[gentoo-user] Re: oracle-jdk-bin-11

[Jörg Schaible]

Hi James,

On Mon, 23 Jul 2018 17:13:14 -0400 james wrote:

> On 07/21/18 11:10, Jörg Schaible wrote:
>> Hi,
>> 
>> is someone aware of the existence of an ebuild for Java 11 e.g.
>> somewhere in the overlays? Or has someone already crafted one and is
>> willing to share it?
>> 
>> Cheers,
>> Jörg
> 
> My response is::
> 
> https://github.com/Obsidian-StudiosInc/os-xtoo/#xtoo-overlay
> 
> I'm just learning/hacking java, so please provide me with robust
> feedback so I know if this offering is great, or needs improvement
> (what/where) and what it is missing, from your point of view. Coding in
> Java is easy; understanding the history, gotcha and devining Oracle's
> future intentions is  often not clear to me. Hence, feedback is greatly
> appreciated, on this list, privately or to a forum is all ok.
> Just help me navigate the java landscape. I'd really like to work with
> java on both gentoo and windows 7/10. I'm starting fresh with Java-11.

Thanks for the interesting pointer and I will certainly look at the ebuild for 
Java 11 to adjust it for my needs. 
However, the overlay is not for me. I use Gentoo as my development platform and 
as library developer I 
have a need for a wide range of JDKs:

$ java-config -L
The following VMs are available for generation-2:
1)  Blackdown JDK 1.4.2.03 [blackdown-jdk-1.4.2]
2)  IBM JDK 1.4.2.13_p9 [ibm-jdk-bin-1.4]
3)  IBM JDK 1.5.0.12_p5 [ibm-jdk-bin-1.5]
4)  IBM JDK 1.6.0.9_p2 [ibm-jdk-bin-1.6]
5)  IBM JDK 1.7.0.5 [ibm-jdk-bin-1.7]
6)  IcedTea JDK 6.1.13.9 [icedtea-bin-6]
7)  IcedTea JDK 7.2.6.11 [icedtea-bin-7]
8)  IcedTea JDK 3.8.0 [icedtea-bin-8]
9)  JamVM JDK 2.0.0 [jamvm]
10) Oracle JDK 1.7.0.80 [oracle-jdk-bin-1.7]
11) Oracle JDK 1.8.0.172 [oracle-jdk-bin-1.8]
*)  Oracle JDK 10.0.1 [oracle-jdk-bin-10]
13) Oracle JDK 9.0.4 [oracle-jdk-bin-9]
14) Sun JDK 1.5.0.22 [sun-jdk-1.5]
15) Sun JDK 1.6.0.45 [sun-jdk-1.6]

Therefore I am not interested in replacing java-config with something else nor 
can I live with with the 
supported JDKs from the overlay, especially since it tries to replace 
dev-java/* from the standard Gentoo 
tree.

Cheers,
Jörg

> 
> 
> hth,
> James

[gentoo-user] oracle-jdk-bin-11

[Jörg Schaible]

Hi,

is someone aware of the existence of an ebuild for Java 11 e.g. somewhere in 
the overlays? Or has someone 
already crafted one and is willing to share it?

Cheers,
Jörg

[gentoo-user] Re: Kernel 4.14.7 no longer switches to VT7

[Jörg Schaible]

Hi Alan,

Am Sat, 30 Dec 2017 02:18:12 +0200 schrieb Alan McKinnon:

[snip]

> It's probably a dodgy kernel point bersion, 4.14 is problematic.

[snip] 
 
> If you want to fix the bugs, then by all means soldier on. But if your
> intent is to have a working system that boots, probably drop using
> 4.14.x and go back to say 4.12.x ?

For my wife's box, I've made 4.12.12 the default, since the TV switch works 
with that kernel.

For my own one I've kept 4.14.8-r1 as default. It works for me apart from the 
manual VT switch. It's just 
annoying. We'll see what happens with the next version of the kernel.

Cheers,
Jörg

[gentoo-user] Re: Kernel 4.14.7 no longer switches to VT7

[Jörg Schaible]

Hello Kai,

Am Sat, 30 Dec 2017 00:48:09 +0100 schrieb Kai Krakow:

[snip]
 
> If I remember right (many months ago), I fixed it by changing one line
> in /etc/sddm.conf:
> 
> [X11]
> ServerArguments=-nolisten tcp -keeptty
>   
>This is where the magic happens
> 
> You may want to add or remove that parameter...

thanks for the suggestion, but for me it had unfortunately no influence on the 
behavior...

Cheers,
Jörg

[gentoo-user] Re: Howto unmask use flag

[Jörg Schaible]

Found it. I have to create my own package.use.mask file in /etc/portage/profile 
that reverts the setting.

Am Thu, 28 Dec 2017 02:10:36 + schrieb Jörg Schaible:

> Hi,
> 
> how can I unmask a use flag masked in profiles/base/package.use.mask? If
> I set the flag for the package in / etc/portage/package.use it is simply
> ignored.
> 
> Concrete example: My /etc/portage/package.use contains the line:
> 
>   dev-java/oracle-jdk-bin:9 doc gentoo-vm source
> 
> However the flag 'gentoo-vm' is always ignored because it is masked in
> profiles/base/package.use.mask with the line:
> 
>   dev-java/oracle-jdk-bin:9 gentoo-vm
> 
> What's the proper procedure to unmask the flag locally?
> 
> (Yes, I know it is experimental and yes I know what I am doing with it.
> I already use eselect for older EA builds of Java 9 with 1.9 slot and
> new EA builds of Java 10 - just not the "official" Java 9 from the
> tree.)
> 
> Cheers,
> Jörg

[gentoo-user] Howto unmask use flag

[Jörg Schaible]

Hi,

how can I unmask a use flag masked in profiles/base/package.use.mask? If I set 
the flag for the package in /
etc/portage/package.use it is simply ignored.

Concrete example: My /etc/portage/package.use contains the line:

  dev-java/oracle-jdk-bin:9 doc gentoo-vm source

However the flag 'gentoo-vm' is always ignored because it is masked in 
profiles/base/package.use.mask with 
the line:

  dev-java/oracle-jdk-bin:9 gentoo-vm

What's the proper procedure to unmask the flag locally?

(Yes, I know it is experimental and yes I know what I am doing with it. I 
already use eselect for older EA builds 
of Java 9 with 1.9 slot and new EA builds of Java 10 - just not the "official" 
Java 9 from the tree.)

Cheers,
Jörg
  

[gentoo-user] Re: Kernel 4.14.7 no longer switches to VT7

[Jörg Schaible]

Hi Raymond,

Am Sat, 23 Dec 2017 22:59:32 -0800 schrieb Raymond Jennings:

> That sounds like a possible issue with your X configuration.
>
> Did you double check /etc/conf.d/xdm and the like to make sure that your
> VT is indeed set to 7.

Content of /etc/conf.d/xdm
=== %< ==
CHECKVT=7
DISPLAYMANAGER="sddm"
rc_use="mysql"
=== %< ==

> Also double check your display manager configuration.
> 
> If your manual VT switch works fine I'd suspect a misbehaving display
> manager possibly being confused by bad configuration

Then, why does it work seamlessly when I boot with the old 4.12.12 kernel?

Since I have this behavior with two desktop machines, I thought others might 
haven been affected as well ...

Cheers,
Jörg

[gentoo-user] Re: How to resume 'emerge -e @world' after grub fails?

[Jörg Schaible]

Am Thu, 21 Dec 2017 13:00:47 +0100 schrieb Marc Joliet:

> Am Donnerstag, 21. Dezember 2017, 10:45:41 CET schrieb Jörg Schaible:
>> Hi,
>> 
>> Am Mon, 18 Dec 2017 11:07:08 -0500 schrieb John Blinka:
>> > On Mon, Dec 18, 2017 at 11:00 AM, Grant Edwards
>> > 
>> > <grant.b.edwa...@gmail.com> wrote:
>> >> How do I skip grub and continue?
>> > 
>> > emerge --skipfirst --resume
>> 
>> This is unfortunately really dangerous, because "emerge --resume" will
>> recalculate the order of the outstanding packages and you have no
>> guarantee that the first one will be the one that failed the last run.
>> In that case you skip an arbitrary package and you may increase your
>> problems.
>> 
>> You can use --skipfirst only if you have restarted emerge with --resume
>> only and you have ensured that it will really continue with the failing
>> package. You may abort the build then with CTRL-C and restart emerge
>> with both options.
> 
> That clashes with my understanding, so I looked it up, and it turns out
> I was right.  From emerge(1):
> 
>>  --skipfirst
>>  
>>   This option is only valid when used with --resume.  It
>>   removes the first package in the resume list.
>>   Dependencies are recalculated for remaining packages and
>>   any that have unsatisfied dependencies or are masked will
>>   be automatically dropped. Also see the related
>>   --keep-going option.
> 
> Note the "remaining dependencies" part.  Otherwise, what would be the
> point of --skipfirst if it were so unpredictable?

Well, that's the difference between theory and practice. I've been bitten more 
than once, but you may do as 
you want, it's your system ...

Cheers,
Jörg

[gentoo-user] Kernel 4.14.7 no longer switches to VT7

[Jörg Schaible]

Hi,

after the update and installation of gentoo-sources-4.14.7 my two machines no 
longer switch to SDDM on 
VT7, it stays on VT1. However, I can switch manually using CTRL-ALT-7 to SDDM 
and login as usual. If I boot 
with the last stable kernel 4.12.12 anything is back to normal and the login 
screen of SDDM appears directly 
while the rest of the modules is loaded in background.

Both machines have older Radeon chips (REDWOOD and CEDAR) and I managed to load 
also their firmware 
with the new kernel 4.14.7, but there's still no automatic switch to VT7 
anymore.

I found nothing obvious in /var/log/messages, dmesg or Xorg.0.log. What may 
cause this weird behavior?

Cheers,
Jörg

[gentoo-user] Re: How to resume 'emerge -e @world' after grub fails?

[Jörg Schaible]

Hi,

Am Mon, 18 Dec 2017 11:07:08 -0500 schrieb John Blinka:

> On Mon, Dec 18, 2017 at 11:00 AM, Grant Edwards
>  wrote:
>>
>> How do I skip grub and continue?
>>
>>
> emerge --skipfirst --resume

This is unfortunately really dangerous, because "emerge --resume" will 
recalculate the order of the 
outstanding packages and you have no guarantee that the first one will be the 
one that failed the last run. In 
that case you skip an arbitrary package and you may increase your problems.

You can use --skipfirst only if you have restarted emerge with --resume only 
and you have ensured that it will 
really continue with the failing package. You may abort the build then with 
CTRL-C and restart emerge with 
both options.

> I had to do that several times in my 17.0 upgrades.

Maybe more times than necessary ;-)

Cheers,
Jörg

[gentoo-user] Re: Re: Re: Re: SMTP on Kmail-4.14.10 does not work without kde-apps/kwalletd-16.04.3-r1

[Jörg Schaible]

Mick wrote:

> On Thursday 08 Jun 2017 16:56:21 Jörg Schaible wrote:
>> Mick wrote:
>> > On Thursday 08 Jun 2017 13:21:56 Jörg Schaible wrote:
>> >> > Yes, this seems to be the problem.  Starting Kmail does not launch
>> >> > kwalletd5 and as a consequence kmail starts asking for each email
>> >> > account password separately.
>> >> > 
>> >> > I guess until kmail:5 is installed I will have to start kwalletd5 by
>> >> > hand.
>> >> 
>> >> My situation is different, since I use kwallet-pam. That one will
>> >> start kwallet5 automatically and I am not asked by KMail for passwords
>> >> (after it now also uses kwallet5).
>> > 
>> > I'm puzzled:  I have kde-plasma/kwallet-pam-5.8.6 installed, but it
>> > will *not*
>> > start kwalletd5.  Bear in mind though, I do not run a full plasma
>> > desktop.
>> 
>> Do you run SDDM? Do you have those two lines in it?
>> 
>> -auth   optionalpam_kwallet5.so
>> -sessionoptionalpam_kwallet5.so auto_start
> 
> Ha!  Thanks for this hint!
> 
> I had these entries in my /etc/pam.d/sddm
> 
> -auth optionalpam_kwallet.so kdehome=.kde4
> -auth optionalpam_kwallet5.so
> -session  optionalpam_kwallet.so
> -session  optionalpam_kwallet5.so auto_start

It seems the only lines required are the kwallet5.so ones.

> but ... sddm has stopped working properly with enlightenment, which is my
> desktop of choice.  So I started using lightdm and forgot to add these
> entries - TBH I thought they were not needed because everything worked as
> it should until now without them.
> 
> So, I added the two lines you suggested and rebooted.  I checked that
> kwalletd5 is running:
> 
> $ ps axf | grep kwallet
>  4515 pts/1SN+0:00  \_ grep --color=auto
>  kwallet
>  4088 ?S  0:00 /usr/bin/kwalletd5 --pam-login 8 10

[snip]

> I think I'll give up on this.  It seems kwallet:4/5 is not working as
> intended
> on my set up.  I hope by the time kdepim has moved from :4 to :5 things
> will
> work as intended.  Until then I will keep the old kwallet:4 installed,
> because at least this works as always did.

First I thought you might have run into an incompatibility between sddm and 
lightdm ...

>> Does your system password match the password of your wallet?
> 
> No, all users have different kwallet and user login passwds.

... but that's it. kwallet-pam requires the same password for wallet and 
login, otherwise it cannot work. Simply adjust the kwallet password or ...

>> > Running Krusader:5 and trying to connect to a remote system starts
>> > kwalletd5 fine, but running kmail:4 it does not.
>> 
>> Does kmail:4 work after krusader:5 started kwallet5?
> 
> Yes, because Krusader:5 calls kwalletd5 and asks for its password.  Then
> this is available for all applications to use, including kmail:4.

... you will need something else to start kwallet5 then. KMail must be fixed 
first to start the new wallet daemon. Maybe you should open a bugzilla 
ticket. I am quite sure you will find an entry in the error log when KMail 
tries to access dbus for the old daemon.

Cheers,
Jörg

[gentoo-user] Re: Re: Re: SMTP on Kmail-4.14.10 does not work without kde-apps/kwalletd-16.04.3-r1

[Jörg Schaible]

Mick wrote:

> On Thursday 08 Jun 2017 13:21:56 Jörg Schaible wrote:
>> > Yes, this seems to be the problem.  Starting Kmail does not launch
>> > kwalletd5 and as a consequence kmail starts asking for each email
>> > account password separately.
>> >
>> > I guess until kmail:5 is installed I will have to start kwalletd5 by
>> > hand.
>> 
>> My situation is different, since I use kwallet-pam. That one will start
>> kwallet5 automatically and I am not asked by KMail for passwords (after
>> it now also uses kwallet5).
> 
> I'm puzzled:  I have kde-plasma/kwallet-pam-5.8.6 installed, but it will
> *not*
> start kwalletd5.  Bear in mind though, I do not run a full plasma desktop.


Do you run SDDM? Do you have those two lines in it?

-auth   optionalpam_kwallet5.so
-sessionoptionalpam_kwallet5.so auto_start

Does your system password match the password of your wallet?

> Running Krusader:5 and trying to connect to a remote system starts
> kwalletd5 fine, but running kmail:4 it does not.

Does kmail:4 work after krusader:5 started kwallet5?
 
>> >> Note, you should install knotify:4 if you want notifications about
>> >> failed mail delivery from KMail. That was removed for me in an
>> >> dependency clean-up, because I had no entry for it in world.
>> >
>> > 
>> >
>> > Hmm ... I thought kde-apps/knotify:4 was replaced with kde-
>> > frameworks/knotifyconfig:5?
>> 
>> I simply recognized following line in the error log:
>> 
>> akonadi_newmailnotifier_agent(6002)/kdeui (KNotification)
>> KNotification::slotReceivedIdError: Error while contacting notify daemon
>> "The name org.kde.knotify was not provided by any .service files"
>> 
>> After installing knotify:4 I suddenly got desktopn messages from KMail
>> again.

[snip]
 
> I think both kwallet:4 and knotify:4 should have been retained as
> dependencies
> until all kde:4 packages are removed from portage.  I can think why
> knotify should be in world, since it ought to be a dependency for kdepim-
> meta/kdepimlibs/kdepim-runtime, all of which are I have installed here.

For me kwallet5 actually replaced kwallet:4 with the latest update. And I am 
glad about it, because the two wallets started to diverge and I had no 
longer a kwalletmanager:4.

Cheers,
Jörg

[gentoo-user] Re: Re: SMTP on Kmail-4.14.10 does not work without kde-apps/kwalletd-16.04.3-r1

[Jörg Schaible]

Hi Mick,

Mick wrote:

> On Thursday 08 Jun 2017 02:04:44 Jörg Schaible wrote:
>> Mick wrote:
>> > On Tuesday 06 Jun 2017 16:35:40 you wrote:
>> >> Hi All,
>> >> 
>> >> I've updated a number of kde (plasma) packages, including kde-
>> >> frameworks/kwallet-5.34.0-r1.  A depclean action wanted to remove
>> >> kde-apps/kwalletd-16.04.3-r1 and I let it do its tha'ng.
>> >> 
>> >> Following a new login I discovered that *every* time I wanted to send
>> >> a
>> >> message I was being asked for the SMTP password.  For some users with
>> >> 6 or
>> >> more email accounts this soon became tiresome.  The pop up advises
>> >> that the password should be saved in kwallet and offers the choice to
>> >> save it in ...
>> >> the configuration file!  It also advises that although it will be
>> >> obfuscated
>> >> it will not be safe in the configuration file.  There are two buttons,
>> >> one to save the password in the configuration file and another to not
>> >> save it (at all).
>> >> 
>> >> So to retain what sanity I may have left, I had to re-install kde-
>> >> apps/kwalletd-16.04.3-r1, which appears to be able to manage SMTP
>> >> passwords without asking each time the user.
>> >> 
>> >> Have you noticed the same?  Is there a fix or workaround for this?
>> > 
>> > Just to clarify, there doesn't seem to be a problem with IMAP4
>> > passwords, only with SMTP.
>> > 
>> > When updating to kde-frameworks/kwallet-5.34.0-r1 I noticed this elog
>> > message, which implies I should no longer need the old kwalletd4:
>> > 
>> > "LOG: postinst
>> > Starting with 5.34.0-r1, kwallet is able to serve applications
>> > that still require old kwalletd4. After migration has finished,
>> > kde-apps/kwalletd can be removed."
>> > 
>> > So, do I have to wait for Kmail:4 to be updated to Kmail:5 before I
>> > give up on kwalletd4 despite the above message?
>> 
>> I had the same problem. But event if kwalletd is installed, it was no
>> longer started.
> 
> Yes, this seems to be the problem.  Starting Kmail does not launch
> kwalletd5 and as a consequence kmail starts asking for each email account
> password separately.
> 
> I guess until kmail:5 is installed I will have to start kwalletd5 by hand.

My situation is different, since I use kwallet-pam. That one will start 
kwallet5 automatically and I am not asked by KMail for passwords (after it 
now also uses kwallet5).

>> Note, you should install knotify:4 if you want notifications about failed
>> mail delivery from KMail. That was removed for me in an dependency
>> clean-up, because I had no entry for it in world.
> 
> Hmm ... I thought kde-apps/knotify:4 was replaced with kde-
> frameworks/knotifyconfig:5?

I simply recognized following line in the error log:

akonadi_newmailnotifier_agent(6002)/kdeui (KNotification) 
KNotification::slotReceivedIdError: Error while contacting notify daemon 
"The name org.kde.knotify was not provided by any .service files"

After installing knotify:4 I suddenly got desktopn messages from KMail 
again.

> A user complained that new messages no longer create a popup.

Seems to match the error log.

Cheers,
Jörg

[gentoo-user] Re: SMTP on Kmail-4.14.10 does not work without kde-apps/kwalletd-16.04.3-r1

[Jörg Schaible]

Mick wrote:

> On Tuesday 06 Jun 2017 16:35:40 you wrote:
>> Hi All,
>> 
>> I've updated a number of kde (plasma) packages, including kde-
>> frameworks/kwallet-5.34.0-r1.  A depclean action wanted to remove
>> kde-apps/kwalletd-16.04.3-r1 and I let it do its tha'ng.
>> 
>> Following a new login I discovered that *every* time I wanted to send a
>> message I was being asked for the SMTP password.  For some users with 6
>> or
>> more email accounts this soon became tiresome.  The pop up advises that
>> the password should be saved in kwallet and offers the choice to save it
>> in ...
>> the configuration file!  It also advises that although it will be
>> obfuscated
>> it will not be safe in the configuration file.  There are two buttons,
>> one to save the password in the configuration file and another to not
>> save it (at all).
>> 
>> So to retain what sanity I may have left, I had to re-install kde-
>> apps/kwalletd-16.04.3-r1, which appears to be able to manage SMTP
>> passwords without asking each time the user.
>> 
>> Have you noticed the same?  Is there a fix or workaround for this?
> 
> Just to clarify, there doesn't seem to be a problem with IMAP4 passwords,
> only with SMTP.
> 
> When updating to kde-frameworks/kwallet-5.34.0-r1 I noticed this elog
> message, which implies I should no longer need the old kwalletd4:
> 
> "LOG: postinst
> Starting with 5.34.0-r1, kwallet is able to serve applications
> that still require old kwalletd4. After migration has finished,
> kde-apps/kwalletd can be removed."
> 
> So, do I have to wait for Kmail:4 to be updated to Kmail:5 before I give
> up on kwalletd4 despite the above message?

I had the same problem. But event if kwalletd is installed, it was no longer 
started. After some time I remembered that I changed the password for my 
mail account. However, that was after the "automated conversion" of the 
wallet from kwallet 4 to 5. So, when I updated my credentials in KMail at 
that time, the old wallet was updated. However, the latest update of kwallet 
seems to take over the APIs of kwallet 4 and redirect them to kwallet 5 
implementation. But that wallet still kept my old password of my mail 
account that was valid at conversion time. All I had to do was to reenter my 
new password in the KMail setup for this mail account.

Note, you should install knotify:4 if you want notifications about failed 
mail delivery from KMail. That was removed for me in an dependency clean-up, 
because I had no entry for it in world.

Cheers,
Jörg

[gentoo-user] Re: Qt-4.8.7 bug

[Jörg Schaible]

Hello Kai,

Kai Krakow wrote:

> Am Mon, 22 May 2017 19:33:55 +0200
> schrieb Jörg Schaible <joerg.schai...@gmx.de>:
> 
>> Peter Humphrey wrote:

[snip]

>> > 
>> > I can only suggest you read bug report 618922 if you haven't
>> > already, including following its reference to bug 595618. It makes
>> > sense to me.
>> 
>> It does not for me. My packages were already compiled with gcc-5.4.0.
>> Those Buzilla issues only talk about (plasma/qt) packages compiled
>> with previous gcc-4.x which are supposed to be incompatible. All of
>> the plasma/qt related packages that have been recompiled, because
>> they were built upon libQtCore.so.4 were already recompiled with
>> gcc5. I've checked my logs.
> 
> Is the problem maybe order of building packages?
> 
> From your description I see one edge case: Plasma could have been
> compiled with gcc-5 but linked to qt still built with gcc-4. Then, qt
> was rebuilt after this and is compiled by gcc-5 now.
> 
> Without reading the bug report I would guess that is what the bug is
> about: Linking plasma against gcc-4 built qt binaries... Even when you
> rebuild qt with gcc-5 after this, you'd need to relink plasma.
> 
> From your logs, you should see the order in which those packages were
> rebuilt and linked.
> 
> revdep-rebuild is not always able to perfectly order packages right for
> emerge, and emerge in turn has no problem with wrong ordering because
> it is rebuilding packages and the dependency constraints are already
> fulfilled (read: runtime and build deps are already there). Portage
> does not consider rebuilds as a hard dependency/precondition for
> rebuilding other packages...
> 
> As far as I understood, "--changed-deps" should fix this and also
> rebuild packages depending on the rebuilt packages _after_ they've been
> rebuilt.
> 
> The "--empty" option would have a similar effect, tho rebuild many more
> packages.
> 
> You could've tried if "revdep-rebuild ... -- --changed-deps" would've
> done anything better. I would be interested...

Too late now for this, but your assumption can be true. The rebuild for the 
gcc5 switch did not work smoothly because emerge failed to build all 515 
packages reported by revdep-rebuild in correct order. Especially the Kontact 
suite was affected, because emerge tried to build kdepim-common-libs before 
some of the dependent packages (happened to me on two different machines). A 
second run rebuilding only the failed stuff succeeded in the end, but that 
might have lead to the linker situation you've described above. I wonder, 
what else might still be affected :-/

Cheers,
Jörg

[gentoo-user] Re: Re: Re: Qt-4.8.7 bug

[Jörg Schaible]

Peter Humphrey wrote:

> On Monday 22 May 2017 09:49:01 Jörg Schaible wrote:
>> Hi Peter,
>> 
>> Peter Humphrey wrote:
>> 
>> [snip]
>> 
>> > Have you seen https://bugs.gentoo.org/show_bug.cgi?id=595618 ? It says
>> > that "Qt plugins compiled with gcc-4 are incompatible with
>> > > > be
>> > expected to anticipate that. On the other hand, some kind of notice
>> > could
>> > be issued, and bug 618922 is pursuing that. (That's the one I started
>> > this thread with.)
>> 
>> well, this does not seem to be the complete truth. When I switched to gcc
>> 5.x I did a revdep-rebuild for anything that was compiled against
>> libstdc++.so.6 just like the according news entry was recommending. And I
>> am quite sure that those Qt plugins were part of my 515 recompiled
>> packages.
>> 
>> Nevertheless, my KDE 4 apps were broken after the update to Qt 4.8.7.
>> Rebuilding anything that was using libQtCore.so.4 solved it, but I fail
>> to see how this is related to the gcc update two weeks ago.
> 
> I can only suggest you read bug report 618922 if you haven't already,
> including following its reference to bug 595618. It makes sense to me.

It does not for me. My packages were already compiled with gcc-5.4.0. Those 
Buzilla issues only talk about (plasma/qt) packages compiled with previous 
gcc-4.x which are supposed to be incompatible. All of the plasma/qt related 
packages that have been recompiled, because they were built upon 
libQtCore.so.4 were already recompiled with gcc5. I've checked my logs.

Cheers,
Jörg

[gentoo-user] Re: Re: Qt-4.8.7 bug

[Jörg Schaible]

Hi Peter,

Peter Humphrey wrote:

[snip]

> Have you seen https://bugs.gentoo.org/show_bug.cgi?id=595618 ? It says
> that "Qt plugins compiled with gcc-4 are incompatible with
>  expected to anticipate that. On the other hand, some kind of notice could
> be issued, and bug 618922 is pursuing that. (That's the one I started this
> thread with.)

well, this does not seem to be the complete truth. When I switched to gcc 
5.x I did a revdep-rebuild for anything that was compiled against 
libstdc++.so.6 just like the according news entry was recommending. And I am 
quite sure that those Qt plugins were part of my 515 recompiled packages.

Nevertheless, my KDE 4 apps were broken after the update to Qt 4.8.7. 
Rebuilding anything that was using libQtCore.so.4 solved it, but I fail to 
see how this is related to the gcc update two weeks ago.

Cheers,
Jörg

[gentoo-user] Re: konqueror:5 - why couldn't it be more like konqueror:3 ?!!

[Jörg Schaible]

Mick wrote:

> OK, I know life moves on, but this move has been a retrograde step for me.
>  My konqueror:5 recently updated seems to have a number of problems and
> 
> features I am not happy with.  Grateful for any pointers to address these.
>  In no particular order.
> 
> 1. The Bookmarks Toolbar will *always* show up when launching Konqueror. 
> I deselect Settings/Toolbars Shown/Bookmark Toolbar and relaunch the
> application, only to find out my deselection will not stick.
> 
> 2. The menu shows no icons, only text; buttons like open new tab/close
> current tab show no icons, making difficult to guess.
> 
> 3. When used as a file manager Konqueror will only open directories or
> files
> if I double click on them.  I have set up in systemsettings5
> Hardware/Input
> Devices/Mouse/Single click to open files and folders.  Konqueror ignores
> it.
> 
> 4. All my years of Bookmarks of Konqueror:4 gone.  I had to import them
> manually.
> 
> 5. Network places, gone.
> 
> 6. Left hand Panels with Places/Devices/Folders ain't thaa'r.
> 
> 7. Konqueror Introduction page, no icons; unless I hover over them,
> hyperlinked titles shown in dark grey over a blue background.  I know my
> eye sight is not as good as it used to be, but this is really akin to
> usability sabotage.

I am missing the support for man: and info: protocols. Do I need now some 
additional package?

Cheers,
Jörg

[gentoo-user] Re: Mysterious encodig problem...

[Jörg Schaible]

Hi,

tu...@posteo.de wrote:

> Hi,
> 
> it seems, that I have an encoding problem...which is triggered
> (only???) somewhere in the chain fetchmail=>procmail=>neomutt with vim.
> 
> ...and it only effects single and double quotes.
> 
> I am living in Non-Ascii-land (germany). Therefore I have
> some "strange" ;) characters on my keyboard...the "Umlauts"
> und the sharp S.
> 
> All those are displayed nicely...in neomutt/vim also (which I use for
> mailing).
> 
> When I receive an email with single quotes (') or double
> quotes ("), they will be changed to "???".
> 
> My settings (settings to English language are for displaying -- for
> example -- manpages in their original version - not the translated ones):
> locale:
> LANG=en_US
> LC_CTYPE="en_US"
> LC_NUMERIC="en_US"
> LC_TIME="en_US"
> LC_COLLATE="en_US"
> LC_MONETARY="en_US"
> LC_MESSAGES="en_US"
> LC_PAPER="en_US"
> LC_NAME="en_US"
> LC_ADDRESS="en_US"
> LC_TELEPHONE="en_US"
> LC_MEASUREMENT="en_US"
> LC_IDENTIFICATION="en_US"
> LC_ALL=

For default locale use en_US.UTF-8 instead. See locale-gen below.

> locale-gen:
>  * Generating locale-archive: forcing # of jobs to 1
>  * Generating 5 locales (this might take a while) with 1 jobs
>  *  (1/5) Generating de_DE.ISO-8859-15@euro ...   
> [ ok ]
>  *  (2/5) Generating de_DE.ISO-8859-1 ... 
> [ ok ]
>  *  (3/5) Generating de_DE.UTF-8 ...  
> [ ok ]
>  *  (4/5) Generating en_US.ISO-8859-1 ... 
> [ ok ]
>  *  (5/5) Generating en_US.UTF-8 ...  
> [ ok ] * Generation complete
> 
> 
> /etc/locale.gen:
> en_US ISO-8859-1
> en_US.UTF-8 UTF-8
> #ja_JP.EUC-JP EUC-JP
> #ja_JP.UTF-8 UTF-8
> #ja_JP EUC-JP
> #en_HK ISO-8859-1
> #en_PH ISO-8859-1
> de_DE.UFT-8 UTF-8
> de_DE ISO-8859-1
> de_DE@euro ISO-8859-15
> #es_MX ISO-8859-1
> #fa_IR UTF-8
> #fr_FR ISO-8859-1
> #fr_FR@euro ISO-8859-15
> #it_IT ISO-8859-1
> 
> 
> Any light sched onto this is very welcome ! :)
> Thanks a lot in advance!

Cheers,
Jörg

[gentoo-user] Re: Re: Flashing hardware via WINE ?

[Jörg Schaible]

tu...@posteo.de wrote:

[snip]

> Hi Kai (that's a rhyme! :)
> 
> I have installed Virtualbox already and use the Linux Image I
> installed there for banking purposes only. Feels more secure.
> 
> I would prefer the WIndows-in-a-(virtual)box-solution) as you
> do -- if I would own a Windows installation disc. But do not.

You might give ReactOS a try: https://www.reactos.org/
Works for me in a VBox, but I have no idea if serial ports are supported.

[snip]

Cheers,
Jörg

[gentoo-user] Re: Re: boost-1.62.0-r1 blocked by nothing ??

[Jörg Schaible]

Dale wrote:

[snip]

> Portage lock?  Sometimes, my brain does that too.  lol

Hehe.
 
> I thought about it after I hit send but figured you would get the
> thought, maybe you had one or the other in a mask/unmask file or
> something that resulted in a conflict?  I was sort of thinking it but
> didn't type it in for some reason.  Still, if you did the same command I
> posted, you would have seen the difference and thought on it.  Generally
> if there is a difference like that, it's because of a local setting, or
> a change in the tree due to different sync time, which would give the
> idea of syncing again.

Again the same issue on another box:

=== %< ==
$ equery l -p boost boost-build
 * Searching for boost ...
[-P-] [  ] dev-libs/boost-1.55.0-r2:0/1.55.0
[IP-] [  ] dev-libs/boost-1.56.0-r1:0/1.56.0
[-P-] [ ~] dev-libs/boost-1.58.0-r1:0/1.58.0
[-P-] [ ~] dev-libs/boost-1.59.0:0/1.59.0
[-P-] [ ~] dev-libs/boost-1.60.0:0/1.60.0
[-P-] [ ~] dev-libs/boost-1.61.0:0/1.61.0
[-P-] [ ~] dev-libs/boost-1.61.0-r1:0/1.61.0
[-P-] [  ] dev-libs/boost-1.62.0-r1:0/1.62.0
[-P-] [ ~] dev-libs/boost-1.63.0:0/1.63.0

 * Searching for boost-build ...
[-P-] [  ] dev-util/boost-build-1.55.0:0
[-P-] [ ~] dev-util/boost-build-1.55.0-r1:0
[IP-] [  ] dev-util/boost-build-1.56.0:0
[-P-] [ ~] dev-util/boost-build-1.58.0:0
[-P-] [ ~] dev-util/boost-build-1.59.0:0
[-P-] [ ~] dev-util/boost-build-1.60.0:0
[-P-] [ ~] dev-util/boost-build-1.61.0:0
[-P-] [  ] dev-util/boost-build-1.62.0-r1:0
[-P-] [ ~] dev-util/boost-build-1.63.0:0
=== %< ==

Portage should be capable of an update.
 
> Anyway, glad it is going.  That's what matters.

Yep, glad that I have a solution for it now.

- Jörg

[gentoo-user] Re: boost-1.62.0-r1 blocked by nothing ??

[Jörg Schaible]

Marc Joliet wrote:

> On Thursday 02 February 2017 00:21:29 Jörg Schaible wrote:
>> Hi,
>> 
>> now I have an emerge mystery myself: It claims boost is blocked by  ...
>> nothing. See yourself:
>> - emerge tells me that boost is for whatever reason blocked
>> - I cannot find any entry for this in /etc/portage/**
>> - eix does not show any kind of mask for 1.62.0-r1
>> - none of the installed packages show a condition preventing an update
>> 
>> Any idea?
>> 
>> Cheers,
>> Jörg
>> 
>> = %< ===
>> $ emerge -uDvta --changed-use --with-bdeps=y --backtrack=50 world
>> 
>> These are the packages that would be merged, in reverse order:
>> 
>> Calculating dependencies... done!
>> 
>> Total: 0 packages, Size of downloads: 0 KiB
>> 
>> !!! The following update has been skipped due to unsatisfied
>> dependencies:
>> 
>> dev-libs/boost:0
>> 
>> !!! All ebuilds that could satisfy "=dev-util/boost-build-1.62*" have
>> been masked.
>> !!! One of the following masked packages is required to complete your
>> request:
>> - dev-util/boost-build-1.62.0-r1::gentoo (masked by: )
>> 
>> (dependency required by "dev-libs/boost-1.62.0-r1::gentoo" [ebuild])
>> For more information, see the MASKED PACKAGES section in the emerge
>> man page or refer to the Gentoo Handbook.
>> 
>> 
>> 
>> Nothing to merge; quitting.
> [...]
> 
> Hah! I had the exact same problem.  And I solved it in the dumbest
> possible way: running the exact same emerge command repeatedly until the
> error went
> away.  I *wish* I knew what the root cause is (actually, no, I don't), but
> that worked on two different systems.  It can take a few retries, though
> (three on my laptop, IIRC, but just one on my desktop).

Actually I had another system that was originally blocked in the same way, 
but after a synchronization the problem was gone too and the two packages 
where part of the update. However, for this system the problem seems more 
persistent... at least I am not the only one facing this strange behavior.

Funny, emerge suddenly can upgrade this package running alone:

emerge -1 boost-build boost

(thanks to Neil's answer)

Cheers,
Jörg

[gentoo-user] Re: boost-1.62.0-r1 blocked by nothing ??

[Jörg Schaible]

the...@sys-concept.com wrote:

> On 02/01/2017 04:21 PM, Jörg Schaible wrote:
>> Hi,
>> 
>> now I have an emerge mystery myself: It claims boost is blocked by  ...
>> nothing. See yourself:
>> - emerge tells me that boost is for whatever reason blocked
>> - I cannot find any entry for this in /etc/portage/**
>> - eix does not show any kind of mask for 1.62.0-r1
>> - none of the installed packages show a condition preventing an update
>> 
>> Any idea?
>> 
>> Cheers,
>> Jörg
>> 
> [snip]
> 
> On gcc-4.8.5 on my system boost-1.62.0-r1 was failing as well
> I switched to gcc-4.9.3 and it emerges just fine.

I have a mystery block by emerge, it has nothing to do with failing builds.

Cheers,
Jörg

[gentoo-user] Re: boost-1.62.0-r1 blocked by nothing ??

[Jörg Schaible]

Hi Dale,

Dale wrote:

[snip]

> If it helps any, this is what I have.
> 
> 
> root@fireball / # equery l -p boost boost-build
>  * Searching for boost ...
> [-P-] [  ] dev-libs/boost-1.55.0-r2:0/1.55.0
> [-P-] [  ] dev-libs/boost-1.56.0-r1:0/1.56.0
> [-P-] [ ~] dev-libs/boost-1.58.0-r1:0/1.58.0
> [-P-] [ ~] dev-libs/boost-1.59.0:0/1.59.0
> [-P-] [ ~] dev-libs/boost-1.60.0:0/1.60.0
> [-P-] [ ~] dev-libs/boost-1.61.0:0/1.61.0
> [-P-] [ ~] dev-libs/boost-1.61.0-r1:0/1.61.0
> [IP-] [  ] dev-libs/boost-1.62.0-r1:0/1.62.0
> [-P-] [ ~] dev-libs/boost-1.63.0:0/1.63.0
> 
>  * Searching for boost-build ...
> [-P-] [  ] dev-util/boost-build-1.55.0:0
> [-P-] [ ~] dev-util/boost-build-1.55.0-r1:0
> [-P-] [  ] dev-util/boost-build-1.56.0:0
> [-P-] [ ~] dev-util/boost-build-1.58.0:0
> [-P-] [ ~] dev-util/boost-build-1.59.0:0
> [-P-] [ ~] dev-util/boost-build-1.60.0:0
> [-P-] [ ~] dev-util/boost-build-1.61.0:0
> [IP-] [  ] dev-util/boost-build-1.62.0-r1:0
> [-P-] [ ~] dev-util/boost-build-1.63.0:0
> root@fireball / #
> 
> 
> Maybe both have to be at the same version??

I assume so, but why did emerge fail to upgrade both at same time?

Cheers,
Jörg

[gentoo-user] Re: Re: boost-1.62.0-r1 blocked by nothing ??

[Jörg Schaible]

Hi Neil,

Neil Bothwick wrote:

> On Thu, 2 Feb 2017 14:47:29 +0200, Nikos Chantziaras wrote:
> 
>> > now I have an emerge mystery myself: It claims boost is blocked
>> > by  ... nothing.
>> 
>> Same here. I don't know why, but the way I solved it is by unmerging
>> boost and then trying the update again.
>> 
>> When I unmerged both boost as well as boost-build, portage wanted to
>> re-install 1.62. The only way I could make it work is keep boost-build
>> 1.62 installed and only unmerge boost.
> 
> All I did was "emerge -1a boost boost-build" and it worked fine, as it
> has done in the past.

Thanks for the idea. Yes, this works. Strange nevertheless.

Cheers,
Jörg

[gentoo-user] boost-1.62.0-r1 blocked by nothing ??

[Jörg Schaible]

Hi,

now I have an emerge mystery myself: It claims boost is blocked by  ... 
nothing. See yourself:
- emerge tells me that boost is for whatever reason blocked
- I cannot find any entry for this in /etc/portage/**
- eix does not show any kind of mask for 1.62.0-r1
- none of the installed packages show a condition preventing an update

Any idea?

Cheers,
Jörg

= %< ===
$ emerge -uDvta --changed-use --with-bdeps=y --backtrack=50 world

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

Calculating dependencies... done!

Total: 0 packages, Size of downloads: 0 KiB

!!! The following update has been skipped due to unsatisfied dependencies:

dev-libs/boost:0

!!! All ebuilds that could satisfy "=dev-util/boost-build-1.62*" have been 
masked.
!!! One of the following masked packages is required to complete your 
request:
- dev-util/boost-build-1.62.0-r1::gentoo (masked by: )

(dependency required by "dev-libs/boost-1.62.0-r1::gentoo" [ebuild])
For more information, see the MASKED PACKAGES section in the emerge
man page or refer to the Gentoo Handbook.



Nothing to merge; quitting.

$ find /etc/portage/ -type f -exec grep boost {} +
/etc/portage/package.use:dev-libs/boost expat
$ eix -I boost
[U] dev-libs/boost
 Available versions:  1.55.0-r2(0/1.55.0)^t 1.56.0-r1(0/1.56.0)^t 
~1.58.0-r1(0/1.58.0)^t ~1.59.0(0/1.59.0)^t ~1.60.0(0/1.60.0)^t 
~1.61.0(0/1.61.0)^t ~1.61.0-r1(0/1.61.0)^t 1.62.0-r1(0/1.62.0)^t 
~1.63.0(0/1.63.0)^t {context debug doc icu mpi +nls python static-libs 
+threads tools ABI_MIPS="n32 n64 o32" ABI_PPC="32 64" ABI_S390="32 64" 
ABI_X86="32 64 x32" PYTHON_TARGETS="python2_7 python3_4 python3_5"}
 Installed versions:  1.56.0-r1^t(11:20:13 AM 12/03/2016)(icu nls 
threads -context -debug -doc -mpi -python -static-libs -tools ABI_MIPS="-n32 
-n64 -o32" ABI_PPC="-32 -64" ABI_S390="-32 -64" ABI_X86="64 -32 -x32" 
PYTHON_TARGETS="python2_7 -python3_4")
 Homepage:http://www.boost.org/
 Description: Boost Libraries for C++

[U] dev-util/boost-build
 Available versions:  1.55.0^t ~1.55.0-r1^t 1.56.0^t ~1.58.0^t ~1.59.0^t 
~1.60.0^t ~1.61.0^t 1.62.0-r1^t ~1.63.0^t {examples python test 
PYTHON_TARGETS="python2_7"}
 Installed versions:  1.56.0^t(07:24:23 PM 07/16/2015)(-examples -python 
-test PYTHON_TARGETS="python2_7")
 Homepage:http://www.boost.org/doc/tools/build/index.html
 Description: A system for large project software construction, 
simple to use and powerful

Found 2 matches
$ equery d boost
 * These packages depend on boost:
app-office/calligra-2.9.11 (dev-libs/boost)
app-office/kmymoney-4.7.2 (dev-libs/boost)
app-office/libreoffice-5.2.3.3-r1 (>=dev-libs/boost-1.55)
app-text/libabw-0.1.1 (>=dev-libs/boost-1.46)
app-text/libebook-0.1.2-r1 (dev-libs/boost)
app-text/libetonyek-0.1.6 (dev-libs/boost)
app-text/libmspub-0.1.2 (dev-libs/boost)
app-text/libmwaw-0.3.8 (dev-libs/boost)
dev-cpp/libcmis-0.5.2_pre20160820 (dev-libs/boost)
dev-cpp/yaml-cpp-0.5.3 (>=dev-libs/boost-1.49.0-r2)
dev-db/mysql-connector-c++-1.1.6 (dev-libs/boost)
dev-db/mysql-workbench-6.3.4-r1 (>=dev-libs/boost-1.55.0[nls])
dev-db/vsqlite++-0.3.13-r1 (>=dev-libs/boost-1.33.1)
dev-libs/libixion-0.11.1 (dev-libs/boost[threads])
dev-libs/liborcus-0.11.2 (dev-libs/boost)
dev-libs/librevenge-0.0.4 (dev-libs/boost)
dev-util/mdds-1.2.2 (dev-libs/boost)
games-puzzle/pingus-0.7.6 (dev-libs/boost)
games-strategy/0ad-0.0.21_alpha (dev-libs/boost)
games-strategy/wesnoth-1.12.6 (>=dev-libs/boost-1.48[nls,threads])
games-strategy/widelands-0.18 (>=dev-libs/boost-1.48)
games-util/grfcodec-6.0.5 (dev-libs/boost)
kde-apps/akonadi-1.13.1_pre20160203-r1 (dev-libs/boost)
kde-apps/gpgmepp-16.08.3 (dev-libs/boost)
kde-apps/kdepim-runtime-4.14.11_pre20160211 (dev-libs/boost)
kde-apps/kdepimlibs-4.14.11_pre20160211-r2 (>=dev-libs/boost-1.35.0-r5)
kde-apps/kget-16.08.3 (dev-libs/boost)
kde-apps/kleopatra-4.14.11_pre20160211 (dev-libs/boost)
kde-frameworks/kactivities-5.29.0 (>=dev-libs/boost-1.54)
kde-frameworks/kactivities-stats-5.29.0 (>=dev-libs/boost-1.54)
kde-plasma/kactivitymanagerd-5.8.5 (>=dev-libs/boost-1.54)
kde-plasma/plasma-desktop-5.8.5-r1 (dev-libs/boost)
media-gfx/digikam-5.3.0-r1 (dev-libs/boost[threads])
media-gfx/enblend-4.1.3-r1 (>=dev-libs/boost-1.31.0)
media-gfx/hugin-2015.0.0-r1 (>=dev-libs/boost-1.49.0-r1)
media-gfx/inkscape-0.91-r3 (>=dev-libs/boost-1.36)
media-libs/chromaprint-1.2 (test ? dev-libs/boost)
media-libs/libcdr-0.1.1 (dev-libs/boost)
media-libs/libpagemaker-0.0.3-r1 (dev-libs/boost)
media-libs/libvisio-0.1.5 (dev-libs/boost)
media-libs/openimageio-1.6.13 (dev-libs/boost[python?])
media-libs/vigra-1.9.0 (>=dev-libs/boost-1.52.0-
r6[python?,python_targets_python2_7(-)?,-python_single_target_jython2_7(-),-
python_single_target_pypy(-),-python_single_target_pypy3(-),-
python_single_target_python3_4(-),-python_single_target_python3_5(-),-
python_single_target_python3_6(-),python_single_target_python2_7(+)])

[gentoo-user] Re: Re: Re: Re: Re: KWallet doesn't recognise my password

[Jörg Schaible]

Mick wrote:

> On Thursday 15 Dec 2016 14:02:39 Jörg Schaible wrote:
>> Mick wrote:
>> > On Wednesday 14 Dec 2016 09:08:11 Jörg Schaible wrote:
>> >> Mick wrote:
>> >> > On Tuesday 13 Dec 2016 11:35:33 Jörg Schaible wrote:
>> >> [snip]
>> >> 
>> >> >> No, that's the point: If you enable it, all kwallet-4 based apps
>> >> >> will fail. At least until 5.7. I've not tested 5.8 yet.
>> >> >> 
>> >> >> Cheers,
>> >> >> Jörg
>> >> > 
>> >> > This is what works here without any problems:
>> >> [snip]
>> >> 
>> >> Well, for me it broke KDE4-based apps on several different machines.
>> >> So, tell me, can you open with Konqueror local files?
>> >> 
>> >> Cheers,
>> >> Jörg
>> > 
>> > Yes, I can open files from Konqueror with a single click, unlike
>> > Dolphin which requires a double click to descent into a directory or
>> > open a file.
>> 
>> Well, for me it broke on several machines for all KDE 4 based apps (e.g.
>> Konqueror) the file protocol (Konqueror only show an error page about an
>> unknown protocol 'file') and the FileOpen dialog no longer works (you
>> cannot open/save files in those apps using the default dialog, e.g.
>> libreoffice, amarok, ...).
>> 
>> As soon as I deactivate kwallet-pam, all apps start working properly.
>> 
>> Cheers,
>> Jörg
> 
> How do you activate/de-activate kwallet-pam?

By setting comments in /etc/pam.d/sddm

=== %< ===
$ cat /etc/pam.d/sddm
#%PAM-1.0

authinclude system-login
account include system-login
passwordinclude system-login
session include system-login
#-auth  optionalpam_kwallet.so kdehome=.kde4
#-auth  optionalpam_kwallet5.so
#-session   optionalpam_kwallet.so
#-session   optionalpam_kwallet5.so auto_start
=== %< ===

> This is what my /etc/pam.d/kde contains, in case yours is different:
> 
> $ cat /etc/pam.d/kde
> #%PAM-1.0
> 
> auth   required pam_nologin.so
> 
> auth   include  system-local-login
> 
> accountinclude  system-local-login
> 
> password   include  system-local-login
> 
> sessioninclude  system-local-login

=== %< ===
$ cat /etc/pam.d/kde
#%PAM-1.0

auth   required pam_nologin.so
auth   include  system-local-login
accountinclude  system-local-login
password   include  system-local-login
sessioninclude  system-local-login
-auth   optionalpam_kwallet.so kdehome=.kde4
-auth   optionalpam_kwallet5.so
-sessionoptionalpam_kwallet.so
-sessionoptionalpam_kwallet5.so auto_start
=== %< ===

It contains obviously still the kwallet-pam entries, but AFAICS only the 
ones for your display manager are relevant. At least this is what the elog 
message indicates if you install it.

Cheers,
Jörg
 

[gentoo-user] Re: Re: Re: Re: KWallet doesn't recognise my password

[Jörg Schaible]

Mick wrote:

> On Wednesday 14 Dec 2016 09:08:11 Jörg Schaible wrote:
>> Mick wrote:
>> > On Tuesday 13 Dec 2016 11:35:33 Jörg Schaible wrote:
>> [snip]
>> 
>> >> No, that's the point: If you enable it, all kwallet-4 based apps will
>> >> fail. At least until 5.7. I've not tested 5.8 yet.
>> >> 
>> >> Cheers,
>> >> Jörg
>> > 
>> > This is what works here without any problems:
>> [snip]
>> 
>> Well, for me it broke KDE4-based apps on several different machines. So,
>> tell me, can you open with Konqueror local files?
>> 
>> Cheers,
>> Jörg
> 
> Yes, I can open files from Konqueror with a single click, unlike Dolphin
> which requires a double click to descent into a directory or open a file.

Well, for me it broke on several machines for all KDE 4 based apps (e.g. 
Konqueror) the file protocol (Konqueror only show an error page about an 
unknown protocol 'file') and the FileOpen dialog no longer works (you cannot 
open/save files in those apps using the default dialog, e.g. libreoffice, 
amarok, ...).

As soon as I deactivate kwallet-pam, all apps start working properly.

Cheers,
Jörg

[gentoo-user] Re: Re: Re: KWallet doesn't recognise my password

[Jörg Schaible]

Mick wrote:

> On Tuesday 13 Dec 2016 11:35:33 Jörg Schaible wrote:

[snip]

>> No, that's the point: If you enable it, all kwallet-4 based apps will
>> fail. At least until 5.7. I've not tested 5.8 yet.
>> 
>> Cheers,
>> Jörg
> 
> This is what works here without any problems:

[snip]

Well, for me it broke KDE4-based apps on several different machines. So, 
tell me, can you open with Konqueror local files?

Cheers,
Jörg

[gentoo-user] Re: Re: KWallet doesn't recognise my password

[Jörg Schaible]

J. Roeleveld wrote:

> On Tuesday, December 13, 2016 11:10:31 AM Jörg Schaible wrote:
>> Peter Humphrey wrote:
>> > Hello list,
>> > 
>> > Until this morning I've had no real problems with KMail and co. for
>> > quite a while, but something's upset the wallet system so that my
>> > password is no longer recognised when I start KMail-2. This is what
>> > I've tried:
>> > 
>> > 1. Re-created a blank /home partition and restored from yesterday's
>> > backup. (Yesterday's setup was working nicely.)
>> > No difference, so:
>> > 
>> > 2. Deleted ~/kde4/share/apps/kwallet (while not running live) and
>> > rebooted. No difference there either.
>> > 
>> > Is there something I can restore from backup to enable me to use the
>> > wallet again - perhaps something in /etc/ssl or /var/tmp? Maybe I need
>> > to remerge the wallet packages, or maybe I'll have to create an
>> > entirely new user for myself. I hoped I'd seen the last of that kind of
>> > masochism.
>> > 
>> > In case it's relevant, the appearance of this problem coincided with a
>> > new kernel, 4.9.0, which I compiled as usual and rebooted. I just got a
>> > blank screen. I had to revert to 4.8.14 and rebuild the associated
>> > modules before I could boot, and then fsck ran to check all the file
>> > systems.
>> 
>> Just in case: Did you activate kwallet-pam?
> 
> I don't think this is a requirement. I have this disabled and need to
> enter the password twice. Once for kwallet-4 and once for kwallet-5.

No, that's the point: If you enable it, all kwallet-4 based apps will fail. 
At least until 5.7. I've not tested 5.8 yet.

Cheers,
Jörg

[gentoo-user] Re: KWallet doesn't recognise my password

[Jörg Schaible]

Peter Humphrey wrote:

> Hello list,
> 
> Until this morning I've had no real problems with KMail and co. for quite
> a while, but something's upset the wallet system so that my password is no
> longer recognised when I start KMail-2. This is what I've tried:
> 
> 1.Re-created a blank /home partition and restored from yesterday's
> backup. (Yesterday's setup was working nicely.)
> No difference, so:
> 
> 2.Deleted ~/kde4/share/apps/kwallet (while not running live) and
> rebooted. No difference there either.
> 
> Is there something I can restore from backup to enable me to use the
> wallet again - perhaps something in /etc/ssl or /var/tmp? Maybe I need to
> remerge the wallet packages, or maybe I'll have to create an entirely new
> user for myself. I hoped I'd seen the last of that kind of masochism.
> 
> In case it's relevant, the appearance of this problem coincided with a new
> kernel, 4.9.0, which I compiled as usual and rebooted. I just got a blank
> screen. I had to revert to 4.8.14 and rebuild the associated modules
> before I could boot, and then fsck ran to check all the file systems.
> 


Just in case: Did you activate kwallet-pam?

Cheers,
Jörg

[gentoo-user] Solved: KDE 5: Broken file protocol for KDE 4 apps caused by kwallet

[Jörg Schaible]

> Jörg Schaible wrote:
> 
>> Hi,
>> 
>> after the update to Plasma 5/KF5, I can no longer open (HTML) files from
>> my local disk with Konqueror. It claims it does no longer know the file
>> protocol. I get a similar error in Amarok when I try to apply a cover to
>> an album from the local disk. It seems all KDE4-based application are
>> affected.
>> 
>> Does anybody what's causing this behavior and know how to solve it?

Jörg Schaible wrote:

> Anyone? After upgrading a second machine to KDE/Plasma 5, I have the same
> behavior there. All KDE-4-based apps fail to interact with the file
> system. Using KMail I can no longer add any attachment to an email nor
> save an existing attachment to disk.

OK, finally I found the culprit. I knew that it had to be something in the 
configuration, because I did not had the problem with a new/unused account 
on the same machine.

After some hours of investigation (deleting and restoring files based on 
their modification time in the local configuration), I found out that the 
file protocol works if I remove the salt file of the KDE4 wallet.

Well, unfortunately this prevented KMail to access the mail servers, because 
the wallet was no longer accessible at all. But I could save attachments ;-)

In the end it turned out that the problem was caused by kwallet-pam. When I 
upgraded KDE 4 to 5 I also upgraded kwallet-pam directly. However, this 
prevents unfortunately the migration of the default wallet from KDE 4 to 5. 
This migration happens normally the first time when the default wallet is 
accessed and the user is asked for the wallet's password.

It seems that the PAM system will access the wallet way to early and the 
migration cannot be performed. I have no idea why this finally breaks the 
file protocol for KDE 4 apps, but this is what happens.

After deactivating the kwallet pam module (in /etc/sddm) and a reboot, the 
migration could be performed after the login and all apps behave properly. 
It is even possible to activate the PAM module again afterwards.

Hope this might help others. At least a big fat warning should be added to 
Gentoo's KDE 5 upgrade guide.

Cheers,
Jörg

[gentoo-user] Re: Re: KDE 5: Broken file protocol for KDE 4 apps

[Jörg Schaible]

Hi,

P Levine wrote:

> On Wed, Oct 12, 2016 at 5:54 PM, Jörg Schaible <joerg.schai...@gmx.de>
> wrote:
>> Anyone? After upgrading a second machine to KDE/Plasma 5, I have the same
>> behavior there. All KDE-4-based apps fail to interact with the file
>> system. Using KMail I can no longer add any attachment to an email nor
>> save an existing attachment to disk.
>>
>> Jörg Schaible wrote:
>>
>>> Hi,
>>>
>>> after the update to Plasma 5/KF5, I can no longer open (HTML) files from
>>> my local disk with Konqueror. It claims it does no longer know the file
>>> protocol. I get a similar error in Amarok when I try to apply a cover to
>>> an album from the local disk. It seems all KDE4-based application are
>>> affected.
>>>
>>> Does anybody what's causing this behavior and know how to solve it?
>>>
>>> Cheers,
>>> Jörg
> 
> Try running the apps from command line, using '--nofork' and/or
> '--debug' where/if applicable.  Try to reproduce the behavior and see
> if something potentially relevant is printed out.

Nothing. No message on the console, no entry in .xsession-errors. I even 
tried to compare strace output for the same call:

  konqueror /usr/shared/docs/some/file.html

One time from a user account where this call behaves correctly, once from 
the account where konqueror fails to open it. Nothing obvious :-/

- Jörg

[gentoo-user] Re: Re: KDE 5: Broken file protocol for KDE 4 apps

[Jörg Schaible]

Michael Mol wrote:

> On Wednesday, October 12, 2016 11:54:48 PM Jörg Schaible wrote:
>> Anyone? After upgrading a second machine to KDE/Plasma 5, I have the same
>> behavior there. All KDE-4-based apps fail to interact with the file
>> system. Using KMail I can no longer add any attachment to an email nor
>> save an existing attachment to disk.
> 
> I'm running KMail (Gentoo doesn't have the KDE5 version in tree yet, so
> KDE4), and I send file attachments all the time. So I can say it's at
> least not *intrinsically* broken...

As already said, it is not KMail, it affects any KDE4-based app ... and 
there are still a lot of them.

> Much of KDE4 and KDE5 wind up installed side by side, FWIW. I'd suggest
> cycling through emerge @preserved-rebuild, revdep-rebuild, depclean, and
> see if that shakes something loose.

My dependency tree is fine.

> The KDE4->KDE5 transition was generally a royal PITA for me, too, though I
> can't remember what all broke...

I did it in the last 4 weeks on 3 machines. Two have this problem, the third 
does not.

However, it has to be something in the local configuration. If I log into 
another (unused) account in this machine, I don't have this problem :-/

Cheers,
Jörg

[gentoo-user] Re: KDE 5: Broken file protocol for KDE 4 apps

[Jörg Schaible]

Anyone? After upgrading a second machine to KDE/Plasma 5, I have the same 
behavior there. All KDE-4-based apps fail to interact with the file system. 
Using KMail I can no longer add any attachment to an email nor save an 
existing attachment to disk.

Jörg Schaible wrote:

> Hi,
> 
> after the update to Plasma 5/KF5, I can no longer open (HTML) files from
> my local disk with Konqueror. It claims it does no longer know the file
> protocol. I get a similar error in Amarok when I try to apply a cover to
> an album from the local disk. It seems all KDE4-based application are
> affected.
> 
> Does anybody what's causing this behavior and know how to solve it?
> 
> Cheers,
> Jörg

[gentoo-user] KDE 5: Broken file protocol for KDE 4 apps

[Jörg Schaible]

Hi,

after the update to Plasma 5/KF5, I can no longer open (HTML) files from my 
local disk with Konqueror. It claims it does no longer know the file 
protocol. I get a similar error in Amarok when I try to apply a cover to an 
album from the local disk. It seems all KDE4-based application are affected.

Does anybody what's causing this behavior and know how to solve it?

Cheers,
Jörg

[gentoo-user] Re: Dependency conflict

[Jörg Schaible]

Dmitry Bogun wrote:

> Hi.
> 
> Perl packages family use extremely annoying dependency definition.
> 
> See example below.
> 
> The question is "what is the correct way to solve this dependency
> conflict?". I don't like to add all "conflicts" into install list in cli. 
> Is there a way to force update of all this "perl sh.t/stuff"? And this is
> not first time when I faced this issue. And usually it happen with perl*
> packages.
> 
> # emerge -a1 bash perl
> 
>  * IMPORTANT: 1 news items need reading for repository 'gentoo'.
>  * Use eselect news read to view new items.
> 
> 
> These are the packages that would be merged, in order:
> 
> Calculating dependencies... done!
> [ebuild U  ] app-shells/bash-4.3_p46-r1 [4.3_p42-r1]
> [ebuild   R] dev-lang/perl-5.20.2
> 
> WARNING: One or more updates/rebuilds have been skipped due to a
> dependency conflict:
> 
> dev-lang/perl:0

[snip]

Not every application is compatible when an important part of the 
infrastructure is about to upgrade. Portage devs try to keep the whole tree 
consistent i.e. in such a case there not only the update of this important 
component (Perl), but also of all application that were incompatible with 
the new version.

However, you did not update world, but Perl only. In consequence it is not 
possible for portage to update also any of the incompatible applications. 
Since these require the old Perl version, you're stuck. Portage cannot 
fulfill your requirement to update Perl only and keep applications that are 
known to be incompatible.

Try: emerge -uDvta --changed-use --with-bdeps=y world

Cheers,
Jörg

[gentoo-user] kactivitymanagerd endless update loop

[Jörg Schaible]

Hi,

my system is still KDE 4 with the help of kde-sunset. However, I face an 
endless loop installing kactivitymanagerd. The package has been in group 
kde-base for KDE 4 and so it is in kde-sunset. However in central portage 
tree it has been moved to group kde-plasma.

The annoying part is now, that portage is constantly moving with each sync 
the installed kde-base/kactivitymanagerd-4.13.3-r1 into group kde-plasma 
with the result that portage tries to install kde-
base/kactivitymanagerd-4.13.3-r1 as new package. Which fails, because 
portage detects that some install files already exist.

So I have to uninstall kde-plasma/kactivitymanagerd-4.13.3-r1 and let 
portage install again kde-base/kactivitymanagerd-4.13.3-r1 until the next 
sync.

How can I stop portage from moving the package into another group?

Cheers,
Jörg

[gentoo-user] Re: Re: Re: Partition of 3TB USB drive not detected

[Jörg Schaible]

Hi Mick

Mick wrote:

> On Sunday 31 Jul 2016 22:38:22 Jörg Schaible wrote:
>> Hi Mick,
>> 
>> Mick wrote:
>> > On Sunday 31 Jul 2016 19:14:45 Jörg Schaible wrote:
>> >> Hi Daniel,
>> >> 
>> >> thanks for your response.
>> >> 
>> >> Daniel Frey wrote:
>> >> 
>> >> [snip]
>> >> 
>> >> > I can only think of two reasons, the kernel on the livecd doesn't
>> >> > support GPT (which is unlikely)
>> >> 
>> >> That would be really strange. However, how can I prove it?
>> > 
>> > If after you boot your systemrescuecd you can list:
>> > 
>> > /sys/firmware/efi
>> > 
>> > you have booted into UEFI mode.  If not, you have booted into legacy
>> > BIOS mode.
>> 
>> This machine has only plain old BIOS. The question is, why one kernel
>> detects the 3TB partition and the the other one does not. How can I prove
>> GPT support for the kernel itself.
> 
> 
> I see.  In this case have a look at /proc/config (it may be compressed) or
> depending on your version of sysrescuecd and kernel choice, have a look
> here:
> 
> https://sourceforge.net/p/systemrescuecd/code/ci/master/tree/
> 
> then compare your configuration to theirs.  The kernel module for GPT is
> 'CONFIG_EFI_PARTITION' and it must be built in, rather than as a separate
> module.

Now it's getting weird.

My normal kernel (4.4.6) does not have that flag set. Nevertheless it 
detects the partition. The two kernels (both 4.4.12) of systemrescue have 
that flag. I've tested another machine with kernel that also has the flag 
and it does not detect this partition also.

However, I have another 6TB USB drive and that one has one big partition 
that is detected by both machines. The funny thing is, it reports to have 
only a MBR with one partition of 6TB (same output on both kernels):

== %< ==
~ # parted /dev/sde print
Model: WD My Book 1230 (scsi)
Disk /dev/sde: 6001GB
Sector size (logical/physical): 4096B/4096B
Partition Table: msdos
Disk Flags: 

Number  Start   End SizeType File system  Flags
 1  8389kB  6001GB  6001GB  primary  ext4

~ # gdisk /dev/sde
GPT fdisk (gdisk) version 1.0.1

Partition table scan:
  MBR: MBR only
  BSD: not present
  APM: not present
  GPT: not present


***
Found invalid GPT and valid MBR; converting MBR to GPT format
in memory. THIS OPERATION IS POTENTIALLY DESTRUCTIVE! Exit by
typing 'q' if you don't want to convert your MBR partitions
to GPT format!
***


Command (? for help): q
== %< ==

AFAICS this partition works fine, fsck does not report any problem. The 
funny thing is, it should not have been possible, because of the 2GB limit 
of MBR.

???

- Jörg

[gentoo-user] Re: Re: Partition of 3TB USB drive not detected

[Jörg Schaible]

james wrote:

> On 07/31/2016 12:56 PM, Jörg Schaible wrote:
>> Jörg Schaible wrote:
>>
>>> Hi Daniel,
>>>
>>> thanks for your response.
>>>
>>> Daniel Frey wrote:
>>>
>>> [snip]
>>>
>>>> I can only think of two reasons, the kernel on the livecd doesn't
>>>> support GPT (which is unlikely)
>>>
>>> That would be really strange. However, how can I prove it?
>>>
>>>> or you're booting a 32-bit kernel live
>>>> USB. I am reasonably certain for drives > 2TB a 64-bit kernel and GPT
>>>> are required.
>>>
>>> No, I've always chosen 64-bit kernels. I wonder what is so special about
>>> this partition ...
>>
>> Currently I wonder, why my system can find the partition at all:
>>
>>  %< 
>> # gdisk -l /dev/sdi
>> GPT fdisk (gdisk) version 1.0.1
>>
>> Partition table scan:
>>   MBR: protective
>>   BSD: not present
>>   APM: not present
>>   GPT: not present
> 
> If you have seen my recent thread,

I saw it, but did not read it in depth, because I had the impression, it is 
mainly about EFI systems. I'll re-read it ...

> much of this automounting during
> boot(strapping) is flaky that is much of what I have been searching out
> is a default (magical) partitioning schema that will eventually lead to
> clear documents on the current state of affairs not only with old versus
> new motherboards (mbr-->efi) and disk (mbr < 2.2T and gpt >2.2T)
> but including all sorts of new arm and other embedded (linux) boards.
> 
> Different forms of Solid State memory are next on my list, with usb (1.x
> --> 3.x) being top of the SS memory mediums. (Sorry I do not have
> more atm).
>
>> Creating new GPT entries.
>> Disk /dev/sdi: 732566646 sectors, 2.7 TiB
>> Logical sector size: 4096 bytes
>> Disk identifier (GUID): 80C04475-9B51-4A44-A52F-1F165AE02695
>> Partition table holds up to 128 entries
>> First usable sector is 6, last usable sector is 732566640
>> Partitions will be aligned on 256-sector boundaries
>> Total free space is 732566635 sectors (2.7 TiB)
>>
>> Number  Start (sector)End (sector)  Size   Code  Name
>>  %< 
>>
>> However, it's mounted successfully, see system logs:
>>
>>  %< 
>> [22735.626752] sd 13:0:0:0: [sdi] 732566646 4096-byte logical blocks:
>> [(3.00
>> TB/2.73 TiB)
>> [22735.629255]  sdi: sdi1
>> [23414.066315] EXT4-fs (sdi1): mounted filesystem with ordered data mode.
>> Opts: (null)
>>  %< 
>>
>> Has anyone ever tried the recovery option of GPT disk to rebuild GPT from
>> MBR?
> 
> I see some sort of 'auto correction' by gpt technology to convert many
> forms of perceived mbr to gpt to be used by the booting process for
> spinning rust. So this issue is not limited to usb medium. I would also
> point out that I'd look deeply into the usb specs for the vendor of your
> usb sticks, as they do some 'funky things' at the firmware level inside
> many of the newer/faster/larger usb devices. It not just dumb memory
> like the early 1.x devices. Many are slanted to Microsoft business
> strategies. I'm not suggesting that is your current issues. I'm merely
> pointing out that some newer usb sticks are systems themselves complete
> with firmware so the devices looks like dumb memory. Furthermore, the
> silicon vendors provide firmware options to usb sticks vendors (like
> Texas Instruments) but also the vendor add to or change the hidden
> firmware as meets their multifaceted business objects. Sadly, the NSA is
> deeply involved here, as are many nation states and large corporations.
> You'd be surprised what youd find in a modern usb stick, should you take
> it into a class 6+ clean-room for analysis. The lower the particle count
> the more fantastic the tools
> to open up silicon and look deeply into what is actually going on.
> This is why folks love those classified research facilities that have
> govt contract and folks hanging around. Lots of very, very cool toys
> you just do not hear about.. Way beyond microscopes built by
> physicist.

Actually it is not that modern. ~5 year old Intenso 2GB. I'd be surprised if 
booting from the stick prevents partition detection of another USB drive, 
but who knows? Maybe I should burn the iso instead and boot that one ;-)

> Prolly not your issue, but still present. Cheap ass usb vendors often
> have corner issues that are unintentional, that is why well recognized
> vendors of SS memory are the best to deal with, for consistency of
> behavior.
> 
> I'd use as many different tools as you can find and read the vendor &
> silicon manufacturer's docs to see what you are really dealing with to
> ferret out this weirdness. (it's a darn time sync, just so you know).
> 
> 
> [1] http://www.cleanroom.byu.edu/particlecount.phtml
> 
> hth,
> James

Thanks,
Jörg

[gentoo-user] Re: Re: Partition of 3TB USB drive not detected

[Jörg Schaible]

Hi Mick,

Mick wrote:

> On Sunday 31 Jul 2016 19:14:45 Jörg Schaible wrote:
>> Hi Daniel,
>> 
>> thanks for your response.
>> 
>> Daniel Frey wrote:
>> 
>> [snip]
>> 
>> > I can only think of two reasons, the kernel on the livecd doesn't
>> > support GPT (which is unlikely)
>> 
>> That would be really strange. However, how can I prove it?
> 
> If after you boot your systemrescuecd you can list:
> 
> /sys/firmware/efi
> 
> you have booted into UEFI mode.  If not, you have booted into legacy BIOS
> mode.


This machine has only plain old BIOS. The question is, why one kernel 
detects the 3TB partition and the the other one does not. How can I prove 
GPT support for the kernel itself.

Cheers,
Jörg

[gentoo-user] Re: Partition of 3TB USB drive not detected

[Jörg Schaible]

Jörg Schaible wrote:

> Hi Daniel,
> 
> thanks for your response.
> 
> Daniel Frey wrote:
> 
> [snip]
>  
>> I can only think of two reasons, the kernel on the livecd doesn't
>> support GPT (which is unlikely)
> 
> That would be really strange. However, how can I prove it?
> 
>> or you're booting a 32-bit kernel live
>> USB. I am reasonably certain for drives > 2TB a 64-bit kernel and GPT
>> are required.
> 
> No, I've always chosen 64-bit kernels. I wonder what is so special about
> this partition ...

Currently I wonder, why my system can find the partition at all:

 %< 
# gdisk -l /dev/sdi
GPT fdisk (gdisk) version 1.0.1

Partition table scan:
  MBR: protective
  BSD: not present
  APM: not present
  GPT: not present

Creating new GPT entries.
Disk /dev/sdi: 732566646 sectors, 2.7 TiB
Logical sector size: 4096 bytes
Disk identifier (GUID): 80C04475-9B51-4A44-A52F-1F165AE02695
Partition table holds up to 128 entries
First usable sector is 6, last usable sector is 732566640
Partitions will be aligned on 256-sector boundaries
Total free space is 732566635 sectors (2.7 TiB)

Number  Start (sector)End (sector)  Size   Code  Name
 %< 

However, it's mounted successfully, see system logs:

 %< 
[22735.626752] sd 13:0:0:0: [sdi] 732566646 4096-byte logical blocks: (3.00 
TB/2.73 TiB)
[22735.629255]  sdi: sdi1
[23414.066315] EXT4-fs (sdi1): mounted filesystem with ordered data mode. 
Opts: (null)
 %< 

Has anyone ever tried the recovery option of GPT disk to rebuild GPT from 
MBR?

- Jörg

[gentoo-user] Re: Partition of 3TB USB drive not detected

[Jörg Schaible]

Hi James,

james wrote:

> On 07/31/2016 08:37 AM, Jörg Schaible wrote:
>> Hi,
>>
>> for my backups I use a 3TB USB drive (one big ext4 partition) without any
>> problems. Just plug in the cable, mount it and perform the backup. The
>> partition (sdi1) is detected an mountable without any problems:
> 
> this tells you the device is valid and working. good.
>>
>> However, when I boot a rescue system from a USB stick, the partition on
>> the USB is not detected. I already tried latest SystemRescueCD (default
>> and alternate kernel), Knoppix and the Gentoo Admin CD. Nothing, the
>> partition is not available.
> 
> So there could be a multitude of reasons. Did thos systems have a bios
> that support booting from a usb device? Are the bios setting set to
> select the bios device correct?

Well, same machine, same bios.

> I recently read somewhere the usb devices only support (2) types of
> image booting, but I cannot find that doc right now. rodsbooks is
> an excellent resource for all the issues around booting and device
> and various hardware/firmware issues. LikeWhoa has made booting the usb
> for gentoo, commonplace so search out those postings on the gentoo
> forums and in the wiki.
> 
> 
>> What's the difference? Why does my kernel find this partition and the
>> other one's do not? It's pretty silly to have a backup drive and cannot
>> access it in question ;-)
> 
> It could be many things. You just have to ferret thru ideas until
> something works for your hardware, and it is a frustrating process.

Definitely annoying.
 
> make sure the image you are trying to boot is on a compatible partition
> and file system that is supported by the boot image.

It is. Thanks for the pointer to rodsbooks.

- Jörg

[gentoo-user] Re: Partition of 3TB USB drive not detected

[Jörg Schaible]

Hi Daniel,

thanks for your response.

Daniel Frey wrote:

[snip]
 
> I can only think of two reasons, the kernel on the livecd doesn't
> support GPT (which is unlikely)

That would be really strange. However, how can I prove it?

> or you're booting a 32-bit kernel live
> USB. I am reasonably certain for drives > 2TB a 64-bit kernel and GPT
> are required.

No, I've always chosen 64-bit kernels. I wonder what is so special about 
this partition ...

Cheers,
Jörg

[gentoo-user] Partition of 3TB USB drive not detected

[Jörg Schaible]

Hi,

for my backups I use a 3TB USB drive (one big ext4 partition) without any 
problems. Just plug in the cable, mount it and perform the backup. The 
partition (sdi1) is detected an mountable without any problems:

=== %< ==
$ ls -l /dev/disk/by-id
total 0
lrwxrwxrwx 1 root root  9 Jul 31 13:17 ata-
Crucial_CT500MX200SSD1_161512468483 -> ../../sda
lrwxrwxrwx 1 root root 10 Jul 31 13:17 ata-
Crucial_CT500MX200SSD1_161512468483-part1 -> ../../sda1
lrwxrwxrwx 1 root root 10 Jul 31 13:17 ata-
Crucial_CT500MX200SSD1_161512468483-part2 -> ../../sda2
lrwxrwxrwx 1 root root 10 Jul 31 13:17 ata-
Crucial_CT500MX200SSD1_161512468483-part3 -> ../../sda3
lrwxrwxrwx 1 root root  9 Jul 31 13:17 ata-HL-DT-ST_DVDRAM_GH41N_K5Q9AN32423 
-> ../../sr0
lrwxrwxrwx 1 root root  9 Jul 31 15:20 ata-ST3000DM001-1CH166_Z1F45VR9 -> 
../../sdi
lrwxrwxrwx 1 root root 10 Jul 31 15:20 ata-ST3000DM001-1CH166_Z1F45VR9-part1 
-> ../../sdi1
lrwxrwxrwx 1 root root  9 Jul 31 13:17 usb-Generic-
_Compact_Flash_2006041309210-0:0 -> ../../sdc
lrwxrwxrwx 1 root root  9 Jul 31 13:17 usb-Generic-_MS_MS-
Pro_2006041309210-0:3 -> ../../sdf
lrwxrwxrwx 1 root root  9 Jul 31 13:17 usb-Generic-
_SD_MMC_2006041309210-0:2 -> ../../sde
lrwxrwxrwx 1 root root  9 Jul 31 13:17 usb-Generic-_SM_xD-
Picture_2006041309210-0:1 -> ../../sdd
lrwxrwxrwx 1 root root  9 Jul 31 13:17 usb-Generic_Flash_HS-
CF_26020128B005-0:0 -> ../../sdg
lrwxrwxrwx 1 root root  9 Jul 31 13:17 usb-Generic_Flash_HS-
COMBO_26020128B005-0:1 -> ../../sdh
lrwxrwxrwx 1 root root  9 Jul 31 13:17 usb-INTENSO_USB_AA0401297518-0:0 
-> ../../sdb
lrwxrwxrwx 1 root root 10 Jul 31 13:17 usb-INTENSO_USB_AA0401297518-0:0-
part1 -> ../../sdb1
lrwxrwxrwx 1 root root  9 Jul 31 15:20 wwn-0x5000c50065531ec5 -> ../../sdi
lrwxrwxrwx 1 root root 10 Jul 31 15:20 wwn-0x5000c50065531ec5-part1 -> 
../../sdi1
lrwxrwxrwx 1 root root  9 Jul 31 13:17 wwn-0x50014800 -> ../../sr0
lrwxrwxrwx 1 root root  9 Jul 31 13:17 wwn-0x500a075112468483 -> ../../sda
lrwxrwxrwx 1 root root 10 Jul 31 13:17 wwn-0x500a075112468483-part1 -> 
../../sda1
lrwxrwxrwx 1 root root 10 Jul 31 13:17 wwn-0x500a075112468483-part2 -> 
../../sda2
lrwxrwxrwx 1 root root 10 Jul 31 13:17 wwn-0x500a075112468483-part3 -> 
../../sda3
=== %< ==
$ ls -l /dev/disk/by-label
total 0
lrwxrwxrwx 1 root root 10 Jul 31 13:17 boot -> ../../sda1
lrwxrwxrwx 1 root root 10 Jul 31 13:17 gentoo -> ../../sda3
lrwxrwxrwx 1 root root 10 Jul 31 15:20 intenso -> ../../sdi1
lrwxrwxrwx 1 root root 10 Jul 31 13:17 swap -> ../../sda2
=== %< ==

However, when I boot a rescue system from a USB stick, the partition on the 
USB is not detected. I already tried latest SystemRescueCD (default and 
alternate kernel), Knoppix and the Gentoo Admin CD. Nothing, the partition 
is not available.

What's the difference? Why does my kernel find this partition and the other 
one's do not? It's pretty silly to have a backup drive and cannot access it 
in question ;-)

- Jörg

[gentoo-user] Re: Re: Update blocked by kdebase-startkde:4

[Jörg Schaible]

Daniel Frey wrote:

> On 07/09/2016 07:08 PM, Peter Humphrey wrote:
>> Thanks Dan. I tried your package.mask and thought I was getting
>> somewhere. But I had to add these to package.use (I have USE=-qt5 in
>> make.conf):
>> 
>> sys-auth/polkit-qt  qt5
>> dev-libs/libdbusmenu-qt qt5
>> media-libs/phonon   qt5
>> media-libs/phonon-vlc   qt5
>> 
>> Then I had to remove >kde-apps/kdebase-runtime-meta-4.15 from
>> package.mask to satisfy "(dependency required by
>> "kde-base/kdebase-startkde-4.11.22::gentoo"
>> [ebuild])". Guess what? Of course - it wanted to install the whole set of
>> [qt5
>> packages.
>> 
>> So I'm still stuck. I don't want to go to KDE-5 until I can find a way to
>> reduce the absurd amount of vertical space occupied by every line of
>> text. It will still be ugly, but at least more manageable.
>> 
>> I've attached screen shots of qt4 and qt5 versions of KMail to show what
>> I mean. The qt5 version is as close as I can get to the qt4.
>> 
> 
> I just tried and no luck here either. I made that list back in April
> when plasma made my machine unusable (hence the "might not work now"
> comment.) It's been several months now but I really don't want to go and
> try plasma again and waste days trying to get my machine usable again.
> I'm just not going to do any updates (and I guess I should do a stage4
> backup in case I have to restore...)
> 
> Right now my machine is nice and stable. What I don't understand is that
> plasma clearly isn't ready for primetime yet (IMHO) yet it seems KDE4 is
> not installable on Gentoo anymore.

It seems so. Really embarassing is however, that the dependencies for even 
already installed packages have been changed under the hood:

= %< ===
$ diff -u `locate kactivities-4.13.3-r2.ebuild`
--- /var/db/pkg/kde-base/kactivities-4.13.3-r2/kactivities-4.13.3-r2.ebuild 
2016-03-09 17:26:34.581846384 +0100
+++ /var/db/portage/gentoo/kde-base/kactivities/kactivities-4.13.3-r2.ebuild
2016-07-08 22:21:51.0 +0200
@@ -10,11 +10,11 @@
 
 DESCRIPTION="KDE Activity Manager"
 
-KEYWORDS="amd64 ~arm ~ppc ~ppc64 x86 ~x86-fbsd ~amd64-linux ~x86-linux"
+KEYWORDS="amd64 ~arm x86 ~x86-fbsd ~amd64-linux ~x86-linux"
 IUSE=""
 
 RDEPEND="
-   || ( $(add_kdebase_dep kactivitymanagerd)  I did try many things trying to get plasma working but everything I
> tried had no results. Plasma would crash even if you didn't do anything
> (no keyboard or mouse input.)
> 
> Dan

Jörg

[gentoo-user] Upgrade guide for ebuilds

[Jörg Schaible]

Hi,

is there somewhere an upgrade guide for ebuilds. I have a bunch of local 
ones with a no longer supported EAPI.

Cheers,
Jörg

[gentoo-user] Re: Emerge is teh FAIL ! ! ! ! !

[Jörg Schaible]

Alan Grimes wrote:

> No, skipfirst only works when it actually tries to compile something.
> Here it is failing in some strange pre-compile stage. I don't know why
> it has to be so anal about this, The normal solution would be to log the
> error, drop the two packages that were affected, and then compile the
> rest...
> 
> but no.
> 
> I had to uninstall wine and winetricks, now about 80% of my packages are
> compiling... I'm probably going to end up with another lengthy list of
> miserable fail

or simply mask 1.9.x versions of wine. Any real reason why to use the 
latest? Personally I use 1.7.55 together with latest pipelight. Newer 
versions of wine broke pipelight for me.

Cheers,
Jörg

[gentoo-user] Re: Switching current java-vm for a single application

[Jörg Schaible]

Hi Leonardo,

Alon Bar-Lev wrote:

> If all what script is doing is executing "java", just add the right
> JRE to your PATH as first element.

Try something along:

PATH=`java-config --select-vm=oracle-jdk-bin-1.8 -o`/bin:$PATH SenchaCmd

Cheers,
Jörg

[gentoo-user] Re: Switching current java-vm for a single application

[Jörg Schaible]

Hi James,

James wrote:

[snip]

> I suggest you file a bug at bugs.gentoo.org. Gentoo's java environment
> has/is undergoing major surgery and stability is a transient experince
> with java on gentoo, for the last decade. If you have ideas how to privide
> a patch (even a miniscule bandaid) I'd suggest you post on
> bugs.gentoo.org, as the gentoo-java team is tremendously understaffed.

If you refer Java applications that are *provided* by Gentoo ebuilds, you 
might be right, but just for managing and switching JDK versions it is 
really perfect and I am using Gentoo here more than a decade.

Cheers,
Jörg

[gentoo-user] Re: java.awt.AWTError

[Jörg Schaible]

Hi Helmut,

Helmut Jarausch wrote:

> Hi,
> 
> when emerging sci-geosciences/josm- I get a Java error :
> 
> 
> java.awt.AWTError: Can't connect to X11 window server using ':0' as the
> value of the DISPLAY variable.
> 
> 
> This is on a local machine using DISPLAY :0
> 
> Would anybody please shed some light on this error.
> 
> (Emerge version 9060 proceeds just fine)
> 
> Many thanks for a hint,
> Helmut

during the ebuild a Java application is running that requests the presence 
of X. So you're either on a machine without X or your JDK has no X support. 
E.g. the icedtea variants have a flag 'headless-awt' wich turns off X suport 
if active (and the name and meaning of the flag recently changed from 'X').

As another alternative you may try to fix the ebuild by running the java 
application in question with system property "-Djava.headless.awt=true". 
This can succeed if the application just initializes AWT, but does not use 
it to display something.

Cheers,
Jörg

[gentoo-user] Re: Switching current java-vm for a single application

[Jörg Schaible]

Hi Leonard,

Leonardo Guilherme wrote:

> Hello.
> 
> I'm using OpenJDK JVM regularly on my machine instead of Oracle's one,
> primarily because of the infinality patches and because I prefer open
> source software.
> 
> There are some applications, though, that do not play ball with it
> (namely, SenchaCmd) and I have to keep switching back and forth between
> installed java-vms just to run it.
> 
> I know nothing about Java or its environment, is there a way to specify
> the java-vm just for this application instead of doing "eselect java-vm
> set user 1; sencha *stuff*; eselect java-vm set user 3" everytime?
> 
> Is there a set of environment variables that can do this? Shall I wrap the
> command in a shell script? Ideas?

Either the application uses already a script to start and you can modify 
JAVA_HOME there or you may wrap it in a shell script. Simply set JAVA_HOME 
using Gentoo's java-config command:

= %< ==
#! /bin/bash
JAVA_HOME=`java-config --select-vm=oracle-jdk-bin-1.8 -o` command
= %< ==

The option -o prints the JRE home, see the --help option to see 
alternatives.

The selection name is the name used in eselect, so you may update the 
package without touching the script.

Cheers,
Jörg

[gentoo-user] Re: Bind stole my /

[Jörg Schaible]

Jarry wrote:

[snip]
 
> vs5-dns ~ # df -a
> Filesystem1K-blocksUsed Available Use% Mounted on
> /dev/sda2 -   - -- /
> proc  0   0 0- /proc
> tmpfs308188 420307768   1% /run
> dev   10240   0 10240   0% /dev
> mqueue0   0 0- /dev/mqueue
> devpts0   0 0- /dev/pts
> shm 1540940   0   1540940   0% /dev/shm
> sysfs 0   0 0- /sys
> cgroup_root   10240   0 10240   0% /sys/fs/cgroup
> openrc0   0 0- /sys/fs/cgroup/openrc
> none1048576   0   1048576   0% /var/tmp/portage
> /etc/bind -   - -- /chroot/dns/etc/bind
> /var/bind -   - -- /chroot/dns/var/bind
> /var/log/named 10138552 2300032   7300460  24% /chroot/dns/var/log/named
> 
> So there *is* /, but strangely it has no size. But it suddenly does
> have its proper size as soon as I stop bind running:
> 
> vs5-dns ~ # df -a
> Filesystem 1K-blocksUsed Available Use% Mounted on
> /dev/sda2  10138552 2300024   7300468  24% /
> proc  0   0 0- /proc
> tmpfs308188 416307772   1% /run
> dev   10240   0 10240   0% /dev
> mqueue0   0 0- /dev/mqueue
> devpts0   0 0- /dev/pts
> shm 1540940   0   1540940   0% /dev/shm
> sysfs 0   0 0- /sys
> cgroup_root   10240   0 10240   0% /sys/fs/cgroup
> openrc0   0 0- /sys/fs/cgroup/openrc
> none1048576   0   1048576   0% /var/tmp/portage
> 
> Jarry

As said, I have the same issue, I also use bind for my nfs4 exports. A 
downgrade to 8.21 or an upgrade to 8.24 solved the issue for me.

Cheers,
Jörg

[gentoo-user] Re: Re: Re: Emerge order not deterministic !?

[Jörg Schaible]

Neil Bothwick wrote:

> On Thu, 12 Nov 2015 09:48:48 +0100, Jörg Schaible wrote:
> 
>>> > Hmmm. And how can you then ever use
>> >> 
>> >>   emerge --resume --skip-fist
>> >> 
>> >> if not even the first build is deterministic? I skip the first
>> >> package anyway only if the problematic package is the first one to
>> >> build after resume, but if I cannot even rely on that?
>> > 
>> > 
>> > Because it re-uses the previous build order, not re-generate a new
>> > one.
>> 
>> That's simply not true. Emerge resume calculates the order again and
>> for me it starts often with a different package.
> 
> Then use emerge --keep-going and portage will take care of skipping
> failing merges for you.

Ah, no, that's not an option. It breaks for a reason. Sometimes I can ignore 
that and look for it later and in this case I skip it, but normally I fix 
the problem first. However, you have to take care, which package you're 
actually skipping. Especially if the build order is different with resume.

Cheers,
Jörg

[gentoo-user] Re: Emerge order not deterministic !?

[Jörg Schaible]

Alan McKinnon wrote:

> On 11/11/2015 21:35, Walter Dnes wrote:
>>   Ongoing installation.  I looked at 2 instances of
>> "emerge -pv x11-base/xorg-server" and the order was somewhat different.
>> Here are a couple of outputs, just a few seconds apart.  Is this a bug
>> or a feature?  See attachments.
>> 
> 
> 
> Emerge order is not deterministic, especially with parallel builds. The
> reason is that it does not need to be according to the dep graph - if
> two packages are at the same level and do not depend on each other, then
> the order they are built in does not affect the final result.
> Practically all parallel processing works this way.
> 
> What is deterministic, is that if you build the same set of packages
> twice and even if portage does them in different order, the binaries
> produced are functionally identical

Hmmm. And how can you then ever use

  emerge --resume --skip-fist

if not even the first build is deterministic? I skip the first package 
anyway only if the problematic package is the first one to build after 
resume, but if I cannot even rely on that?

Cheers,
Jörg

[gentoo-user] Re: Re: Emerge order not deterministic !?

[Jörg Schaible]

Alan McKinnon wrote:

> On 12/11/2015 10:29, Jörg Schaible wrote:
>> Alan McKinnon wrote:
>> 
>>> On 11/11/2015 21:35, Walter Dnes wrote:
>>>>   Ongoing installation.  I looked at 2 instances of
>>>> "emerge -pv x11-base/xorg-server" and the order was somewhat different.
>>>> Here are a couple of outputs, just a few seconds apart.  Is this a bug
>>>> or a feature?  See attachments.
>>>>
>>>
>>>
>>> Emerge order is not deterministic, especially with parallel builds. The
>>> reason is that it does not need to be according to the dep graph - if
>>> two packages are at the same level and do not depend on each other, then
>>> the order they are built in does not affect the final result.
>>> Practically all parallel processing works this way.
>>>
>>> What is deterministic, is that if you build the same set of packages
>>> twice and even if portage does them in different order, the binaries
>>> produced are functionally identical
>> 
>> Hmmm. And how can you then ever use
>> 
>>   emerge --resume --skip-fist
>> 
>> if not even the first build is deterministic? I skip the first package
>> anyway only if the problematic package is the first one to build after
>> resume, but if I cannot even rely on that?
> 
> 
> Because it re-uses the previous build order, not re-generate a new one.

That's simply not true. Emerge resume calculates the order again and for me 
it starts often with a different package.

Cheers,
Jörg

[gentoo-user] Re: Re: Re: Emerge order not deterministic !?

[Jörg Schaible]

Alan McKinnon wrote:

> On 12/11/2015 10:48, Jörg Schaible wrote:
>> Alan McKinnon wrote:
>> 
>>> On 12/11/2015 10:29, Jörg Schaible wrote:
>>>> Alan McKinnon wrote:

[snip]

>>>> Hmmm. And how can you then ever use
>>>>
>>>>   emerge --resume --skip-fist
>>>>
>>>> if not even the first build is deterministic? I skip the first package
>>>> anyway only if the problematic package is the first one to build after
>>>> resume, but if I cannot even rely on that?
>>>
>>>
>>> Because it re-uses the previous build order, not re-generate a new one.
>> 
>> That's simply not true. Emerge resume calculates the order again and for
>> me it starts often with a different package.
> 
> I've never noticed that. For me --skip-first has always skipped the
> correct first package (the one that previously failed).

That's what I always did originally also, until my build suddenly broke at 
the same package again and I had to notice that it skipped a completely 
different.

> As long as a known build failure is not in the --resume list, I don't
> care what the build order is because it is irrelevant. The only time it
> becomes relevant is when an ebuild has a bug such as a missing dep. But
> that's a bug in the ebuild and is fixed there.

Well, normally I don't care about the sequence either, except when skipping 
the first ;-)

Cheers,
Jörg

[gentoo-user] Re: LINGUAS issue

[Jörg Schaible]

Hi Francisco,

Francisco Ares wrote:

> Hi, all.
> 
> My locale language is "pt_BR" (Brazilian Portuguese), and many
> applications now support native translations.
> 
> And there is the "pt" possible LINGUAS entry, and there is no "pt_PT"
> (Portugal spoken Portuguese), for instance, neither any derivatives for
> other Portuguese speaking countries, which possibly have their own
> regional differences.
> 
> There are a few applications that do not distinguish "pt_BR" from "pt" and
> treat Portuguese language as simply "pt". An example is the OCR program
> "tesseract", that builds language specifics according to the LINGUAS
> environment variable.
> 
> Is there a way for specifying particular "LINGUAS" for individual
> packages?  I would not like to have to build dozens of applications to
> include "pt" to my "LINGUAS" definition just to have "tesseract" to
> include
> my native language support.  I've found some old messages about this on
> the net, but did not get any real solution.
> 
> Or should I ask the "tesseract" package maintainer to add "pt_BR" to the
> available options?

You can set this individually for every package. My LINGUAS is set to "en" 
only (I hate localized man pages), but I have entries in my package.use file 
to turn on linguas individually. e.g.:

app-office/libreoffice-l10n linguas_de linguas_en_GB linguas_fr offlinehelp


Cheers,
Jörg

[gentoo-user] Re: Bind stole my /

[Jörg Schaible]

Hi Jarry,

Jarry wrote:

> On 08-Nov-15 17:58, Mike Gilbert wrote:
>> On Fri, Nov 6, 2015 at 12:19 PM, Jarry  wrote:
>>> Hi Gentoo-users,
>>>
>>> I noted one strange thing today: It seems one of my servers lost "/"!
>>>
>>> vs5-dns ~ # df
>>> Filesystem1K-blocksUsed Available Use% Mounted on
>>> /var/log/named 10138552 2223148   7377344  24% /chroot/dns/var/log/named
>>> tmpfs308196 420307776   1% /run
>>> dev   10240   0 10240   0% /dev
>>> shm 1540968   0   1540968   0% /dev/shm
>>> cgroup_root   10240   0 10240   0% /sys/fs/cgroup
>>> none1048576   0   1048576   0% /var/tmp/portage
>>
>> Is your /etc/mtab a regular file, or is it a symlink to
>> /proc/self/mounts? The latter is recommended.
> 
> It is regular file. I never changed it...
> 
> vs5-dns ~ # ls -l /etc/mtab
> -rw-r--r-- 1 root root 908 Nov  9 19:14 /etc/mtab
> 
>> Anyway, please have a look at the contents of /etc/mtab,
>> /proc/self/mounts, and proc/self/mountinfo while named is running and
>> when it is stopped. If you pastebin them we can take a look for key
>> differences.
> 
> With bind running:
> http://pastebin.com/wkTW6xAY
> 
> without bind:
> http://pastebin.com/JG5FPNDW
> 
> While I can see some differences there, I still do not understand
> why is "/" missing in "df" output. BTW I can not proove it, but this
> was not the case all the time. At least when I was tuning monitoring
> software, I'm pretty sure "/" was there...

Actually I have the same problem, but I do not run named. /etc/mtab is a 
regular file on my system, too (since years). I am actually missing two 
entries:

 /
 /var/share/storage

Compare with contents of mtab. Both entries were originally there.

Do you have also nfs exports?

=== %< =
joehni@bobbel ~ $ df -hl
Filesystem Size  Used Avail Use% Mounted on
udev10M 0   10M   0% /dev
/var/spool/cups-pdf892G  261G  586G  31% /export/cups-pdf
tmpfs  396M  572K  395M   1% /run
none   2.0G 0  2.0G   0% /dev/shm
cgroup_root 10M 0   10M   0% /sys/fs/cgroup
/dev/sda6  926M  269M  594M  32% /var/log
/var/share/storage/Video/Film  3.6T  2.3T  1.2T  65% 
/export/storage/Video/Film
joehni@bobbel ~ $ cat /etc/mtab
proc /proc proc rw,nosuid,nodev,noexec,relatime 0 0
udev /dev devtmpfs rw,nosuid,relatime,size=10240k,nr_inodes=505255,mode=755 
0 0
devpts /dev/pts devpts rw,relatime,gid=5,mode=620 0 0
sysfs /sys sysfs rw,nosuid,nodev,noexec,relatime 0 0
/dev/sda5 / ext4 rw,relatime,data=ordered 0 0
tmpfs /run tmpfs rw,nodev,relatime,size=404508k,mode=755 0 0
mqueue /dev/mqueue mqueue rw,nosuid,nodev,noexec,relatime 0 0
none /dev/shm tmpfs rw,nosuid,nodev,noexec,relatime,size=2097152k 0 0
cgroup_root /sys/fs/cgroup tmpfs 
rw,nosuid,nodev,noexec,relatime,size=10240k,mode=755 0 0
fusectl /sys/fs/fuse/connections fusectl rw,nosuid,nodev,noexec,relatime 0 0
openrc /sys/fs/cgroup/openrc cgroup 
rw,nosuid,nodev,noexec,relatime,release_agent=/lib64/rc/sh/cgroup-release-
agent.sh,name=openrc 0 0
cpuset /sys/fs/cgroup/cpuset cgroup rw,nosuid,nodev,noexec,relatime,cpuset 0 
0
cpu /sys/fs/cgroup/cpu cgroup rw,nosuid,nodev,noexec,relatime,cpu 0 0
cpuacct /sys/fs/cgroup/cpuacct cgroup 
rw,nosuid,nodev,noexec,relatime,cpuacct 0 0
blkio /sys/fs/cgroup/blkio cgroup rw,nosuid,nodev,noexec,relatime,blkio 0 0
memory /sys/fs/cgroup/memory cgroup rw,nosuid,nodev,noexec,relatime,memory 0 
0
devices /sys/fs/cgroup/devices cgroup 
rw,nosuid,nodev,noexec,relatime,devices 0 0
freezer /sys/fs/cgroup/freezer cgroup 
rw,nosuid,nodev,noexec,relatime,freezer 0 0 
  
net_cls /sys/fs/cgroup/net_cls cgroup 
rw,nosuid,nodev,noexec,relatime,net_cls 0 0 
  
perf_event /sys/fs/cgroup/perf_event cgroup 
rw,nosuid,nodev,noexec,relatime,perf_event 0 0  
net_prio /sys/fs/cgroup/net_prio cgroup 
rw,nosuid,nodev,noexec,relatime,net_prio 0 0
debug /sys/fs/cgroup/debug cgroup rw,nosuid,nodev,noexec,relatime,debug 0 0 

/dev/sda6 /var/log ext4 rw 0 2  

/dev/sdb1 /var/share/storage/Video/Film ext4 rw 0 2 

/var/db/portage /export/portage none rw,bind 0 0

/var/share/storage /export/storage none rw,bind 0 0 

/var/spool/cups-pdf /export/cups-pdf none rw,bind 0 0   

/var/share/storage/Video/Film /export/storage/Video/Film none 

[gentoo-user] Re: Re: dev-qt/qtwebkit-5.4.0

[Jörg Schaible]

Stefan G. Weichinger wrote:

 On 05.02.2015 17:59, Michael Palimaka wrote:
 On 04/02/15 08:07, Stefan G. Weichinger wrote:
 Am 03.02.2015 um 20:30 schrieb Jörg Schaible:

 Consider a memcheck. Arbitrary failures while the CPU is high is often
 because some component starts dying. Sometimes cleaning the fans work
 wonders.

 Good suggestion, will check tmrw and clean the fans as well.

 It gave internal compiler error afai remember.
 
 How much free memory do you have, and are you building with debug
 symbols? qtwebkit:5 is exceptionally hungry, and I've seen it hit by the
 kernel OOM killer a lot.
 
 I removed one of the four DIMMs with 4 GB each .. so I am now working
 with 12 GB of RAM.
 
 When I re-emerge qtwebkit:5 I see 5-7 GB of *free* RAM ... and this with
 /var/tmp/portage as tmpfs (for the first minutes ... )
 
 The removed DIMM threw errors ... ok, so on with only 12 gigs.

;-)

Good that you could locate it!

Cheers,
Jörg 

[gentoo-user] Re: dev-qt/qtwebkit-5.4.0

[Jörg Schaible]

Hi Stefan,

Stefan G. Weichinger wrote:

 Am 03.02.2015 um 12:27 schrieb Alec Ten Harmsel:
 I've never had qtwebkit fail to build (at least not recently), what's
 the exact error? Can you post the output of `emerge --info
 dev-qt/qtwebkit` as well?
 
 hmmm, yes, I would like to ... but now it worked!
 I had it fail 3 times already ... maybe the shutdown did some magic
 change inbetween.

Consider a memcheck. Arbitrary failures while the CPU is high is often 
because some component starts dying. Sometimes cleaning the fans work 
wonders.

Cheers,
Jörg

[gentoo-user] Re: akonadi-server upgrade desaster

[Jörg Schaible]

Hi,

Mick wrote:

 On Wednesday 17 Dec 2014 19:13:03 Jörg Schaible wrote:
 Hi folks,
 
 it seems there's no way for me to upgrade my akonadi-server 1.11.0 to
 1.12.x or 1.13.x. I am using an external MySQL for years, but it fails to
 upgrade the tables nor will it recreate them without errors if I drop
 them all. All I can do is to downgrade to 1.11.0 again and restore the DB
 schema from a backup.

[snip]
 
 So, what now? I am out of ideas ...
 
 - Jörg
 
 
 I'm sure I've seen a bug reported in KDE about it when I suffered some
 similar error, but can't find the link just now.
 
 You may want to dump the database so that you have a back up and convert
 it from MyISAM to InnoDB before you try to update akonadi.

Actually I already tried that, but it failed again. I replaced all 
occurrences of MyISAM with InnoDB in them dumb file, dropped the DB and 
tried to restore it, but again with strange failures that the table allready 
existed. Something is weird with the InnoDB part of MySQL, but I dunno what.

- Jörg

[gentoo-user] akonadi-server upgrade desaster

[Jörg Schaible]

Hi folks,

it seems there's no way for me to upgrade my akonadi-server 1.11.0 to 
1.12.x or 1.13.x. I am using an external MySQL for years, but it fails to 
upgrade the tables nor will it recreate them without errors if I drop them 
all. All I can do is to downgrade to 1.11.0 again and restore the DB schema 
from a backup.

When I start akonadi manually with the DB restored for version 1.11.0, 
I get:

=== % 
~ $ akonadictl start
Starting Akonadi Server... 
   done. 
Connecting to deprecated signal 
QDBusConnectionInterface::serviceOwnerChanged(QString,QString,QString)
~ $ search paths:  (/usr/local/bin, /usr/bin, /bin, /opt/bin, 
/usr/x86_64-pc-linux-gnu/gcc-bin/4.8.3, /usr/games/bin, 
/home/joehni/bin, /home/joehni/bin, /usr/sbin, /usr/local/sbin, 
/usr/local/libexec, /usr/libexec, /opt/mysql/libexec, 
/opt/local/lib/mysql5/bin, /opt/mysql/sbin) 
Found mysql_install_db:   
Found mysqlcheck:  /usr/bin/mysqlcheck 
QSqlDatabasePrivate::removeDatabase: connection 'initConnection' is still 
in use, all queries will cease to work.
Database akonadi opened using driver QMYSQL 
DbInitializer::run() 
checking table  SchemaVersionTable 
checking table  ResourceTable 
checking table  CollectionTable 
ALTER TABLE CollectionTable ADD COLUMN enabled BOOL NOT NULL DEFAULT true 
ALTER TABLE CollectionTable ADD COLUMN syncPref TINYINT DEFAULT 2 
ALTER TABLE CollectionTable ADD COLUMN displayPref TINYINT DEFAULT 2 
ALTER TABLE CollectionTable ADD COLUMN indexPref TINYINT DEFAULT 2 
ALTER TABLE CollectionTable ADD COLUMN referenced BOOL NOT NULL DEFAULT 
false 
ALTER TABLE CollectionTable ADD COLUMN queryAttributes VARBINARY(255) 
ALTER TABLE CollectionTable ADD COLUMN queryCollections VARBINARY(255) 
checking table  MimeTypeTable 
checking table  PimItemTable 
checking table  FlagTable 
checking table  PartTypeTable 
CREATE TABLE PartTypeTable (id BIGINT NOT NULL AUTO_INCREMENT PRIMARY KEY, 
name VARBINARY(255) NOT NULL, ns VARBINARY(255) NOT NULL)  
COLLATE=utf8_general_ci DEFAULT CHARSET=utf8 

Sql error: Table '`akonadi`.`PartTypeTable`' already exists QMYSQL: Unable 
to execute query
Query: CREATE TABLE PartTypeTable (id BIGINT NOT NULL AUTO_INCREMENT 
PRIMARY 
KEY, name VARBINARY(255) NOT NULL, ns VARBINARY(255) NOT NULL)  
COLLATE=utf8_general_ci DEFAULT CHARSET=utf8
Unable to initialize database.
[== skipped stack trace ==]
ProcessControl: Application 'akonadiserver' returned with exit code 255 
(Unknown error)
search paths:  (/usr/local/bin, /usr/bin, /bin, /opt/bin, 
/usr/x86_64-pc-linux-gnu/gcc-bin/4.8.3, /usr/games/bin, 
/home/joehni/bin, /home/joehni/bin, /usr/sbin, /usr/local/sbin, 
/usr/local/libexec, /usr/libexec, /opt/mysql/libexec, 
/opt/local/lib/mysql5/bin, /opt/mysql/sbin) 
Found mysql_install_db:   
Found mysqlcheck:  /usr/bin/mysqlcheck 
QSqlDatabasePrivate::removeDatabase: connection 'initConnection' is still 
in use, all queries will cease to work.
Database akonadi opened using driver QMYSQL 
DbInitializer::run() 
checking table  SchemaVersionTable 
checking table  ResourceTable 
checking table  CollectionTable 
checking table  MimeTypeTable 
checking table  PimItemTable 
checking table  FlagTable 
checking table  PartTypeTable 
CREATE TABLE PartTypeTable (id BIGINT NOT NULL AUTO_INCREMENT PRIMARY KEY, 
name VARBINARY(255) NOT NULL, ns VARBINARY(255) NOT NULL)  
COLLATE=utf8_general_ci DEFAULT CHARSET=utf8 

Sql error: Can't create table 'akonadi.PartTypeTable' (errno: -1) QMYSQL: 
Unable to execute query
Query: CREATE TABLE PartTypeTable (id BIGINT NOT NULL AUTO_INCREMENT 
PRIMARY 
KEY, name VARBINARY(255) NOT NULL, ns VARBINARY(255) NOT NULL)  
COLLATE=utf8_general_ci DEFAULT CHARSET=utf8
Unable to initialize database.
[== skipped stack trace ==]
ProcessControl: Application 'akonadiserver' returned with exit code 255 
(Unknown error)
=== % 

It tries to create the (new) table PartTypeTable, fails with the obscure 
table already exists error and fails on any subsequent try with this 
errno 
-1.

One problem seems to be that all my akonadi tables are based on the MyISAM 
engine (executed before the failed upgrade attempt):

=== % 
mysql SELECT TABLE_NAME,ENGINE FROM information_schema.TABLES WHERE 
TABLE_SCHEMA='akonadi';
+++
| TABLE_NAME | ENGINE |
+++
| CollectionAttributeTable   | MyISAM |
| CollectionMimeTypeRelation | MyISAM |
| CollectionPimItemRelation  | MyISAM |
| CollectionTable| MyISAM |
| FlagTable  | MyISAM |
| MimeTypeTable  | MyISAM |
| PartTable  | MyISAM |
| PimItemFlagRelation| MyISAM |
| PimItemTable   | MyISAM |
| ResourceTable  | MyISAM |
| SchemaVersionTable | MyISAM |
+++
11 rows in set (0.00 sec)
=== % 

[gentoo-user] Re: Checking the reason for (-useflags) in brackets

[Jörg Schaible]

man emerge would have been even faster ;-)

meino.cra...@gmx.de wrote:

 Alan McKinnon alan.mckin...@gmail.com [14-06-21 12:36]:
 On 21/06/2014 11:19, meino.cra...@gmx.de wrote:
  Hi,
  
  for some applications I want to activate some USE flags, which are
  disabled by default.
  
  Some of those USE flags are set in () brackets. From searching the
  internet I learned, that this may be due to unresolveable dependencies
  or settings in the make.profile or
  
  Is there a way to exactly pin point the reason why a certain USE flags
  gets (diabled) and whether it is possible to resolve the problem?
  
  How can I figure out that?
 
 
 From the emerge man page:
 
 
  ()   circumfix   forced, masked, or removed
 
 
 So the answer is usually one of
 
 - flag not in ebuild anymore. Look in the ebuild
 - masked in profile. For these I usually search the profile directory
 recursively for the flag and figure it out that way
 
 
 What is it that you are trying to find out? A disabled flag is disabled
 and can't work, what further detail do you need?
 
 
 
 --
 Alan McKinnon
 alan.mckin...@gmail.com
 
 
 
 Thanks for your help. In the internet I have already found the answer
 in the meanwhile.
 
 
 Best regards,
 mcc

[gentoo-user] Re: How to control SLOT?

[Jörg Schaible]

Mick wrote:

 On Friday 16 May 2014 13:44:37 Alan McKinnon wrote:
 On 16/05/2014 14:38, Mick wrote:
  After all these years I still can't find my way around SLOT management
  needed for enlightenment.
  
  I want to emerge the latest available in the tree, which right now
  happens to be (~) 0.18.7.  So I added this in my keywords:
=x11-wm/enlightenment-0.18.7:0.17 ~amd64
 
 Either drop the = or drop the SLOT :0.17
 
 = always means a specific version is to be specified and a SLOT is not a
 version
 
 OK, dropping = still wants to install e16.
 
 Dropping  still wants to install e16.
 
 Dropping the SLOT :0.17 still wants to install e16.
 
 
 There must be some way of bringing in any =0.18.7:0.17, but not 'x11-
 wm/enlightenment-1.0.11:0'
 
 What else could I try?

x11-wm/enlightenment:0.17 ~amd64

- Jörg

[gentoo-user] Re: HP scanner is no longer found

[Jörg Schaible]

Hi Dale,

the scanner part of my Brother DCP stopped suddenly also one month ago. My 
troubles were caused by the upgrade from sane-backends-1.0.23 to 1.0.24-r1. 
Downgrading immediately led to a working scanner again.

[snip]

Cheers,
Jörg

[gentoo-user] Re: Delays while building Libre Office.

[Jörg Schaible]

Alan Mackenzie wrote:

 Hi, Gentoo.
 
 I've just built libreoffice-3.6.6.2 and it took 2 hours 10 minutes on my
 2.6 GHz quad core Athlon 2.  It used to take about an hour.
 
 Watching the build, it became evident that the first 50 minutes or so
 was taken up by several hundred mkdir operations (more precisely, mkdir
 -p long path).  Some of these mkdir's would take, perhaps, a minute to
 execute.  All the while, top showed make taking 100% of one core.
 
 There seems to be something suboptimal here.  Has anybody else seen
 this, or does anybody have any ideas how to fix the problem?

Just to test the file system, you may copy a large tree with cp -l within 
the same partition, since it creates only hard links, no data is actually 
copied.

- Jörg

[gentoo-user] Re: Eth0 interface not found - udev that little slut!!!!!

[Jörg Schaible]

Jarry wrote:

 On 06-Apr-13 19:10, Alan Mackenzie wrote:

 STOP SPREADING THIS FUD

 It did not happen to pretty much everybody. It happened to people who
 blindly updated thignsd and walked away, who did not read the news
 announcement, who did not read the CLEARLY WORDED wiki article at
 freedesktop.org or alternatively went into mod-induced panic and started
 making shit up in their heads.

 Steady on, old chap!  By it I was meaning the general inconvenience
 all round occasioned by the changes between udev-{197,200}.  Not
 everybody encountered this.  For example Dale, and Walt D. didn't have
 to do anything.  But pretty much everybody else did.
 
 The problem is, news item is not correct! I followed it
 and yet finished with server having old network name (eth0).
 Problem was the point 4. in news item, which is not quite clear:
 
 -
 4. predictable network interface names:
 If /etc/udev/rules.d/80-net-name-slot.rules is an empty file
 or a symlink to /dev/null, the new names will be disabled and
 the kernel will do all the interface naming...
 -
 
 Well, in my case 80-net-names-slot.rules was neither empty,
 nor symlink to dev null, but FULL OF COMMENTS AND NOTING ELSE,
 which basically did the same thing as empty file: disabled
 new network names. Unfortunatelly, I found it just after
 screwed reboot. But I did everything I found in news item:
 checked and verified that file was not symlink to /dev/null
 and that it was not empty (1667 bytes does not seem to me
 to be empty file).
 
 As I wrote previously, I am pretty sure I never created this
 file manually so it must have been created by som previous
 udev-version. So I finished up with similar problem as OP:
 after rebooting I did not find interface I expected. The
 only difference is I expected already interface with new
 name, and OP is probably the old one...

You're not alone, this happened for me on all my 4 machines.

 
 So I must add my point to complaining about news item
 not beeing quite clear. And this happens quite often...

- Jörg

[gentoo-user] Re: Eth0 interface not found - udev that little slut!!!!!

[Jörg Schaible]

Volker Armin Hemmann wrote:

 Am 06.04.2013 21:33, schrieb Mick:
 On Saturday 06 Apr 2013 20:03:15 Volker Armin Hemmann wrote:
 Am 06.04.2013 17:57, schrieb Alan Mackenzie:
 Hi, Nick.

 On Sat, Apr 06, 2013 at 10:51:42AM -0400, Nick Khamis wrote:
 After updating our systems we lost network connectivity to the
 servers. When trying to start net.eth0 we got the following message:
 /ib64/rc/net/wpa_supplicant.sh: line 68: _is wireless command not
 found /etc/init.d/net.eth0: line 548: _exists command not found
 Errror: Interface eth0 does not exist
 Ensure that you have loaded the correct kernel modules for your
 hardware
 # lsmod
 module used by
 tg3   0
 lbphytg3
 eth0
 flags=4098broadcast,multicast mtu 1500
 
 interrupt=16

 lo
 flags=73UP,LOOPBACK,RUNNING mtu 16436
 inet 127.0.0.1 BROADCAST 255.255.255.0
 inet6 ::1 prefixlen 128 scopeid 0x10 host

 Please excuse me, I am running back and forth from the servers and
 typing the error message here. Did our configuration get switched to
 IP6? These are our DB servers and why me!!! Why ME!
 No, it's not just you, it's happened to pretty much everybody. 
 udev-200 now renames eth0, eth1,  to something else, dependent upon
 complicated rules.  In my case eth0 has become p6p1, though many people
 seem to have got longer names.

 Have a look in /sys/class/net and see if your new name is there.  If
 so, edit all your config files containing eth0, switching to the new
 name.

 Once you got that done and things work again, take a deep breath and
 have a look at the most recent Gentoo news item ($ eselect news read)
 which
 explains it all, more or less.  Then decide whether the above is a long
 term solution, and if not start reading docs about writing udev rules.

 Yes, it's a pain in the backside.  But at least with Gentoo, you've a
 good chance of fixing things like this quickly.

 Your help is greatly appreciated,
 Nick
 in my case it is still eth0:
 ifconfig
 eth0: flags=4163UP,BROADCAST,RUNNING,MULTICAST  mtu 1500
 inet 192.168.178.21  netmask 255.255.255.0  broadcast
 192.168.178.255
 inet6 fe80::1e6f:65ff:fe87:6f6a  prefixlen 64  scopeid
 0x20link
 ether 1c:6f:65:87:6f:6a  txqueuelen 1000  (Ethernet)
 RX packets 4647305  bytes 6693078055 (6.2 GiB)
 RX errors 0  dropped 0  overruns 0  frame 0
 TX packets 2943816  bytes 226871998 (216.3 MiB)
 TX errors 0  dropped 1 overruns 0  carrier 0  collisions 0

 sys-fs/udev
  Available versions:  (~)168-r2[1] [M]171-r10 197-r8^t{tbz2}
 (~)198-r6^t{tbz2} (~)199-r1^t{tbz2} 200^t{tbz2} **^t {acl
 action_modeswitch build debug doc edd extras +firmware-loader floppy
 gudev hwdb introspection keymap +kmod +openrc +rule_generator selinux
 static-libs test}
  Installed versions:  200^t{tbz2}(18:30:31
 29.03.2013)(firmware-loader gudev hwdb keymap kmod openrc -acl -doc
 -introspection -selinux -static-libs)

 I did keep net.eth0
 Is your eth0 NIC a module (modprobed), or built in the kernel?
 r8169  41918  0
 module

For me its built in.

- Jörg

[gentoo-user] Re: Re: Udev update and persistent net rules changes

[Jörg Schaible]

Neil Bothwick wrote:

 On Wed, 3 Apr 2013 16:38:28 + (UTC), Grant Edwards wrote:
 
  Have you read the news item?
 
 Yes.  I found it rather confusing.
 
 It refers to a new format for rules, but the examples use the exact
 same format as the old rules.
 
 Poor choice of terminology there, the format is the same only the chosen
 namespace is different.
 
 It talks about how 80-net-name-slot.rules needs to be either an empty
 file or a synmlink to /dev/null if you want to disable the new naming
 scheme -- but that doesn't seem to be right.  After the upgrade my
 80-net-name-slot.rules file was neither an empty file nor a symlink to
 /dev/null, but I'm still getting the same old names.
 
 Do you have a 70-persistent-net.rules file? That would override to give
 the old names, which is why the news item tells you to change it

 If the system still has old network interface renaming rules in
 /etc/udev/rules.d, like 70-persistent-net.rules, those will need
 to be either modified or removed.

I don't have any rules except the 80-* one installed by new udev and I still 
have the old names - and this has been the case now for 3 machines and I 
upgrade a 4th right now.

  It explains why the file should be renamed and also why you should
  change the names in the rules to not use ethN.
 
 The only explanation I found was the old way is now deprecated.
 
 My bad, I thought that was covered in the news item, but it is left to
 one of the linked pages to explain it.

- Jörg

[gentoo-user] Re: Udev update and persistent net rules changes

[Jörg Schaible]

Hi,

Grant Edwards wrote:

 On 2013-04-03, Neil Bothwick n...@digimed.co.uk wrote:
 
 Have you read the news item?
 
 Yes.  I found it rather confusing.
 
 It refers to a new format for rules, but the examples use the exact
 same format as the old rules.
 
 It talks about how 80-net-name-slot.rules needs to be either an empty
 file or a synmlink to /dev/null if you want to disable the new naming
 scheme -- but that doesn't seem to be right.  After the upgrade my
 80-net-name-slot.rules file was neither an empty file nor a symlink to
 /dev/null, but I'm still getting the same old names.

same for me. I followed the upgrade guide and removed any 70-* files, 
renamed the net.eth0 link to the new scheme net.enp0s1 just to to find out 
that the kernel could not bring up a network with the such a device. The 
machine booted fine after using eth0 instead again. One a second machine I 
kept eth0 immediately and it booted without problems afterwards.
 
 It explains why the file should be renamed and also why you should
 change the names in the rules to not use ethN.
 
 The only explanation I found was the old way is now deprecated.

And the new name simply did not work.

- Jörg

[gentoo-user] Re: problem starting openvpn after recenet upgrade

[Jörg Schaible]

Joseph wrote:

 I'm having problem starting openvpn after recent upgrade.
 
 When I try to start it I get a normal respond:
 
 openvpn.client_clinic_atom start
   * Starting openvpn.client_clinic_atom ...[ ok ]
   * WARNING: openvpn.client_clinic_atom has started, but is inactive
 
 but ifconfig is not showing this connection in addition it looks strange,
 not as it used to:
 
 ifconfig
 eth1: flags=4163UP,BROADCAST,RUNNING,MULTICAST  mtu 1500
  inet 10.0.0.100  netmask 255.255.255.0  broadcast 10.0.0.255
  inet6 fe80::21b:21ff:feb8:27b6  prefixlen 64  scopeid 0x20link
  ether 00:1b:21:b8:27:b6  txqueuelen 1000  (Ethernet)
  RX packets 16129  bytes 9087618 (8.6 MiB)
  RX errors 0  dropped 0  overruns 0  frame 0
  TX packets 15573  bytes 2667489 (2.5 MiB)
  TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
  device interrupt 16  memory 0xfddc-fdde
 
 lo: flags=73UP,LOOPBACK,RUNNING  mtu 16436
  inet 127.0.0.1  netmask 255.0.0.0
  inet6 ::1  prefixlen 128  scopeid 0x10host
  loop  txqueuelen 0  (Local Loopback)
  RX packets 1218  bytes 1433926 (1.3 MiB)
  RX errors 0  dropped 0  overruns 0  frame 0
  TX packets 1218  bytes 1433926 (1.3 MiB)
  TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
 

Known problem. Rebuild openvpn and it works again. The build adds some hard 
coded paths for some net-tools utilities and the location of ifconfig has 
changed :-/

- Jörg

[gentoo-user] Re: Re: problem starting openvpn after recenet upgrade

[Jörg Schaible]

Matthias Hanft wrote:

 Joseph wrote:

 Yes, you are correct. Rebuilding openvpn worked; thank you.
 
 As far as I have seen, you can set USE=iproute2 for OpenVPN which
 seems to use iproute2 instead of net-tools - should avoid your problem,
 too.

If iprout2 is relocated, it will fail for openvpn exactly the same miserably 
way.

- Jörg

[gentoo-user] xserver update: tilde no longer dead key

[Jörg Schaible]

Hi folks,

it seems the latest xorg-server update also changed the German standard 
keyboard layout, since the tilde is no longer a dead key. I verified that 
the standard variant is still used, I have no ~/.Xkbmap or /etc/X11/Xkbmap 
or (Xmodmap or Xresources), so the standard must have changed.

Where is this defined and how do I get back the normal behavior?

If you're used to a tilde working as dead key, the original key sequence 
like produced for e.g. this annoying example:

rm -rf ~ /tmp

I wand my dead key back!
Anyone?

Cheers,
Jörg

[gentoo-user] Re: xserver update: tilde no longer dead key

[Jörg Schaible]

Jörg Schaible wrote:

 Hi folks,
 
 it seems the latest xorg-server update also changed the German standard
 keyboard layout, since the tilde is no longer a dead key. I verified that
 the standard variant is still used, I have no ~/.Xkbmap or /etc/X11/Xkbmap
 or (Xmodmap or Xresources), so the standard must have changed.
 
 Where is this defined and how do I get back the normal behavior?
 
 If you're used to a tilde working as dead key, the original key sequence
 like produced for e.g. this annoying example:
 
 rm -rf ~ /tmp
 
 I wand my dead key back!
 Anyone?

= % ===
$ diff -u /mnt/backup/floh/system/20121006/tree/usr/share/X11/xkb/symbols/de 
/usr/share/X11/xkb/symbols/de
--- /mnt/backup/floh/system/20121006/tree/usr/share/X11/xkb/symbols/de  
2012-06-23 16:25:07.0 +0200
+++ /usr/share/X11/xkb/symbols/de   2012-11-17 16:22:16.070643609 +0100
@@ -26,7 +26,7 @@
 key AD03 { [ e,  E, EuroSign, EuroSign ] };
 key AD06 { [ z,  Z,leftarrow,  yen ] };
 key AD11 { [udiaeresis, Udiaeresis, dead_diaeresis, dead_abovering ] 
};
-key AD12 { [  plus,   asterisk,   dead_tilde,  dead_macron ] };
+key AD12 { [  plus,   asterisk,   asciitilde,  macron ]  };

   
 key AC02  { [ s,  S,U017F, U1E9E
] };
 key AC07  { [ j,  J,dead_belowdot, 
dead_abovedot   ] };
= % ===

Comparing the current standard with an older backup, explains this change. 
Who is responsible for that?

- Jörg

[gentoo-user] Re: xserver update: tilde no longer dead key

[Jörg Schaible]

Jörg Schaible wrote:

 Jörg Schaible wrote:
 
 Hi folks,
 
 it seems the latest xorg-server update also changed the German standard
 keyboard layout, since the tilde is no longer a dead key. I verified that
 the standard variant is still used, I have no ~/.Xkbmap or
 /etc/X11/Xkbmap or (Xmodmap or Xresources), so the standard must have
 changed.
 
 Where is this defined and how do I get back the normal behavior?
 
 If you're used to a tilde working as dead key, the original key sequence
 like produced for e.g. this annoying example:
 
 rm -rf ~ /tmp
 
 I wand my dead key back!
 Anyone?
 
 = % ===
 $ diff -u
 /mnt/backup/floh/system/20121006/tree/usr/share/X11/xkb/symbols/de
 /usr/share/X11/xkb/symbols/de ---
 /mnt/backup/floh/system/20121006/tree/usr/share/X11/xkb/symbols/de
 2012-06-23 16:25:07.0 +0200
 +++ /usr/share/X11/xkb/symbols/de   2012-11-17 16:22:16.070643609
 +0100 @@ -26,7 +26,7 @@
  key AD03 { [ e,  E, EuroSign, EuroSign ] };
  key AD06 { [ z,  Z,leftarrow,  yen ] };
  key AD11 { [udiaeresis, Udiaeresis, dead_diaeresis, dead_abovering
  ]
 };
 -key AD12 { [  plus,   asterisk,   dead_tilde,  dead_macron ] };
 +key AD12 { [  plus,   asterisk,   asciitilde,  macron ]  };
   
  
  key AC02  { [ s,  S,U017F,
  U1E9E
 ] };
  key AC07  { [ j,  J,dead_belowdot,
 dead_abovedot   ] };
 = % ===
 
 Comparing the current standard with an older backup, explains this change.
 Who is responsible for that?

OK, it has been the update of xkeyboard-config from 2.6 to 2.7, change was 
caused due to this issue:
https://bugs.freedesktop.org/show_bug.cgi?id=9752

Masking the version now. Thanks to Gentoo for this possibility!

- Jörg

[gentoo-user] Re: Brother Scanner

[Jörg Schaible]

Hi Silvio,

Silvio Siefke wrote:

 Hello,
 
 i have a Brother MFC 7320 Printer, Scanner and Fax. Printer is installed
 perfectly, scanner want not work. I try as user and as root, but xsane
 found not the scanner.
 
 gentoo-desk ~ # sane-find-scanner -q
 found USB scanner (vendor=0x04f9, product=0x01eb) at libusb:002:008
 
 gentoo-desk ~ # cat /etc/udev/rules.d/70-libsane.rules
 ATTRS{idVendor}==04f9, ATTRS{idProduct}==01eb, MODE=0664,
 GROUP=scanner, ENV{libsane_matched}=yes
 
 gentoo-desk ~ # cat /etc/group | grep scanner
 scanner:x:102:siefke
 
 Has someone a idea, where i can search the mistake?

I have a Brother DCP-9010CN (also an USB PSC) connected to my server and can 
typically print and scan over network (using cupsd and saned). So I guess 
you should be able to do the same.

Typically you have to install a driver directly from Brother: 
http://welcome.solutions.brother.com/bsc/public_s/id/linux/en/instruction_scn1.html

You can install this RPM on Gentoo with rmp2targz. I have to use for my PSC 
the driver generation 3 and had to install it afterwards with 
/usr/local/Brother/sane/setupSaneScan3 for configuration and 
/usr/bin/brsaneconfig3 -i adds the driver to the SANE configuration. 
Something similar should exist for your driver.

Hope this helps,
Jörg

[gentoo-user] Re: accented characters in KDE apps

[Jörg Schaible]

Philip Webb wrote:

 120226 Volker Armin Hemmann wrote:
 Am Sonntag, 26. Februar 2012, 03:32:58 schrieb Philip Webb:
 In Gvim I can enter 'e-umlaut' via the keystrokes 'control-k e : ', ie 
 ë . Kate has a Vi mode, which seems to reproduce Vim fairly well,
 but there's no sign of 'control-k' or any substitute.
 KDE System Settings has a menu for setting a compose key,
 but tests with 'left-control'  'pause' in Konsole  Kate did nothing.
 I tried in Kate with Luxi Mono  Courier(IBM), Konsole with LM 
 Fixed(Misc). I don't have the KDE desktop pkgs installed, but use many
 apps  it shouldn't make a difference.
 Does anyone know if  how accented characters cb entered in KDE apps ?
 google is your friend.
 http://userbase.kde.org/Tutorials/ComposeKey
 
 Yes, I saw that, but ...
 
 Typing Macrons, Umlauts, Accents, ...
 The compose key will be now whatever you have configured it to be,
 e. g. right logo.
 
 ... as I said above, it doesn't work.
 
 Thanks for trying, but please read the full msg before doing so (smile).

On a German keyboard layout I type AltGr+ü and then the letter: äöüïëÿ

Cheers,
Jörg

[gentoo-user] Re: fonts in gitview

[Jörg Schaible]

Jörg Schaible wrote:

 Hi,
 
 can anybody tell me, what font is used by gitview in the diff pane and how
 to configure it? I did not use gitview for some time and cannot say, when
 this started, but currently I get some kind of strange script font that is
 hardly readable. I am normally working in a KDE4 environment.

Ping. Anyone? I face the same unreadable font in Eclipse's Javadoc View for 
all mono-spaced elements. Where is this system font coming from?

- Jörg

[gentoo-user] Re: KMail

[Jörg Schaible]

Mick wrote:

 On Saturday 11 Feb 2012 07:47:41 Volker Armin Hemmann wrote:
 Am Donnerstag, 9. Februar 2012, 18:33:12 schrieb Daniel D Jones:
  I was one of those bitten by the upgrade to KMail back in October. 
  After spending nearly two days without access to email while trying to
  get the upgrade and migration to work, I gave up and pinned KMail to
  version 4.4.11.1.
  
  With the recent upgrade to KDE, I'm curious if anyone knows whether the
  bugs have been worked out and the KMail upgrade will go smoother now?
 
 I switched a long time ago, so I can't say anything about 'smoother' but
 it does work, as long as you remove all mail ressources and add them back
 one after another.
 
 Unfortunately, it does not work here on my test x86 stable box.  I have
 removed and added each resource more than a dozen times.  POP3 seems to
 broadly work (but it creates duplicate messages which once deleted are
 recreated) and IMAP4 does not show any messages in the Inbox.  The
 kaddressbook does not show the addresses.
 
 I've tried deleting akonadi databases, adding the resources or importing
 them and starting again many, many times to come to the conclusion that in
 my set
 up it just won't work.  I hope that code progress eventually restores
 functionality which I consider absolutely basic - i.e. to be able to read
 my messages and use my addresses.

I successfully imported two installations with IMAP accounts. On both system 
the automated migration failed.

At one installation IMAP server was Courier and the mails were kept all on 
the server, here I had to delete and recreate the accounts.

The other installation used IMAP to access MS Exchange server and my account 
has typically no mail on the server, but all locally. After recreating my 
account, none of my local mail was available anymore. However I simply had 
to move (well, with backup) the contents of the old maildir folders to the 
new one and all I had to do in KMail was to refresh the folders manually. 
That was really fast - especially compared to that lengthly and unsuccessful 
migration attempt.

Local addresses resources had to be recreated, an LDAP resource worked after 
migration.

Nevertheless, it often feels like a 1.0 version, the old 4.4.x line had less 
bugs.

 Having wasted far too much time for no positive result I'm now in hope
 that trinity desktop will come up with a solution that I was sorry to see
 left behind - KDE 3.5.

Yes, some apps in KDE4 are still not at the same level, however, a lot are 
better. Unfortunately it seems those not-yet-at-same level are the more 
essential ones.

Cheers,
Jörg

[gentoo-user] fonts in gitview

[Jörg Schaible]

Hi,

can anybody tell me, what font is used by gitview in the diff pane and how 
to configure it? I did not use gitview for some time and cannot say, when 
this started, but currently I get some kind of strange script font that is 
hardly readable. I am normally working in a KDE4 environment.

Thanks,
Jörg

[gentoo-user] Re: Tcl in your system...

[Jörg Schaible]

Albert W. Hopkins wrote:

 On Tue, 2011-11-29 at 11:33 +0700, Pandu Poluan wrote:
 Just wondering if anyone here ever use Tcl for scripting (i.e.,
 automating repetitive procedures) or even *gasp* serious programming.
 
 At a previous job we had licensed some software that was written in
 TCL.. and even had an API for the system.  I can't recall the name of
 the software at the moment, but it was very specialized so not cheap.
 This was during the late 90s/early 2000s.
 
 I have no idea if the company is still around of if they still sell the
 software or if it's still tcl-based (my memory fails me).

Vignette? Was at that time a major player in CMS market ;-)
Acquired by OpenText in 2009.

- Jörg

[gentoo-user] Re: Fonts.

[Jörg Schaible]

Stayvoid wrote:

 Bad, works immediately: Put them in /usr/share/fonts
 That didn't work for me.

/usr/local/share/fonts

 
 Good: I'm sure that there is a setting for per-user font dirs in ${HOME}
 Could you be more specific?

[gentoo-user] Re: experience with rsnapshot

[Jörg Schaible]

cov...@ccs.covici.com wrote:

 I am using rdiff-backup which is no longer maintained, but still seems
 to work, but I was thinking to use rsnapshot instead which seems like a
 nice way to do this, but this seems not to have been maintained for a
 while, either, so I was wondering if anyone is using it and how it works
 for you?
 
 Thanks in advance for any ideas.

I am using dirvish for several years.

- Jörg

[gentoo-user] Re: what is /var/log/wtmp ?

[Jörg Schaible]

Philip Webb wrote:

 110325 Alex Schuster wrote:
 Philip Webb writes:
 In  /var/log/  there is a file  wtmp , which is  24 MB   owned by  utmp
 Can anyone explain what it's for  whether it cb safely deleted ?
 It tracks logins, you can use the 'last' command to show its contents.
 If wou want to get the space, I suggest compressing the file,
 bzip2 -v /var/log/wtmp
 
 Yes! -- it fell from  24 MB  to  444 KB ! -- most of it is repetitious.
 It's just started a new 'wtmp' after a reboot, good for another  4,5 yr .

Simply install logrotate. It delivers already a setup to clean it up once a 
month automatically.

- Jörg 

[gentoo-user] Re: Re: Strange problem with audio CDs

[Jörg Schaible]

Joerg Schilling wrote:

 Jörg Schaible joerg.schai...@gmx.de wrote:
 
  There is a set of red-book deviations that is called cactus
  datashield
  
  cdparanoia will not extract more than 40 seconds from such a CD
  regardless of the drive you are using.
  
  With cdda2wav you will be able to extract the whole CD as long as the
  drive is not cunfused by the media and as long as you don't run hostile
  software like hald or similar that helps to confuse the drive.

 At least from konsole I can use cdda2wav with normal speed, while the
 ripping KDE apps all seem to fail. Since KDE is still based on hald,
 about what kind of effects you're talking?
 
 The defects seen in cactus datashield will cause many drives to go into
 an endless loop in firmware in case you send the READ TOC command to the
 drive.
 
 Hald was written by people who know too few about CD-ROM handling and as
 hald and other programs (e.g. libGStreamer and cdparanoia) send a READ TOC
 to the drive, they may trigger this problem. The only way out of this
 problem is to mechanically reload the media and not to let any other
 program access the drive except for cdda2wav.

Thanks for taking the time.

- Jörg

[gentoo-user] Re: Strange problem with audio CDs

[Jörg Schaible]

Hi Joe,

Jake Moe wrote:

 On 01/13/11 20:48, J. Roeleveld wrote:
 On Thursday 13 January 2011 11:33:09 Jake Moe wrote:
 On 01/13/11 18:12, J. Roeleveld wrote:

[snip]

 Can you actually play that wav-file? Or is it just a collection of
 garbage?

 As far as I know, CD-Paranoia access the cd-drive a bit more directly
 then other tools. Eg. it approaches it like a CD-ROM, rather then
 CD-Audio.

 The error messages appear as soon as you put the CD into the drive?
 Am wondering if some auto-mounting tool is trying to access it and is
 causing problems here.
 Do you also get those messages when you disable all KDE/Gnome/X/... and
 related stuff?

 Personally, I tend to use cdparanoia and other tools to generate OGG or
 MP3 files and store them on a fileserver and play them from there.

 --
 Joost

 Yeah, the wav file played fine.  At least, it started out fine; I only
 listened to the first 15 - 30 seconds to make sure it sounded ok, and
 then assumed the rest was fine, since nothing else had even gotten that
 far.
 
 And yeah, the errors start as soon as I put the CD in the drive.  What
 automounting tool might I have in FVWM?  I use a pretty basic config
 (which is why I like FVWM, not many frills to muck things up :-P).

My situation improved a bit after upgrading to the latest udev available. 
With 164-r1 I don't have these errors in the log anymore. Unfortunately when 
I try to rip the CD with KDE stuff, the process slows still down until it 
aborts.

However, the situation improved with the CLI tools (even running from with 
KDE). Cdda2Wav takes now 2,5 mins for a 45min CD, cdparanoia takes nearly 
twice, but succeeds also. After reconfiguring abcde to take cdda2wav I can 
at least rip CDs into MP3s in a quite normal time.

- Jörg

[gentoo-user] Re: Strange problem with audio CDs

[Jörg Schaible]

Hi Jörg,

Joerg Schilling wrote:

 Jake Moe jakesaddr...@gmail.com wrote:
 
 Yeah, the wav file played fine.  At least, it started out fine; I only
 listened to the first 15 - 30 seconds to make sure it sounded ok, and
 then assumed the rest was fine, since nothing else had even gotten that
 far.
 
 Be careful!
 
 There is a set of red-book deviations that is called cactus datashield
 
 cdparanoia will not extract more than 40 seconds from such a CD
 regardless of the drive you are using.
 
 With cdda2wav you will be able to extract the whole CD as long as the
 drive is not cunfused by the media and as long as you don't run hostile
 software like hald or similar that helps to confuse the drive.

At least from konsole I can use cdda2wav with normal speed, while the 
ripping KDE apps all seem to fail. Since KDE is still based on hald, about 
what kind of effects you're talking?

- Jörg

[gentoo-user] Re: Re: Strange problem with audio CDs

[Jörg Schaible]

Jake Moe wrote:

 On 01/11/11 04:38, Jörg Schaible wrote:
 Hi Jake,

 Jake Moe wrote:

 I can't seem to get audio CDs to work with my drive.  Data CDs work
 fine, I can mount the filesystem and read them.  Data and Video DVDs
 seem to work fine as well.  But when I try to listen to an audio CD, I
 get the attached errors in log.bz2.  I've tried using things from KsCD
 to cdplay; everything gives the same errors.  Googling seems to indicate
 that there might be a problem with udev somehow, but most of those that
 I find have the fix as update to the latest udev using apt/rpm/other
 binary distro package tool, which obviously won't work for Gentoo.
 Other solutions seem to be update to libATA, but I'm already using
 that.

 I've gone through and tried to check anything obvious in my kernel
 config, but I can't see anything that'd affect it like this.  Also, if I
 reboot into Windows (this laptop is a work computer as well), it plays
 and rips the same CDs just fine.

 Hardware is an HP EliteBook nc6930p laptop.  CD/DVD drive is /dev/sr0.
 Controller is:

 00:1f.2 SATA controller: Intel Corporation ICH9M/M-E SATA AHCI
 Controller (rev 03) (prog-if 01 [AHCI 1.0])
 Subsystem: Hewlett-Packard Company Device 30dc
 Flags: bus master, 66MHz, medium devsel, latency 0, IRQ 46
 I/O ports at 8118 [size=8]
 I/O ports at 813c [size=4]
 I/O ports at 8110 [size=8]
 I/O ports at 8138 [size=4]
 I/O ports at 8000 [size=32]
 Memory at d8426000 (32-bit, non-prefetchable) [size=2K]
 Capabilities: [80] MSI: Enable+ Count=1/16 Maskable- 64bit-
 Capabilities: [70] Power Management version 3
 Capabilities: [a8] SATA HBA ?
 Capabilities: [b0] PCI Advanced Features
 Kernel driver in use: ahci

 Oddly, if I open Konqueror and type in audiocd:/, it lists the tracks,
 and has the FLAC, MP3, Ogg, etc folders.  But it won't play or copy the
 files; it gives the error in error.gif.

 Any other info you need, please let me know.  This is driving me nuts.
 Same for me: http://forums.gentoo.org/viewtopic.php?p=6372251#6372251

 I still have my old box around just because of this problem :-/

 00:1f.2 SATA controller: Intel Corporation 82801JI (ICH10 Family) SATA
 AHCI Controller (prog-if 01 [AHCI 1.0])
 Subsystem: Acer Incorporated [ALI] Device 0198
 Flags: bus master, 66MHz, medium devsel, latency 0, IRQ 64
 I/O ports at c880 [size=8]
 I/O ports at c800 [size=4]
 I/O ports at c480 [size=8]
 I/O ports at c400 [size=4]
 I/O ports at c080 [size=32]
 Memory at fbcfc000 (32-bit, non-prefetchable) [size=2K]
 Capabilities: [80] MSI: Enable+ Count=1/16 Maskable- 64bit-
 Capabilities: [70] Power Management version 3
 Capabilities: [a8] SATA HBA ?
 Capabilities: [b0] PCI Advanced Features
 Kernel driver in use: ahci

 When I rip a CD it typically starts to read it slow permanently down and
 after ~ the 6th song the process is not profgressing anymore ...

 You're also running 64-bit ?

 - Jörg
 Well, mine is a bit different.

Not convinced ;-)

 I typically run FVWM from a SLIM logon,
 so there's no KDE or Gnome auto-anything running.  I only used Konqueror
 as an example of another way of accessing the CDs that might have
 worked, but didn't.   I can even stop XDM, log in from a console prompt
 with no X running, and try to play a CD with cdplay or dcd, and I'll get
 the same results.  And with me, it doesn't start to work and then slow
 down; it never works.  It can only read track listings, but not any of
 the music.

As I said in the forum, I have these log entries running from a pure console 
(no X started at all) even with a stopped hal. It's enough to put an audio 
CD into the drive. Happens also with vanilla kernel. Since 2.6.35 I have the 
message only once though, in the previous two kernels (34+35) they are 
repeated permanently.

 And no, I'm on 32-bit stable Gentoo, with only unstable packages
 being ones that don't have stable ebuilds.

Same for me, just using 64-bit.

 Thanks for trying, though.  :-)  Anyone else have any ideas?

Me, no - unfortunately.

- Jörg

[gentoo-user] Re: Strange problem with audio CDs

[Jörg Schaible]

Hi Jake,

Jake Moe wrote:

 I can't seem to get audio CDs to work with my drive.  Data CDs work
 fine, I can mount the filesystem and read them.  Data and Video DVDs
 seem to work fine as well.  But when I try to listen to an audio CD, I
 get the attached errors in log.bz2.  I've tried using things from KsCD
 to cdplay; everything gives the same errors.  Googling seems to indicate
 that there might be a problem with udev somehow, but most of those that
 I find have the fix as update to the latest udev using apt/rpm/other
 binary distro package tool, which obviously won't work for Gentoo.
 Other solutions seem to be update to libATA, but I'm already using that.
 
 I've gone through and tried to check anything obvious in my kernel
 config, but I can't see anything that'd affect it like this.  Also, if I
 reboot into Windows (this laptop is a work computer as well), it plays
 and rips the same CDs just fine.
 
 Hardware is an HP EliteBook nc6930p laptop.  CD/DVD drive is /dev/sr0.
 Controller is:
 
 00:1f.2 SATA controller: Intel Corporation ICH9M/M-E SATA AHCI
 Controller (rev 03) (prog-if 01 [AHCI 1.0])
 Subsystem: Hewlett-Packard Company Device 30dc
 Flags: bus master, 66MHz, medium devsel, latency 0, IRQ 46
 I/O ports at 8118 [size=8]
 I/O ports at 813c [size=4]
 I/O ports at 8110 [size=8]
 I/O ports at 8138 [size=4]
 I/O ports at 8000 [size=32]
 Memory at d8426000 (32-bit, non-prefetchable) [size=2K]
 Capabilities: [80] MSI: Enable+ Count=1/16 Maskable- 64bit-
 Capabilities: [70] Power Management version 3
 Capabilities: [a8] SATA HBA ?
 Capabilities: [b0] PCI Advanced Features
 Kernel driver in use: ahci
 
 Oddly, if I open Konqueror and type in audiocd:/, it lists the tracks,
 and has the FLAC, MP3, Ogg, etc folders.  But it won't play or copy the
 files; it gives the error in error.gif.
 
 Any other info you need, please let me know.  This is driving me nuts.

Same for me: http://forums.gentoo.org/viewtopic.php?p=6372251#6372251

I still have my old box around just because of this problem :-/

00:1f.2 SATA controller: Intel Corporation 82801JI (ICH10 Family) SATA AHCI 
Controller (prog-if 01 [AHCI 1.0])
Subsystem: Acer Incorporated [ALI] Device 0198
Flags: bus master, 66MHz, medium devsel, latency 0, IRQ 64
I/O ports at c880 [size=8]
I/O ports at c800 [size=4]
I/O ports at c480 [size=8]
I/O ports at c400 [size=4]
I/O ports at c080 [size=32]
Memory at fbcfc000 (32-bit, non-prefetchable) [size=2K]
Capabilities: [80] MSI: Enable+ Count=1/16 Maskable- 64bit-
Capabilities: [70] Power Management version 3
Capabilities: [a8] SATA HBA ?
Capabilities: [b0] PCI Advanced Features
Kernel driver in use: ahci

When I rip a CD it typically starts to read it slow permanently down and 
after ~ the 6th song the process is not profgressing anymore ...

You're also running 64-bit ?

- Jörg

[gentoo-user] Re: Re: Re: Re: Changing boot device with 2.6.36

[Jörg Schaible]

Hi Andrea,

Andrea Conti wrote:

 Had to boot this morning 5 times, since the root device switched
 arbitrarily between sde3 and sdg3
 
 Try disabling CONFIG_SCSI_SCAN_ASYNC (Asynchronous SCSI scanning under
 SCSI options). While it is not a solution, this might somewhat reduce
 the randomness you are experiencing.

I turned of the flag on the kernel line, but it did not help, root is still 
at sde3.

Anyway, thanks for the suggestion,
Jörg

[gentoo-user] Re: Re: Re: Re: Re: Changing boot device with 2.6.36

[Jörg Schaible]

Hi Joost,

J. Roeleveld wrote:

 On Friday 07 January 2011 09:47:28 Jörg Schaible wrote:
 Hi Dale,
 
 Dale wrote:
  Jörg Schaible wrote:
  that approves my tests ... :-/
  
  Had to boot this morning 5 times, since the root device switched
  arbitrarily between sde3 and sdg3 and I've chosen by bad luck always
  the wrong one. It seems there is also some timing involved regarding
  the initialization of the available devices ... sigh.
  
  - Jörg
  
  I had to reboot last night and was in my BIOS looking for other things
  but did notice this feature.  I have a setting in my BIOS that tells it
  what drive to look for to boot first.  It's above the part where you
  tell it to boot CDROM, hard drive, floppy, ZIP and other bootable
  things.  You may want to check and see if you have the same thing. 
  Mine
  is called hard disk boot priority.  I'm not sure this will help but
  it couldn't hurt to check I guess.
 
 The first device to try is my HD and as alternative I can only select the
 CD drive anyway (which is deactivated). At boot time I can still switch
 into a boot menu of the BIOS to select something else.
 
 This will not affect the order the Linux kernel will identify and label
 the devices.
 It will only affect where the BIOS will look for boot-code.
 
 Simply put, the following happens when a PC boots:
 
 1) BIOS goes through its self-check
 
 2) BIOS looks for boot-code on the devices it found in the order
 configured in the BIOS (BIOS -Boot Order)
 
 3) BIOS runs boot-code
 
 4) boot-code starts the boot-loader (GRUB)
 
 5) GRUB loads kernel into memory
 
 6) starts kernel
 
 7) kernel detects drives and assigns them names in order of finding them
 
 At this point, it goes wrong as the drivers are not always identified in
 the same order. From what it looks like, on the OPs system, the
 USB-subsystem is scanned before the SATA-controller.
 The easiest solution to this problem would be to ensure that the
 USB-subsystem is not scanned before the boot-device is identified by the
 kernels boot- process.
 
 This can be achieved by configuring the USB-mass-storage support as a
 module.

This is what I did now and it seems the only setup that actually brings back 
my root on sda3.

 Another option would be to patch the kernel to either support Labels
 natively or to have it include a scan harddisks in following order:
 option which lists which harddisk-drivers (sata/ide/usb) are scanned and
 in which order.

Yep. Maybe LABELs are supported in future ... it would definitely improve 
the situation.

Tanks for your help,
hJörg

[gentoo-user] Re: Re: Re: Re: Changing boot device with 2.6.36

[Jörg Schaible]

Hi Dale,

Dale wrote:

 Jörg Schaible wrote:
 that approves my tests ... :-/

 Had to boot this morning 5 times, since the root device switched
 arbitrarily between sde3 and sdg3 and I've chosen by bad luck always the
 wrong one. It seems there is also some timing involved regarding the
 initialization of the available devices ... sigh.

 - Jörg
 
 I had to reboot last night and was in my BIOS looking for other things
 but did notice this feature.  I have a setting in my BIOS that tells it
 what drive to look for to boot first.  It's above the part where you
 tell it to boot CDROM, hard drive, floppy, ZIP and other bootable
 things.  You may want to check and see if you have the same thing.  Mine
 is called hard disk boot priority.  I'm not sure this will help but it
 couldn't hurt to check I guess.

The first device to try is my HD and as alternative I can only select the CD 
drive anyway (which is deactivated). At boot time I can still switch into a 
boot menu of the BIOS to select something else.
 
 I also noticed something else that may not be related.  I enabled AHCI a
 while back and noticed it will not boot from a CD/DVD when in that
 mode.  I have to set it back to IDE for it to be able to boot from other
 than the hard drive.  Maybe if you set yours to AHCI, it will skip the
 external stuff like USB?

All of this is set to AHCI. And I rather not set this to IDE.

 I'm seriously pulling it out my butt here.  I hope one of these will
 help.  ;-)

What really annoys me is, that I had no problems with this in kernels before 
2.6.35 :-/

Anyway, thanks for your help,
Jörg

[gentoo-user] Re: Re: Re: Re: Re: Changing boot device with 2.6.36

[Jörg Schaible]

Hi Neil,

Neil Bothwick wrote:

 On Fri, 07 Jan 2011 09:47:28 +0100, Jörg Schaible wrote:
 
 What really annoys me is, that I had no problems with this in kernels
 before 2.6.35 :-/
 
 Have you tried diffing the configs to look for a potential cause?

Yes, but I could not detect anything related. It was a simple make 
oldconfig and the diff displays nothing unexpected.

- Jörg

[gentoo-user] Re: Re: Re: Changing boot device with 2.6.36

[Jörg Schaible]

Hi,

Dale wrote:

 Alan McKinnon wrote:
 Apparently, though unproven, at 15:18 on Tuesday 04 January 2011,
 Stroller did opine thusly:


 I found numerous references to this syntax going back to 2005 or
 so, and some major distros seem to use it as the default way of
 describing root= to the kernel.

 http://www.linuxforums.org/forum/redhat-fedora-linux-help/23010-root-
label-
 grub-conf.html
 http://linux.derkeiler.com/Mailing-Lists/RedHat/2005-01/0026.html

 However:
 http://old.nabble.com/Re%3A-Using--%22root%3DLABEL%3D%22-in-
grub.conf-p
 21909347.html http://tinyurl.com/2u4srg4

 Stroller.
  

 All the major distros I've seen it on also use initrds though (rare in
 gentoo- land). I have no idea how it all works, I just know how to type
 it on a RHEL box.

 Elsewhere in the thread someone mentioned that this syntax relies on an
 initrd, and I suspect he may be correct.
 
 I tried using labels with the old grub a while back and it didn't work.
 Labels in fstab works fine tho.  We may have to wait on the new grub to
 get finished.

that approves my tests ... :-/

Had to boot this morning 5 times, since the root device switched arbitrarily 
between sde3 and sdg3 and I've chosen by bad luck always the wrong one. It 
seems there is also some timing involved regarding the initialization of the 
available devices ... sigh.

- Jörg

[gentoo-user] Re: Re: Re: Changing boot device with 2.6.36

[Jörg Schaible]

Hi Alex,

Alex Schuster wrote:

 Dale writes:
 
 Alan McKinnon wrote:
 Apparently, though unproven, at 15:18 on Tuesday 04 January 2011,
 Stroller did opine thusly:

 I found numerous references to this syntax going back to 2005 or
 so, and some major distros seem to use it as the default way of
 describing root= to the kernel.

 http://www.linuxforums.org/forum/redhat-fedora-linux-help/23010-root-
label-
 grub-conf.html
 http://linux.derkeiler.com/Mailing-Lists/RedHat/2005-01/0026.html

 However:
 http://old.nabble.com/Re%3A-Using--%22root%3DLABEL%3D%22-in-
grub.conf-p
 21909347.html http://tinyurl.com/2u4srg4

 Stroller.

 All the major distros I've seen it on also use initrds though (rare in
 gentoo- land). I have no idea how it all works, I just know how to type
 it on a RHEL box.
 
 I am using an initrd, I need it since my root partition is encrypted.
 It's generated and copied to /boot with 'genkernel --install --luks
 --lvm all', but you have to have CLEAN=no in /etc/genkernel.conf or
 genkernel will create its own .config.
 
 Elsewhere in the thread someone mentioned that this syntax relies on an
 initrd, and I suspect he may be correct.
 
 And Stroller's 3rd link also does this.
 
 I tried using labels with the old grub a while back and it didn't work.
 Labels in fstab works fine tho.  We may have to wait on the new grub to
 get finished
 
 I would be surprised if it had this feature. AFAIK grub is already done
 at this stage, the kernel has taken over. And I guess it does not know
 about the LABEL= syntax, and has no code to scan all devices for file
 system labels.

I fear so, too. Grub finds the boot device properly, it's the kernel 
complaining about the value in the root option.

 With an initramfs, the kernel runs an init script which can do various
 stuff, like probing all devices for file system labels.

I never had the need for an initrd.

- Jörg

[gentoo-user] Re: Re: Changing boot device with 2.6.36

[Jörg Schaible]

Hi Stroller,

Stroller wrote:

 
 On 3/1/2011, at 7:36pm, Jörg Schaible wrote:
 ...
 And how does this help the kernel to find the root device where
 /etc/fstab is located ?
 
 The kernel doesn't. You leave that to GRUB.
 
 I'm not saying this helps solve your problem, I'm just sayin'.
 
 BTW: Yes, I will boot next time with a LABEL entry in the kernels boot
 option, but I still don't want a kernel that assigns devices in random
 order.
 
 As long as you can boot, you should seriously stop caring.

That's it: I cannot! At least not always.

 If you're concerned about mounting USB sticks or memory cards then use
 udev rules to distinguish them.

 Regarding the booting, and having to change what's in your grub.conf, I'd
 assume this is a one-off change - you'll change grub.conf to point to the
 new /dev/sdX and that will require no maintenance in the forseeable
 future.

But that's the point: What is X? It is constantly changing with 2.6.36!

 GRUB can do labels, but it needs an initrd or initramfs, I think.

I could not get that working until now ...

- Jörg

[gentoo-user] Re: Changing boot device with 2.6.36

[Jörg Schaible]

Hi Paul,

Paul Hartman wrote:

 On Mon, Jan 3, 2011 at 11:43 AM, Jörg Schaible joerg.schai...@gmx.de
 wrote:
 Hi,

 starting with the 2.6.36-r5 kernel of the Gentoo sources my boot device
 changes. With 2.6.35 and below it is alway /dev/sda3, with the new kernel
 it seems that anything that is internally connected with USB is assigned
 a device first. Since my computer has an internal media bay (and my
 monitor has such a thing also) the first HD moves - it I take care it is
 now /dev/sde3. However, if I forget to switch on the monitor and do this
 later or if an USB stick is already plugged in at boot time, the HD gets
 a different device number again.

 Can somebody else confirm such a behaviour with the 2.6.36 kernel and how
 can this brought back to normal operation?
 
 Is it possible that your BIOS is changing device order? Do you have
 USB device set to boot before HDD device?

No, HD is first. But I'll recheck.

- Jörg

[gentoo-user] Re: Re: Changing boot device with 2.6.36

[Jörg Schaible]

Hi Alan,

Alan McKinnon wrote:

 Apparently, though unproven, at 21:36 on Monday 03 January 2011, Jörg
 Schaible did opine thusly:
 
 Nikos Chantziaras wrote:
  On 01/03/2011 07:43 PM, Jörg Schaible wrote:
  Hi,
  
  starting with the 2.6.36-r5 kernel of the Gentoo sources my boot
  device changes. With 2.6.35 and below it is alway /dev/sda3, with the
  new kernel it seems that anything that is internally connected with
  USB is assigned a device first. Since my computer has an internal
  media bay (and my monitor has such a thing also) the first HD moves -
  it I take care it is now /dev/sde3. However, if I forget to switch on
  the monitor and do this later or if an USB stick is already plugged in
  at boot time, the HD gets a different device number again.
  
  Can somebody else confirm such a behaviour with the 2.6.36 kernel and
  how can this brought back to normal operation?
  
  This has been solved long ago: Label your filesystems and mount them by
  label.
 
 I did this long ago, therefore I can switch between the old and new
 kernel easily.
 
  For example, don't put /dev/sda3 in your fstab, but label that
  filesystem with a name like root_fs and use
  /dev/disk/by-label/root_fs in fstab.
  
  Ext2/3/4 filesystem can be labeled with the e2label tool.  For
  example:
 e2label /dev/sda3 root_fs
  
  After that, modify your fstab accordingly.
 
 And how does this help the kernel to find the root device where
 /etc/fstab is located ?
 
 Does
 
 boot=LABEL=boot_device_label
 
 in grub config work for you?

I hoped so, but actually no. Grub complains at boot time not finding the 
root device. Is this available in the grub-0.97 series at all?

- Jörg

[gentoo-user] Re: Changing boot device with 2.6.36

[Jörg Schaible]

Hi Joost,

J. Roeleveld wrote:

 On Monday 03 January 2011 18:43:28 Jörg Schaible wrote:
 Hi,
 
 starting with the 2.6.36-r5 kernel of the Gentoo sources my boot device
 changes. With 2.6.35 and below it is alway /dev/sda3, with the new kernel
 it seems that anything that is internally connected with USB is assigned
 a device first. Since my computer has an internal media bay (and my
 monitor has such a thing also) the first HD moves - it I take care it is
 now /dev/sde3. However, if I forget to switch on the monitor and do this
 later or if an USB stick is already plugged in at boot time, the HD gets
 a different device number again.
 
 Can somebody else confirm such a behaviour with the 2.6.36 kernel and how
 can this brought back to normal operation?
 
 - Jörg
 
 One way to avoid USB-devices to be picked up before the kernel picks its
 boot- device is to put the USB-stuff as modules and have them loaded
 later.
 
 I haven't found a way to delay usb-device detection yet.

If nothing else helps ... :-/

- Jörg