Re: Removing libsystemd0 from a non-systemd system

[tomas]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Sun, May 06, 2018 at 02:44:16AM +, David Griffith wrote:
> 
> Have any advances been made in figuring out just how to remove
> libsystemd0 from a Debian 9 machine that's running sysvinit?  The
> ongoing presence of libsystemd0 has caused slowly-progressing
> trouble with several machines of mine culminating in complete
> failure a couple days ago.  Initially I thought this was unrelated
> to systemd, but now I tracked it down to systemd's remnants and the
> problem is progressing much faster with freshly-installed machines.

Could you be a bit more precise?

At least one example where libsystemd0 is causing problems would
be enlightening. Pointers welcome.

Cheers
- -- tomás
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlruogQACgkQBcgs9XrR2katOgCfUJNU4QWrXkzsn4AJWEsNITfX
x8wAn0sszJyo1Qnxze12dVd83An17598
=wLyA
-END PGP SIGNATURE-

Re: Distinguishing among [unmount, Safely Remove Drive, Eject]

[Richard Hector]

On 06/05/18 07:35, Brian wrote:
> On Sat 05 May 2018 at 11:06:25 -0500, Richard Owlett wrote:
> 
>> What are the distinguishing features of unmount, Safely Remove Drive, and
>> Eject?
> 
> There are none. The device is either unmounted or it isn't. It cannnot
> be half-unmounted.

Hmm. Is there not a point where it's been made inaccessible to the
filesystem, but caches are not yet flushed?

Richard




signature.asc
Description: OpenPGP digital signature

Removing libsystemd0 from a non-systemd system

[David Griffith]

Have any advances been made in figuring out just how to remove libsystemd0 
from a Debian 9 machine that's running sysvinit?  The ongoing presence of 
libsystemd0 has caused slowly-progressing trouble with several machines of 
mine culminating in complete failure a couple days ago.  Initially I 
thought this was unrelated to systemd, but now I tracked it down to 
systemd's remnants and the problem is progressing much faster with 
freshly-installed machines.



--
David Griffith
d...@661.org

A: Because it fouls the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

Re: KDE 4.4.3 in unstable

[Martin Steigerwald]

Am Donnerstag 06 Mai 2010 schrieb Modestas Vainius:
> Hello,
> 
> On ketvirtadienis 06 Gegužė 2010 23:48:43 Boyd Stephen Smith Jr. wrote:
[...]
> > >  YOU assume that your truth is an ultimate one.
> > 
> > Not really.  I made objective statement about KMail based on
> > observable facts. I also voiced an opinion that I based on that
> > statement.
> 
> What facts? That kdepim/kmail needs akonadi? That's hardly news.
> Akonadi integration is not stable enough? There is still time to
> improve it. That's what Ana's blog was about.
> 
> > >  If YOU have so
> > >  many problems with particular piece of software, look for better
> > >  options or read the first part of this mail again.
> > 
> > I have problems with a very narrow selection of selection of the
> > software. Specifically, I don't want to need MySQL installed in
> > order to use KMail effectively in Debian stable.
> 
> It is not like every mail client on the market suddenly needs MySQL.
> Actually, kmail is probably unique in this area.

Again. AFAIK KMail doesn't use Akonadi at all. Its the new KAddressBook 
that based on Akonadi. And since KMail uses it for address completion...

For the user it looks as if KMail uses Akonadi, but to the best of my 
knowledge it doesn't - at least not directly. Otherwise there wouldn't be 
much point in the akonadi- / nepomuk-based rewrite of KMail as KMail 2.

> > There are a number of solutions to this.  Newer Akonadi should run on
> > non- MySQL data stores.  Older KMail doesn't talk to Akonadi. 
> > Patches could be applied to either.  Stable could include software
> > from multiple KDE releases, as has been done before.
> 
> What I tried to say to you all this time, if Akonadi renders Kmail
> unusable to you, switch the client! Akonadi by itself is not a bug, it
> is not going away! Take it or switch to something else, simple as
> that.

And I want to add: It does not render it unusable for everyone.

Even when Akonadi does not start up cleanly, which happens on first start 
of Kontact for example as I reported already, I still *can* use KMail. Its 
just that the address book function do not work properly then. Although I 
didn't even verify this. Cause even when I get the Akonadi startup errors, 
KAddressBook shows the list of the people in my address book, its just 
ghosted out and not clickable. So maybe address completion would still 
work. But even if it doesn't, KMail as a whole actually does still work 
then.

Anyway, most important is to file and an all bugs with Akonadi in KDEPIM 
upstream ASAP. And that is where I suggest that people - including you 
Boyd - put their energy now.

Ciao,
-- 
Martin 'Helios' Steigerwald - http://www.Lichtvoll.de
GPG: 03B0 0D6C 0040 0710 4AFA  B82F 991B EAAC A599 84C7


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

Re: color unstable under jessie

[Gene Heskett]

On Saturday 05 May 2018 18:11:18 tom arnall wrote:

> Greetings!
>
> over the last few weeks i've been having trouble on my system with
> color. the color on the screen becomes very unstable, and keeps going
> into a state where most of the color is pink. has anyone else had this
> problem.
>
> Regards,
>
> Tom Arnall

This, generally, is not a debian problem. Speaking as a C.E.T. with 60+ 
years in tv service, and broadcast engineering, first, I'd disconnect 
the monitor and reconnect it to see if theres a bad connection there.

Failing that, I'd remove and reseat the video card if its not part of the 
motherboard. While you have it out, make sure the fan spins freely, and 
verify it runs when re-installed & powered up. Third, try a different 
monitor. Somewhere in that chain, between the digital to initial rgb 
conversion, all the way to the crt or lcd, is a flaky connection. Could 
even be a microscopic crack in a solder joint on the motherboard. Those 
are hell to find.



-- 
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

Re: color unstable under jessie

[Abdullah Ramazanoğlu]

On Sat, 5 May 2018 15:11:18 -0700 tom arnall said:

> the color on the screen becomes very unstable, and keeps going into a state 
> where most of the color is pink.

My monitor sometimes tends to get fluctuating, weak, or lost blue signal, which
results in yellowish tendency. It is probably a connector issue in the casing,
though I haven't opened it up as yet. I solve the issue by a couple of taps at
the back of the monitor, for another 3 months without issues. :D

Regards
-- 
Abdullah Ramazanoğlu

Re: color unstable under jessie

[David Wright]

On Sat 05 May 2018 at 15:11:18 (-0700), tom arnall wrote:
> over the last few weeks i've been having trouble on my system with
> color. the color on the screen becomes very unstable, and keeps going
> into a state where most of the color is pink. has anyone else had this
> problem.

My only experience of this happening was caused by a poor contact
where the VGA cable plugs into the computer. Diagnosed by wiggling it.

Cheers,
David.

Re: color unstable under jessie

[Felix Miata]

tom arnall composed on 2018-05-05 15:11 (UTC-0700):

> over the last few weeks i've been having trouble on my system with
> color. the color on the screen becomes very unstable, and keeps going
> into a state where most of the color is pink. has anyone else had this
> problem.

IME that's almost certainly a hardware problem, either

1-poor cable connection
2-deteriorated or otherwise defective cable
3-bad solder joint on display's connector
4-other (internal) display device failure

When was the display made? What brand? What model?

It could be something a $0.40 part would fix. http://www.badcaps.net/
-- 
"Wisdom is supreme; therefore get wisdom. Whatever else you
get, get wisdom." Proverbs 4:7 (New Living Translation)

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/

Re: color unstable under jessie

[Patrick Wiseman]

On Sat, May 5, 2018, 6:11 PM tom arnall  wrote:

> Greetings!
>
> over the last few weeks i've been having trouble on my system with
> color. the color on the screen becomes very unstable, and keeps going
> into a state where most of the color is pink. has anyone else had this
> problem.
>

I had that happen some years ago, before my screen failed completely. That
laptop has since been a headless server!

Patrick


>

color unstable under jessie

[tom arnall]

Greetings!

over the last few weeks i've been having trouble on my system with
color. the color on the screen becomes very unstable, and keeps going
into a state where most of the color is pink. has anyone else had this
problem.

Regards,

Tom Arnall

Re: messed up release in apt

[Anil Duggirala]

I have removed the /etc/apt/trusted.gpg and that removed the error from apt 
update.
I will try now adding correct stretch-backports and installing the packages I  
require,
thanks a lot, 

On Tue, May 1, 2018, at 12:29 PM, Brian wrote:
> On Tue 01 May 2018 at 09:18:56 -0500, David Wright wrote:
> 
> > On Tue 01 May 2018 at 08:11:28 (-0500), Anil Duggirala wrote:
> > > > If it were my machine (so that if I sank it I would be the only one
> > > > to
> > > > go down with the ship), I might run:
> > > > 
> > > >  'apt-key update'
> > > > 
> > > 
> > > When running that command I am getting :
> > > Warning: 'apt-key update' is deprecated and should not be used anymore!
> > > Note: In your distribution this command is a no-op and can therefore be
> > > removed safely.
> > > 
> > > > after removing 
> > > > 
> > > >  '/etc/apt/trusted.gpg' 
> > > > 
> > > When you say removing you mean :
> > > rm /etc/apt/trusted.gpg ?
> > > 
> > > I appreciate any other alternative procedure to correct this,
> > > thanks,
> > 
> > I don't know if this rather long thread would help. It does appear
> > that /etc/apt/trusted.gpg is no longer used, in favour of
> > /etc/apt/trusted.gpg.d/ but if the former exists, it can cause
> > problems. There's a new user on stretch called _apt but I don't
> > know its function.
> 
> A security enhancement. _apt is an unprivileged user, allowing
> sandboxed downloading. The user exists to protect against bugs
> in the http protocol handler, ssl libraries, compressors, etc.
> It also protects against permission issues elsewhere.
> 
> -- 
> Brian.
> 

Re: Distinguishing among [unmount, Safely Remove Drive, Eject]

[Brian]

On Sat 05 May 2018 at 11:06:25 -0500, Richard Owlett wrote:

> What are the distinguishing features of unmount, Safely Remove Drive, and
> Eject?

There are none. The device is either unmounted or it isn't. It cannnot
be half-unmounted.

-- 
Brian.

Re: Distinguishing among [unmount, Safely Remove Drive, Eject]

[Reco]

Hi.

On Sat, May 05, 2018 at 08:41:21PM +0200, Thomas Schmitt wrote:
> Hi,
> 
> i wrote:
> > > CAP_DESTROY_HARDWARE and CAP_ENDANGER_BYSTANDERS.
> 
> Reco wrote:
> > Please. Surely you heard of SATA/SAS hotplug? Or SATA rack?
> 
> Plus a robot arm which performs the mechanical part ?

EMC does it. Price could be lower though, as AFAIK you have to buy
Symmetrix (sp?) for the feature.


> But i rather thought of a 3.5 inch steel frisbee hopping out of its
> casing at 7200 rpm.

It's either glass or aluminium these days. Could still cause some damage
if fractured before the ejection ☺.

Reco

Re: pkg-config: required as dep so often - How and where to report?

[deloptes]

Alexander Traud wrote:

> Or am I on the complete wrong track and this is not even a minor bug?

IMO it is not really a bug, because I do not need pkg-config, to build some
peace of code. pkg-config is needed in case you use the build scripts that
require pkg-config, but the developer could not know which way you will go,
so I think it is OK to not have it anywhere.
Installing a *dev package suggest that you will use the header file. It does
not suggest that you will use build scripts that require pkg-config. For
example cmake projects offer macros that obsolete pkg-config and the
developer do not know what exactly you will use to build your project.

I see some developers put it in as suggested, depends or alike, but what is
exactly the problem with not having it somewhere. The fun with Linux is
that it is an adventure that never ends.
You can of course request putting it in as suggested for example.

regards

Re: Distinguishing among [unmount, Safely Remove Drive, Eject]

[Thomas Schmitt]

Hi,

i wrote:
> > CAP_DESTROY_HARDWARE and CAP_ENDANGER_BYSTANDERS.

Reco wrote:
> Please. Surely you heard of SATA/SAS hotplug? Or SATA rack?

Plus a robot arm which performs the mechanical part ?

But i rather thought of a 3.5 inch steel frisbee hopping out of its
casing at 7200 rpm. And even if it's a mechanically harmless SSD:
Who will sweep up all the popped-off memory chips ?


Have a nice day :)

Thomas

Re: Distinguishing among [unmount, Safely Remove Drive, Eject]

[Reco]

Hi.

On Sat, May 05, 2018 at 05:59:15PM +0100, Brad Rogers wrote:
> On Sat, 5 May 2018 11:06:25 -0500
> Richard Owlett  wrote:
> 
> Hello Richard,
> 
> >Why?
> 
> Probably not an exhaustive list
> 
> Ever tried to eject an HD?

Yup. Did it a week ago last time, I was changing yet another NAS SATA
drive.

> (Un)mount is the only thing that makes sense here.

Actually, no. For instance, this particular PC says to me that it's
possible to remove two HDs from the running kernel (powering down it
effectively):

$ ls -al /sys/block/sd*/device/delete
--w--- 1 root root 4096 May  5 21:00 /sys/block/sda/device/delete
--w--- 1 root root 4096 May  5 21:00 /sys/block/sdb/device/delete

But, considering *who* designed DE in question, it may be for the best
that users are oblivious about this particular feature ;).


> Safely Remove is to ensure writes to flash drives etc. are actually
> completed as opposed to being in progress from buffers.  Only then is it
> safe to remove the memory card.

In other words, perform umount(2). How is it different from Unmount is
anyone's guess.

Reco

Re: Distinguishing among [unmount, Safely Remove Drive, Eject]

[Reco]

Hi.

On Sat, May 05, 2018 at 07:04:08PM +0200, Thomas Schmitt wrote:
> Hi,
> 
> Richard Owlett wrote:
> > unmount, Safely Remove Drive, and Eject?
> > [...] For a hard disk partition, only the first is given.
> 
> To eject a hard disk you need the undocumented privileges CAP_DESTROY_HARDWARE
> and CAP_ENDANGER_BYSTANDERS.

Please. Surely you heard of SATA/SAS hotplug? Or SATA rack?

It's true that it's usually impossible to *eject* the harddrive.
Does not mean that it's impossible to replace a harddrive without
powering down. 

Reco

Re: Distinguishing among [unmount, Safely Remove Drive, Eject]

[Brad Rogers]

On Sat, 5 May 2018 11:06:25 -0500
Richard Owlett  wrote:

Hello Richard,

>Why?

Probably not an exhaustive list

Ever tried to eject an HD?  (Un)mount is the only thing that makes sense
here.

Safely Remove is to ensure writes to flash drives etc. are actually
completed as opposed to being in progress from buffers.  Only then is it
safe to remove the memory card.

Eject will open a CD/DVD/BluRay drive tray, or actually eject the disk
from a slot drive.

-- 
 Regards  _
 / )   "The blindingly obvious is
/ _)radnever immediately apparent"
What will you do when the gas taps turn?
The Gasman Cometh - Crass


pgpBi1Wz4wbCp.pgp
Description: OpenPGP digital signature

Re: Distinguishing among [unmount, Safely Remove Drive, Eject]

[Thomas Schmitt]

Hi,

Richard Owlett wrote:
> unmount, Safely Remove Drive, and Eject?
> [...] For a hard disk partition, only the first is given.

To eject a hard disk you need the undocumented privileges CAP_DESTROY_HARDWARE
and CAP_ENDANGER_BYSTANDERS.


Have a nice day :)

Thomas

Distinguishing among [unmount, Safely Remove Drive, Eject]

[Richard Owlett]

What are the distinguishing features of unmount, Safely Remove Drive, 
and Eject?


The question is prompted by observing that for partitions on USB flash 
drives which have been auto-mounted, one or both of the last two are 
listed when clicking on the icon associated with a partition. For a hard 
disk partition, only the first is given.


Why?

Re: Debian 9 t: Update Gstreamer Base plugins package to resolve a Gstreamer bug

[Liam O'Toole]

On 2018-05-05, Henrique de Moraes Holschuh  wrote:
> On Fri, 04 May 2018, Dinesh Iyer wrote:
>> "You'll have to talk to whoever is providing you your older version of
>> GStreamer. They will have to backport the fix to that old version, but it
>> should just cleanly apply to the older versions."
>> 
>> The patch that needs to be applied is:
>> https://github.com/GStreamer/gst-plugins-base/commit/9f9000e693694ea33c21607140ffc29aa1734062#diff-acf6f1abae3d918458c7a91ce8a0b3ad
>> 
>> Is this something that would be possible to do? This would really help me
>> as this stall is a show stopper for my application.
>
> Talking in the general way, yes, it is possible.  I am not one of the
> DDs responsible for gstream, though.  So all I can tell you is how it
> would go for one of the packages I am responsible for:
>
> Since a stable update is needed, one has to go the long way to get the
> Debian stable release manager to approve it (it doesn't depend only on
> the DD responsible for the package).
>
> First, you get the fix into unstable (either through a patch, or by
> ensuring an already fixed version is present).  And wait for it to
> migrate to testing (usually, five-seven days).

[...]

At this point wouldn't it be easier to get the fix into
stretch-backports?

Re: Debian 9 t: Update Gstreamer Base plugins package to resolve a Gstreamer bug

[Henrique de Moraes Holschuh]

On Fri, 04 May 2018, Dinesh Iyer wrote:
> "You'll have to talk to whoever is providing you your older version of
> GStreamer. They will have to backport the fix to that old version, but it
> should just cleanly apply to the older versions."
> 
> The patch that needs to be applied is:
> https://github.com/GStreamer/gst-plugins-base/commit/9f9000e693694ea33c21607140ffc29aa1734062#diff-acf6f1abae3d918458c7a91ce8a0b3ad
> 
> Is this something that would be possible to do? This would really help me
> as this stall is a show stopper for my application.

Talking in the general way, yes, it is possible.  I am not one of the
DDs responsible for gstream, though.  So all I can tell you is how it
would go for one of the packages I am responsible for:

Since a stable update is needed, one has to go the long way to get the
Debian stable release manager to approve it (it doesn't depend only on
the DD responsible for the package).

First, you get the fix into unstable (either through a patch, or by
ensuring an already fixed version is present).  And wait for it to
migrate to testing (usually, five-seven days).

Then, you need to also test the fix throughoutly, as applied to the
Debian stable codebase:  you'd get the current source packages for
gstreamer in Debian stable, apply the required patches, and do a lot of
testing to ensure it caused no regressions _and_ fixed the issue
properly.

Then you file a bug, severity important or above, with a full
description of the issue, how to recreate it, why it is really important
to have it fixed.

You also describe in that bug report all the testing the change has seen
(how much time since it is upstream, how much time it has been on Debian
unstable and testing, etc).  You describe all the testing you did on the
modified Debian stable package with the fix.  You describe the results
of this testing.

This bug report is *IMPORTANT*, and yes, it is a lot of work.  But don't
shortcut it: how well you do this is directly related to the chances of
the stable update being approved by the release manager.  I know that
from experience (it is no easier for a DD to get a stable update
approved, we need to do all that work as well).

Please title the bug report with "stable update request: " or
something like that.  The maintainer can then provide extra info to the
bug report, retitle and reassign it to the stable release team as
required per Debian procedures.

-- 
  Henrique Holschuh

Re: pkg-config: required as dep so often - How and where to report?

[Alexander Traud]

>> libsrtp2-dev needs pkg-config only when I compile it myself.
> The libsrtp2-dev package surely both depends and build-depends on
> pkg-config.

Yes, the package description (debian/control) does. The created files do
not. I can compile, link, and run with libsrtp2.so without pkg-config.

Actually for this package, the result of pkg-config is wrong (--cflags
is empty, --libs lists -lpcap unnecessarily). However, that is another
issue (although it might be the cause why it was declared as Depends).

> [For libgtk-3-dev y]ou cannot really expect anyone to get this right
> without pkg-config, and so libgtk-3-dev depends on pkg-config.

Yes, my assumption might not be right for every package in that list.
Therefore, I would love to see someone more experienced taking over this
issue. Your argument is a strong argument for at least Suggests.
However, even in this example, pkg-config is not required. In such a
case, one can argue for Recommends instead of Suggests. Nevertheless,
I would not go for Depends.

That is what I am about. Perhaps someone sees a pattern and comes up
with a general rule. For example: "Every package with a pkgconfig/*.pc
file has to Suggest pkg-config†. If the shared library does not declare
its dependencies itself, and additional libraries are required to run
the shared library, pkg-config has to be Recommends. pkg-config should
never be Depends because the developer might have other means to build."
Again, the last sentence could be wrong if there is a dedicated config
tool in the package itself, which actually relies on pkg-config.

Another example: libxml2-dev » libicu-dev » libicu-le-hb-dev »
libharfbuzz-dev » libglib2.0-dev » pkg-config
Only because of the dependency graph, pkg-config must be installed.
Again, pkg-config is not required to compile/link against libxml2.
Actually, the package itself declares pkg-config just as Suggests.
Perhaps this is a bad example as well, because there is a bug in the
dependency graph. Anyway from that 250 packages, I have just looked
in-depth into libxml2 and libsrtp2, yet.

† That part of "my" rule would affect another several hundred packages.

Re: Question about Running rsnapshot

[Andrew McGlashan]

Hi,

On 05/05/18 07:45, Martin McCormick wrote:
>   I just realized that I goofed when I wrote the name of
> the application that combines multiple drives in to one large
> drive.  I meant
> mhddfs for example:
> mhddfs /rsnapshot1,/rsnapshot2 /var/cache/rsnapshot -o mlimit=100M 
> >/dev/null 2>&1 

Okay, that explains it.  It seems that the underlying file systems are
independent from each other in most respects.  The virtual presentation
is, interesting, but I wouldn't be using this setup for backups even
though it might be useful for other types of use.


https://romanrm.net/mhddfs

Cheers
A.



signature.asc
Description: OpenPGP digital signature