

Alan Grimes' e-mail address seems to be from Verizon, which is, if I
understand correctly, Yahoo Mail.

If that is the case, I guess Gmail is not marking these messages as
SPAM, but rejecting them as a result of the DMARC policy, and the
message you get is from the mailing list software at gentoo.

-- 
Nuno Silva

[gentoo-user] Re: sneak peek at next update.

[nunojsilva]

On 2019-10-12, Neil Bothwick wrote:

> On Sat, 12 Oct 2019 02:14:51 -0400, Alan Grimes wrote:
>
>> right now the next update is looking **BAD**:
>> 
>> 
>> 
>> tortoise ~ # emerge --update --verbose portage --backtrack=30 
>> --verbose-conflicts --pretend
>
> What are you emerging here, @system or @world?

portage

Alan, what if you try updating gentoolkit and portage at the same time?

>> These are the packages that would be merged, in order:
>> 
>> Calculating dependencies... done!
>> [ebuild U?? ] sys-apps/portage-2.3.76::gentoo [2.3.65::gentoo] 
>> USE="(ipc) native-extensions rsync-verify xattr -build -doc -epydoc 
>> -gentoo-dev (-selinux)" PYTHON_TARGETS="python2_7 python3_6 python3_7 
>> -pypy -python3_5" 1,010 KiB
>> [blocks B?? ] > ("
> Unmerge gentoolkit, then install the later version after portage if
> updated.

I don't think it's necessary to unmerge, just to update. I might be
wrong, but I believe the reason why portage isn't solving this block by
itself is because it was told to update portage only.

-- 
Nuno Silva

[gentoo-user] Re: Emerging glib-2.60.6 failed

[nunojsilva]

On 2019-09-08, Hartmut Figge wrote:

> my latest update 'emerge -q -uDN @world' failed at glib-2.60.6.
> Continuing with --skipfirst succeeded. Today, after a new 'emerge
> --sync', 'emerge -q -uDN @system' still failed with the same error.
>
> My Gentoo is mostly stable. The relevant info can be found at
> http://www.triffids.de/pub/gentoo/glib-2.60.6/


http://www.triffids.de/pub/gentoo/glib-2.60.6/build.log.txt gives me
HTTP 403 (Forbidden).

-- 
Nuno Silva

[gentoo-user] Re: switch from gnome/systemd to xfce/openrc borked my system - I GIVE UP

[nunojsilva]

On 2019-08-20, Raffaele Belardi wrote:

> Raffaele Belardi wrote:
[...]
> - I grub-loaded a different kernel, one built for systemd. It stops in
> the exact same place as the openrc-built one.

What are the kernel command lines for both kernels?

-- 
Nuno Silva

[gentoo-user] Re: escape from i3lock

[nunojsilva]

On 2019-07-12, Ian Zimmerman wrote:

> On 2019-07-11 21:28, Nuno Silva wrote:
>
>> vlock -n -a
>
> Does vlock work from an XWindow session?  Or would I have to use it on
> top of whatever I do to lock the XWindow session - xscreensaver/i3lock
> etc?

It does work from inside X11 here. I can, for example, run it inside a
terminal emulator or through the window manager.

(You will probably need to add your user to the "vlock" group.)


> (I browsed to the vlock README page on github but it doesn't answer this
> question.)

-- 
Nuno Silva

[gentoo-user] Re: escape from i3lock

[nunojsilva]

On 2019-07-10, François-Xavier CARTON wrote:

> On 7/10/19 7:03 PM, Ian Zimmerman wrote:
>> Here is my next "low information" question, haha.
>>
>> I use i3lock which is like Xscreensaver but much much simpler; it plays
>> no movies or games, just blanks the screen with a configured color or
>> image.  To unlock it you have to type your password.
>>
>> It bothers me that even when i3lock has locked the X session, I can
>> still switch to other Linux virtual consoles with Alt-Control-F ,
>> without typing the password.  It so happens that on one of the other
>> virtual consoles there is often an interactive root shell :-P
>>
>> So, is it possible to prevent virtual console switching while the X
>> screen is locked, but still allow it at other times?  Looks like
>> something the locker program would have to do, not the X server; but
>> again I don't know much about this stuff.
>>
>
> Not a direct answer to your question, but as a workaround you can use
> tmux sessions, and simply detach them and logout when you lock your
> computer.
>
> Also, if this is just a shell to start the X server, you can launch it
> as "startx & bg; disown" and then logout.
>

Another workaround: If you can't find a better solution, give vlock a
try:

vlock -n -a

-- 
Nuno Silva

[gentoo-user] Re: fdm fails during ./configure: libssl not found

[nunojsilva]

On 2019-06-24, Bryant Morrow wrote:

> ---
>
> fdm-1.9 consistently fails to build, with configure citing a lack of
> libssl as the reason. I have openssl installed, and
> /usr/lib64/libssl.so (symlink to /usr/lib64/libssl.so.1.1) exists.
> I've looked for similar problems online, with no luck. It originally
> seemed to be related to a broken version of openssl being unmasked
> along with the use of ~amd64 in ACCEPT_KEYWORDS, but resolving that
> made no difference. It seems to me that fdm's config scripts simply
> can't find libssl, but they appear to be looking in all of the correct
> places. I had it installed and working in the past, but it was removed
> due to a stupid (on my part) accident with emerge --depclean and now
> fails to reinstall.
>
> Output of emerge -pqv 'net-mail/fdm-1.9::gentoo':
> http://dpaste.com/2B1YBPZ //emerge -pqv
>
> Output of emerge --info 'net-mail/fdm-1.9::gentoo':
> http://dpaste.com/0D1PK0E //emerge --info
>
> Contents of build.log: http://dpaste.com/1CFZF87 //build.log
>
> Contents of portage environment file: http://dpaste.com/2XJD461
> //environment

This looks like bug 677484 - the problem would indeed be the way this
config script tries to detect libssl.

I don't use fdm, but from what I read in the bug report, I suppose you
could try downgrading openssl to 1.0.*, or change the fdm ebuild to
apply the patch from comment 9.

https://bugs.gentoo.org/show_bug.cgi?id=677484

-- 
Nuno Silva

[gentoo-user] Re: x11-libs/guile-gtk-2.1-r3 fails to compile

[nunojsilva]

On 2019-05-06, Jack wrote:

> On 2019.05.06 13:26, Stefan Schmiedl wrote:
[...]
>> I got curious, so I tried to emerge guile-gtk, too.
>> 
>> After the ebuild failed, the output got dumped to the terminal
>> and showed lots of magenta warnings and cyan notes
>> and somewhere in there three red blobs appear:
>> 
>> In file included from /usr/include/guile/2.2/libguile.h:31,
>>  from  
>> /var/tmp/portage/x11-libs/guile-gtk-2.1-r3/work/guile-gtk-2.1/gtk-glue.c:4:
>> /var/tmp/portage/x11-libs/guile-gtk-2.1-r3/work/guile-gtk-2.1/gtk-glue.c:  
>> In function ‘sgtk_gtk_input_add_full’:
>> /var/tmp/portage/x11-libs/guile-gtk-2.1-r3/work/guile-gtk-2.1/gtk-glue.c:3579:63:
>>   
>> error: ‘scm_tc16_fport’ undeclared (first use in this function); did  
>> you mean ‘scm_t_fport’?
>>SCM_ASSERT ((SCM_NIMP (p_source) && SCM_TYP16 (p_source) ==  
>> scm_tc16_fport && SCM_OPPORTP (p_source)), p_source, SCM_ARG1,  
>> s_gtk_input_add_full);
>> 
>> ^~
>> 
>> namely: "error", "scm_tc16_fport" and the accompanying wiggles right  
>> below.
>> 
>> So, as Jack correctly claimed, the problem appears in gtk-glue.c.
>> 
>> While I don't have a clue how to fix what is really wrong here,
>> I just wanted to point out that the colorful output of gcc is very  
>> helpful,
>> if you get to see it.
> Dale,
>
> I'll offer a wild guess - what version of guile do you have installed?   
> The latest stable is 2.0.14-r3, but perhaps (noting this version of  
> guild-gtk is ~) it requires a newer version - perhaps 2.2.3 or 2.2.4?   
> Actually, I see from the snippet of output that it looks like you do  
> have guile 2.2.   Oddly, I do see scm_tc16_fport defined in  
> .../libguile/fports.h and libguil.h includes "#include  
> "libguile/fports.h" although that is in guile 2.0 that I have  
> installed.  I wonder if libguile.h in 2.2 omits that further include?
>
> Jack

That error looks similar to the one reported in bug 641864[0]. According
to the bug report, a workaround is to downgrade to guile 2.0.

[0] https://bugs.gentoo.org/show_bug.cgi?id=641864

-- 
Nuno Silva

[gentoo-user] Re: Libreoffice and copying web pages

[nunojsilva]

On 2019-04-30, Dale wrote:

> Howdy,
>
> As some know, I got a printer.  Now I'm trying to get some info and
> print it using LOo for the most part.  This is the way I do this.  I go
> to a web page, sites will vary, and I highlight what I want and copy it
> to the clipboard.  I then go to LOo and paste it as HTML, since that is
> what it is.  At that point, LOo fetches things like pics and such to
> place on the document.  It takes a little time and I avoid copying
> videos since it can't print a video.  For the most part, this works
> great.  It takes a minute or so to fetch the pics and such and
> everything looks fine.  I remove anything I don't want such as ads and
> such.  Basically, it looks like the web page but I can edit it to make
> fonts larger etc.  However, sometimes it doesn't work.
>
> When it doesn't work, this is usually what happens.  I paste the
> document and LOo gets very slow or doesn't respond at all.  Sometimes it
> gets to the point I have to literally kill the thing to get it to stop. 
> On the occasions that it is just slow, scrolling up or down to make
> edits is very slow.  I'm talking I move the mouse wheel and a minute or
> two later it scrolls up a couple lines.  In a document that is several
> pages long, it takes forever to scroll to the top after pasting it in. 
> In older versions of LOo it didn't have this problem.  It would copy and
> paste and even while fetching the pics, it would respond very well and
> was pretty fast.  The pics sometimes would show up as empty boxes tho
> until they were fetched.  I would wait until that was done before saving
> the docs.  About a year or so ago, it started this very slow to
> downright won't respond thing. 
>
> Is there a LOo guru that has a better idea on how to correct this or how
> to do this differently so that it responds even while fetching pics and
> such?  Maybe there is a setting somewhere I can change to make it work
> better. 
>
> Things I've considered but didn't like the results of.  Selecting what I
> want to print and printing it as a pdf file or just printing directly
> without having a copy to edit.  For obvious reasons, that doesn't work
> well because I sometimes change fonts and sizes for these old eyes. 
> Doing this in LOo makes things better if I can get it working right. 
> I've also tried the print friendly version when available but that
> removes some things I do want to have, pics mostly.


Here are two ideas that won't solve the problem, but might be useful as
workarounds:

You could try OpenOffice (only available as openoffice-bin?) and see if
it works better in that situation.

Depending on the web browser you're using (Seamonkey?), you might have
an "Inspector" (context menu, "Inspect Element"), which can be used to
remove undesired elements. It works when the page is saved (elements
which were removed using the inspector are not present in the saved
version), hopefully it will also work with copy-paste (but I have never
tried to do that).

Unfortunately, depending on the web page, the inspector itself can
sometimes be slow or even freeze.

If you're using Seamonkey as a web browser, here's another idea I didn't
try but which might help: ctrl+E will open the current web page for
editing in Seamonkey's Composer.

-- 
Nuno Silva

[gentoo-user] Re: Quad UART PCIe Adapter (Oxford-Chipset) seems (not?) to work? Check?

[nunojsilva]

On 2019-03-24, tu...@posteo.de wrote:

> Hi,
>
> In my PC there is a quad uart PCIe-adapter with Oxford Chipset.
> With lspci it is listed as:
> 04:00.0 Serial controller: Oxford Semiconductor Ltd OX16PCI954 (Quad 16950 
> UART) function 0 (Uart)
> and 
> 04:00.1 Bridge: Oxford Semiconductor Ltd OX16PCI954 (Quad 16950 UART) 
> function 1 (8bit bus)
>
> I have a DAUP9-cable, which connects the serial port to the FBUS/MBUS
> of my NOKIA 3310 (2001).
>
> I cannot speak to my NOKI handy via gammu/gnokii: Timeout
>
> To minimize the number of variables in this equitation I would like to
> ensure, that at least the UART adapter is working.
>
> The modules loaded by now are:
[...]
> Do I miss a driver?
> Is there any kind of check or test (without using another serial
> thingy, which I do not have), to test the UART adapter is working as
> expected?

The output of "lspci -k" is probably useful to check whether the device
was picked up by some driver. It will also show drivers which are not
modules.

Does it show any driver name for that UART controller?

-- 
Nuno Silva

[gentoo-user] Re: Ssh problem : half-solved

[nunojsilva]

On 2019-03-11, Philip Webb wrote:

> 190311 Neil Bothwick wrote:
>> Have you run ssh with -v
>> to see what configuration options it is reading from where.
>> Bear in mind that ssh stops at the first matching host definition,
>> so if you have a "host *" in your config, it must be last.
>
> This is what I get :
>
>   522: ~> ssh -v 
[...]

What is ?

Are you using the same hostname or address that is present in the "Host"
line you added?

>   Unable to negotiate with  port 22: no matching key exchange
> method found. Their offer:
> diffie-hellman-group-exchange-sha1,diffie-hellman-group1-sha1
>
> Is that any help ?

-- 
Nuno Silva

[gentoo-user] Re: mac-fdisk with different block size

[nunojsilva]

On 2019-01-13, Andrew Udvare wrote:

>> On Jan 13, 2019, at 09:24, (Nuno Silva)  wrote:
>> 
>> On 2019-01-13, Andrew Udvare wrote:
>> 
 On 2019-01-13, at 07:49, (Nuno Silva)  wrote:
 
 I am trying to create an Apple partition map with a block size of 4096
 bytes, but I can't find an option to change the block size in mac-fdisk,
 which defaults to 512 bytes.
 
 Does anybody know of a utility that can create and modify such partition
 maps under Gentoo?
>>> 
>>> 
>>> Quick look and it seems that for mac-fdisk the 512 size is hard-coded:
>>> 
>>> https://github.com/glaubitz/mac-fdisk-debian/blob/bda743065fa2c75a83fec60166bc2e317059ef7a/io.h#L32
>> 
>> 
>> That code appears to be version 0.4 from early 1997. Under Gentoo, the
>> README file installed at /usr/share/doc/mac-fdisk-0.1_p18/ mentions 0.4
>> as well.
>> 
>> According to a changelog at apple.com[1], variable block size support
>> was added after that, and should be present in version 0.5.
>
> On actual macOS latest, there's a -b option for block size.

GNU parted does not seem to have an option to *change* the block size,
but it will use the drive's block size when creating a new Apple
partition table.

-- 
Nuno Silva

[gentoo-user] Re: mac-fdisk with different block size

[nunojsilva]

On 2019-01-13, Andrew Udvare wrote:

>> On 2019-01-13, at 07:49, (Nuno Silva)  wrote:
>> 
>> I am trying to create an Apple partition map with a block size of 4096
>> bytes, but I can't find an option to change the block size in mac-fdisk,
>> which defaults to 512 bytes.
>> 
>> Does anybody know of a utility that can create and modify such partition
>> maps under Gentoo?
>
>
> Quick look and it seems that for mac-fdisk the 512 size is hard-coded:
>
> https://github.com/glaubitz/mac-fdisk-debian/blob/bda743065fa2c75a83fec60166bc2e317059ef7a/io.h#L32


That code appears to be version 0.4 from early 1997. Under Gentoo, the
README file installed at /usr/share/doc/mac-fdisk-0.1_p18/ mentions 0.4
as well.

According to a changelog at apple.com[1], variable block size support
was added after that, and should be present in version 0.5.

[1] https://opensource.apple.com/source/pdisk/pdisk-9/HISTORY.auto.html

If there is no other utility that can do this, I'll have a look at the
newer source code from apple.com...

-- 
Nuno Silva

[gentoo-user] mac-fdisk with different block size

[nunojsilva]

I am trying to create an Apple partition map with a block size of 4096
bytes, but I can't find an option to change the block size in mac-fdisk,
which defaults to 512 bytes.

Does anybody know of a utility that can create and modify such partition
maps under Gentoo?

-- 
Nuno Silva

[gentoo-user] Re: SystemRescueCD with nonm

[nunojsilva]

On 2019-01-01, Mick wrote:

> Which package do I need to emerge for isohybrid?

sys-boot/syslinux

-- 
Nuno Silva

[gentoo-user] Re: Is it that hard to add a package, or am I doing wrong?

[nunojsilva]

On 2018-12-20, YUE Daian wrote:

> On 2018-12-20 03:50, Nils Freydank  wrote:
>> Hi Danny,
>>
>> first I want to thank you for submitting your ebuild, and I'm quite sorry to 
>> see another contributor who doesn't get responses for a long while. This is 
>> no 
>> evil intention, just a lack of manpower and the lack of someone maintaining
>> your "new" package. (This was what jstein meant with his response[1]).
>>
> I do understand the situation of lacking manpower, also I realized made
> some mistakes in my ebuild file, so you do not have to apologize. :-)
>
>> Additionally bugzilla is seen as too impractical to use for new packages 
>> that 
>> many don't get much attention there, only on github.com.
>>
> Well the Gentoo Wiki https://wiki.gentoo.org/wiki/Submitting_ebuilds
> suggested that new ebuilds should be submitted via Bugzilla.
>
> Could you please tell me if it is still the recommended way?
> If not, IMHO it is better to change Wiki as well to prevent further
> misunderstanding.

I would like to ask again for a clarification about this. Last month, I
asked if there was some rule against using bugzilla, but there were no
replies:

https://marc.info/?l=gentoo-user=154318918422492=2

I do understand lack of manpower can affect new package requests. But
there are also bug reports with patches that have had zero feedback so
far. Of course these will also be affected by a manpower shortage, but
should be easier to handle than new package requests?

-- 
Nuno Silva

[gentoo-user] Re: trouble with my normal emerge --update @world

[nunojsilva]

On 2018-12-02, allan gottlieb wrote:

> On one of my stable amd64 systems, I just ran
>
>  emerge --update --changed-use --with-bdeps=n --deep @world
>
> and received a list of 65 packages to merge (many gnome).
>
> The update completed and I was told
>
> !!! existing preserved libs:
 package: dev-libs/libcdio-2.0.0-r1
>  *  - /usr/lib64/libcdio.so.16
>  *  - /usr/lib64/libcdio.so.16.0.0
>  *  used by /usr/bin/libcdio-paranoia (dev-libs/libcdio-paranoia-0.93_p1)
>  *  used by /usr/lib64/libcdio_cdda.so.2.0.0 
> (dev-libs/libcdio-paranoia-0.93_p1)
>  *  used by /usr/lib64/libcdio_paranoia.so.2.0.0 
> (dev-libs/libcdio-paranoia-0.93_p1)
>  *  used by /usr/libexec/gvfsd-cdda (gnome-base/gvfs-1.32.2)
> Use emerge @preserved-rebuild to rebuild packages using these libraries
>
> When I ran emerge --pretend @preserved-rebuild I was told to run
> merge dev-qt/qtcore-5.11.1 which is ~amd64, which seems wrong.
> The full output is below.
> Can someone please explain what is happening.
>
> thanks,
> allan
>
> E6430 ~ # emerge --pretend @preserved-rebuild
> These are the packages that would be merged, in reverse order:
>
> Calculating dependencies... done!
> [ebuild   R] gnome-base/gvfs-1.32.2 
> [ebuild   R]  dev-libs/libcdio-paranoia-0.93_p1 
> [nomerge   ] gnome-base/gvfs-1.32.2 
> [nomerge   ]  sys-fs/udisks-2.7.4-r1 
> [nomerge   ]   sys-libs/libblockdev-2.14-r1 
> [nomerge   ]dev-libs/volume_key-0.3.11 
> [nomerge   ] app-crypt/gpgme-1.11.1 
> [nomerge   ]  dev-qt/qttest-5.9.6 
> [ebuild UD~]   dev-qt/qtcore-5.9.6-r1 [5.11.1-r1]
>
> !!! Multiple package instances within a single package slot have been pulled
> !!! into the dependency graph, resulting in a slot conflict:
>
> dev-qt/qtcore:5
>
>   (dev-qt/qtcore-5.11.1-r1:5/5.11::gentoo, installed) pulled in by
> ~dev-qt/qtcore-5.11.1 required by 
> (dev-qt/qtconcurrent-5.11.1:5/5.11::gentoo, installed)
> ^  ^^ 
>  
> (and 9 more with the same problem)
>
>   (dev-qt/qtcore-5.9.6-r1:5/5.9::gentoo, ebuild scheduled for merge) pulled 
> in by
> ~dev-qt/qtcore-5.9.6 required by (dev-qt/qttest-5.9.6:5/5.9::gentoo, 
> installed)
> ^  ^
[...]

I would try upgrading qttest first (it is the one pulling
qtcore-5.9.6{,-r1}, according to the output). Do you have any qttest
5.11.* version which is keyworded amd64 (and not ~amd64)?

Here and on https://packages.gentoo.org/packages/dev-qt/qttest,
qttest-5.11.1 is amd64. Maybe the tree was in an inconsistent state when
you last synced it?

If you have qttest 5.11.1 in ~amd64 and you need a very quick fix, try
adding it to package.accept_keywords and upgrading it. Otherwise, you
might prefer to sync the portage tree again first.

-- 
Nuno Silva

[gentoo-user] Re: net-nntp/inn - This package is masked and could be removed soon!

[nunojsilva]

On 2018-11-25, R0b0t1 wrote:

> On Sun, Nov 25, 2018 at 1:28 PM Grant Taylor
>  wrote:
>>
>> Hi,
>>
>> I happily use net-nntp/inn on my server and was surprised to find that
>> it is now masked and apparently up for removal.  It looks like
>> maintenance has dropped off on the package.
>>
>> I've never maintained a portage overlay or otherwise contributed to
>> Gentoo (save for mailing lists).  As such I don't know what I can do to
>> help.
>>
>> I did skim the Proxy Maintainers page [1] and don't know that I'm ready
>> to tackle that much responsibility.  Is there something else that I can
>> do to help avoid the removal of the net-nntp/inn package?  Possibly at
>> least keep it around as a masked package?
>>
>> Does anyone have any recommendations before blindly diving head first
>> into something I'll regret by assuming responsibility that I'm not sure
>> I'm ready for?
>>
>> Thanks for any pointers in advance.
>>
>
> It depends why it is up for removal. Fix that issue and submit a pull
> requests via GitHub or via email to gentoo-dev. If using gentoo-dev
> there is the possibility that it will never be allowed through the
> filter, so perhaps ask about it on IRC as well.

Why not bugzilla? Is there some new rule suggesting that bugzilla
shouldn't be used?

You might have just forgotten to mention it, that's okay -- I'm just
asking because bugzilla bugs 601032, 660966 and 663432 have made me
wonder if there is something going on, and I want to be sure I didn't
miss anything.

Especially 601032, which is about to turn 1 year old.

> In the rare chance that the package is just being removed because it's
> old, making gentoo-dev aware that you use it should be enough.
> Otherwise bump the version.

-- 
Nuno Silva

[gentoo-user] Re: copy text file to clipboard on startup

[nunojsilva]

On 2018-11-22, the...@sys-concept.com wrote:

> On 11/22/2018 10:02 AM, Andrew Savchenko wrote:
>> On Thu, 22 Nov 2018 09:35:42 -0700 the...@sys-concept.com wrote:
>>> I have a simple text file (with few lines in it) and using XFCE.
>>>
>>> How do I copy text from that file to "clipboard" so that user can past
>>> it with "ctrl-v"
>>> I would like to that text to be in a clipboard after XFCE started.
>> 
>> Use x11-misc/xclip:
>>   xclip -in filename_with_paste
>> 
>> Add this script to you XFCE autostart. This can be done either by:
>> 
>> 1) GUI: Settings -> Session and Startup -> Application Autostart
>> https://docs.xfce.org/xfce/xfce4-session/preferences#application_autostart
>> 
>> 2) Custom run hook:
>> Edit ~/.config/xfce4/xinitrc properly (call xclip, then
>> default xfce4 xinitrc)
>> https://unix.stackexchange.com/a/267238
>> 
>> Best regards,
>> Andrew Savchenko
>
> Hmm... I tried it from the command line and restarting the XFCE; nothing
> in the clipboard, empty. Nothing to paste.
>
> xclip -in test.txt

xclip defaults to the X11 primary selection. To use the "clipboard"
selection, try

xclip -selection clipboard -in test.txt

To make sure this command works, you can run it in a terminal emulator
and then check if ctrl+V pastes what you want in another program.  But
for the automation part, you must put the command somewhere else (such
as the autostart feature Andrew mentioned).

-- 
Nuno Silva

[gentoo-user] Re: Firefox and Thunderbird compile issue

[nunojsilva]

On 2018-11-18, Daniel Frey wrote:
[...]
> It's really unfortunate that on massive builds distcc is not an option.
> Maybe I should consider -bin instead.

If you do happen to have a more powerful machine running Gentoo on a
compatible architecture[1], you could also try Gentoo binary packages
(that is, not -bin, but binary packages made from www-client/firefox and
mail-client/thunderbird on that machine, either automatically with
FEATURES=buildpkg, or manually using quickpkg or emerge --buildpkg).

[1] I guess it should also be possible to cross-compile, but that will
require additional steps, with which I am not acquainted.

Do you know Seamonkey? If you like it and it works for everything you
use Firefox and Thunderbird with, another idea to consider would be
replacing Firefox and Thunderbird with Seamonkey Navigator and Seamonkey
Mail  That way, it would be "just" one massive build instead of
two.

-- 
Nuno Silva

[gentoo-user] Re: Firefox and Thunderbird compile issue

[nunojsilva]

On 2018-11-18, Daniel Frey wrote:

> On 11/18/18 02:47, Alarig Le Lay wrote:
>> Hi Daniel,
>> 
>> Didi you tried to remove the temporary directory (inside /var/tmp) and
>> re-emerge id again? It looks like an incorrectly decompressed archive.
>> 
>
> I just tried this, to no avail.
>
> I'm trying to rebuild the installed packages I have under dev-python/*
> but I'm not sure it will help.

Do you, by any chance, have distcc enabled?:

https://bugs.gentoo.org/show_bug.cgi?id=646062
https://forums.gentoo.org/viewtopic-p-8275494.html

-- 
Nuno Silva

[gentoo-user] Re: PSI+ compile error

[nunojsilva]

On 2018-09-09, Klaus Ethgen wrote:

> Hi,
>
> I try to compile psi+ as I depend on OTR.

You might be also interested in the following bug report
https://bugs.gentoo.org/show_bug.cgi?id=651248

-- 
Nuno Silva

[gentoo-user] Re: PSI+ compile error

[nunojsilva]

On 2018-09-09, Klaus Ethgen wrote:

> I created a ebuild patch that fixes the issue. Unfortunatelly I do not
> know where to report that error in gentoo.
>
> There is no reportbug binary and I do not know a mail address where to
> post that to.
>
> Does anybody have an idea?

Bugzilla at https://bugs.gentoo.org/

-- 
Nuno Silva

[gentoo-user] Re: libreoffice 6

[nunojsilva]

On 2018-09-04, james wrote:

> I recently upgraded to LO 6.0.6.2; the upgrade went fine.
>
>
> When saving docs, I had it default to ".doc" as the most
> common format and I mostly use libreoffice for doz based
> folks.
>
>
> Now the save option does not list out all of those choices to save
> a file as, like ::pdf .doc  .docx, etc etc. all I see now is
> 3 types of 'odg'
>
> : All formats
> odg drawing
> odg drawing template
> fiat xml odg dreawing

I don't use LibreOffice myself (openoffice-bin here), but those look
like native formats for LibreOffice Draw.

Any chance you opened LibreOffice Draw by mistake? (Or that they changed
the UI so that what you used to do to open a blank text document now
opens a "draw document"?)

-- 
Nuno Silva

[gentoo-user] Re: Xorg on really old PC

[nunojsilva]

On 2018-08-30, François-Xavier CARTON wrote:

> Hi,
>
> I'm installing Gentoo on a really old PC (with a Pentium III and a
> i815 graphic card). I've installed a minimal Gentoo system, and I'm
> trying to get Xorg working.
>
> I've searched on the Gentoo wiki, and found that I should use an old
> version of mesa (<8.0) to get OpenGL support [1]. So I have masked
> recent versions of mesa and some Xorg packages in order to install the
> last version of mesa that is below 8.0. My package.mask is:
>>media-libs/mesa-7.10.3
>>=app-eselect/eselect-opengl-1.2.6
>>x11-proto/glproto-1.4.15
>>x11-base/xorg-server-1.12.4-r5
> I have no xorg.conf.

Perhaps try an older linux kernel, from the same time as the versions of
X11 and mesa that you are trying to run.

Running a system like this is going to be harder now, though, because
the X11 headers were reorganized, and packages which use X11 now depend
on the newer headers, which will most likely pull in the new
xorg-server.

> When I run startx, the screen goes black and the Xorg server
> segfaults, according to the log file [2]. Nothing works after that,
> including the keyboard, so I cannot go back to a linux console. Maybe
> it is a configuration issue, or maybe I'm using a buggy version. Has
> anyone any suggestion on what configuration and versions I should use?
>
> Thanks,
> François-Xavier Carton
>
> [1] https://wiki.gentoo.org/index.php?title=Intel=599748=599742
> [2] Xorg.0.log: http://sprunge.us/ZdWJNH
>
>

-- 
Nuno Silva

[gentoo-user] Re: python targets problem

[nunojsilva]

On 2018-08-25, Philip Webb wrote:

> For a long time, several pkgs have been refusing to install,
> apparently due to a conflict re python targets :
>
> root:595 ~> emerge -pv certifi file pyblake2 meson setuptools

What if you add dev-python/pygments to this list? This might be a stupid
guess, though... I'm too low on coffee to parse that emerge output.

(Also, do not forget --oneshot/-1, unless you really want to add those
to the world file.)

-- 
Nuno Silva

[gentoo-user] Re: Update circle

[nunojsilva]

On 2018-08-22, Zoltán Kócsi wrote:

> I have a Gentoo machine, which has not been updated for a while. Quite
> a long while, actually.
>
> Now I needed to install a package, and I did so. It went up, no
> problem. Except that it relies on a newer version of openrc than
> what's on the system. From here it's all downhill:
>
> emerge openrc:
> - Fails, as the current portage is EAPI 5 (whatever that is), thus all
>   EAPI 6 packages are masked. Must upgrade portage.
>
> emerge portage:
> - Fails, as it needs tar-1.30 and the installed tar is 1.27. Must update
>   tar.
>
> emerge tar:
> - Fails, as tar-1.30 needs EAPI 6. Must upgrade portage.
>
> It seems that I'm kind of stuck. Wiping the disk and rebuilding the
> system from scratch is absolutely not an option, the existing (and
> running) system must be updated somehow.
>
> I would appreciate any advice, I'm in complete darkness about the
> internal workings of portage (would be happy to find some detailed docs
> about that, by the way, preferably in PDF).
>
> Thanks in advance,
>
> Zoltan

First, and before anything else: it can be done. You will have to deal
with issues that are perhaps a bit more complex than the usual conflicts
and issues that show up once in a while even on a stable system which is
updated frequently. But it's definitely not impossible.

The biggest hurdle in your case is that you need to update tar. Do what
was already mentioned elsewhere in the thread, fetch the older EAPI 5
ebuild, put it in the local overlay and let portage grab that.

If you're lucky, that might be the only problem you'll have. If you're
not so lucky, you will have a couple packages that were since removed
from the tree, and which must be either uninstalled or added to a local
overlay and adjusted as required. (Because they might depend on older
versions of other packages which do not exist anymore.)

You will probably find some actual bugs too, for example, some packages
may fail because their dependencies are not correct.

When was the last time this system was fully updated/upgraded? (Not the
last time you synced the tree, but the last time you did
emerge -DuN world or the like.) This might be useful to figure out which
issues are you going to find.

Some recent issues were a change in the C++ ABI and the build system
change in certain X11-related packages that pulls python3 as a build
dependency.

-- 
Nuno Silva

[gentoo-user] Re: Running 3rd party apps intended for Ubuntu/RedHat

[nunojsilva]

On 2018-08-21, Grant Edwards wrote:

> There are a handful of 3rd party, closed-source apps that I run on my
> Gentoo systems.  Often they're available for RedHat or Ubuntu,
> sometimes for "generic" Linux.
>
> The apps for "generic" Linux usually run without too much trouble,
> since they tend to include most of the libraries they need bundled
> with the package or linked statically with the executable.
>
> Apps packaged for RedHat or Ubuntu tend to rely on the host for far
> more libraries (e.g. Qt or Gtk and underlying X11 stuff).  Sometimes I
> have to copy some libraries from a RedHat or Ubuntu system and set
> LD_LIBRARY_PATH to point to that set of "private" libraries to get
> these apps running.  Manually figuring out which libraries are
> required is a time-consuming and error-prone process.  One of the ones
> I use regularly is going to stop working one of these days because it
> depends on qtwebkit-4.8, which has been EOL'ed on Gentoo.  [So I'll
> have to grab one more library from an Ubuntu system.]
>
> I've been thinking about trying to automate this by installing the app
> on an Ubuntu or RedHat system and then running a bash script that uses
> ldd et alia to find and bundle up the set of required library files.
> (How deep to recurse in the tree of library dependencies will be a big
> question.)
>
> If I understand what containers are (never used them), it occurs to me
> that if I bundle up everything all the way down to libc and libgcc, I
> might as well be using a container, right?
>
> Is this a good use case for containers, or is there some other way to
> do this?

This is something I would like to do as well (grabbing all the required
libraries for a given binary). Unfortunately, so far I didn't have
enough time to look into it. But here's a link from my bookmarks, which
might (or not...) be useful to you:

http://www.ucc.asn.au/~dagobah/~dagobah/things/make-static.html

-- 
Nuno Silva

[gentoo-user] Re: Is this a portage bug?

[nunojsilva]

On 2018-08-15, Peter Humphrey wrote:

> Hello list,
>
> While trying to get a USB printer recognised on an Atom box, I found this,
> immediately after emerge --sync && eix-update:
>
> (atom) peak / # eix -c kyocera
> [N] net-print/kyocera-1x2x-mfp-driver (--): Printer descriptions (PPDs) and 
> filters for Kyocera 1x2x MFP
> [N] net-print/kyocera-mita-ppds ((~)8.4-r1{tbz2}): PPD description files for 
> (some) Kyocera Mita Printers
> Found 2 matches
>
> (atom) peak / # emerge -pv net-print/kyocera-1x2x-mfp-driver 
> net-print/kyocera-mita-ppds
>
> These are the packages that would be merged, in order:
>
> Calculating dependencies  ... done!
>
> !!! All ebuilds that could satisfy "net-print/kyocera-1x2x-mfp-driver" have 
> been masked.
> !!! One of the following masked packages is required to complete your request:
> - net-print/kyocera-1x2x-mfp-driver-1.1203-r1::gentoo (masked by: missing 
> keyword)
>
> (atom) peak / # grep kyocera /etc/portage/package.keywords
> net-print/kyocera-1x2x-mfp-driver
> net-print/kyocera-mita-ppds
>
> I wonder whether the "1x2x" is confusing portage.

There is no confusion. The package isn't available for your
architecture/keyword.

Portage finds it, and says it is masked by "missing keyword".

"(--)" in the eix output is what eix -c shows when there's no version
available for your system.

You could unmask it - you can run

emerge -pv --autounmask net-print/kyocera-1x2x-mfp-driver

to see what files need to be changed and how (this won't make any
changes by itself).

But keep in mind, from portage(5) ("man 5 portage"):

«Additional Note: If you encounter the -* KEYWORD, this indicates
that the package is known to be broken on all systems which are not
otherwise listed in KEYWORDS.  For example, a binary only package
which is built for x86 will look like:

games-fps/quake3-demo-1.11.ebuild:KEYWORDS="-* x86"»


-- 
Nuno Silva

[gentoo-user] emerge: custom formatting for USE flags in -pv/-av output

[nunojsilva]

Hello,

I see there is a way to customize colors in the output of emerge, but
is there a way to customize the other formatting attributes, short of
editing the source itself?

What I would like to do is to change the way USE flags are shown,
possibly

- Removing the bold attribute for disabled USE flags;

- Using bold for USE flags which are set but will not change;

- Using reverse video for USE flags that will change.

-- 
Nuno Silva

[gentoo-user] Re: [Bulk] Re: Back to openrc from systemd

[nunojsilva]

On 2013-03-22, Kevin Chadwick wrote:

  If you don't need user session monitoring for anything (which is what
  ConsoleKit and logind provides), nor interactive privilege granting
  (which is what polkit provides), then I believe you will have no  
 
 Thanks. Now *that* is what I call explaining something in a nutshell :-)
 
  problems switching OpenRC and systemd withouth needing to recompile
  anything. However, that means no upower and no udisks at least; GNOME
  cannot run without any of those. XFCE needs them if the udev USE flag
  is enabled, which is enabled by default in Gentoo desktop profiles,
  and in KDE the three of them are optional dependencies turned on by
  default. You can turn them of in XFCE and KDE, but you kinda lose
  functionality without them.  
 
 I do indeed remember having to fight the KDE use flags so that I could
 pull kdelibs without pulling the whole set of u* things someone decided
 that were required for a desktop environment (the fun thing being that I
 wasn't even using KDE as a DE).
 
 But I hope you don't mean the GNOME *libs* will be requiring
 logind/Consolekit/... in the near future? That would cause me some
 trouble, as I rely on evince a lot.

 A good overview though I don't agree with If you don't 'need'

 Did your desktop really fail to run at all?

I don't need any of this u* or other things for my desktop computer to
work. Maybe this is related to the fact that I don't run a desktop
environment, even if I use linux for desktop computing and run X.

 Why are dependencies suddenly getting a lot worse (ignoring konquerorFM
 without kde) when for so long dependencies were understood to be a big
 problem that must be fixed. It can only be bad design if a desktop does
 not work at all because  1% of the functionality is missing and may
 well have been replaced in every case above by alternative and in some
 cases superior (permissions) that may override others (sessions you
 don't use), choices of functionality.

Bad design, bad choices by developers, people who don't want to accept
binary distros aren't the only thing around, and people who don't have a
taste for simplicity.

 Is it really a freedesktop when almost all the rest are free-er?

freedesktop seems to be doing a good job at creating standards where
there were no standards, the problem is that (IMHO) they seem to be
coming up with bad standards to fill the gap (see clipboard/selection
handling), and sometimes it is actually good to have things done in
different ways.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: Back to openrc from systemd

[nunojsilva]

On 2013-03-22, Canek Peláez Valdés wrote:

 On Thu, Mar 21, 2013 at 7:45 PM, João Matos jaon...@gmail.com wrote:
 Hi list,

 do you know some guide to switch form systemd to openrc, or keep both? I
 googled and I didn't find.

 The motivation is that I'm studing many server stuff, and I'm tired of
 search for alternatives to systemd (that is really good). I also set up some
 servers, using openrc on them, so, sometimes I like to reproduce the
 configuration o my machine.

 If possible, I prefer to keep both. If not, I'll switch back to openrc.

 I've enable the openrc user flag, updated the system, I created a grub
 entrace, and everything seems to work pretty well on openrc, but I cant
 start the X. no screens found, but dbus, udev and consolekit are started
 without error.

 Everything is working with systemd.

 For server stuff, you should have no problem. If the machine where
 you want to have both systemd and OpenRC also works as a desktop
 workstation, right now that is not possible; there are several desktop
 packages that cannot decide at run time if they use systemd (actually,
 logind) or ConsoleKit (polkit being the most obvious).

Are these packages essential or the like? I don't think my desktop
systems have dependencies either on systemd or polkit/consolekit.

What is logind used for?

 So you'll need to remove the systemd USE flag, add the consolekit one,
 and recompile the necessary packages to get back to a systemd-less
 desktop. Be aware that ConsoleKit is basically dead; it has no
 upstream, no new features are being developed for it, and I don't
 think even basic security bugs are actively fixed. Everything that
 depended on  ConsoleKit has switched or is considering switching to
 logind, which right now is provided only by systemd (Canonical is
 working in an alternative implementation, and I believe some *BSD guys
 were also looking into the matter).

 When the logind alternative implementations are ready, maybe it will
 be possible to again boot to both OpenRC and systemd with the same
 binaries; right now is not possible.

 Regards.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: Back to openrc from systemd

[nunojsilva]

On 2013-03-22, Canek Peláez Valdés wrote:

 On Fri, Mar 22, 2013 at 4:09 AM, Nuno Silva nunojsi...@ist.utl.pt wrote:
 On 2013-03-22, Canek Peláez Valdés wrote:

 On Thu, Mar 21, 2013 at 7:45 PM, João Matos jaon...@gmail.com wrote:
 Hi list,

 do you know some guide to switch form systemd to openrc, or keep both? I
 googled and I didn't find.

 The motivation is that I'm studing many server stuff, and I'm tired of
 search for alternatives to systemd (that is really good). I also
 set up some
 servers, using openrc on them, so, sometimes I like to reproduce the
 configuration o my machine.

 If possible, I prefer to keep both. If not, I'll switch back to openrc.

 I've enable the openrc user flag, updated the system, I created a grub
 entrace, and everything seems to work pretty well on openrc, but I cant
 start the X. no screens found, but dbus, udev and consolekit are started
 without error.

 Everything is working with systemd.

 For server stuff, you should have no problem. If the machine where
 you want to have both systemd and OpenRC also works as a desktop
 workstation, right now that is not possible; there are several desktop
 packages that cannot decide at run time if they use systemd (actually,
 logind) or ConsoleKit (polkit being the most obvious).

 Are these packages essential or the like? I don't think my desktop
 systems have dependencies either on systemd or polkit/consolekit.

 If you don't need user session monitoring for anything (which is what
 ConsoleKit and logind provides), nor interactive privilege granting
 (which is what polkit provides), then I believe you will have no

Thanks. Now *that* is what I call explaining something in a nutshell :-)

 problems switching OpenRC and systemd withouth needing to recompile
 anything. However, that means no upower and no udisks at least; GNOME
 cannot run without any of those. XFCE needs them if the udev USE flag
 is enabled, which is enabled by default in Gentoo desktop profiles,
 and in KDE the three of them are optional dependencies turned on by
 default. You can turn them of in XFCE and KDE, but you kinda lose
 functionality without them.

I do indeed remember having to fight the KDE use flags so that I could
pull kdelibs without pulling the whole set of u* things someone decided
that were required for a desktop environment (the fun thing being that I
wasn't even using KDE as a DE).

But I hope you don't mean the GNOME *libs* will be requiring
logind/Consolekit/... in the near future? That would cause me some
trouble, as I rely on evince a lot.

 What is logind used for?

 User session monitoring, as ConsoleKit did, only better:

 http://www.freedesktop.org/wiki/Software/systemd/logind

 Regards.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: AMD64, Firefox = Java Plugin?

[nunojsilva]

On 2013-03-17, walt wrote:

 On 03/16/2013 06:15 AM, meino.cra...@gmx.de wrote:
 Hi,
 
  I am looking for a useable howto/tutorial which describes
  howto install the java plugin for the current fireox.

 I see your question is already answered, but I'll add that there are
 a zillion open security bugs for java and many are actively being
 exploited in the wild recently.

 If you must use java in Firefox (as I do) then make sure that your
 version of Firefox is the latest one (19-something) because it will
 ask your permission before running any java applets.  Firefox will
 allow you to override the warnings for selected websites so it won't
 become irritating at sites you trust.

 IIUC there are exploitable bugs in java7 that don't exist in java6,
 so I'm sticking to java6 for now, though 6 isn't perfect either.

Or use NoScript or some other extension that lets you block embedded
content by default, like some people have been doing for years now.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: I guess it is time to update udev from 171-r10 to 197-r8...

[nunojsilva]

On 2013-03-17, Tanstaafl wrote:

 On 2013-03-17 2:17 PM, Neil Bothwick n...@digimed.co.uk wrote:
 On Sun, 17 Mar 2013 13:46:39 -0400, Tanstaafl wrote:

 Also, should I manually fix the blockers:

 [blocks B  ] sys-apps/module-init-tools
 (sys-apps/module-init-tools is blocking sys-apps/kmod-12-r1)
 [blocks B  ] sys-apps/kmod (sys-apps/kmod is blocking
 sys-apps/module-init-tools-3.16-r2)

 by doing emerge -C module-init-tools  emerge kmod *before*
 upgrading udev?

 No, because that adds kmod to world. Just unmerge module-init-tools and
 then emerge world, letting portage install what it needs

 Ah, ok... but as for the rest... I should be able to safely upgrade
 udev, with a reasonable (I know there are no guarantees) expectation
 of everything 'just working' (ie, my lvm managed /usr partition
 shouldn't be an issue like it would have been earlier on in this
 process)?

From what I know (no LVM experience here), if you had it working with
171, it will work with a newer udev. There were no changes regarding how
stuff from /usr is used between 171 and the newer udevs.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: Gentoo speed comparison to other distros

[nunojsilva]

On 2013-03-15, Dale wrote:

 I didn't miss anything.  I get what some are saying.  The reason for my
 question is this.  Gentoo allows a person to customize the OS to the
 specific hardware it is being run on.  Redhat and other binary distros
 don't allow this, unless you compile your own packages which is no
 longer really a binary install. 

 So, if I install Redhat on my machine, would it be less efficient than
 my Gentoo install which is customized for my hardware?  Has someone else
 tested this and made it public? 

 If people can't get this, never mind. 

I have not tested this nor seen data on this, but I'd look for
comparisons on the efficiency and gains from gcc optimizations. These
would be what benefits source-based distros on a specific system
compared to binary distros, and a benchmark made with gcc will be
simpler and easier to deal with than an os-wide benchmark.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] udev rules and hibernation

[nunojsilva]

Hello,

Currently, I decided to rely on udev to give me predictable network
interface names. Strangely, the rules I am using do not seem to work
after resuming from hibernation.

For example, for one of my interfaces, I have

SUBSYSTEM==net, DRIVERS==?*, ATTR{address}==xx:xx:xx:xx:xx:xx,
NAME=lan0

(Where I obfuscated the MAC address.)

Now this rule *does* work, on a cold boot or if I rmmod and modprobe
after resuming, but, for some reason, it still comes up as ethX after
resuming. It seems the rule is not triggered at all when resuming.

Any ideas of what may be going on?

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: Acroread

[nunojsilva]

On 2013-02-17, Nilesh Govindrajan wrote:
 On Feb 17, 2013 3:04 PM, Yohan Pereira yohan.pere...@gmail.com wrote:

 On 17/02/13 at 02:44pm, Nilesh Govindrajan wrote:
  That's what I did. But I tend to switch browsers too often (chromium and
  Firefox).
  Firefox has a pdf.js addon, doesn't work reliably many times.

 If you use KDE try this. You can then use okular(among other things)
 in your browsers.

  www-plugins/kpartsplugin
  http://www.unix-ag.uni-kl.de/~fischer/kpartsplugin/
  Description: Plugin using KDE's KParts technology to embed file
  viewers into non-KDE browsers

 That sounds interesting. Will try it out. Thanks.

 But nobody replied if Adobe still supports acroread?

AFAIK there was no annoucement regarding end of life for acroread, so I
don't see any reason to expect otherwise.

Flash is a separate thing.

But keep in mind that Adobe Acrobat Reader is one of the worst, most
bloated and most heavy PDF viewers out there. The only thing it may be
worthy for is some kind of bleeding edge PDF feature libpoppler and the
like don't have yet.

Also, I actually had to try running it recently. I was trying to print a
document with annotations -- spoiler: it didn't work, not even with
acroread, closest I got was generating a postscript file using acroread
in the commandline after manually hacking the acroread settings to
enable annotation printing, and even then part of the annotations don't
show up or are covered, and there's no mapping between annotations and
their icons. But the interface was *really* slow, almost unusable. I
wonder why. I possibly overlooked something.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: udev-197: what to do -- S0LVED

[nunojsilva]

On 2013-02-15, Stefan G. Weichinger wrote:

 Am 28.01.2013 00:00, schrieb Allan Gottlieb:

 Thanks for all the suggestions.  I did the following, which worked.
 
 1. Built and installed kernel with CONFIG_DEVTMPFS=y
 
 2. Moved udev-postmount back to /etc/init.d (I had moved it to /tmp).
rc-update add udev-postmount default.
 3. Reboot with new kernel (udev unchanged).  Success.
 
 4. Changed NAME=eth0 to NAME=net0 in 70-persistent-net.rules and
eliminated clauses so have only (on one line)
  SUBSYSTEM==net, ACTION==add, ATTR{address}==00:1e:c9:48:f9:a0,
  NAME=net0
Corresponding changes to /etc/init.d /etc/runlevels/default
 
 5. Emerge update world to get new udev (just -1 udev has blocks)
 
 6. Change kernel configs as per udisks emerge output
 
 7. /usr/lib/udev already empty (due to make world?) so nothing to do
 
 8. Reboot with new kernel.  Success

 As I prepare/consider to upgrade a remote gentoo server later this
 evening I prefer to ask twice:

 the running kernel 3.5.7 does not have CONFIG_DEVTMPFS=y

 I built a new kernel (upgrading it to 3.6.11 btw) with CONFIG_DEVTMPFS=y
 and plan to reboot the server at first.

 After a hopefully correct reboot (can't access that server physically) I
 plan to upgrade udev ... (I won't enable the new networking naming, btw).

If you depend in the network device order in any way, and you used names
like the ones the kernel uses, you *have* to do something about the
network device naming.

For example, if you have eth0 and eth1 and you rely on eth0 being A and
eth1 B, you can't  do that anymore with plain udev, even if the rules
are still in place. eth0 may become B and eth1 A.

 Right now I upgrade lvm2 in advance as it doesn't pull in udev yet.

 Pls comment or correct my plans ;-)  thanks, Stefan




-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: Delayed update semantics

[nunojsilva]

On 2013-02-14, James wrote:

 So, my latest ideas is to sync up and then wait one week
 before acutally installing those new packages. This would
 allow the fodder that the good folks on this list catch,
 bitch about (um, I mean file bug reports) and fix, to 
 occur first; then I can complete the package update
 cautiously avoiding an emerge sync.

The fun thing is that this is supposed to be why we do have unstable
keywords. Some of the breakage which happened and hit the stable tree
with no timely news item was actually discussed for some time before in
the -devel list. You may want to follow that list too.

That sync up you talk about is effectively equivalent to having ~amd64
for amd64.

 But when you emerge sync if to do the updates immediately, they'll be
 the latest packages. If I do a emerge sync and wait
 7 days to begin updating the packages, I'll be delayed
 by one week, and have a one week  of buffered fixes for added
 problem filtering. But those fixes might not be available
 without a fresh emerge sync?


 When time permits I CAN CHOOSE to emerge sync and then immediately
 update the packages and parse through the issues mostly. Call
 this the stable-stable approach to gentoo updates.

 Does anyone see any problems or a better way to stay one-week-delayed ?

Packages with problems simply shouldn't hit stable. I'd rather just do
--sync and perform world upgrades, and I would just keep an eye on the
triggered updates, in order to spot any potentially problematic package
(like c++-based libraries).

 I'm increasingly managing more Gentoo systems, particularly embedded
 and server based gentoo systems and that is the source that compounds these
 time-sink-issues for me.  Maybe some external-integrated management approach
 such as CFengine is my answer?

 Your comments and thoughts are most welcome.

Your problem does not seem to be emerge --sync, but that you run
commands that cause existing upgrades to be applied on a system-wide
basis.

If there is a fix you need/want, re-emerge that specific package, and
let the others wait. Only the strictly needed dependencies should be
pulled this way. You're not forced to use -DuN world or even -DuN
every time you want to upgrade something.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: No server profile anymore???

[nunojsilva]

On 2013-02-10, Alan McKinnon wrote:

 On 10/02/2013 19:25, Jarry wrote:

 !!! Your current profile is deprecated and not supported anymore.
 !!! Use eselect profile to update your profile.
 !!! Please upgrade to the following profile if possible:
   default/linux/amd64/13.0
[...]
 So is server-profile not suported anymore??? I hope devs had
 good reason for this, but anyway a change like *this* should
 definitely be communicated with users in advance...
[...]
 [1] OK, the news items, or more specifically the lack of timeous news
 items in advance. This is the second occurrence in recent times where
 devs have had to do some back-pedalling, the first was udev with it's
 TMPDEVFS fiasco. I myself am getting a teeny bit pissed off with this
 now. I think a large collection of user should pen a nice polite letter
 to whomever deals with such things asking for more attention to be paid
 to QA matters like this.

+1

I have no doubts that devs have lots of work to do, but it's a rather
serious situation if the difference between unstable and stable land is
*not* used as an advantage when it comes to deal with situations like
this and udev's kernel requirements and network rules.

I guess a good rule of thumb would be: if a stabilization/profile change
or introduced error message will require users to change their settings
by hand, change their kernel config to match new requirements in order
to have an usable system or to treat some packages/flags in a different
way, this should not go forward until a news item has been prepared to
notify users about it.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: No server profile anymore???

[nunojsilva]

On 2013-02-10, Jarry wrote:

 Hi gentoo-users,

 today after syncing portage tree I tried to update my system,
 but I was greeted with message:

 !!! Your current profile is deprecated and not supported anymore.
 !!! Use eselect profile to update your profile.
 !!! Please upgrade to the following profile if possible:
   default/linux/amd64/13.0

 So I reverted to previous snapshot just to see what I had before
 (eselect profile list did not show any selected). Till today
 I had default/linux/amd64/10.0/server. To my big surprise,
 there is no server profile for default/linux/amd64/13.0.

 So is server-profile not suported anymore??? I hope devs had
 good reason for this, but anyway a change like *this* should
 definitely be communicated with users in advance...

 Right now I'm not sure if Gentoo is not supported for server
 at all or 13.0 is substitution for 10.0/server. But before
 there were pure 10.0 and 10.0/server profiles. So it seems
 to me server profile was simply removed without any substitution.

No, the only reason the server profile existed was to have the name
server on it. Debate on the -dev list quickly led to the conclusion
that there was no point in keeping a separate server profile (we
already have the base profiles which are quite minimalistic, so they
*are* what the server profile should be, anyway).

In fact, if you are going to ask they why did we have two different
profiles, well, that's how the discussion got to the point where the
decision was to ditch the /server profile.

From what I remember, differences were only a *couple* USE flags. So
just sitch to the base profile. 

Now this could have used a news item, yes.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: error: Cg/cg.h: No such file or directory

[nunojsilva]

On 2013-02-03, João Matos wrote:

 Dear list,

 I'm trying to build dolphin-emu from gamerlay, and I got the following
 error:


 error: Cg/cg.h: No such file or directory

 But the file actually exists:

 /opt/nvidia-cg-toolkit/include/Cg/cg.h

 Should it be a problem that I can solve myself?

 I thought so, than I've made a symlink for
 /opt/nvidia-cg-toolkit/include/Cg at /usr/include. The compilation took a
 little longer, but later, I've got the following:

It took a little longer because it compiled, in fact the error occured
so early that it failed without doing that much, I guess.


 /usr/lib/gcc/x86_64-pc-linux-gnu/4.6.3/../../../../x86_64-pc-linux-gnu/bin/ld:
 cannot find -lCg
 /usr/lib/gcc/x86_64-pc-linux-gnu/4.6.3/../../../../x86_64-pc-linux-gnu/bin/ld:
 cannot find -lCgGL

It is some library detail: when configuring the makefiles to build some
piece of source, the configure script needs to find the libraries and
set *two* things from these libraries:

- The include directories, to which it should append the
  /opt/nvidia-cg-toolkit/include path

- The library directories, to which it should append a directory where
  nvidia-cg-toolkit has its libs

This, at least for some packages, is done with pkg-config, see, for
example, if you have gtk+-2.x on your system:

   pkg-config --cflags gtk+-2.0

This generates the right flags to pass to gcc when you want to compile a
gtk+-2.x application. Likewise, --libs gives you the linker flags.

Now what happened: you *did* provide it the header file, which is enough
to compile the code part that depends on the nvidia library
(compilation, which is modular, you can just go for parts of the code
and information on other parts are on the header files), but you still
need the library itself when you are assembling the final binary, so
that the binary can be extended with information on where to get the
library (or with the library itself, in the case of static linking).

Although here --libs only gives me -l* for gtk, there is also a
companion -L flag that, like -I for includes, sets the directories where
the libraries can be found. I assue that, if -L is not given, that just
means gtk+ is in the default locations.


If you can figure out how to tell the configure script where to look for
the libs, that would be the easiest thing to do; next to that, you can
manually link the files using the right -L parameter or perhaps
appending it to LD_LIBRARY_PATH when compiling.


Perhaps try reinstalling nvidia-cg-toolkit after resyncing: some
searches tell me there was yet another compilation breakage that got
fixed recently: https://bugs.gentoo.org/show_bug.cgi?id=443546

But one person noted it does not work in amd64... I wonder if linking
lib64 to lib there would work.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: Binary chrome - is it safe in terms of dependencies?

[nunojsilva]

On 2013-01-31, Nilesh Govindrajan wrote:

 On Thu, Jan 31, 2013 at 7:05 AM, Mike Gilbert flop...@gentoo.org wrote:
 On Wed, Jan 30, 2013 at 1:09 PM, Yohan Pereira
 yohan.pere...@gmail.com wrote:
 On 30/01/13 at 11:09pm, Nilesh Govindrajan wrote:
 Since Gentoo updates libraries very quickly, I'm wondering if it is
 safe to use the binary version? Has anyone faced library breakages on
 this?

 Chromium is easily recompiled with new libraries and you don't have a
 broken browser, which won't really be the case with the binary
 version.

 I've used the binary version (google-chrome) for a while and never
 had any breakages. I guess if there's a library update that could
 potentially break google-chrome the gentoo devs would add a blocker so
 you wont be able to install the 2 at the same time.


 Or I can just bundle a copy of the necessary libraries, similar to
 what I have done for libudev.so.0.


 Sounds good. I guess I'll switch to binary chrome then.

Also, I suppose that, if there were library incompatibilities, the
package would never go stable, or would at least, like Yohan said, lead
to a block/version dependency.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: Binary chrome - is it safe in terms of dependencies?

[nunojsilva]

On 2013-01-31, Nilesh Govindrajan wrote:

On Jan 31, 2013 5:38 PM, Nuno Silva nunojsi...@ist.utl.pt wrote:

 Also, I suppose that, if there were library incompatibilities, the
 package would never go stable, or would at least, like Yohan said, lead
 to a block/version dependency.

 Well, many times you can't really anticipate everything.

 I had my libreoffice-bin pdf import broken for two months because some
 shared library had got upgraded against which it wasn't linked.

I guess that sometimes this kind of issues may be harder to spot, or
require harder fixes (see for example the current state of LISP where
some packages require ASDF 2 but the stable one is ASDF 1, AFAIK the
stabilization of ASDF 2 is pending because some eclasses have to be
changed), but even then I'd suppose this is what the unstable arches are
for.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/

[gentoo-user] Re: can not print pdf w/e-document viewer

[nunojsilva]

On 2013-01-24, Joseph wrote:

 On 01/24/13 11:25, Bruce Hill wrote:
On Thu, Jan 24, 2013 at 09:43:06AM -0700, Joseph wrote:
 I have a document letter size in landscape mode and I'm trying to
 print it with e-document viewer
 4-pages per side and it will not print.  Some documents prints OK
 but this one will not print it.

 Are there better programs in Linux for printing pdf files?

 pdfinfo Biol_321_2013_Lec_06_Biogeography_1_per.pdf
 Title: Microsoft PowerPoint - Biol 321 2013 Lec 06
 Biogeography.pptx
 Author: hproctor
 Creator:PScript5.dll Version 5.2.2
 Producer:   Acrobat Distiller 7.0.5 (Windows)
 CreationDate:   Wed Jan 23 17:43:54 2013
 ModDate:Wed Jan 23 17:43:54 2013
 Tagged: no
 Pages:  18
 Encrypted:  no
 Page size:  612 x 792 pts (letter)
 File size:  961318 bytes
 Optimized:  yes
 PDF version:1.4

lp still does a good job for me:

mingdao@workstation ~ $ lpstat -a
Officejet_Pro_8500_A910 accepting requests since Wed 23 Jan 2013
 02:52:15 PM CST
mingdao@workstation ~ $ lp -d Officejet_Pro_8500_A910 -o scaling=75
 HOW-TO/apcupsd.pdf
request id is Officejet_Pro_8500_A910-19 (1 file(s))

Bruce

 The document prints OK from windows but Linux drivers are not up to
 standard :-/

 I'm trying lpr but the following command does not print the pages I
 want, it prints all pages instead of 1-8

 lpr -o media=Letter -o landscape -o number-up=4 -o page-ranges=1-8 -o
 number-up-layout=btlr 

 I think -o number-up=4 can not be combine with: -o page-ranges=1-8

Try using something like pdfnup (app-text/pdfjam) to generate an n-up
version of the pdf before sending it to the printer. 

If everything else fails, try rewriting the pdf, either using pdftops
then ps2pdf, or by using ghostscript directly.

And I don't think you can compare the two things directly: in Windows,
IIRC, the applications print using GDI. lpr sends the PDF as is directly
to the CUPS server. If the PDF lacks builtin fonts, for example, those
won't appear even if your PDF viewer  can view them (think, fonts under
your home directory, a printing server in a different machine...).

Some PDF or PostScript features can hit ghostscript bugs or other
issues.

But, if you want my two cents, look at psnup and pdfnup. At least then
you can be sure that the 4-per-page part is done. lpr options are
quite simplistic; do also have a look at pdftk if you need, for example
to rotate PDF pages, or to concatenate PDFs without rewriting their
contents (keep the code as-is, unlike what would happen if you just fed
ghostscript several pdfs, where it would rewrite the PDF code).


If your issue is with a single PDF, the problem is likely some issue
between a PDF feature used by that PDF and the incarnation of
ghostscript you are using. Try pdftops and ps2pdf and see if the result
is printable.

-- 
Nuno Silva (aka njsg)
http://njsg.sdf-eu.org/