Re: questions about cron.daily

[Max Nikulin]

On 09/04/2023 14:54, Michel Verdier wrote:

Le 8 avril 2023 Max Nikulin a écrit :

There is ready to use one: /usr/lib/systemd/user/emacs.service Perhaps there
is no such file in buster.


/usr/lib/systemd/user is for global system running. If you want to change
something in the service you copy it in ~/.config/systemd to supersede
global one.


For minor tuning (adding new parameters or overwriting specific ones) it 
is possible to create a drop-in file, e.g. 
~/.config/systemd/user/emacs.service.d/override.conf (or in /etc for all 
users).


systemctl --user edit emacs.service

See systemd.unit(5). Examples are usually given for system, not user units:
- https://wiki.debian.org/systemd#Creating_or_altering_services
- https://wiki.archlinux.org/title/systemd#Editing_provided_units
- 
https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/8/html/configuring_basic_system_settings/assembly_working-with-systemd-unit-files_configuring-basic-system-settings#proc_extending-the-default-unit-configuration_assembly_working-with-systemd-unit-files

Re: questions about cron.daily

[David Wright]

On Mon 10 Apr 2023 at 17:39:57 (+0200), zithro wrote:
> On 10 Apr 2023 03:23, David Wright wrote:
> > On Sun 09 Apr 2023 at 21:48:22 (+0200), zithro wrote:
> > > > IOW, while I run crontab -e on bookworm, inside my emacs session,
> > > > I want a subshell to run crontab -l, but the latter has to run on
> > > > bullseye in order to pick up the old crontab. I'm not sure how
> > > > I would do that.
> > > 
> > > Try running :
> > > ssh user@bullseye crontab -l
> > > 
> > > It will locally list the crontab from remote user "user".
> > > 
> > > Note I've never used emacs, so dunno if ssh is allowed !
> > 
> > In case it's not clear, bullseye and bookworm are Debian distribution
> > codenames, not hostnames.
> 
> In case it's not clear, to distinguish hosts in help messages, it's
> easy to refer to a host using its distro/codename. I have no idea how
> you name your hosts, nor is it useful for the conversation. I thought
> you could do the name translation by yourself.

In case it's not clear, to distinguish root filesystems in help
messages, it's easy to refer to a rootfs using its distro/codename.

For hostnames, I tend to follow RFCs 1178 and 2100. :)

Cheers,
David.

Re: questions about cron.daily

[David Wright]

On Mon 10 Apr 2023 at 08:31:16 (+0200), Michel Verdier wrote:
> Le 10 avril 2023 David Wright a écrit :
> 
> > In case it's not clear, bullseye and bookworm are Debian distribution
> > codenames, not hostnames. I can't edit my crontab on a newly installed
> > bookworm system while simultaneously listing my old crontab on the old
> > bullseye system on the same computer.
> >
> > The machine is set up to dual boot (currently bullseye and buster),
> > but not simultaneously!
> 
> You can boot on one system and mount the other system partition to
> easily compare both.

You're right, though technically that depends on your friendly
sysadmin to mount it. (Of course, typically, that's me.) But
it doesn't address the other advantages of a separate file.

Cheers,
David.

Re: questions about cron.daily

[zithro]

On 10 Apr 2023 03:23, David Wright wrote:

On Sun 09 Apr 2023 at 21:48:22 (+0200), zithro wrote:

IOW, while I run crontab -e on bookworm, inside my emacs session,
I want a subshell to run crontab -l, but the latter has to run on
bullseye in order to pick up the old crontab. I'm not sure how
I would do that.


Try running :
ssh user@bullseye crontab -l

It will locally list the crontab from remote user "user".

Note I've never used emacs, so dunno if ssh is allowed !


In case it's not clear, bullseye and bookworm are Debian distribution
codenames, not hostnames.


In case it's not clear, to distinguish hosts in help messages, it's easy 
to refer to a host using its distro/codename. I have no idea how you 
name your hosts, nor is it useful for the conversation. I thought you 
could do the name translation by yourself.



I can't edit my crontab on a newly installed
bookworm system while simultaneously listing my old crontab on the old
bullseye system on the same computer.
The machine is set up to dual boot (currently bullseye and buster),
but not simultaneously!


Missed that information. So it's even easier.
Mount the /var partition of bullseye on bookworm.
Go to /MOUNTPOINT/var/spool/cron/crontabs
Done.

Re: questions about cron.daily

[Michel Verdier]

Le 10 avril 2023 David Wright a écrit :

> In case it's not clear, bullseye and bookworm are Debian distribution
> codenames, not hostnames. I can't edit my crontab on a newly installed
> bookworm system while simultaneously listing my old crontab on the old
> bullseye system on the same computer.
>
> The machine is set up to dual boot (currently bullseye and buster),
> but not simultaneously!

You can boot on one system and mount the other system partition to
easily compare both.

Re: questions about cron.daily

[davidson]

On Sun, 9 Apr 2023 David Wright wrote:

On Sun 09 Apr 2023 at 21:48:22 (+0200), zithro wrote:

[Previously David "Between-the-Lines" Wright wrote:]

IOW, while I run crontab -e on bookworm, inside my emacs session,
I want a subshell to run crontab -l, but the latter has to run on
bullseye in order to pick up the old crontab. I'm not sure how I
would do that.


"So it's a good thing I don't need to, since I've got all the
materials I need in my home directory, under ~/.cron"


Try running :
ssh user@bullseye crontab -l

It will locally list the crontab from remote user "user".

Note I've never used emacs, so dunno if ssh is allowed !


I too saw a plea for halp at first. At first I was derping out a reply
"Well David, since the bullseye system isn't running, you just
mount..."

But nope. Mirage!

Just rhetorical puzzlement implying an unspoken conclusion: "And
*that's* why we keep all the cron things on our home partition."

--
Hackers are free people. They are like artists. If they are in a good
mood, they get up in the morning and begin painting their pictures.
-- Vladimir Putin

Re: questions about cron.daily

[David Wright]

On Sun 09 Apr 2023 at 21:48:22 (+0200), zithro wrote:
> > IOW, while I run crontab -e on bookworm, inside my emacs session,
> > I want a subshell to run crontab -l, but the latter has to run on
> > bullseye in order to pick up the old crontab. I'm not sure how
> > I would do that.
> 
> Try running :
> ssh user@bullseye crontab -l
> 
> It will locally list the crontab from remote user "user".
> 
> Note I've never used emacs, so dunno if ssh is allowed !

In case it's not clear, bullseye and bookworm are Debian distribution
codenames, not hostnames. I can't edit my crontab on a newly installed
bookworm system while simultaneously listing my old crontab on the old
bullseye system on the same computer.

The machine is set up to dual boot (currently bullseye and buster),
but not simultaneously!

Even for the same username, the crontab on one computer differs from
that on another, as the machines have different roles.

Cheers,
David.

Re: questions about cron.daily

[zithro]

IOW, while I run crontab -e on bookworm, inside my emacs session,
I want a subshell to run crontab -l, but the latter has to run on
bullseye in order to pick up the old crontab. I'm not sure how
I would do that.


Try running :
ssh user@bullseye crontab -l

It will locally list the crontab from remote user "user".

Note I've never used emacs, so dunno if ssh is allowed !

Re: questions about cron.daily

[David Wright]

On Thu 06 Apr 2023 at 18:54:31 (-0400), Greg Wooledge wrote:
> On Thu, Apr 06, 2023 at 05:45:08PM -0500, David Wright wrote:
> > Users (including root) write their crontabs anywhere they like,
> > typically in a directory like ~/.cron/.
> 
> Is that... normal?  I can't say I've ever seen anyone keep a private
> copy of their crontab in their home directory like that.

Well, it's pretty normal if you use the first form of the command,
though there's no special need for it to be either in the home
directory or private.

> Most people just use "crontab -e" to edit the system's copy of their
> personal crontab...
> 
> > They then have to be installed
> > with crontab, which copies them into /var/spool/cron/crontabs/.
> 
> ... which lives there.

That's the workflow I might have used thirty years ago when I was
a plain old user of a university unix system with likely zero to
one lines of crontab and no need to think about backups. But that's
the old normal, which I'd find rather inflexible now. For example,
if you're busy editing crontab and it's time to go home, then you
either save it and it immediately becomes the active copy, or you
abandon editing, or you save it to some file before abandoning.
Say what? Isn't that a private copy?

Next month, or whenever, I'll be installing bookworm into the other
root partition on this machine. I'll want to copy (and preen, maybe)
my personal crontab from bullseye into bookworm.

IOW, while I run crontab -e on bookworm, inside my emacs session,
I want a subshell to run crontab -l, but the latter has to run on
bullseye in order to pick up the old crontab. I'm not sure how
I would do that.

Before very long, I'll be travelling again, taking one of my
laptops with me. It's roles at home and away are completely
different, so I switch crontabs for the duration, not with
crontab -e but with crontab ~/.cron/.
And that command is in a script that takes care of other
changes that need to be made for its travelling role.

I populate my ~/.cron directory with anything to do with cron,
like the scripts¹ that some crontab entries run, the crontabs
that I push to my other machines, and a copy of root's crontab
for when the occasion might arise to migrate a personal job to
being a systemwide one. And as it's all under my home directory,
it gets backed up too. (I don't backup /var/spool/.)

I also keep there the ephemeral files that control unattended
recording of live radio (via LineInput), because that system is
driven by a cron job running each minute. So an empty file like
.cron/2023-04-09-06-55-rk2wav-190 would record three hours on
Sunday morning. (I've been running this for over two decades,
though it doesn't get a fraction of the use that it did in the
days before BBC iplayer.)

¹ ISTR you had misgivings about that a couple of years ago.

Cheers,
David.

Re: questions about cron.daily

[Michel Verdier]

Le 8 avril 2023 Max Nikulin a écrit :

> On 08/04/2023 22:17, Kushal Kumaran wrote:
>>> Have you ever actually *made* a systemd --user unit file?  If so, for
>>> what purpose?
>> I have one.  It starts emacs server for me when I login.
>
> There is ready to use one: /usr/lib/systemd/user/emacs.service Perhaps there
> is no such file in buster.

/usr/lib/systemd/user is for global system running. If you want to change
something in the service you copy it in ~/.config/systemd to supersede
global one.

Re: questions about cron.daily

[Michel Verdier]

Le 8 avril 2023 Greg Wooledge a écrit :

>> systemd user files can be put in ~/.config/systemd/user/ where you can
>> use git directly
>
> Have you ever actually *made* a systemd --user unit file?  If so, for
> what purpose?

$ find .config/systemd/
.config/systemd/
.config/systemd/user
.config/systemd/user/xsession.target.requires
.config/systemd/user/xsession.target.requires/dwm.service
.config/systemd/user/dwm.service
.config/systemd/user/xsession.target
.config/systemd/user/default.target.wants
.config/systemd/user/default.target.wants/pipewire.service
.config/systemd/user/default.target.wants/pipewire-pulse.service
.config/systemd/user/sockets.target.wants
.config/systemd/user/sockets.target.wants/pipewire.socket
.config/systemd/user/sockets.target.wants/pipewire-pulse.socket
.config/systemd/user/pulseaudio.service
.config/systemd/user/pipewire.service.wants
.config/systemd/user/pipewire.service.wants/wireplumber.service
.config/systemd/user/fetchmail.service

Re: questions about cron.daily

[Max Nikulin]

On 08/04/2023 22:17, Kushal Kumaran wrote:

Have you ever actually *made* a systemd --user unit file?  If so, for
what purpose?


I have one.  It starts emacs server for me when I login.


There is ready to use one: /usr/lib/systemd/user/emacs.service Perhaps 
there is no such file in buster.

Re: questions about cron.daily

[Kushal Kumaran]

On Sat, Apr 08 2023 at 08:39:14 AM, Greg Wooledge  wrote:
> On Sat, Apr 08, 2023 at 11:16:51AM +0200, Michel Verdier wrote:
>> Le 8 avril 2023 Andrew M. A. Cater a écrit :
>> 
>> > Likewise for creating systemd unit files - NEVER "just start editing over
>> > the top" always have an example to work from and save it. You can then
>> > commit the series to git if you want to record exact changes.
>> 
>> systemd user files can be put in ~/.config/systemd/user/ where you can
>> use git directly
>
> Have you ever actually *made* a systemd --user unit file?  If so, for
> what purpose?

I have one.  It starts emacs server for me when I login.

-- 
regards,
kushal

Re: questions about cron.daily

[Max Nikulin]

On 08/04/2023 19:39, Greg Wooledge wrote:


Have you ever actually *made* a systemd --user unit file?  If so, for
what purpose?


For LXC unprivileged containers that are stopped on logout.

Do you mean it is exceptional case when default user units need 
adjustment? /usr/lib/systemd/user directory has enough entries when a 
desktop environment is installed. Even emacs server may be started as a 
systemd service.

Re: questions about cron.daily

[Greg Wooledge]

On Sat, Apr 08, 2023 at 11:16:51AM +0200, Michel Verdier wrote:
> Le 8 avril 2023 Andrew M. A. Cater a écrit :
> 
> > Likewise for creating systemd unit files - NEVER "just start editing over
> > the top" always have an example to work from and save it. You can then
> > commit the series to git if you want to record exact changes.
> 
> systemd user files can be put in ~/.config/systemd/user/ where you can
> use git directly

Have you ever actually *made* a systemd --user unit file?  If so, for
what purpose?

Re: questions about cron.daily

[Michel Verdier]

Le 8 avril 2023 Andrew M. A. Cater a écrit :

> Likewise for creating systemd unit files - NEVER "just start editing over
> the top" always have an example to work from and save it. You can then
> commit the series to git if you want to record exact changes.

systemd user files can be put in ~/.config/systemd/user/ where you can
use git directly

Re: questions about cron.daily

[Andrew M.A. Cater]

On Sat, Apr 08, 2023 at 11:45:50AM +1200, Alex King wrote:
> See man crontab.
> 
> There are 2 ways of maintaining your crontab:
> 
> crontab [ -u user ] file
> ...
> The  first  form  of this command is used to install a new crontab from some
> named file
> 
> I.e. you can keep a file in your home directory (or anywhere,) update it and
> install it when changed using "crontab filename".  This is especially useful
> if you keep your home directory under version control and want your crontab
> under version control as well.  But I guess hardly anyone does this.
> 

This is *exactly* what I do when I'm editing crontabs, especially on a 
remote machine. 

crontab -l

lists any crontabs for that user.

 crontab -l > crontab.old

gives you a version of the crontab that you can keep.

cp crontab.old crontab.new ; vi crontab.new ; diff crontab.old crontab.new

will allow you to add new lines / change times or whatever.

crontab crontab.new will put the new crontab in - but you've already got
the old one if you need to revert changes. If you want to be fancy, you
can do crontab. and have them sorted.

Likewise for creating systemd unit files - NEVER "just start editing over
the top" always have an example to work from and save it. You can then
commit the series to git if you want to record exact changes.

> 
> I believe it actually causes the current crontab to be copied to a temporary
> file, where is edited, then it is installed after a successful editing
> session and the temporary file is deleted.
> 

Yes,

> Like others, I've hardly ever met people who do things the first way, but it
> likely was the original way of doing things.
> 
> Thanks,
> Alex
> 
>

All the very best, as ever,

Andy 

Re: questions about cron.daily

[Alex King]

See man crontab.

There are 2 ways of maintaining your crontab:

crontab [ -u user ] file
...
The  first  form  of this command is used to install a new crontab from 
some named file


I.e. you can keep a file in your home directory (or anywhere,) update it 
and install it when changed using "crontab filename".  This is 
especially useful if you keep your home directory under version control 
and want your crontab under version control as well.  But I guess hardly 
anyone does this.


The second way:

crontab [ -u user ] [ -i ] { -e | -l | -r }
...
The -e option is used to edit the  current  crontab  using  the  editor
specified  by  the  VISUAL  or EDITOR environment variables.  After you
exit from the editor, the modified crontab will be installed  automati‐
cally.

I believe it actually causes the current crontab to be copied to a 
temporary file, where is edited, then it is installed after a successful 
editing session and the temporary file is deleted.


Like others, I've hardly ever met people who do things the first way, 
but it likely was the original way of doing things.


Thanks,
Alex

On 7/04/23 17:08, to...@tuxteam.de wrote:

On Thu, Apr 06, 2023 at 06:54:31PM -0400, Greg Wooledge wrote:

On Thu, Apr 06, 2023 at 05:45:08PM -0500, David Wright wrote:

Users (including root) write their crontabs anywhere they like,
typically in a directory like ~/.cron/.

Is that... normal?  I can't say I've ever seen anyone keep a private
copy of their crontab in their home directory like that.

It's news to me, too. I olny knew about the /var/spool/cron/crontabs
location.

Cheers

Re: questions about cron.daily

[Cindy Sue Causey]

On 4/7/23, Anssi Saari  wrote:
> Greg Wooledge  writes:
>
>> On Thu, Apr 06, 2023 at 05:45:08PM -0500, David Wright wrote:
>>> Users (including root) write their crontabs anywhere they like,
>>> typically in a directory like ~/.cron/.
>>
>> Is that... normal?  I can't say I've ever seen anyone keep a private
>> copy of their crontab in their home directory like that.
>
> I don't know if it's normal but sounds like a good practice, to have a
> backup of your crontab. I've been bitten by this sometime when my old
> shell provider retired a system and I had no copy of my crontab. My home
> dir was not affected by that retirement since those were all NFS mounted
> from a different server.


I like mine there. I haven't tried crontab yet, but I've put other
things at that location. It's more easily transferable without having
to look for or remember any personalizations. I think it was building
via npm that made it a comfortable, memorable CHOICE.

The .local directory is coming to mind quicker for me these days.
/opt, too, thanks to [upstream?] Chrome educating me that it's a nice,
empty, less trafficked location to quickly peek at specifically
installed packages.

There's also /usr/local that makes it easier to more quickly remember
packages that are specifically installed favorites for whatever
reason. Among /usr/local's potential plusses is that it's about root
permissions whereas putting its same contents in .local means mixing
root and user.

Mixing directories of varying permissions is no biggie for seasoned
users but can quickly mangle things for newer ones. I mangled mine a
long time ago when I got the bright idea to "chown" my user's entire
~/. We.. try not to do that anymore. What a mess. :)

It's all about available cognitive abilities while computing for my usage case..

Cindy :)

N.B. because it might catch the eye of a curious newbie: I don't do
npm these days. I had fun while the experience lasted. There were some
scary security issues a couple years back because it's such a wide
open space out there. NPM's security was serious enough that it was a
US-CERT advisement. The original is still sitting in my inbox:

https://www.cisa.gov/news-events/alerts/2021/10/22/malware-discovered-popular-npm-package-ua-parser-js

-- 
Talking Rock, Pickens County, Georgia, USA
* runs with birdseed *

Re: questions about cron.daily

[Richard Hector]

On 7/04/23 10:54, Greg Wooledge wrote:

On Thu, Apr 06, 2023 at 05:45:08PM -0500, David Wright wrote:

Users (including root) write their crontabs anywhere they like,
typically in a directory like ~/.cron/.


Is that... normal?  I can't say I've ever seen anyone keep a private
copy of their crontab in their home directory like that.

Most people just use "crontab -e" to edit the system's copy of their
personal crontab...


Perhaps if they want to keep it in version control?

Richard

Re: questions about cron.daily

[Anssi Saari]

Greg Wooledge  writes:

> On Thu, Apr 06, 2023 at 05:45:08PM -0500, David Wright wrote:
>> Users (including root) write their crontabs anywhere they like,
>> typically in a directory like ~/.cron/.
>
> Is that... normal?  I can't say I've ever seen anyone keep a private
> copy of their crontab in their home directory like that.

I don't know if it's normal but sounds like a good practice, to have a
backup of your crontab. I've been bitten by this sometime when my old
shell provider retired a system and I had no copy of my crontab. My home
dir was not affected by that retirement since those were all NFS mounted
from a different server.

I think they did dig out my crontab from tape when I asked.

Re: questions about cron.daily

[tomas]

On Thu, Apr 06, 2023 at 06:54:31PM -0400, Greg Wooledge wrote:
> On Thu, Apr 06, 2023 at 05:45:08PM -0500, David Wright wrote:
> > Users (including root) write their crontabs anywhere they like,
> > typically in a directory like ~/.cron/.
> 
> Is that... normal?  I can't say I've ever seen anyone keep a private
> copy of their crontab in their home directory like that.

It's news to me, too. I olny knew about the /var/spool/cron/crontabs
location.

Cheers
-- 
t


signature.asc
Description: PGP signature

Re: questions about cron.daily

[Dan Ritter]

Tom Furie wrote: 
> On Fri, Apr 07, 2023 at 08:05:18AM +0800, k...@openmbox.net wrote:
> > Are the time format in /etc/crontab just random? why they are 6:25, 6:47
> > etc?
> 
> They aren't *random*, though they are somewhat arbitrary. The daily tasks
> run at 6:25, a time chosen by someone somewhere back in the mists of time as
> a time that the system is likely to be quiet. The daily tasks still run on
> the days the weekly tasks are scheduled, so some time displacement is
> factored in to give the daily job time to finish. Likewise with the monthly
> allowing for both the daily and weekly.

It's also the case that if you have a dozen, hundred or ten
thousand machines, you might not want them all doing the same
thing at precisely the same time -- especially if it's disk,
power or network intensive.

So spreading out jobs is often a good idea. Automated deployment
systems usually come with a way of randomizing the initial
install of a cronjob.


-dsr-

Re: questions about cron.daily

[Tom Furie]

On Fri, Apr 07, 2023 at 08:05:18AM +0800, k...@openmbox.net wrote:
> Are the time format in /etc/crontab just random? why they are 6:25, 6:47
> etc?

They aren't *random*, though they are somewhat arbitrary. The daily tasks
run at 6:25, a time chosen by someone somewhere back in the mists of time as
a time that the system is likely to be quiet. The daily tasks still run on
the days the weekly tasks are scheduled, so some time displacement is
factored in to give the daily job time to finish. Likewise with the monthly
allowing for both the daily and weekly.

If any of the times are inappropriate for your system, you're at liberty to
change them to something more suitable.

Cheers,
Tom

-- 
I am not now, nor have I ever been, a member of the demigodic party.
-- Dennis Ritchie


signature.asc
Description: PGP signature

Re: questions about cron.daily

[ken]

On 2023-04-07 05:20, davidson wrote:

 25 6* * *   roottest -x /usr/sbin/anacron || ( cd / && 
run-parts --report /etc/cron.daily )


Are the time format in /etc/crontab just random? why they are 6:25, 6:47 
etc?



--
Ken Peng
https://kenpeng.pages.dev/

Re: questions about cron.daily

[Greg Wooledge]

On Thu, Apr 06, 2023 at 05:45:08PM -0500, David Wright wrote:
> Users (including root) write their crontabs anywhere they like,
> typically in a directory like ~/.cron/.

Is that... normal?  I can't say I've ever seen anyone keep a private
copy of their crontab in their home directory like that.

Most people just use "crontab -e" to edit the system's copy of their
personal crontab...

> They then have to be installed
> with crontab, which copies them into /var/spool/cron/crontabs/.

... which lives there.

Re: questions about cron.daily

[David Wright]

On Thu 06 Apr 2023 at 12:28:05 (-0700), Fred wrote:
> On 4/6/23 09:44, Pierre Frenkiel wrote:
> > 
> > crontab -l
> > 
> > On Thu, 6 Apr 2023, Fred wrote:
> > > I also would like to know when cron.daily scripts run.  Greg's
> > > command does not appear to reveal the time for that script.  I
> > > ran Greg's command and got the same result.
> 
> fred@ragnok:~$ crontab -l
> no crontab for fred
> 
> root@ragnok:/home/fred# crontab -l
> no crontab for root

Users (including root) write their crontabs anywhere they like,
typically in a directory like ~/.cron/. They then have to be installed
with crontab, which copies them into /var/spool/cron/crontabs/.
Those are the files that crontab -l lists. Why? Because cron isn't
going to poke around in your home directory looking for a crontab
there. It wants them somewhere it has control over.

The system crontab is different, and is found in /etc/. It is an
active file, and doesn't require any installing. It also has to
contain an extra field: the user under which each item should be
run—usually, but not always, root. Other parts of the system can
also have their own crontabs, with similar properties, which are
found in /etc/cron.d/.

Cheers,
David.

Re: questions about cron.daily

[davidson]

On Thu, 6 Apr 2023 Fred wrote:
[trimmed]
I also would like to know when cron.daily scripts run.  Greg's command does 
not appear to reveal the time for that script.  I ran Greg's command and got 
the same result.


 $ grep -FA7 "Example of job definition" /etc/crontab ; grep daily /etc/crontab
 # Example of job definition:
 # . minute (0 - 59)
 # |  .- hour (0 - 23)
 # |  |  .-- day of month (1 - 31)
 # |  |  |  .--- month (1 - 12) OR jan,feb,mar,apr ...
 # |  |  |  |  . day of week (0 - 6) (Sunday=0 or 7) OR 
sun,mon,tue,wed,thu,fri,sat
 # |  |  |  |  |
 # *  *  *  *  * user-name command to be executed
 25 6* * *   roottest -x /usr/sbin/anacron || ( cd / && run-parts 
--report /etc/cron.daily )

--
You shall also keep a trowel in your equipment. With it, when you go
outside to ease nature, you shall first dig a hole and afterward cover
up your excrement. Since the Lord, your God, journeys along within
your camp [and would be displeased to step in it] -- Deuteronomy 23

Re: questions about cron.daily

[Fred]

On 4/6/23 09:44, Pierre Frenkiel wrote:


crontab -l

Pierre Frenkiel

On Thu, 6 Apr 2023, Fred wrote:


On 4/6/23 08:33, Andy Smith wrote:

Hello,

On Thu, Apr 06, 2023 at 07:33:26PM +0800, cor...@free.fr wrote:

For scripts put under /etc/cron.daily, which special time will they be
implemented?


Greg already showed you how to check this on your own systems.

If you need something to run daily but at a specific time, consider
instead either creating a custom cron job in /etc/cron.d/ (or a
user's crontab) or else using a systemd timer.


I found some services like apache2, chkrootkit will put the scripts in
this dir automatically. are they for system cleaning purpose?


The purpose of cron.daily is literally to run scripts daily for
*whatever purpose each script has*. If you have questions about what
a specific script does, try looking at it, as it's just an
executable script (usually sh or bash). If you're still confused
then ask a specific question about a specific script.

Cheers,
Andy

I also would like to know when cron.daily scripts run.  Greg's command 
does not appear to reveal the time for that script.  I ran Greg's 
command and got the same result.


Best regards,
Fred


fred@ragnok:~$ crontab -l
no crontab for fred

root@ragnok:/home/fred# crontab -l
no crontab for root

best regards,
Fred

Re: questions about cron.daily

[Alexis Grigoriou]

On Thu, 2023-04-06 at 19:05 +0200, Thomas Schmitt wrote:
> Hi,
> 
> 
> Pierre Frenkiel wrote:
> > crontab -l
> 
> Plus:
>   man 5 crontab
> which has in its section "EXAMPLE CRON FILE"
> 
>    # run five minutes after midnight, every day
>    5 0 * * *   $HOME/bin/daily.job >> $HOME/tmp/out 2>&1
> 
> and explains further up
> 
>   The  format of a cron command is very much the V7 standard,
> with a num‐
>   ber of upward-compatible extensions.  Each line has five time
> and  date
>   fields,  followed by a command, followed by a newline character
> ('\n').
>   ...
>  field  allowed values
>   -  --
>   minute 0-59
>   hour   0-23
>   day of month   1-31
>   month  1-12 (or names, see below)
>   day of week    0-7 (0 or 7 is Sun, or use names)
> 
> So the command shown by Greg Wooledge runs at 6:25 am the comand
> 
>   test -x /usr/sbin/anacron || ( cd / && run-parts --report
> /etc/cron.daily )
> 
> which runs the programs in /etc/cron.daily (by run-parts(8)) if not
> /usr/sbin/anacron exists and is executable.
> 

https://crontab.guru has a very nice explaining also.

Re: questions about cron.daily

[Greg Wooledge]

On Thu, Apr 06, 2023 at 01:42:33PM -0400, Stefan Monnier wrote:
> > I also would like to know when cron.daily scripts run.  Greg's command does
> > not appear to reveal the time for that script.  I ran Greg's command and got
> > the same result.
> 
> As explained, his command's output does show the actual time, but
> I don't think it's the whole story: AFAIK the actual time can be
> different for example if you installed `anacron`, and I hope that's also
> true if you installed `systemd-cron`.
> 
> I say "I hope", because some of my machines I basically always OFF at
> that time.

In that case, I'd recommend changing the start time in /etc/crontab to
some time when your machines are likely to be on, but idle.  If there's
no such time, well, then you're looking at the alternatives like anacron
which are designed for such situations.

Re: questions about cron.daily

[Stefan Monnier]

> I also would like to know when cron.daily scripts run.  Greg's command does
> not appear to reveal the time for that script.  I ran Greg's command and got
> the same result.

As explained, his command's output does show the actual time, but
I don't think it's the whole story: AFAIK the actual time can be
different for example if you installed `anacron`, and I hope that's also
true if you installed `systemd-cron`.

I say "I hope", because some of my machines I basically always OFF at
that time.


Stefan

Re: questions about cron.daily

[Thomas Schmitt]

Hi,

Greg Wooledge wrote:
> > > unicorn:~$ grep daily /etc/crontab
> > > 25 6* * *   roottest -x /usr/sbin/anacron || ( cd / && run-parts 
> > > --report /etc/cron.daily )

Fred wrote:
> > I also would like to know when cron.daily scripts run.  Greg's command does
> > not appear to reveal the time for that script.  I ran Greg's command and got
> > the same result.

Pierre Frenkiel wrote:
> crontab -l

Plus:
  man 5 crontab
which has in its section "EXAMPLE CRON FILE"

   # run five minutes after midnight, every day
   5 0 * * *   $HOME/bin/daily.job >> $HOME/tmp/out 2>&1

and explains further up

  The  format of a cron command is very much the V7 standard, with a num‐
  ber of upward-compatible extensions.  Each line has five time and  date
  fields,  followed by a command, followed by a newline character ('\n').
  ...
 field  allowed values
  -  --
  minute 0-59
  hour   0-23
  day of month   1-31
  month  1-12 (or names, see below)
  day of week0-7 (0 or 7 is Sun, or use names)

So the command shown by Greg Wooledge runs at 6:25 am the comand

  test -x /usr/sbin/anacron || ( cd / && run-parts --report /etc/cron.daily )

which runs the programs in /etc/cron.daily (by run-parts(8)) if not
/usr/sbin/anacron exists and is executable.


Have a nice day :)

Thomas

Re: questions about cron.daily

[Michel Verdier]

Le 6 avril 2023 Fred a écrit :

> I also would like to know when cron.daily scripts run.  Greg's command does
> not appear to reveal the time for that script.  I ran Greg's command and got
> the same result.

Greg shows it:
unicorn:~$ grep daily /etc/crontab
25 6* * *   roottest -x /usr/sbin/anacron || ( cd / && run-parts 
--report /etc/cron.daily )

which means runs at 6:25 each day

to get meanings of each col:
man 5 crontab

Re: questions about cron.daily

[Tom Furie]

On Thu, Apr 06, 2023 at 09:02:15AM -0700, Fred wrote:
> I also would like to know when cron.daily scripts run.  Greg's command does
> not appear to reveal the time for that script.  I ran Greg's command and got
> the same result.

Then you need to read the documentation for cron. I'd suggest beginning with
'man 5 crontab' for the details of a crontab entry.

Cheers,
Tom

-- 
Here there be tygers.


signature.asc
Description: PGP signature

Re: questions about cron.daily

[Pierre Frenkiel]

crontab -l

Pierre Frenkiel

On Thu, 6 Apr 2023, Fred wrote:


On 4/6/23 08:33, Andy Smith wrote:

Hello,

On Thu, Apr 06, 2023 at 07:33:26PM +0800, cor...@free.fr wrote:

For scripts put under /etc/cron.daily, which special time will they be
implemented?


Greg already showed you how to check this on your own systems.

If you need something to run daily but at a specific time, consider
instead either creating a custom cron job in /etc/cron.d/ (or a
user's crontab) or else using a systemd timer.


I found some services like apache2, chkrootkit will put the scripts in
this dir automatically. are they for system cleaning purpose?


The purpose of cron.daily is literally to run scripts daily for
*whatever purpose each script has*. If you have questions about what
a specific script does, try looking at it, as it's just an
executable script (usually sh or bash). If you're still confused
then ask a specific question about a specific script.

Cheers,
Andy

I also would like to know when cron.daily scripts run.  Greg's command does 
not appear to reveal the time for that script.  I ran Greg's command and got 
the same result.


Best regards,
Fred

Re: questions about cron.daily

[Fred]

On 4/6/23 08:33, Andy Smith wrote:

Hello,

On Thu, Apr 06, 2023 at 07:33:26PM +0800, cor...@free.fr wrote:

For scripts put under /etc/cron.daily, which special time will they be
implemented?


Greg already showed you how to check this on your own systems.

If you need something to run daily but at a specific time, consider
instead either creating a custom cron job in /etc/cron.d/ (or a
user's crontab) or else using a systemd timer.


I found some services like apache2, chkrootkit will put the scripts in
this dir automatically. are they for system cleaning purpose?


The purpose of cron.daily is literally to run scripts daily for
*whatever purpose each script has*. If you have questions about what
a specific script does, try looking at it, as it's just an
executable script (usually sh or bash). If you're still confused
then ask a specific question about a specific script.

Cheers,
Andy

I also would like to know when cron.daily scripts run.  Greg's command 
does not appear to reveal the time for that script.  I ran Greg's 
command and got the same result.


Best regards,
Fred

Re: questions about cron.daily

[Andy Smith]

Hello,

On Thu, Apr 06, 2023 at 07:33:26PM +0800, cor...@free.fr wrote:
> For scripts put under /etc/cron.daily, which special time will they be
> implemented?

Greg already showed you how to check this on your own systems.

If you need something to run daily but at a specific time, consider
instead either creating a custom cron job in /etc/cron.d/ (or a
user's crontab) or else using a systemd timer.

> I found some services like apache2, chkrootkit will put the scripts in
> this dir automatically. are they for system cleaning purpose?

The purpose of cron.daily is literally to run scripts daily for
*whatever purpose each script has*. If you have questions about what
a specific script does, try looking at it, as it's just an
executable script (usually sh or bash). If you're still confused
then ask a specific question about a specific script.

Cheers,
Andy

-- 
https://bitfolk.com/ -- No-nonsense VPS hosting

Re: questions about cron.daily

[Greg Wooledge]

On Thu, Apr 06, 2023 at 07:33:26PM +0800, cor...@free.fr wrote:
> For scripts put under /etc/cron.daily, which special time will they be
> implemented?

unicorn:~$ grep daily /etc/crontab
25 6* * *   roottest -x /usr/sbin/anacron || ( cd / && run-parts 
--report /etc/cron.daily )

Re: questions on iptables

[Piperみかこ]

Thanks John.

The following rules seem to work for me now.

sudo /usr/sbin/iptables -F
sudo /usr/sbin/iptables -A INPUT -m state --state RELATED,ESTABLISHED -j
ACCEPT
sudo /usr/sbin/iptables -A INPUT -p icmp -j ACCEPT
sudo /usr/sbin/iptables -A INPUT -i lo -j ACCEPT
sudo /usr/sbin/iptables -A INPUT -p tcp --dport 22 -j ACCEPT
sudo /usr/sbin/iptables -A INPUT -p tcp --dport 80 -j ACCEPT
sudo /usr/sbin/iptables -A INPUT -j REJECT --reject-with
icmp-admin-prohibited


are there any problems with my setting above?

Thanks

Re: questions on iptables

[John Conover]

John Conover writes:
> =?UTF-8?B?UGlwZXLjgb/jgYvjgZM=?= writes:
> > 
> > sudo /usr/sbin/iptables -F
> > sudo /usr/sbin/iptables -A INPUT -s 127.0.0.1 -j ACCEPT
> > sudo /usr/sbin/iptables -A INPUT -s xx.xx.xx.xx -j ACCEPT  # my server
> > public IP
> > sudo /usr/sbin/iptables -A INPUT -p tcp --dport 22 -j ACCEPT
> > sudo /usr/sbin/iptables -A INPUT -p tcp --dport 80 -j ACCEPT
> > sudo /usr/sbin/iptables -A INPUT -p tcp -j DROP
> >
> 
> iptables -F
> iptables -P INPUT ACCEPT
> iptables -P FORWARD ACCEPT
> iptables -P OUTPUT ACCEPT
> iptables -A INPUT -m state --state RELATED,ESTABLISHED -j ACCEPT
> iptables -A INPUT -p icmp -j ACCEPT
> iptables -A INPUT -i lo -j ACCEPT
> iptables -A INPUT -p tcp --dport 22 -j ACCEPT
> iptables -A INPUT -p tcp --dport 80 -j ACCEPT
> iptables -A INPUT -j REJECT --reject-with icmp-admin-prohibited
> iptables -A FORWARD -j REJECT --reject-with icmp-admin-prohibited
> iptables-save > /etc/iptables/rules.v4
> 
> ip6tables -F 
> ip6tables -P INPUT ACCEPT
> ip6tables -P FORWARD ACCEPT
> ip6tables -P OUTPUT ACCEPT
> ip6tables -A INPUT -m state --state RELATED,ESTABLISHED -j ACCEPT
> ip6tables -A INPUT -p ipv6-icmp -j ACCEPT
> ip6tables -A INPUT -i lo -j ACCEPT
> ip6tables -A INPUT -p tcp -—dport 22 -j ACCEPT
> ip6tables -A INPUT -p tcp -—dport 80 -j ACCEPT
> ip6tables -A INPUT -j REJECT --reject-with icmp6-adm-prohibited
> ip6tables -A FORWARD -j REJECT --reject-with icmp6-adm-prohibited
> ip6tables-save > /etc/iptables/rules.v6
> 
> And, look in /etc/iptables to make sure, and reboot; then check for
> outbound connections with a browser. (Note how icmp is handled-it is
> required.)
> 

Sorry, cut from my machine docs. The two ipv6 statement should
obviously be:

ip6tables -A INPUT -p tcp --dport 22 -j ACCEPT
ip6tables -A INPUT -p tcp --dport 80 -j ACCEPT

Also, if you have root access on another machine, for assurance:

nmap -4 -Pn -sS -v -v IPV4_ADDRESS
nmap -4 -Pn -sU -v -v IPV4_ADDRESS
nmap -6 -Pn -sS -v -v IPV6_ADDRESS
nmap -6 -Pn -sU -v -v IPV6_ADDRESS

will take a lot of time to run, and should only find the two open
ports.

John

--


John Conover, cono...@panix.com, http://www.johncon.com/

Re: questions on iptables

[John Conover]

=?UTF-8?B?UGlwZXLjgb/jgYvjgZM=?= writes:
> 
> sudo /usr/sbin/iptables -F
> sudo /usr/sbin/iptables -A INPUT -s 127.0.0.1 -j ACCEPT
> sudo /usr/sbin/iptables -A INPUT -s xx.xx.xx.xx -j ACCEPT  # my server
> public IP
> sudo /usr/sbin/iptables -A INPUT -p tcp --dport 22 -j ACCEPT
> sudo /usr/sbin/iptables -A INPUT -p tcp --dport 80 -j ACCEPT
> sudo /usr/sbin/iptables -A INPUT -p tcp -j DROP
>

iptables -F
iptables -P INPUT ACCEPT
iptables -P FORWARD ACCEPT
iptables -P OUTPUT ACCEPT
iptables -A INPUT -m state --state RELATED,ESTABLISHED -j ACCEPT
iptables -A INPUT -p icmp -j ACCEPT
iptables -A INPUT -i lo -j ACCEPT
iptables -A INPUT -p tcp --dport 22 -j ACCEPT
iptables -A INPUT -p tcp --dport 80 -j ACCEPT
iptables -A INPUT -j REJECT --reject-with icmp-admin-prohibited
iptables -A FORWARD -j REJECT --reject-with icmp-admin-prohibited
iptables-save > /etc/iptables/rules.v4

ip6tables -F 
ip6tables -P INPUT ACCEPT
ip6tables -P FORWARD ACCEPT
ip6tables -P OUTPUT ACCEPT
ip6tables -A INPUT -m state --state RELATED,ESTABLISHED -j ACCEPT
ip6tables -A INPUT -p ipv6-icmp -j ACCEPT
ip6tables -A INPUT -i lo -j ACCEPT
ip6tables -A INPUT -p tcp -—dport 22 -j ACCEPT
ip6tables -A INPUT -p tcp -—dport 80 -j ACCEPT
ip6tables -A INPUT -j REJECT --reject-with icmp6-adm-prohibited
ip6tables -A FORWARD -j REJECT --reject-with icmp6-adm-prohibited
ip6tables-save > /etc/iptables/rules.v6

And, look in /etc/iptables to make sure, and reboot; then check for
outbound connections with a browser. (Note how icmp is handled-it is
required.)

John

-- 

John Conover, cono...@panix.com, http://www.johncon.com/

Re: questions on iptables

[Piperみかこ]

Hello

I have a server box who has public IP in the DC.
My requirements,

1. allow access to ssh and http port on the server from external.
2. allow access to any services from the server to external.

The #1 can be done by my rules already set.
But #2 doesn't work. for instance, when rsync connection to another server
from that box, it won't work (connection timeout).

So how to ?

Thanks

On Sun, Dec 25, 2022 at 10:28 AM jeremy ardley  wrote:

>
> On 25/12/22 10:14, Piperみかこ wrote:
> > Hello,
> >
> > I have setup the following iptables on linux server,
> >
> > sudo /usr/sbin/iptables -F
> > sudo /usr/sbin/iptables -A INPUT -s 127.0.0.1 -j ACCEPT
> > sudo /usr/sbin/iptables -A INPUT -s xx.xx.xx.xx -j ACCEPT  # my server
> > public IP
> > sudo /usr/sbin/iptables -A INPUT -p tcp --dport 22 -j ACCEPT
> > sudo /usr/sbin/iptables -A INPUT -p tcp --dport 80 -j ACCEPT
> > sudo /usr/sbin/iptables -A INPUT -p tcp -j DROP
> >
> > These rules work for incoming connections.
> > But if I made a request from this server box to the external server,
> > such as rsync to another remote server, it won't work.
> >
> > So how can I set up the outgoing rules as well?
> >
> > Thanks and happy holidays~
> >
> > Piper
>
>
> It's not quite clear what you are doing. I guess your linux box is in
> your LAN and you have an external server on the internet?
>
> Anyway, it's usual to have a line like this to handle handshaking with
> external systems.
>
> -A INPUT -m state --state RELATED,ESTABLISHED -j ACCEPT
>
> --
> Jeremy
>
>

Re: questions on iptables

[Dan Ritter]

Piperみかこ wrote: 
> Hello,
> 
> I have setup the following iptables on linux server,
> 
> sudo /usr/sbin/iptables -F
> sudo /usr/sbin/iptables -A INPUT -s 127.0.0.1 -j ACCEPT
> sudo /usr/sbin/iptables -A INPUT -s xx.xx.xx.xx -j ACCEPT  # my server
> public IP
> sudo /usr/sbin/iptables -A INPUT -p tcp --dport 22 -j ACCEPT
> sudo /usr/sbin/iptables -A INPUT -p tcp --dport 80 -j ACCEPT
> sudo /usr/sbin/iptables -A INPUT -p tcp -j DROP
> 
> These rules work for incoming connections.
> But if I made a request from this server box to the external server, such
> as rsync to another remote server, it won't work.
> 
> So how can I set up the outgoing rules as well?

At boot time, there are three chains:

INPUT
FORWARD
OUTPUT

each has a policy which determines what happens if no other rule
applies.

FORWARD is only interesting if you are routing packets.

INPUT controls incoming packets.

OUTPUT controls outgoing packets.

You can specify IPs, interfaces, protocols, ports, and a bunch
of other things.

-dsr-

Re: questions on iptables

[jeremy ardley]

On 25/12/22 10:14, Piperみかこ wrote:

Hello,

I have setup the following iptables on linux server,

sudo /usr/sbin/iptables -F
sudo /usr/sbin/iptables -A INPUT -s 127.0.0.1 -j ACCEPT
sudo /usr/sbin/iptables -A INPUT -s xx.xx.xx.xx -j ACCEPT  # my server 
public IP

sudo /usr/sbin/iptables -A INPUT -p tcp --dport 22 -j ACCEPT
sudo /usr/sbin/iptables -A INPUT -p tcp --dport 80 -j ACCEPT
sudo /usr/sbin/iptables -A INPUT -p tcp -j DROP

These rules work for incoming connections.
But if I made a request from this server box to the external server, 
such as rsync to another remote server, it won't work.


So how can I set up the outgoing rules as well?

Thanks and happy holidays~

Piper



It's not quite clear what you are doing. I guess your linux box is in 
your LAN and you have an external server on the internet?


Anyway, it's usual to have a line like this to handle handshaking with 
external systems.


-A INPUT -m state --state RELATED,ESTABLISHED -j ACCEPT

--
Jeremy

Re: Questions about installing Debian on a laptop

[Georgi Naplatanov]

On 12/5/22 11:57, Timothy M Butterworth wrote:



On Mon, Dec 5, 2022 at 4:15 AM Georgi Naplatanov > wrote:


On 12/5/22 10:47, Timothy M Butterworth wrote:
 >
 >
 > On Mon, Dec 5, 2022 at 3:37 AM Timothy M Butterworth
 > mailto:timothy.m.butterwo...@gmail.com>
 > >> wrote:
 >
 >
 >
 >     On Sun, Dec 4, 2022 at 2:41 PM David Christensen
 >     mailto:dpchr...@holgerdanske.com>
>> wrote:
 >
 >         On 12/4/22 05:52, Gabor Urban wrote:
 >          > Hi,
 >          >
 >          > I am planning to install Debian on a laptop the first
 >         time.That will not be
 >          > my first installing but I never used notebooks for that. I
 >         have found a lot
 >          > of useful information but I would like to have some
guidance
 >         at the start.
 >          >
 >          > What are the most important issues selecting a laptop I
 >         should be mindful
 >          > about?
 >          >
 >          > Thanks in advance,
 >
 > I have a Ryzen 7 4700U 8 core processor with Radeon Graphics. I am
 > running Debian Testing because Debian Stable does not have
working HDMI
 > sound drivers
 >
 >
https://smile.amazon.com/gp/product/B08PC72314/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8=1 
 
>
 >
 > This Laptop requires non-free firmware for: WiFi, Sound, Video
etc. As
 > long as you do not mind running binary blobs I recommend it. The
price
 > is good for the performance.
 >

Hi Timothy,

in case this Radeon video card is integrated in CPU, would you share
your experience with it as usability and performance?

Kind regards
Georgi


The integrated Radeon Graphics performs great.  I play openMW, TES IV 
Oblivion and TES V Skyrim with CrossOver Office on the highest video 
settings and they run like a champ! Without the Non-Free blobs you only 
get a resolution of 1024x768. With the non-free binary blobs you get 
1920x1080. HDMI Audio does not function on Debian 11 due to the kernel 
and drivers being too old. Everything works well with Debian Testing and 
Non-Free Drivers. I transcode a lot of video into Matroska Theora format 
and the transcoding speed is great.




Thank you for the provided information!

Kind regards
Georgi

Re: Questions about installing Debian on a laptop

[Timothy M Butterworth]

On Mon, Dec 5, 2022 at 4:15 AM Georgi Naplatanov  wrote:

> On 12/5/22 10:47, Timothy M Butterworth wrote:
> >
> >
> > On Mon, Dec 5, 2022 at 3:37 AM Timothy M Butterworth
> >  > > wrote:
> >
> >
> >
> > On Sun, Dec 4, 2022 at 2:41 PM David Christensen
> > mailto:dpchr...@holgerdanske.com>>
> wrote:
> >
> > On 12/4/22 05:52, Gabor Urban wrote:
> >  > Hi,
> >  >
> >  > I am planning to install Debian on a laptop the first
> > time.That will not be
> >  > my first installing but I never used notebooks for that. I
> > have found a lot
> >  > of useful information but I would like to have some guidance
> > at the start.
> >  >
> >  > What are the most important issues selecting a laptop I
> > should be mindful
> >  > about?
> >  >
> >  > Thanks in advance,
> >
> > I have a Ryzen 7 4700U 8 core processor with Radeon Graphics. I am
> > running Debian Testing because Debian Stable does not have working HDMI
> > sound drivers
> >
> >
> https://smile.amazon.com/gp/product/B08PC72314/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8=1
> <
> https://smile.amazon.com/gp/product/B08PC72314/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8=1
> >
> >
> > This Laptop requires non-free firmware for: WiFi, Sound, Video etc. As
> > long as you do not mind running binary blobs I recommend it. The price
> > is good for the performance.
> >
>
> Hi Timothy,
>
> in case this Radeon video card is integrated in CPU, would you share
> your experience with it as usability and performance?
>
> Kind regards
> Georgi
>
>
The integrated Radeon Graphics performs great.  I play openMW, TES IV
Oblivion and TES V Skyrim with CrossOver Office on the highest video
settings and they run like a champ! Without the Non-Free blobs you only get
a resolution of 1024x768. With the non-free binary blobs you get 1920x1080.
HDMI Audio does not function on Debian 11 due to the kernel and drivers
being too old. Everything works well with Debian Testing and Non-Free
Drivers. I transcode a lot of video into Matroska Theora format and the
transcoding speed is great.

-- 
⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org/
⠈⠳⣄⠀⠀

Re: Questions about installing Debian on a laptop

[Georgi Naplatanov]

On 12/5/22 10:47, Timothy M Butterworth wrote:



On Mon, Dec 5, 2022 at 3:37 AM Timothy M Butterworth 
> wrote:




On Sun, Dec 4, 2022 at 2:41 PM David Christensen
mailto:dpchr...@holgerdanske.com>> wrote:

On 12/4/22 05:52, Gabor Urban wrote:
 > Hi,
 >
 > I am planning to install Debian on a laptop the first
time.That will not be
 > my first installing but I never used notebooks for that. I
have found a lot
 > of useful information but I would like to have some guidance
at the start.
 >
 > What are the most important issues selecting a laptop I
should be mindful
 > about?
 >
 > Thanks in advance,

I have a Ryzen 7 4700U 8 core processor with Radeon Graphics. I am 
running Debian Testing because Debian Stable does not have working HDMI 
sound drivers


https://smile.amazon.com/gp/product/B08PC72314/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8=1
 


This Laptop requires non-free firmware for: WiFi, Sound, Video etc. As 
long as you do not mind running binary blobs I recommend it. The price 
is good for the performance.




Hi Timothy,

in case this Radeon video card is integrated in CPU, would you share 
your experience with it as usability and performance?


Kind regards
Georgi

Re: Questions about installing Debian on a laptop

[Timothy M Butterworth]

On Mon, Dec 5, 2022 at 3:37 AM Timothy M Butterworth <
timothy.m.butterwo...@gmail.com> wrote:

>
>
> On Sun, Dec 4, 2022 at 2:41 PM David Christensen <
> dpchr...@holgerdanske.com> wrote:
>
>> On 12/4/22 05:52, Gabor Urban wrote:
>> > Hi,
>> >
>> > I am planning to install Debian on a laptop the first time.That will
>> not be
>> > my first installing but I never used notebooks for that. I have found a
>> lot
>> > of useful information but I would like to have some guidance at the
>> start.
>> >
>> > What are the most important issues selecting a laptop I should be
>> mindful
>> > about?
>> >
>> > Thanks in advance,
>
>
I have a Ryzen 7 4700U 8 core processor with Radeon Graphics. I am running
Debian Testing because Debian Stable does not have working HDMI sound
drivers

https://smile.amazon.com/gp/product/B08PC72314/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8=1

This Laptop requires non-free firmware for: WiFi, Sound, Video etc. As long
as you do not mind running binary blobs I recommend it. The price is good
for the performance.


-- 
> ⢀⣴⠾⠻⢶⣦⠀
> ⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
> ⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org/
> ⠈⠳⣄⠀⠀
>


-- 
⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org/
⠈⠳⣄⠀⠀

Re: Questions about installing Debian on a laptop

[David Christensen]

On 12/4/22 05:52, Gabor Urban wrote:

Hi,

I am planning to install Debian on a laptop the first time.That will not be
my first installing but I never used notebooks for that. I have found a lot
of useful information but I would like to have some guidance at the start.

What are the most important issues selecting a laptop I should be mindful
about?

Thanks in advance,



As other readers have mentioned, the Linux kernel in Debian Stable is 
typically a year old or more.  So, the latest hardware might no be 
supported.  My newest laptop is a 2019 Dell Latitude 5400 with an Intel 
Core i7-8665U.  My oldest laptop is a 2007 Dell Inspiron E1505 with an 
Intel Core 2 T7400 (upgraded).  Both have Intel integrated graphics 
(only).  Both run Debian.  Both needed non-free Wi-Fi firmware packages.



I agree with the suggestion to test with Debian Live prior to laptop 
purchase or Debian installation.  I would use an image without 
proprietary firmware, so that I could identify firmware issues.  I use 
the Xfce desktop; choose the desktop you prefer, if any:


https://cdimage.debian.org/cdimage/archive/11.4.0-live/amd64/iso-hybrid/

debian-live-11.4.0-amd64-xfce.iso


I typically go one step further and install Debian onto a portable drive 
of some sort (e.g. USB flash drive, SATA SSD with adapter, etc.).  I 
then install whatever drivers, firmware, etc., are needed for my various 
machines; plus my favorite sysadmin/ devops tools.



If you have a make/ model/ series/ range of laptop in mind, I suggest 
posting a message to this list asking if anyone has experience running 
Debian on that laptop.



David

Re: Questions about installing Debian on a laptop

[Charles Curley]

On Sun, 4 Dec 2022 14:52:22 +0100
Gabor Urban  wrote:

> What are the most important issues selecting a laptop I should be
> mindful about?

The first thing I would do, even before buying the thing, is consult
Hardware for Linux to see what experience others have reported.
https://linux-hardware.org

And when you finish your installation, successfully or not, please
contribute back. https://linux-hardware.org/?view=howto

-- 
Does anybody read signatures any more?

https://charlescurley.com
https://charlescurley.com/blog/

Re: Questions about installing Debian on a laptop

[Alexander V. Makartsev]

On 04.12.2022 18:52, Gabor Urban wrote:

Hi,

I am planning to install Debian on a laptop the first time.That will 
not be my first installing but I never used notebooks for that. I have 
found a lot of useful information but I would like to have some 
guidance at the start.


What are the most important issues selecting a laptop I should be 
mindful about?


Probably the most problematic point of a Linux laptop is a wireless 
connectivity, both WiFi and Bluetooth.
It's always better to check beforehand if wireless adapter that comes 
preinstalled is supported natively by Linux kernel [1], or would work 
with external drivers (kernel modules) supplied by device manufacturer.
Another thing to keep in mind, is to get a laptop with a MUX switch (AKA 
Advanced Optimus) to be able to select between CPU internal and discrete 
VGAs.
While classic Optimus technology could work out of the box, it is 
usually a headache to make it work, because hardware implementation is 
different for every laptop manufacturer.
So it is better to just select discrete VGA in BIOS as primary, and MUX 
switch allows you to do that.


If you already have the laptop handy you can test how it would work with 
Debian Live [2] before removing preinstalled OS.
It's better to choose from images with non-free firmware included to 
save yourself some time from extra troubleshooting while trying make 
things work.



[1] https://wireless.wiki.kernel.org/en/users/drivers
[2] https://www.debian.org/CD/live/

--
With kindest regards, Alexander.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀https://www.debian.org
⠈⠳⣄

Re: Questions about installing Debian on a laptop

[Andrew M.A. Cater]

On Sun, Dec 04, 2022 at 02:52:22PM +0100, Gabor Urban wrote:
> Hi,
> 
> I am planning to install Debian on a laptop the first time.That will not be
> my first installing but I never used notebooks for that. I have found a lot
> of useful information but I would like to have some guidance at the start.
> 
> What are the most important issues selecting a laptop I should be mindful
> about?
> 
> Thanks in advance,
> -- 
> Urbán Gábor
> 
> Linux is like a wigwam: no Gates, no Windows and an Apache inside.

Use the firmware .iso - 
https://cdimage.debian.org/cdimage/unofficial/non-free/cd-including-firmware/11.5.0+nonfree/amd64/iso-cd/firmware-11.5.0-amd64-netinst.iso

If the laptop is really new, do you need to keep whatever is on there already?

A VERY new laptop from the very latest ranges might need to be ruuning 
something later than Debian stable, but in general, that shouldn't be a problem.

All best, as ever,

Andy C.

Re: Questions Python

[Pierre ESTREM]

Bonjour Marc et le Monde,

La réponse est ici :
https://www.developpez.net/forums/d1474088/autres-langages/python/gui/tkinter/obtenir-nom-widget-focus/

Mais j'ai fait un peu autrement.
On utilise la propriété "_name" qu'il faut donc fixer à la création du 
widget.


# Un extrait

# Une fenêtre principale
master=TK()

# du code ...

widget=mater.focus_get()

bouton1=Button(master)
bouton1._name="bouton1"
bouton1.pack()

#  et plein d'autres widgets !

# Afficherait "bouton1" si il a le focus
print(widget._name)

master.mailoop()

# Fin

Et ceci m'ouvre des perspectives énormes...

a+
pierre estrem


Le 31/08/2022 à 07:45, Marc Chantreux a écrit :

salut Pierre,

On Tue, Aug 30, 2022 at 07:31:58PM +0200, Pierre ESTREM wrote:

   widget = master.focus_get()
    print(widget()) # Aucun retour

vraiment: je t'encourage à reposer la question sur le forum de l'afpy.

* ce tu trouveras des gens plus motivés et connaisseurs pour répondre à cette 
question
* je pense que ce genre de questions est vraiment hors sujet ici (en
   tout cas pour ma part je ne suis plus sur les listes python justement
   parce que ce langage me ... laisse dubitatif :))


bonne chance.

Re: Questions Python

[Marc Chantreux]

salut Pierre,

On Tue, Aug 30, 2022 at 07:31:58PM +0200, Pierre ESTREM wrote:
>   widget = master.focus_get()
>    print(widget()) # Aucun retour

vraiment: je t'encourage à reposer la question sur le forum de l'afpy.

* ce tu trouveras des gens plus motivés et connaisseurs pour répondre à cette 
question
* je pense que ce genre de questions est vraiment hors sujet ici (en
  tout cas pour ma part je ne suis plus sur les listes python justement
  parce que ce langage me ... laisse dubitatif :))


bonne chance.

-- 
Marc Chantreux
Pôle de Calcul et Services Avancés à la Recherche (CESAR)
http://annuaire.unistra.fr/p/20200

Re: Questions Python

[Pierre ESTREM]

Bonjour,

La variable "widget" reçoit (pointe au sens de  C ?) l'objet qui a le 
focus (à ce stade je ne sais pas quel objet aurait ce focus).


""" Mon code

#!/usr/bin/env python3
# coding: utf-8

 from tkinter import *
from tkinter.ttk import *
import os

master = Tk()

def focus(event):
  widget = master.focus_get()
   print(widget()) # Aucun retour

"""

J'ai testé différentes propriétés mais j'obtiens uniquement la classe de 
l'objet (ou erreurs) et jamais le nom de la variable pointée.

Ca doit se faire ! :)

Pour ipython3 je vois cela, merci.

a+
pierre estrem


Le 30/08/2022 à 13:09, Marc Chantreux a écrit :

Bonjour Pierre,


Pour obtenir une liste des méthodes d'un objet de classe Canvas, je lis
qu'on peut exécuter :
print dir (Canvas())

c'est de la syntaxe python2 (ce qui me fait me demander quelle doc. tu
lis: vraiment, si tu n'as pas de code historique à gérer, oublie
python2).


J'obtiens une erreur ; quelle est la bonne syntaxe ?

c'était la bonne syntaxe... il y a longtemps. il faut maintenant mettre
des parenthèses autour des arguments de print.


D'autre part je ne trouve pas la méthode qui retourne le nom d'un widget
pointé mais j'obtiens son type, en faisant :
  print (widget)
L'instance widget a le focus et j'ai besoin de connaître son type, mais de
plus son petit nom (ex: listbox1, entry2 etc).

de quoi tu parles quand tu dis "le nom"? si c'est le nom de la variable
qui y fait référence dans ton programme (ce que je crois comprendre en lisant 
l'exemple),
ce soit etre possible en comparant le la référence à toutes les clefs du
namespace standard. je ne crois pas que python aie une fonction builtin
pour ca vu le peu d'intéret de la chose.


Merci de m'aider.

Avec plaisir. Je me permet 2 conseils:

* reposer ces questions aurprès des gens de l'afpy (ils avaient une
   liste de diffusion naguère mais ils ont une vision bien a eux de la
   modernité et sont passés sur discourse:https://discuss.afpy.org/):
   python est exactement le sujet de leurs fora, tu y trouveras des gens
   calés, passionnés et sympa.
* si tu souhaites explorer les objets, découvrir leurs methodes, les
   valeurs et tout ça (on appelle ça l'introspection), il existe ipython3
   qui devrait de faciliter la vie.

cordialement,

Re: Questions Python

[Pierre ESTREM]

Bonjour JF et le Monde,

Effectivement en Python3 il me fallait encadrer "dir" avec des parenthèses.
On obtient bien une liste des propriétés et méthodes de la classe passée 
en paramètre de "dir".


> Pour le widget, cela dépendpeut-être de sa nature; as-tu essayé:
>   print (widget())

Avec cette écriture, aucun retour.

a+

pierre estrem



Le 30/08/2022 à 09:47, Jean-François Colas a écrit :

Bonjour Pierre,

Avec Python 3, il faudra plutôt écrire:
   print (dir (Canvas))

La syntaxe que tu utilises pourrait être valide avec Python 2, en 
fonction de ce qu'est Canvas.


Pour le widget, cela dépendpeut-être de sa nature; as-tu essayé:
   print (widget())

J.-F.


Le 29/08/2022 à 23:14, Pierre ESTREM a écrit :

Bonjour la liste,

Dans mon apprentissage de Python, je progresse et sèche sur des points.

Pour obtenir une liste des méthodes d'un objet de classe Canvas, je 
lis qu'on peut exécuter :

   print dir (Canvas())

J'obtiens une erreur ; quelle est la bonne syntaxe ?

D'autre part je ne trouve pas la méthode qui retourne le nom d'un 
widget pointé mais j'obtiens son type, en faisant :

   print (widget)

L'instance widget a le focus et j'ai besoin de connaître son type, 
mais de plus son petit nom (ex: listbox1, entry2 etc).


Merci de m'aider.
--
Pierre ESTREM

Re: Questions Python

[Jean-François Colas]

Bonjour Pierre,

Avec Python 3, il faudra plutôt écrire:
   print (dir (Canvas))

La syntaxe que tu utilises pourrait être valide avec Python 2, en 
fonction de ce qu'est Canvas.


Pour le widget, cela dépendpeut-être de sa nature; as-tu essayé:
   print (widget())

J.-F.


Le 29/08/2022 à 23:14, Pierre ESTREM a écrit :

Bonjour la liste,

Dans mon apprentissage de Python, je progresse et sèche sur des points.

Pour obtenir une liste des méthodes d'un objet de classe Canvas, je lis 
qu'on peut exécuter :

   print dir (Canvas())

J'obtiens une erreur ; quelle est la bonne syntaxe ?

D'autre part je ne trouve pas la méthode qui retourne le nom d'un widget 
pointé mais j'obtiens son type, en faisant :

   print (widget)

L'instance widget a le focus et j'ai besoin de connaître son type, mais 
de plus son petit nom (ex: listbox1, entry2 etc).


Merci de m'aider.
--
Pierre ESTREM

Re: Questions Python

[Marc Chantreux]

Bonjour Pierre,

> Pour obtenir une liste des méthodes d'un objet de classe Canvas, je lis
> qu'on peut exécuter :
> print dir (Canvas())

c'est de la syntaxe python2 (ce qui me fait me demander quelle doc. tu
lis: vraiment, si tu n'as pas de code historique à gérer, oublie
python2).

> J'obtiens une erreur ; quelle est la bonne syntaxe ?

c'était la bonne syntaxe... il y a longtemps. il faut maintenant mettre
des parenthèses autour des arguments de print.

> D'autre part je ne trouve pas la méthode qui retourne le nom d'un widget
> pointé mais j'obtiens son type, en faisant :
>  print (widget)
> L'instance widget a le focus et j'ai besoin de connaître son type, mais de
> plus son petit nom (ex: listbox1, entry2 etc).

de quoi tu parles quand tu dis "le nom"? si c'est le nom de la variable
qui y fait référence dans ton programme (ce que je crois comprendre en lisant 
l'exemple),
ce soit etre possible en comparant le la référence à toutes les clefs du
namespace standard. je ne crois pas que python aie une fonction builtin
pour ca vu le peu d'intéret de la chose.

> Merci de m'aider.

Avec plaisir. Je me permet 2 conseils:

* reposer ces questions aurprès des gens de l'afpy (ils avaient une
  liste de diffusion naguère mais ils ont une vision bien a eux de la
  modernité et sont passés sur discourse: https://discuss.afpy.org/):
  python est exactement le sujet de leurs fora, tu y trouveras des gens
  calés, passionnés et sympa.
* si tu souhaites explorer les objets, découvrir leurs methodes, les
  valeurs et tout ça (on appelle ça l'introspection), il existe ipython3
  qui devrait de faciliter la vie.

cordialement,

-- 
Marc Chantreux
Pôle de Calcul et Services Avancés à la Recherche (CESAR)
http://annuaire.unistra.fr/p/20200

Re: Questions Python

[Yahoo]

Bonjour,

pour lister les méthodes d'une "class" en python3 il faut mettre dir 
dans les parenthèses


print(dir(Canvas))

Cdt

Loïc

Le 29/08/2022 à 23:14, Pierre ESTREM a écrit :

Bonjour la liste,

Dans mon apprentissage de Python, je progresse et sèche sur des points.

Pour obtenir une liste des méthodes d'un objet de classe Canvas, je 
lis qu'on peut exécuter :

  print dir (Canvas())

J'obtiens une erreur ; quelle est la bonne syntaxe ?

D'autre part je ne trouve pas la méthode qui retourne le nom d'un 
widget pointé mais j'obtiens son type, en faisant :

  print (widget)

L'instance widget a le focus et j'ai besoin de connaître son type, 
mais de plus son petit nom (ex: listbox1, entry2 etc).


Merci de m'aider.
--
Pierre ESTREM

Re: Questions Python

[Sébastien NOBILI]

Bonjour,

(Réponse partielle)

Le 2022-08-29 23:14, Pierre ESTREM a écrit :

Pour obtenir une liste des méthodes d'un objet de classe Canvas, je
lis qu'on peut exécuter :
  print dir (Canvas())


En Python 3 il faut mettre des parenthèses à "print". Ça n'est pas ça
ton erreur ?

Sébastien

Re: Questions de débutant sur un annuaire LDAP

[Olivier]

En effet, ça a l'air assez complet.

Merci de l'info

Le mer. 23 oct. 2019 à 19:27, ajh-valmer  a écrit :

> On Wednesday 23 October 2019 18:06:26 Olivier wrote:
> > 1. Pourriez-vous me conseiller un annuaire LDAP avec les propriétés
> > suivantes:
> > - empaqueté dans Stretch et Buster
> > - adapté pour gérer une liste de 1000 contacts commerciaux ou de
> > collaborateurs
> > - l'ajout, la modification ou la suppression des contacts peut s'opérer
> via
> > une appli web accessible à des non-informaticiens
> > - chaque contact possède plusieurs "champs téléphoniques" (le numéro de
> > ligne directe, le numéro de poste interne, le numéro mobile).
> > 2. Supposons que je veuille que chaque contact de mon annuaire ait un
> > champs Toto.
> > Quel annuaire me permet de pouvoir étendre à la fois le schéma des
> données
> > conservées et l'appli d'édition des données ?
>
>
> LDAP-Studio (ldap en mode graphique) pourrait t'aider.
>
> https://directory.apache.org/studio/
>
> A. Valmer
>
>

Re: Questions de débutant sur un annuaire LDAP

[ajh-valmer]

On Wednesday 23 October 2019 18:06:26 Olivier wrote:
> 1. Pourriez-vous me conseiller un annuaire LDAP avec les propriétés
> suivantes:
> - empaqueté dans Stretch et Buster
> - adapté pour gérer une liste de 1000 contacts commerciaux ou de
> collaborateurs
> - l'ajout, la modification ou la suppression des contacts peut s'opérer via
> une appli web accessible à des non-informaticiens
> - chaque contact possède plusieurs "champs téléphoniques" (le numéro de
> ligne directe, le numéro de poste interne, le numéro mobile).
> 2. Supposons que je veuille que chaque contact de mon annuaire ait un
> champs Toto.
> Quel annuaire me permet de pouvoir étendre à la fois le schéma des données
> conservées et l'appli d'édition des données ?


LDAP-Studio (ldap en mode graphique) pourrait t'aider.

https://directory.apache.org/studio/

A. Valmer

Re: Questions about VRF function in /etc/network/interfaces

[Reco]

Hi.

On Sat, Dec 29, 2018 at 01:31:02PM +0800, Simon Jones wrote:
> Hi all,
> 
> This is my OS:
> 
> > # uname -a
> > Linux dut211 4.9.0-7-amd64 #1 SMP Debian 4.9.110-3+deb9u2 (2015-12-19)
> > x86_64 GNU/Linux

That's not a kernel 4.3, for starters. And it's outdated, consider
upgrading. Does not affect your problem though.


> Now I have to rewrite /etc/network/interfaces to implement this function,
> but I got errors, so I want to know if there is demo about how to define
> VRF interface and implement VRF function in /etc/network/interfaces.
> 
> As I follow your man file, I don't know how to do, and gots errors.

Usual debugging of interfaces(5) involves 'ifup -v' and 'ifdown -v'.


> This is my try on this feature, rewrite /etc/network/interfaces like this
> 
> iface eth0 inet static
> > address 172.18.8.211
> > netmask 255.255.255.0
> > ## management network policy routing rules
> > # management port up rules
> > up ip -4 link add mgmtvrf type vrf table 10
> > up ip -4 link set dev mgmtvrf up
> > up ip -4 link set dev eth0 master mgmtvrf
> > up ip -4 route add default via 172.18.8.1 dev eth0 table 10
> > up ip -4 route add 172.18.8.0/24 dev eth0 table 10
> > up ip -4 rule add from 172.18.8.211/32 table 10
> > post-up sysctl -w net.ipv4.tcp_l3mdev_accept=1
> > # management port down rules
> > down ip -4 route delete default via 172.18.8.1 dev eth0 table 10
> > down ip -4 route delete 172.18.8.0/24 dev eth0 table 10
> > down ip -4 rule delete from 172.18.8.211/32 table 10
> > down ip -4 link set dev eth0 nomaster

'-4' is redundant here (you either modify L2 entities or it can be
guessed from the context), you might remove it as well.

> This is errors I got
> 
> Dec 29 02:38:48 dut211 ifup[8690]: RTNETLINK answers: File exists

This. Everything else in your log is useless.
A simple test shows that:

$ ifup -v eth0

ifup: configuring interface eth0=eth0 (inet)
/bin/run-parts --exit-on-error --verbose /etc/network/if-pre-up.d
/bin/ip addr add 172.18.8.211/255.255.255.0 broadcast 172.18.8.255
dev eth0 label eth0
/bin/ip link set dev eth0   up

ip -4 link add mgmtvrf type vrf table 10
ip -4 link set dev mgmtvrf up
ip -4 link set dev eth0 master mgmtvrf
ip -4 route add default via 172.18.8.1 dev eth0 table 10
ip -4 route add 172.18.8.0/24 dev eth0 table 10
RTNETLINK answers: File exists
ifup: failed to bring up eth0

So, it's all good until you try to add an additional route to
172.18.8.0/24, because this route is there already:

$ ip ro l table 10
broadcast 172.18.8.0 dev eth0 proto kernel scope link src 172.18.8.211
172.18.8.0/24 dev eth0 proto kernel scope link src 172.18.8.211
local 172.18.8.211 dev eth0 proto kernel scope host src 172.18.8.211
broadcast 172.18.8.255 dev eth0 proto kernel scope link src 172.18.8.211

And you've got your 'down' rules wrong, you should delete your custom
'mgmtvrf' interface:

# ifdown eth0
# ip a l dev mgmtvrf
5: mgmtvrf:  mtu 65536 qdisc noqueue state UP
group default qlen 1000
link/ether 4a:dc:f1:71:c7:00 brd ff:ff:ff:ff:ff:ff

And, of course, there's a leftover kernel knob:

# /sbin/sysctl net.ipv4.tcp_l3mdev_accept
net.ipv4.tcp_l3mdev_accept = 1

Summing all this up:

iface eth0 inet static
 address 172.18.8.211
 netmask 255.255.255.0
 ## management network policy routing rules
 # management port up rules
 up ip link add mgmtvrf type vrf table 10
 up ip link set dev mgmtvrf up
 up ip link set dev eth0 master mgmtvrf
 up ip route add default via 172.18.8.1 dev eth0 table 10
 up ip rule add from 172.18.8.211/32 table 10
 post-up sysctl -qw net.ipv4.tcp_l3mdev_accept=1
 # management port down rules
 down ip -4 route delete default via 172.18.8.1 dev eth0 table 10
 down ip -4 route delete 172.18.8.0/24 dev eth0 table 10
 down ip -4 rule delete from 172.18.8.211/32 table 10
 down ip -4 link set dev eth0 nomaster
 down ip -4 link del mgmtvrf
 post-down sysctl -qw net.ipv4.tcp_l3mdev_accept=0

Reco

Re: Questions about upgrading from stretch to testing

[Shane Dev]

On Wed, 14 Nov 2018 at 16:04, john doe  wrote:

>
> Any reasons why you want package from testing and not from Stretch
> backports?
>
> because I didn't know Stretch backports existed - now I do. It seems to
offer to the stability of the stretch destribution with the ability to
upgrade certain packages to more up to date versions. Thanks for the tip!

Re: Questions about upgrading from stretch to testing

[john doe]

On 11/14/2018 12:58 PM, Shane Dev wrote:
> Hello,
> 
> I downloaded firmware-9.6.0-amd64-netinst.iso and successfully installed a
> minimal debian stretch on my UEFI/GPT disk. I then I replaced all
> references to "stretch" with "testing" in /etc/apt/sources.list and
> executed sudo apt update; sudo apt install gnome-core firefox-esr.
>

Any reasons why you want package from testing and not from Stretch
backports?

-- 
John Doe

Re: Questions about upgrading from stretch to testing

[songbird]

Shane Dev wrote:
...
> 3. Last time I tried apt upgrade, my grub menu was replaced with the grub
> command prompted. If there any way to avoid this happening again?

  while testing has been fairly decent, once in a while
there are issues you may have to resolve via other means.

  if you are new enough to linux and Debian i do not 
recommend using testing because once in a while you
may be stuck.

  are you running this system and expecting it to 
always be available?

  what i do to make sure i have a booting system is
that i keep a stable partition and i also keep a 
bootable USB stick.

  i did have to use these this past month or so.


  songbird

Re: Questions about upgrading from stretch to testing

[Dan Ritter]

Shane Dev wrote: 
> Hello,
> 
> I downloaded firmware-9.6.0-amd64-netinst.iso and successfully installed a
> minimal debian stretch on my UEFI/GPT disk. I then I replaced all
> references to "stretch" with "testing" in /etc/apt/sources.list and
> executed sudo apt update; sudo apt install gnome-core firefox-esr.
> Everything works as expected but I have the following questions which I
> posted on the debian-testing mailing list. Since no-one answered, I thought
> I would try here -

apt update loads in new packages to your database; the apt
install installed those particular packages and their
dependencies.

You now have a system which is mostly stretch and partially
testing, and will become more testing over time.

If you want to get everything to testing, do

sudo apt dist-upgrade

> 1. according to uname -r, I am running 4.9.0-7-amd64 which appears to be
> the stretch kernel. Is it safe to run the stretch kernel with programs /
> apps / utilities from testing?

Yes. If a testing version needs a later kernel, that will be a
dependency and it will be installed. (And you will need to
reboot.)

> 2. I noticed the intel-microcode package is not installed. If I install it,
> could this testing version conflict with the stretch kernel?

If it depends on a certain version, see above answer.


> 3. Last time I tried apt upgrade, my grub menu was replaced with the grub
> command prompted. If there any way to avoid this happening again?

Not enough information to tell what happened.

-dsr-

Re: Questions sur les outils de gestion d'un dépôt privé apt

[randy11]

> From: "Olivier" 
> To: "ML Debian User French" 
> Sent: Wednesday, 13 December, 2017 6:08:08 PM
> Subject: Questions sur les outils de gestion d'un dépôt privé apt

> J'ai survolé la description de d'aptly qui a l'air de bien répondre
> mais un retour d'expérience serait le bienvenu.

Bonjour, 
J'ai mis en place un Aptly pour Jessie, Stretch et Ubuntu Xenial. Un vrai 
bonheur ! 

J'utilise la version du site d'origine et pas celle de la distribution qui est 
plus 
ancienne. Un seul bémol : les sources ne sont pas gérés quand la 
compression est faite avec "bzip". 

Le scénario classique : 
- aptly repo create ... (à faire une seule fois) 
- aptly repo add ... 
- aptly snapshot create ... from repo ... 
- aptly publish snapshot  (à faire une seule fois) 
- aplty publish switch   

Si je ne fais pas le ménage dans mon dépôt, je conserve toutes les 
version de mes paquets. 
Les snapshots sont utiles quand tu t’aperçois qu'une erreur a été 
commise et qu'il faut très rapidement faire marche arrière. Dans 
ce cas : 
- aptly snapshot list 
- aptly publish switch   

> Le point qui me semble le plus obscur est la maîtrise, avec aptly ou
> autre chose, de la version des paquets installés:
> 1. le dépôt de Debian/Stetch héberge le logiciel toto.1.1,
> 2. tant bien que mal, je crée un paquet 1.1.3 que je pousse dans mon
> dépôt privé,
> 3. un apt-get upgrade sur un serveur utilisant exclusivement ce dépôt
> privé, installe le paquet toto-1.1.3
> 4. après une commande magique je peux obtenir un retour arrière si
> besoin (au besoin en falsifiant un paquet 1.1.4 qui aurait dans les
> faits le même contenu qu'un paquet 1.1.X antérieur
> 5. comment se prémunir d'une mise à jour dans le dépôt officiel de
> Debian

> Qu'en pensez-vous ?

> Slts

Mon utilisation pour l'instant consiste à installer des paquets 
"maison", donc avec un nom absent des paquets Debian. Aucun 
risque de conflit. 

Comme cela a été proposé, s'inspirer de ce qui est fait pour "backports" 
et l'utilisation de "/etc/apt/preferences" est une très bonne idée. 
Mais cela dépend avant tout de ce que tu veux faire et non pas 
de l'outil (Aptly) pour le miroir. 

Si tu te contente de rajouter une source qui est ton miroir privé dans 
"/etc/apt/sources.list.d/" et changeant seulement le numéro de version, 
quand un nouveau paquet avec un numéro de version supérieur sera 
disponible; il sera toujours installé. 

En jouant avec "/etc/apt/preferences", tu peux (à choisir) mettre ton 
dépôt comme source prioritaire ou des paquets (en les nommants) 
comme ne pouvant être mis à jour que par ton dépôt. 

Ton problème de mise à jour existera tant qu'un paquet aura le même 
nom dans plusieurs dépôt. 

Par exemple, les "backports" utilise l'extension "bpo" dans le nom de 
version. La comparaison par défaut se fait sur une comparaison de 
chaîne de caractère classique : 1.2.3 < 1.2.3.aaa < 1.2.3.aab 

Re: Questions sur les outils de gestion d'un dépôt privé apt

[Gabriel Moreau]

Ça ne devrait pas se produire car les versions des logiciels dans la branche
stable ne changent pas (ça c’est l’ancienne théorie, car dans la pratique, il
arrive que certains changent, Firefox par exemple).


Ou le noyau qui a pété il y a quelques jours une bonne partie des 
machines bi-proc !


gaby
--
Gabriel Moreau - IR CNRShttp://www.legi.grenoble-inp.fr
LEGI (UMR 5519) Laboratoire des Ecoulements Geophysiques et Industriels
Domaine Universitaire, CS 40700, 38041 Grenoble Cedex 9, France
mailto:gabriel.mor...@legi.grenoble-inp.fr  tel:+33.476.825.015

Re: Questions sur les outils de gestion d'un dépôt privé apt

[Sébastien NOBILI]

Le mercredi 13 décembre 2017 à 19:08, Olivier a écrit :
> 2. de centraliser l'accès à des dépôts apt de plusieurs serveurs afin que
> ces serveurs n'aient plus directement besoin d'accéder à Internet pour
> installer de nouveaux paquets ou mettre des paquets anciens (un mirroir me
> semble bien répondre à ce besoin puisque seul le mirroir a besoin
> d'Internet pour remplir son office).

« apt-cacher-ng » peut peut-être répondre à ce besoin.

> Le point qui me semble le plus obscur est la maîtrise, avec aptly ou autre
> chose, de la version des paquets installés:
> 1. le dépôt de Debian/Stetch héberge le logiciel toto.1.1,
> 2. tant bien que mal, je crée un paquet 1.1.3 que je pousse dans mon dépôt
> privé,
> 3. un apt-get upgrade sur un serveur utilisant exclusivement ce dépôt
> privé, installe le paquet toto-1.1.3
> 4. après une commande magique je peux obtenir un retour arrière si besoin
> (au besoin en falsifiant un paquet 1.1.4 qui aurait dans les faits le même
> contenu qu'un paquet 1.1.X antérieur
> 5. comment se prémunir d'une mise à jour dans le dépôt officiel de Debian

Ça ne devrait pas se produire car les versions des logiciels dans la branche
stable ne changent pas (ça c’est l’ancienne théorie, car dans la pratique, il
arrive que certains changent, Firefox par exemple).

Sébastien

Re: Questions sur les outils de gestion d'un dépôt privé apt

[yamo']

Salut,
Olivier a écrit le 13/12/2017 à 19:10 :

> Qu'en pensez-vous ?


Est-ce que la solution décrite sur la page ci-dessous répond au besoin?


-- 
Stéphane

Re: Questions sur les outils de gestion d'un dépôt privé apt

[Daniel Caillibaud]

Le 13/12/17 à 19:08, Olivier  a écrit :
O> Bonjour,
O> 
O> J'ai un double besoin :
O> 
O> 1. empaqueter puis distribuer pour Stretch des logiciels, en version un
O> peu plus récente, que celle actuellement dans les dépôts publics de
O> Debian,

Je suppose que tu connais stretch-backports et que ça ne répond pas à ton
besoin…

O> 2. de centraliser l'accès à des dépôts apt de plusieurs serveurs afin que
O> ces serveurs n'aient plus directement besoin d'accéder à Internet pour
O> installer de nouveaux paquets ou mettre des paquets anciens (un mirroir
O> me semble bien répondre à ce besoin puisque seul le mirroir a besoin
O> d'Internet pour remplir son office).
O> 
O> Quel outils/architecture recommander pour couvrir ces 2 besoins ?
O> 
O> J'ai survolé la description de d'aptly qui a l'air de bien répondre mais
O> un retour d'expérience serait le bienvenu.

Dsl connaît pas…

O> Le point qui me semble le plus obscur est la maîtrise, avec aptly ou
O> autre chose, de la version des paquets installés:
O> 1. le dépôt de Debian/Stetch héberge le logiciel toto.1.1,
O> 2. tant bien que mal, je crée un paquet 1.1.3 que je pousse dans mon
O> dépôt privé,
O> 3. un apt-get upgrade sur un serveur utilisant exclusivement ce dépôt
O> privé, installe le paquet toto-1.1.3
O> 4. après une commande magique je peux obtenir un retour arrière si besoin
O> (au besoin en falsifiant un paquet 1.1.4 qui aurait dans les faits le
O> même contenu qu'un paquet 1.1.X antérieur

Tu peux forcer la réinstallation d'un paquet antérieur (en précisant sa
version quand tu l'installe) et le bloquer à cette version (hold).

O> 5. comment se prémunir d'une mise à jour dans le dépôt officiel de Debian
O> 
O> Qu'en pensez-vous ?

Tu peux pour tout ça jouer avec apt-preferences, mettre une priorité plus
élevée pour ton dépôt privé (cherche aussi "debian package pinning" sur le
net pour voir les façons de faire et les risques à maîtriser).

Mais pour éviter une màj auto dans une version antérieure mise à jour sur
le dépôt debian officiel, le mieux est d'adapter pour tes paquets privés une
numérotation qui suit la version des sources, ce que fait debian (en
ajoutant éventuellement un suffixe), ainsi un n° de version de paquet plus
élevé correspondra toujours à une version plus récente du source (donc
créer une 1.1.4 qui serait une 1.1.2 pour forcer un downgrade de 1.1.3 me
semble une mauvaise idée)

man apt_preferences
man apt-cache # commande policy pour vérifier que tes préférences donnent le
  # résultat attendu

-- 
Daniel

Il est vrai qu'on ne peut trouver la pierre philosophale, 
mais il est bon qu'on la cherche.
Bernard Fontenelle

Re: Questions on nonfree software

[Brian]

On Sat 01 Jul 2017 at 11:12:44 -0700, Charles Chambers wrote:

> Is the firmware-9.0.0-amd64-DVD-1 the right media to use during
> installation if I have hardware that requires the nonfree repository?

Yes.

Re: Questions after doing update and upgrade on Stretch

[Dejan Jocic]

On 27-05-17, Dejan Jocic wrote:
> On 26-05-17, Fungi4All wrote:
> > From: jode...@gmail.com
> > As for reason why old one is not removed, it is because Debian keeps not
> > just newly installed kernel, but also one previously installed. One
> > before that was removed by apt-get autoremove.
> > 
> > Unless different desktops have different autoremove behavior in my recent
> > experience kernels can only be removed manually. If you get a current
> > today's image of 8.8 and update/upgrade, then switch to testing and upd/upg,
> > then unstable and upd/upg and autoremove as many times as you like,
> > you will end up with 3 kernels as a choice to boot unstable from.
> > I think in the past it did not work this way.
> 
> Please, do not replay to me and cc it to the list, you mess things up.
> No need to cc me either, I'm subscribed to the list. Thank you in
> advance for that.
> 
> As for subject itself, no, it is not desktop related. For your specific
> case of upgrading from stable to unstable via testing, have no idea,
> never did it that way. 

Just to clarify it a bit, what happens with your old kernel packages
should be set with:

/etc/kernel/postinst.d/apt-auto-removal

which generates file:

/etc/apt/apt.conf.d/01autoremove-kernels

Re: Questions after doing update and upgrade on Stretch

[Dejan Jocic]

On 26-05-17, Fungi4All wrote:
> From: jode...@gmail.com
> As for reason why old one is not removed, it is because Debian keeps not
> just newly installed kernel, but also one previously installed. One
> before that was removed by apt-get autoremove.
> 
> Unless different desktops have different autoremove behavior in my recent
> experience kernels can only be removed manually. If you get a current
> today's image of 8.8 and update/upgrade, then switch to testing and upd/upg,
> then unstable and upd/upg and autoremove as many times as you like,
> you will end up with 3 kernels as a choice to boot unstable from.
> I think in the past it did not work this way.

Please, do not replay to me and cc it to the list, you mess things up.
No need to cc me either, I'm subscribed to the list. Thank you in
advance for that.

As for subject itself, no, it is not desktop related. For your specific
case of upgrading from stable to unstable via testing, have no idea,
never did it that way. 

Re: Questions after doing update and upgrade on Stretch

[Fungi4All]

From: jode...@gmail.com
As for reason why old one is not removed, it is because Debian keeps not
just newly installed kernel, but also one previously installed. One
before that was removed by apt-get autoremove.

Unless different desktops have different autoremove behavior in my recent
experience kernels can only be removed manually. If you get a current
today's image of 8.8 and update/upgrade, then switch to testing and upd/upg,
then unstable and upd/upg and autoremove as many times as you like,
you will end up with 3 kernels as a choice to boot unstable from.
I think in the past it did not work this way.

Re: Questions after doing update and upgrade on Stretch

[Dejan Jocic]

On 26-05-17, David Wright wrote:
> On Mon 22 May 2017 at 11:12:02 (+0200), Dejan Jocic wrote:
> > On 21-05-17, Brian wrote:
> > > On Sun 21 May 2017 at 22:18:11 +0200, Dejan Jocic wrote:
> > > 
> > > > On 21-05-17, David Wright wrote:
> > > > > On Sun 21 May 2017 at 16:31:55 (+0200), Dejan Jocic wrote:
> > > > > > > 
> > > > > > As for number 1 can't say much about it, I do not get it either. 
> > > > > > But 2
> > > > > > happens because you've used apt-get upgrade instead of apt-get
> > > > > > dist-upgrade. Packages that will uninstall some packages already
> > > > > > installed on your system and that will change some dependencies 
> > > > > > require dist-upgrade.
> > > > > 
> > > > > Agreed.
> > > > > 
> > > > > > It happens always in case of linux-image packages. 
> > > > > 
> > > > > Is this¹ new with stretch? My linux-images upgrade just like any other
> > > > > package; here's the penultimate occasion for jessie:
> > > > > 
> > > > > Start-Date: 2017-03-08  19:20:34
> > > > > Commandline: apt-get upgrade
> > > > > Upgrade: linux-source-3.16:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > > > linux-headers-3.16.0-4-586:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > > > linux-image-3.16.0-4-586:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > > > linux-libc-dev:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > > > linux-compiler-gcc-4.8-x86:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > > > linux-headers-3.16.0-4-common:i386 (3.16.39-1+deb8u1, 
> > > > > 3.16.39-1+deb8u2)
> > > > > End-Date: 2017-03-08  19:22:50
> > > > > 
> > > > > (The last one's log was rather larger.)
> > > > 
> > > > Ehh, sorry not sure if it is new with Stretch, can't remember for
> > > > Jessie. I'm certain that it was like that on Stretch and on Ubuntu
> > > > 16.04.
> > > 
> > > Rather than just a contrast, I was rather hoping to hear how David
> > > Wright's observations (which I agree with)fit in with yours.
> > > 
> > Sorry, not sure what I can add to it. 
> > 
> > > > > > It will leave your previous working linux-image on though, but will 
> > > > > > uninstall one older than that, so you will always end up with 
> > > > > > chance to 
> > > > > > boot in working kernel, if new one messes up some things.
> > > > > 
> > > > > Same question. My wheezy system has had at least 28 linux-image
> > > > > upgrades (3.2.57-3+deb7u2→3.2.60-1+deb7u1 to 3.2.86-1→3.2.88-1)
> > > > > but there's still only one kernel image on the system:
> > > > > 
> > > > > -rw-r--r-- 1 root root  134839 Apr 27 16:52 config-3.2.0-4-686-pae
> > > > > drwxr-xr-x 3 root root   12288 Apr 28 07:44 grub
> > > > > -rw-r--r-- 1 root root 2672854 Apr 28 07:44 initrd.img-3.2.0-4-686-pae
> > > > > -rw-r--r-- 1 root root 1674268 Apr 27 16:52 System.map-3.2.0-4-686-pae
> > > > > -rw-r--r-- 1 root root 2709184 Apr 27 16:51 vmlinuz-3.2.0-4-686-pae
> > > > > 
> > > > > (I have to notice these upgrades myself because they overwrite
> > > > > my edited version of /boot/grub/grub.cfg which I then replace.)
> > > > > 
> > > > > Cheers,
> > > > > David.
> > > > > 
> > > > > ¹ I'm not disagreeing that something is holding back the upgrade
> > > > > on this specific occasion, but this is unusual.
> > 
> > 
> > This is what I have in /boot and, as stated above, usual outcome of
> > upgrades, both in Stretch and in Ubuntu 16.04:
> > 
> > -rw-r--r-- 1 root root   186695 Mar 30 03:16 config-4.9.0-2-amd64
> > -rw-r--r-- 1 root root   186380 May  2 17:21 config-4.9.0-3-amd64
> > drwxr-xr-x 5 root root 4096 May 19 08:55 grub
> > -rw-r--r-- 1 root root 19660713 May 17 16:43 initrd.img-4.9.0-2-amd64
> > -rw-r--r-- 1 root root 19534447 May 18 08:40 initrd.img-4.9.0-3-amd64
> > -rw-r--r-- 1 root root  3169870 Mar 30 03:16 System.map-4.9.0-2-amd64
> > -rw-r--r-- 1 root root  3176652 May  2 17:21 System.map-4.9.0-3-amd64
> > -rw-r--r-- 1 root root  4193832 Mar 30 18:43 vmlinuz-4.9.0-2-amd64
> > -rw-r--r-- 1 root root  4204320 May  2 17:21 vmlinuz-4.9.0-3-amd64
> > 
> > As you can see, 2 kernels. All i do is my morning routine which consists
> > of apt-get update && apt-get upgrade, followed by apt-get dist-upgrade
> > in case of need(as stated before, usually case of need is kernel
> > upgrade). Of course, it is followed by apt-get autoremove too.I do not
> > have anything unusual on this system.  It was fairly recent install of
> > jessie(few months ago), followed by upgrade to stretch right after
> > install. Was installed from unofficial net-install cd with firmware on
> > it. Also, nothing was changed in apt preferences and stuff like that.
> 
> OK, I've just been looking at kernel changelogs and that explains
> what's going on here. There's nothing different about kernels in
> particular, and the reason you've got two kernels is that they
> are different kernels, so different Debian packages.
> 
> I installed wheezy no earlier than May 2013 (ie not as a release
> candidate) and 3.2.0-4 came out in Sept 2012, so as I said
> I've had at least 28 upgrades of the one 

Re: Questions after doing update and upgrade on Stretch

[David Wright]

On Mon 22 May 2017 at 11:12:02 (+0200), Dejan Jocic wrote:
> On 21-05-17, Brian wrote:
> > On Sun 21 May 2017 at 22:18:11 +0200, Dejan Jocic wrote:
> > 
> > > On 21-05-17, David Wright wrote:
> > > > On Sun 21 May 2017 at 16:31:55 (+0200), Dejan Jocic wrote:
> > > > > > 
> > > > > As for number 1 can't say much about it, I do not get it either. But 2
> > > > > happens because you've used apt-get upgrade instead of apt-get
> > > > > dist-upgrade. Packages that will uninstall some packages already
> > > > > installed on your system and that will change some dependencies 
> > > > > require dist-upgrade.
> > > > 
> > > > Agreed.
> > > > 
> > > > > It happens always in case of linux-image packages. 
> > > > 
> > > > Is this¹ new with stretch? My linux-images upgrade just like any other
> > > > package; here's the penultimate occasion for jessie:
> > > > 
> > > > Start-Date: 2017-03-08  19:20:34
> > > > Commandline: apt-get upgrade
> > > > Upgrade: linux-source-3.16:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > > linux-headers-3.16.0-4-586:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > > linux-image-3.16.0-4-586:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > > linux-libc-dev:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > > linux-compiler-gcc-4.8-x86:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > > linux-headers-3.16.0-4-common:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2)
> > > > End-Date: 2017-03-08  19:22:50
> > > > 
> > > > (The last one's log was rather larger.)
> > > 
> > > Ehh, sorry not sure if it is new with Stretch, can't remember for
> > > Jessie. I'm certain that it was like that on Stretch and on Ubuntu
> > > 16.04.
> > 
> > Rather than just a contrast, I was rather hoping to hear how David
> > Wright's observations (which I agree with)fit in with yours.
> > 
> Sorry, not sure what I can add to it. 
> 
> > > > > It will leave your previous working linux-image on though, but will 
> > > > > uninstall one older than that, so you will always end up with chance 
> > > > > to 
> > > > > boot in working kernel, if new one messes up some things.
> > > > 
> > > > Same question. My wheezy system has had at least 28 linux-image
> > > > upgrades (3.2.57-3+deb7u2→3.2.60-1+deb7u1 to 3.2.86-1→3.2.88-1)
> > > > but there's still only one kernel image on the system:
> > > > 
> > > > -rw-r--r-- 1 root root  134839 Apr 27 16:52 config-3.2.0-4-686-pae
> > > > drwxr-xr-x 3 root root   12288 Apr 28 07:44 grub
> > > > -rw-r--r-- 1 root root 2672854 Apr 28 07:44 initrd.img-3.2.0-4-686-pae
> > > > -rw-r--r-- 1 root root 1674268 Apr 27 16:52 System.map-3.2.0-4-686-pae
> > > > -rw-r--r-- 1 root root 2709184 Apr 27 16:51 vmlinuz-3.2.0-4-686-pae
> > > > 
> > > > (I have to notice these upgrades myself because they overwrite
> > > > my edited version of /boot/grub/grub.cfg which I then replace.)
> > > > 
> > > > Cheers,
> > > > David.
> > > > 
> > > > ¹ I'm not disagreeing that something is holding back the upgrade
> > > > on this specific occasion, but this is unusual.
> 
> 
> This is what I have in /boot and, as stated above, usual outcome of
> upgrades, both in Stretch and in Ubuntu 16.04:
> 
> -rw-r--r-- 1 root root   186695 Mar 30 03:16 config-4.9.0-2-amd64
> -rw-r--r-- 1 root root   186380 May  2 17:21 config-4.9.0-3-amd64
> drwxr-xr-x 5 root root 4096 May 19 08:55 grub
> -rw-r--r-- 1 root root 19660713 May 17 16:43 initrd.img-4.9.0-2-amd64
> -rw-r--r-- 1 root root 19534447 May 18 08:40 initrd.img-4.9.0-3-amd64
> -rw-r--r-- 1 root root  3169870 Mar 30 03:16 System.map-4.9.0-2-amd64
> -rw-r--r-- 1 root root  3176652 May  2 17:21 System.map-4.9.0-3-amd64
> -rw-r--r-- 1 root root  4193832 Mar 30 18:43 vmlinuz-4.9.0-2-amd64
> -rw-r--r-- 1 root root  4204320 May  2 17:21 vmlinuz-4.9.0-3-amd64
> 
> As you can see, 2 kernels. All i do is my morning routine which consists
> of apt-get update && apt-get upgrade, followed by apt-get dist-upgrade
> in case of need(as stated before, usually case of need is kernel
> upgrade). Of course, it is followed by apt-get autoremove too.I do not
> have anything unusual on this system.  It was fairly recent install of
> jessie(few months ago), followed by upgrade to stretch right after
> install. Was installed from unofficial net-install cd with firmware on
> it. Also, nothing was changed in apt preferences and stuff like that.

OK, I've just been looking at kernel changelogs and that explains
what's going on here. There's nothing different about kernels in
particular, and the reason you've got two kernels is that they
are different kernels, so different Debian packages.

I installed wheezy no earlier than May 2013 (ie not as a release
candidate) and 3.2.0-4 came out in Sept 2012, so as I said
I've had at least 28 upgrades of the one kernel version.

OTOH the only stretch system I have is currently a basic system,
with just 248 packages (minimal install from RC3). The kernel
version is linux-image-4.9.0-2 at Debian version 4.9.18-1 and that
version is obsolete. Currently we seem 

Re: Questions after doing update and upgrade on Stretch

[Larry Dighera]

On Sun, 21 May 2017 09:35:55 -0500, you wrote:

>> Check with blkid that sda5 has this UUID.
>
>Gparted reports that UUID for /dev/sda5

'lsblk --fs' provides a human-friendly "graphical" tree view that
includes UUID and LABEL of each partition on each disk.

Re: Questions after doing update and upgrade on Stretch

[Brian]

On Mon 22 May 2017 at 05:31:26 -0500, Richard Owlett wrote:

> On 05/21/2017 09:31 AM, Dejan Jocic wrote:
> >>
> >As for number 1 can't say much about it, I do not get it either. But 2
> >happens because you've used apt-get upgrade instead of apt-get
> >dist-upgrade. Packages that will uninstall some packages already
> >installed on your system and that will change some dependencies
> >require dist-upgrade. It happens always in case of linux-image packages.
> >It will leave your previous working linux-image on though, but will
> >uninstall one older than that, so you will always end up with chance
> >to  boot in working kernel, if new one messes up some things.
> >
> 
> If I had problems after doing apt-get dist-upgrade,
>   1. how would I distinguish a kernel problem from other problems?

Far too general a question. Better would be a description of the actual
problem.

>   2. how would I boot with the previous kernel?

Replace the kernel version in the linux and initrd lines within the GRUB
menu. For permanency, use /etc/grub.d/40_custom.

>   3. is there some specific documentation I should be reading?

On what? In general, at this stage of the freeze, dist-upgrade shouldn't
give any problem.

-- 
Brian

Re: Questions after doing update and upgrade on Stretch

[Dejan Jocic]

On 22-05-17, Richard Owlett wrote:
> On 05/21/2017 09:31 AM, Dejan Jocic wrote:
> > On 21-05-17, Richard Owlett wrote:
> > > [snip]
> > > 
> > > My questions:
> > > 
> > > 1. In the first run, I don't understand:
> > >  Processing triggers for initramfs-tools (0.130) ...
> > >  update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
> > >  I: The initramfs will attempt to resume from /dev/sda5
> > >  I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
> > >  I: Set the RESUME variable to override this.
> > >  As /dev/sda5 is my SWAP.
> > > 
> > > 2. I don't understand any implications of:
> > >  The following packages have been kept back:
> > >linux-image-686-pae xorg xserver-xorg
> > > 
> > > TIA
> > > 
> > > 
> > > 
> > > 
> > As for number 1 can't say much about it, I do not get it either. But 2
> > happens because you've used apt-get upgrade instead of apt-get
> > dist-upgrade. Packages that will uninstall some packages already
> > installed on your system and that will change some dependencies
> > require dist-upgrade. It happens always in case of linux-image packages.
> > It will leave your previous working linux-image on though, but will
> > uninstall one older than that, so you will always end up with chance
> > to  boot in working kernel, if new one messes up some things.
> > 
> 
> If I had problems after doing apt-get dist-upgrade,
>   1. how would I distinguish a kernel problem from other problems?
>   2. how would I boot with the previous kernel?
>   3. is there some specific documentation I should be reading?
> 
> I've done online upgrades before having relied on purchased DVD sets of
> point releases due to bandwidth constraints.
> 
> Thank you.
> 
> 

1. That would depend on problem itself, right? As for kernel problems, I
was unusually lucky with them for years and have yet to run on problem
with upgraded kernel. Must be that new kernels really loved my hardware.

2. It is easy task, as long as there is previous kernel present in your
/boot. If everything works as expected you should have it. From debian
administrator handbook:

8.11.1. Features of a Debian Kernel Package

A Debian kernel package installs the kernel image (vmlinuz-version), its
configuration (config-version) and its symbols table
(System.map-version) in /boot/. The symbols table helps developers
understand the meaning of a kernel error message; without it, kernel
“oopses” (an “oops” is the kernel equivalent of a segmentation fault for
user-space programs, in other words messages generated following an
invalid pointer dereference) only contain numeric memory addresses,
which is useless information without the table mapping these addresses
to symbols and function names. The modules are installed in the
/lib/modules/version/ directory.

The package's configuration scripts automatically generate an initrd
image, which is a mini-system designed to be loaded in memory (hence the
name, which stands for “init ramdisk”) by the bootloader, and used by
the Linux kernel solely for loading the modules needed to access the
devices containing the complete Debian system (for example, the driver
for SATA disks). Finally, the post-installation scripts update the
symbolic links /vmlinuz, /vmlinuz.old, /initrd.img and /initrd.img.old
so that they point to the latest two kernels installed, respectively, as
well as the corresponding initrd images.

Most of those tasks are offloaded to hook scripts in the
/etc/kernel/*.d/ directories. For instance, the integration with grub
relies on /etc/kernel/postinst.d/zz-update-grub and
/etc/kernel/postrm.d/zz-update-grub to call update-grub when kernels are
installed or removed. 

As you can see there, if everything works as intended, you will still
have your old kernel installed, which you can choose from grub menu.
Think that it is under advanced options, but did not use it recently and
am not in mood to reboot now to check. Anyway, as long as you do not use
apt-get autoremove, even your older than previous kernel packages should
be around, though I did not have need for that.

3. https://debian-handbook.info/ and usual man pages for grub, kernel,
apt-get and friends, I guess.

Pleasure :)

Re: Questions after doing update and upgrade on Stretch

[Richard Owlett]

On 05/21/2017 09:31 AM, Dejan Jocic wrote:

On 21-05-17, Richard Owlett wrote:

[snip]

My questions:

1. In the first run, I don't understand:
 Processing triggers for initramfs-tools (0.130) ...
 update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
 I: The initramfs will attempt to resume from /dev/sda5
 I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
 I: Set the RESUME variable to override this.
 As /dev/sda5 is my SWAP.

2. I don't understand any implications of:
 The following packages have been kept back:
   linux-image-686-pae xorg xserver-xorg

TIA





As for number 1 can't say much about it, I do not get it either. But 2
happens because you've used apt-get upgrade instead of apt-get
dist-upgrade. Packages that will uninstall some packages already
installed on your system and that will change some dependencies
require dist-upgrade. It happens always in case of linux-image packages.
It will leave your previous working linux-image on though, but will
uninstall one older than that, so you will always end up with chance
to  boot in working kernel, if new one messes up some things.



If I had problems after doing apt-get dist-upgrade,
  1. how would I distinguish a kernel problem from other problems?
  2. how would I boot with the previous kernel?
  3. is there some specific documentation I should be reading?

I've done online upgrades before having relied on purchased DVD sets of 
point releases due to bandwidth constraints.


Thank you.

Re: Questions after doing update and upgrade on Stretch

[Brian]

On Mon 22 May 2017 at 11:12:02 +0200, Dejan Jocic wrote:

> On 21-05-17, Brian wrote:
> > On Sun 21 May 2017 at 22:18:11 +0200, Dejan Jocic wrote:
> > 
> > > Sorry, but you are doing it wrong way. Grub 2 should not be customized
> > > by editing /boot/grub/grub.cfg , but by editing /etc/default/grub and
> > > files in /etc/grub.d/. Reason is obvious, your customization is lost
> > > whenever something related to linux-image is upgraded. Just saying :)
> > 
> > "Wrong" isn't quite the right way to put. For most people in most
> > circumstances editing grub.cfg and using update-grub is a wise procedure
> > and to be advocated. But a hand-crafted grub.cfg can be very useful.
> > update-grub can be prevented from getting its hands on it with
> > dpkg-divert,
> > 
> > -- 
> > Brian.
> > 
> 
> Well, I was always under assumption that in case of grub2 you can change
> anything in grub.cfg by editing /etc/default/grub and files in
> /etc/grub.d/. Perhaps I was wrong about it, in which case I do
> apologise.

No need to apologise. Your assumption is correct and your advice sound;
100% of Debian users should thank you for it. The 0.1% who wander from
the straight and narrow presumably know what they are doing.

-- 
Brian.

Re: Questions after doing update and upgrade on Stretch

[Dejan Jocic]

On 21-05-17, Brian wrote:
> On Sun 21 May 2017 at 22:18:11 +0200, Dejan Jocic wrote:
> 
> > On 21-05-17, David Wright wrote:
> > > On Sun 21 May 2017 at 16:31:55 (+0200), Dejan Jocic wrote:
> > > > > 
> > > > As for number 1 can't say much about it, I do not get it either. But 2
> > > > happens because you've used apt-get upgrade instead of apt-get
> > > > dist-upgrade. Packages that will uninstall some packages already
> > > > installed on your system and that will change some dependencies 
> > > > require dist-upgrade.
> > > 
> > > Agreed.
> > > 
> > > > It happens always in case of linux-image packages. 
> > > 
> > > Is this¹ new with stretch? My linux-images upgrade just like any other
> > > package; here's the penultimate occasion for jessie:
> > > 
> > > Start-Date: 2017-03-08  19:20:34
> > > Commandline: apt-get upgrade
> > > Upgrade: linux-source-3.16:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > linux-headers-3.16.0-4-586:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > linux-image-3.16.0-4-586:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > linux-libc-dev:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > linux-compiler-gcc-4.8-x86:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > > linux-headers-3.16.0-4-common:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2)
> > > End-Date: 2017-03-08  19:22:50
> > > 
> > > (The last one's log was rather larger.)
> > 
> > Ehh, sorry not sure if it is new with Stretch, can't remember for
> > Jessie. I'm certain that it was like that on Stretch and on Ubuntu
> > 16.04.
> 
> Rather than just a contrast, I was rather hoping to hear how David
> Wright's observations (which I agree with)fit in with yours.
> 
Sorry, not sure what I can add to it. 

> > > > It will leave your previous working linux-image on though, but will 
> > > > uninstall one older than that, so you will always end up with chance to 
> > > > boot in working kernel, if new one messes up some things.
> > > 
> > > Same question. My wheezy system has had at least 28 linux-image
> > > upgrades (3.2.57-3+deb7u2→3.2.60-1+deb7u1 to 3.2.86-1→3.2.88-1)
> > > but there's still only one kernel image on the system:
> > > 
> > > -rw-r--r-- 1 root root  134839 Apr 27 16:52 config-3.2.0-4-686-pae
> > > drwxr-xr-x 3 root root   12288 Apr 28 07:44 grub
> > > -rw-r--r-- 1 root root 2672854 Apr 28 07:44 initrd.img-3.2.0-4-686-pae
> > > -rw-r--r-- 1 root root 1674268 Apr 27 16:52 System.map-3.2.0-4-686-pae
> > > -rw-r--r-- 1 root root 2709184 Apr 27 16:51 vmlinuz-3.2.0-4-686-pae
> > > 
> > > (I have to notice these upgrades myself because they overwrite
> > > my edited version of /boot/grub/grub.cfg which I then replace.)
> > > 
> > > Cheers,
> > > David.
> > > 
> > > ¹ I'm not disagreeing that something is holding back the upgrade
> > > on this specific occasion, but this is unusual.


This is what I have in /boot and, as stated above, usual outcome of
upgrades, both in Stretch and in Ubuntu 16.04:

-rw-r--r-- 1 root root   186695 Mar 30 03:16 config-4.9.0-2-amd64
-rw-r--r-- 1 root root   186380 May  2 17:21 config-4.9.0-3-amd64
drwxr-xr-x 5 root root 4096 May 19 08:55 grub
-rw-r--r-- 1 root root 19660713 May 17 16:43 initrd.img-4.9.0-2-amd64
-rw-r--r-- 1 root root 19534447 May 18 08:40 initrd.img-4.9.0-3-amd64
-rw-r--r-- 1 root root  3169870 Mar 30 03:16 System.map-4.9.0-2-amd64
-rw-r--r-- 1 root root  3176652 May  2 17:21 System.map-4.9.0-3-amd64
-rw-r--r-- 1 root root  4193832 Mar 30 18:43 vmlinuz-4.9.0-2-amd64
-rw-r--r-- 1 root root  4204320 May  2 17:21 vmlinuz-4.9.0-3-amd64

As you can see, 2 kernels. All i do is my morning routine which consists
of apt-get update && apt-get upgrade, followed by apt-get dist-upgrade
in case of need(as stated before, usually case of need is kernel
upgrade). Of course, it is followed by apt-get autoremove too.I do not
have anything unusual on this system.  It was fairly recent install of
jessie(few months ago), followed by upgrade to stretch right after
install. Was installed from unofficial net-install cd with firmware on
it. Also, nothing was changed in apt preferences and stuff like that.

> > 
> > Sorry, but you are doing it wrong way. Grub 2 should not be customized
> > by editing /boot/grub/grub.cfg , but by editing /etc/default/grub and
> > files in /etc/grub.d/. Reason is obvious, your customization is lost
> > whenever something related to linux-image is upgraded. Just saying :)
> 
> "Wrong" isn't quite the right way to put. For most people in most
> circumstances editing grub.cfg and using update-grub is a wise procedure
> and to be advocated. But a hand-crafted grub.cfg can be very useful.
> update-grub can be prevented from getting its hands on it with
> dpkg-divert,
> 
> -- 
> Brian.
> 

Well, I was always under assumption that in case of grub2 you can change
anything in grub.cfg by editing /etc/default/grub and files in
/etc/grub.d/. Perhaps I was wrong about it, in which case I do
apologise.

Re: Questions after doing update and upgrade on Stretch

[David Wright]

On Sun 21 May 2017 at 22:18:11 (+0200), Dejan Jocic wrote:
> On 21-05-17, David Wright wrote:
> > -rw-r--r-- 1 root root  134839 Apr 27 16:52 config-3.2.0-4-686-pae
> > drwxr-xr-x 3 root root   12288 Apr 28 07:44 grub
> > -rw-r--r-- 1 root root 2672854 Apr 28 07:44 initrd.img-3.2.0-4-686-pae
> > -rw-r--r-- 1 root root 1674268 Apr 27 16:52 System.map-3.2.0-4-686-pae
> > -rw-r--r-- 1 root root 2709184 Apr 27 16:51 vmlinuz-3.2.0-4-686-pae
> > 
> > (I have to notice these upgrades myself because they overwrite
> > my edited version of /boot/grub/grub.cfg which I then replace.)
> > 
> 
> Sorry, but you are doing it wrong way. Grub 2 should not be customized
> by editing /boot/grub/grub.cfg , but by editing /etc/default/grub and
> files in /etc/grub.d/. Reason is obvious, your customization is lost
> whenever something related to linux-image is upgraded. Just saying :)

Yes, it would be nice to use the Debian Way. But there is not enough
flexibility in their scripts to do what I want.

a) the logic of
 # Uncomment if you don't want GRUB to pass "root=UUID=xxx" parameter to 
Linux
 #GRUB_DISABLE_LINUX_UUID=true
   is odd and, unfortunately, there's no facility for
 GRUB_ENABLE_LINUX_LABEL=true
   (I did have a shot at modifying the scripts, but it's less trivial
   than it would seem at first sight).

b) they insert entries for sysvinit and recovery but not for fsck.
   I modify the first subentry, using the ids fsck and fsck so I can use
   grub-reboot 'fsck>fsck'
   to invoke it from a script that also logs a timestamp. I refuse to
   type monstrosities like
   grub-reboot 'Advanced options for Debian 
GNU/Linux>gnulinux-advanced-7ccc1c1c-a690-418b-96c0-edcce6ebd3c1'

So it's far easier to write one python script to parse /run/udev/data,
automatically make the changes I want, and keep the original and
edited files as /boot/grub/grub.cfg-{uuids,edited}. After any upgrade,
if the former matches /boot/grub/grub.cfg, it gets overwritten by the
latter.

Cheers,
David.

Re: Questions after doing update and upgrade on Stretch

[Brian]

On Sun 21 May 2017 at 22:18:11 +0200, Dejan Jocic wrote:

> On 21-05-17, David Wright wrote:
> > On Sun 21 May 2017 at 16:31:55 (+0200), Dejan Jocic wrote:
> > > > 
> > > As for number 1 can't say much about it, I do not get it either. But 2
> > > happens because you've used apt-get upgrade instead of apt-get
> > > dist-upgrade. Packages that will uninstall some packages already
> > > installed on your system and that will change some dependencies 
> > > require dist-upgrade.
> > 
> > Agreed.
> > 
> > > It happens always in case of linux-image packages. 
> > 
> > Is this¹ new with stretch? My linux-images upgrade just like any other
> > package; here's the penultimate occasion for jessie:
> > 
> > Start-Date: 2017-03-08  19:20:34
> > Commandline: apt-get upgrade
> > Upgrade: linux-source-3.16:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > linux-headers-3.16.0-4-586:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > linux-image-3.16.0-4-586:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > linux-libc-dev:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > linux-compiler-gcc-4.8-x86:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> > linux-headers-3.16.0-4-common:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2)
> > End-Date: 2017-03-08  19:22:50
> > 
> > (The last one's log was rather larger.)
> 
> Ehh, sorry not sure if it is new with Stretch, can't remember for
> Jessie. I'm certain that it was like that on Stretch and on Ubuntu
> 16.04.

Rather than just a contrast, I was rather hoping to hear how David
Wright's observations (which I agree with)fit in with yours.

> > > It will leave your previous working linux-image on though, but will 
> > > uninstall one older than that, so you will always end up with chance to 
> > > boot in working kernel, if new one messes up some things.
> > 
> > Same question. My wheezy system has had at least 28 linux-image
> > upgrades (3.2.57-3+deb7u2→3.2.60-1+deb7u1 to 3.2.86-1→3.2.88-1)
> > but there's still only one kernel image on the system:
> > 
> > -rw-r--r-- 1 root root  134839 Apr 27 16:52 config-3.2.0-4-686-pae
> > drwxr-xr-x 3 root root   12288 Apr 28 07:44 grub
> > -rw-r--r-- 1 root root 2672854 Apr 28 07:44 initrd.img-3.2.0-4-686-pae
> > -rw-r--r-- 1 root root 1674268 Apr 27 16:52 System.map-3.2.0-4-686-pae
> > -rw-r--r-- 1 root root 2709184 Apr 27 16:51 vmlinuz-3.2.0-4-686-pae
> > 
> > (I have to notice these upgrades myself because they overwrite
> > my edited version of /boot/grub/grub.cfg which I then replace.)
> > 
> > Cheers,
> > David.
> > 
> > ¹ I'm not disagreeing that something is holding back the upgrade
> > on this specific occasion, but this is unusual.
> 
> Sorry, but you are doing it wrong way. Grub 2 should not be customized
> by editing /boot/grub/grub.cfg , but by editing /etc/default/grub and
> files in /etc/grub.d/. Reason is obvious, your customization is lost
> whenever something related to linux-image is upgraded. Just saying :)

"Wrong" isn't quite the right way to put. For most people in most
circumstances editing grub.cfg and using update-grub is a wise procedure
and to be advocated. But a hand-crafted grub.cfg can be very useful.
update-grub can be prevented from getting its hands on it with
dpkg-divert,

-- 
Brian.

Re: Questions after doing update and upgrade on Stretch

[Dejan Jocic]

On 21-05-17, David Wright wrote:
> On Sun 21 May 2017 at 16:31:55 (+0200), Dejan Jocic wrote:
> > On 21-05-17, Richard Owlett wrote:
> > > I had done:
> > >   apt-get update
> > >   apt-get upgrade
> > > The tail end of the output was:
> > > ...
> > > Setting up libkde3support4 (4:4.14.26-2) ...
> > > Setting up libktexteditor4 (4:4.14.26-2) ...
> > > Setting up libkdewebkit5 (4:4.14.26-2) ...
> > > Setting up libkhtml5 (4:4.14.26-2) ...
> > > Setting up libplasma3 (4:4.14.26-2) ...
> > > Setting up kdelibs5-plugins (4:4.14.26-2) ...
> > > Processing triggers for initramfs-tools (0.130) ...
> > > update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
> > > I: The initramfs will attempt to resume from /dev/sda5
> > > I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
> > > I: Set the RESUME variable to override this.
> > > Processing triggers for libc-bin (2.24-10) ...
> > > root@stretch-2nd:/home/richard#
> > > 
> > > Before asking for confirmation to do the "upgrade" it said 3 packages 
> > > would
> > > not be upgraded.
> > > If it said which packages, I didn't spot it.
> > > I then reran with following result.
> > > 
> > > root@stretch-2nd:/home/richard# apt-get upgrade
> > > Reading package lists... Done
> > > Building dependency tree
> > > Reading state information... Done
> > > Calculating upgrade... Done
> > > The following packages were automatically installed and are no longer
> > > required:
> > >   dconf-cli gir1.2-notify-0.7 libconfig9 libgtkspell3-3-0 libindicator3-7
> > > mate-indicator-applet
> > >   mate-indicator-applet-common python3-psutil python3-setproctitle
> > > Use 'apt autoremove' to remove them.
> > > The following packages have been kept back:
> > >   linux-image-686-pae xorg xserver-xorg
> > > 0 upgraded, 0 newly installed, 0 to remove and 3 not upgraded.
> > > root@stretch-2nd:/home/richard#
> > > 
> > > My questions:
> > > 
> > > 1. In the first run, I don't understand:
> > >  Processing triggers for initramfs-tools (0.130) ...
> > >  update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
> > >  I: The initramfs will attempt to resume from /dev/sda5
> > >  I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
> > >  I: Set the RESUME variable to override this.
> > >  As /dev/sda5 is my SWAP.
> > > 
> > > 2. I don't understand any implications of:
> > >  The following packages have been kept back:
> > >linux-image-686-pae xorg xserver-xorg
> > > 
> > > TIA
> > > 
> > > 
> > > 
> > > 
> > As for number 1 can't say much about it, I do not get it either. But 2
> > happens because you've used apt-get upgrade instead of apt-get
> > dist-upgrade. Packages that will uninstall some packages already
> > installed on your system and that will change some dependencies 
> > require dist-upgrade.
> 
> Agreed.
> 
> > It happens always in case of linux-image packages. 
> 
> Is this¹ new with stretch? My linux-images upgrade just like any other
> package; here's the penultimate occasion for jessie:
> 
> Start-Date: 2017-03-08  19:20:34
> Commandline: apt-get upgrade
> Upgrade: linux-source-3.16:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> linux-headers-3.16.0-4-586:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> linux-image-3.16.0-4-586:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> linux-libc-dev:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> linux-compiler-gcc-4.8-x86:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
> linux-headers-3.16.0-4-common:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2)
> End-Date: 2017-03-08  19:22:50
> 
> (The last one's log was rather larger.)

Ehh, sorry not sure if it is new with Stretch, can't remember for
Jessie. I'm certain that it was like that on Stretch and on Ubuntu
16.04.

> 
> > It will leave your previous working linux-image on though, but will 
> > uninstall one older than that, so you will always end up with chance to 
> > boot in working kernel, if new one messes up some things.
> 
> Same question. My wheezy system has had at least 28 linux-image
> upgrades (3.2.57-3+deb7u2→3.2.60-1+deb7u1 to 3.2.86-1→3.2.88-1)
> but there's still only one kernel image on the system:
> 
> -rw-r--r-- 1 root root  134839 Apr 27 16:52 config-3.2.0-4-686-pae
> drwxr-xr-x 3 root root   12288 Apr 28 07:44 grub
> -rw-r--r-- 1 root root 2672854 Apr 28 07:44 initrd.img-3.2.0-4-686-pae
> -rw-r--r-- 1 root root 1674268 Apr 27 16:52 System.map-3.2.0-4-686-pae
> -rw-r--r-- 1 root root 2709184 Apr 27 16:51 vmlinuz-3.2.0-4-686-pae
> 
> (I have to notice these upgrades myself because they overwrite
> my edited version of /boot/grub/grub.cfg which I then replace.)
> 
> Cheers,
> David.
> 
> ¹ I'm not disagreeing that something is holding back the upgrade
> on this specific occasion, but this is unusual.
> 

Sorry, but you are doing it wrong way. Grub 2 should not be customized
by editing /boot/grub/grub.cfg , but by editing /etc/default/grub and
files in /etc/grub.d/. Reason is obvious, your customization is lost
whenever something related to linux-image is 

Re: Questions after doing update and upgrade on Stretch

[David Wright]

On Sun 21 May 2017 at 16:31:55 (+0200), Dejan Jocic wrote:
> On 21-05-17, Richard Owlett wrote:
> > I had done:
> >   apt-get update
> >   apt-get upgrade
> > The tail end of the output was:
> > ...
> > Setting up libkde3support4 (4:4.14.26-2) ...
> > Setting up libktexteditor4 (4:4.14.26-2) ...
> > Setting up libkdewebkit5 (4:4.14.26-2) ...
> > Setting up libkhtml5 (4:4.14.26-2) ...
> > Setting up libplasma3 (4:4.14.26-2) ...
> > Setting up kdelibs5-plugins (4:4.14.26-2) ...
> > Processing triggers for initramfs-tools (0.130) ...
> > update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
> > I: The initramfs will attempt to resume from /dev/sda5
> > I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
> > I: Set the RESUME variable to override this.
> > Processing triggers for libc-bin (2.24-10) ...
> > root@stretch-2nd:/home/richard#
> > 
> > Before asking for confirmation to do the "upgrade" it said 3 packages would
> > not be upgraded.
> > If it said which packages, I didn't spot it.
> > I then reran with following result.
> > 
> > root@stretch-2nd:/home/richard# apt-get upgrade
> > Reading package lists... Done
> > Building dependency tree
> > Reading state information... Done
> > Calculating upgrade... Done
> > The following packages were automatically installed and are no longer
> > required:
> >   dconf-cli gir1.2-notify-0.7 libconfig9 libgtkspell3-3-0 libindicator3-7
> > mate-indicator-applet
> >   mate-indicator-applet-common python3-psutil python3-setproctitle
> > Use 'apt autoremove' to remove them.
> > The following packages have been kept back:
> >   linux-image-686-pae xorg xserver-xorg
> > 0 upgraded, 0 newly installed, 0 to remove and 3 not upgraded.
> > root@stretch-2nd:/home/richard#
> > 
> > My questions:
> > 
> > 1. In the first run, I don't understand:
> >  Processing triggers for initramfs-tools (0.130) ...
> >  update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
> >  I: The initramfs will attempt to resume from /dev/sda5
> >  I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
> >  I: Set the RESUME variable to override this.
> >  As /dev/sda5 is my SWAP.
> > 
> > 2. I don't understand any implications of:
> >  The following packages have been kept back:
> >linux-image-686-pae xorg xserver-xorg
> > 
> > TIA
> > 
> > 
> > 
> > 
> As for number 1 can't say much about it, I do not get it either. But 2
> happens because you've used apt-get upgrade instead of apt-get
> dist-upgrade. Packages that will uninstall some packages already
> installed on your system and that will change some dependencies 
> require dist-upgrade.

Agreed.

> It happens always in case of linux-image packages. 

Is this¹ new with stretch? My linux-images upgrade just like any other
package; here's the penultimate occasion for jessie:

Start-Date: 2017-03-08  19:20:34
Commandline: apt-get upgrade
Upgrade: linux-source-3.16:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
linux-headers-3.16.0-4-586:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
linux-image-3.16.0-4-586:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
linux-libc-dev:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
linux-compiler-gcc-4.8-x86:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2), 
linux-headers-3.16.0-4-common:i386 (3.16.39-1+deb8u1, 3.16.39-1+deb8u2)
End-Date: 2017-03-08  19:22:50

(The last one's log was rather larger.)

> It will leave your previous working linux-image on though, but will 
> uninstall one older than that, so you will always end up with chance to 
> boot in working kernel, if new one messes up some things.

Same question. My wheezy system has had at least 28 linux-image
upgrades (3.2.57-3+deb7u2→3.2.60-1+deb7u1 to 3.2.86-1→3.2.88-1)
but there's still only one kernel image on the system:

-rw-r--r-- 1 root root  134839 Apr 27 16:52 config-3.2.0-4-686-pae
drwxr-xr-x 3 root root   12288 Apr 28 07:44 grub
-rw-r--r-- 1 root root 2672854 Apr 28 07:44 initrd.img-3.2.0-4-686-pae
-rw-r--r-- 1 root root 1674268 Apr 27 16:52 System.map-3.2.0-4-686-pae
-rw-r--r-- 1 root root 2709184 Apr 27 16:51 vmlinuz-3.2.0-4-686-pae

(I have to notice these upgrades myself because they overwrite
my edited version of /boot/grub/grub.cfg which I then replace.)

Cheers,
David.

¹ I'm not disagreeing that something is holding back the upgrade
on this specific occasion, but this is unusual.

Re: Questions after doing update and upgrade on Stretch

[Dejan Jocic]

On 21-05-17, Richard Owlett wrote:
> I had done:
>   apt-get update
>   apt-get upgrade
> The tail end of the output was:
> ...
> Setting up libkde3support4 (4:4.14.26-2) ...
> Setting up libktexteditor4 (4:4.14.26-2) ...
> Setting up libkdewebkit5 (4:4.14.26-2) ...
> Setting up libkhtml5 (4:4.14.26-2) ...
> Setting up libplasma3 (4:4.14.26-2) ...
> Setting up kdelibs5-plugins (4:4.14.26-2) ...
> Processing triggers for initramfs-tools (0.130) ...
> update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
> I: The initramfs will attempt to resume from /dev/sda5
> I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
> I: Set the RESUME variable to override this.
> Processing triggers for libc-bin (2.24-10) ...
> root@stretch-2nd:/home/richard#
> 
> Before asking for confirmation to do the "upgrade" it said 3 packages would
> not be upgraded.
> If it said which packages, I didn't spot it.
> I then reran with following result.
> 
> root@stretch-2nd:/home/richard# apt-get upgrade
> Reading package lists... Done
> Building dependency tree
> Reading state information... Done
> Calculating upgrade... Done
> The following packages were automatically installed and are no longer
> required:
>   dconf-cli gir1.2-notify-0.7 libconfig9 libgtkspell3-3-0 libindicator3-7
> mate-indicator-applet
>   mate-indicator-applet-common python3-psutil python3-setproctitle
> Use 'apt autoremove' to remove them.
> The following packages have been kept back:
>   linux-image-686-pae xorg xserver-xorg
> 0 upgraded, 0 newly installed, 0 to remove and 3 not upgraded.
> root@stretch-2nd:/home/richard#
> 
> My questions:
> 
> 1. In the first run, I don't understand:
>  Processing triggers for initramfs-tools (0.130) ...
>  update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
>  I: The initramfs will attempt to resume from /dev/sda5
>  I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
>  I: Set the RESUME variable to override this.
>  As /dev/sda5 is my SWAP.
> 

Actually, after bit of digging, that part became much more clear,
because resume option is used to specify partition device for software
suspend, and swap is logical choice for that.

Re: Questions after doing update and upgrade on Stretch

[Pascal Hambourg]

Le 21/05/2017 à 16:35, Richard Owlett a écrit :

On 05/21/2017 09:23 AM, Pascal Hambourg wrote:


Check with blkid that sda5 has this UUID.


Gparted reports that UUID for /dev/sda5


Then it is fine.


I'll hold off on that until I can double-check on packages that were
reported as "automatically installed" but no longer needed. I thought
that one was installed manually.


Don't worry. These packages won't be automatically removed unless you 
run apt-get autoremove.

Re: Questions after doing update and upgrade on Stretch

[songbird]

Richard Owlett wrote:
> I had done:
>apt-get update
>apt-get upgrade
> The tail end of the output was:
> ...
> Setting up libkde3support4 (4:4.14.26-2) ...
> Setting up libktexteditor4 (4:4.14.26-2) ...
> Setting up libkdewebkit5 (4:4.14.26-2) ...
> Setting up libkhtml5 (4:4.14.26-2) ...
> Setting up libplasma3 (4:4.14.26-2) ...
> Setting up kdelibs5-plugins (4:4.14.26-2) ...
> Processing triggers for initramfs-tools (0.130) ...
> update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
> I: The initramfs will attempt to resume from /dev/sda5
> I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
> I: Set the RESUME variable to override this.
> Processing triggers for libc-bin (2.24-10) ...
> root@stretch-2nd:/home/richard#

  last bit is new message from initramfs upgrade.
i edited my /etc/default/grub file to make sure
there is no resume going on at all by changing the
line:

GRUB_CMDLINE_LINUX="root=LABEL=ROOT_00 noresume"

your own line may look different.  :)


> Before asking for confirmation to do the "upgrade" it said 3 packages 
> would not be upgraded.
> If it said which packages, I didn't spot it.

  you'd need dist-upgrade to grab the new kernel version
that is all that is about.


> I then reran with following result.
>
> root@stretch-2nd:/home/richard# apt-get upgrade
> Reading package lists... Done
> Building dependency tree
> Reading state information... Done
> Calculating upgrade... Done
> The following packages were automatically installed and are no longer 
> required:
>dconf-cli gir1.2-notify-0.7 libconfig9 libgtkspell3-3-0 
> libindicator3-7 mate-indicator-applet
>mate-indicator-applet-common python3-psutil python3-setproctitle
> Use 'apt autoremove' to remove them.
> The following packages have been kept back:
>linux-image-686-pae xorg xserver-xorg
> 0 upgraded, 0 newly installed, 0 to remove and 3 not upgraded.
> root@stretch-2nd:/home/richard#
>
> My questions:
>
> 1. In the first run, I don't understand:
>   Processing triggers for initramfs-tools (0.130) ...
>   update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
>   I: The initramfs will attempt to resume from /dev/sda5
>   I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
>   I: Set the RESUME variable to override this.
>   As /dev/sda5 is my SWAP.

  'splained above...


> 2. I don't understand any implications of:
>   The following packages have been kept back:
> linux-image-686-pae xorg xserver-xorg

  the difference between apt-get upgrade and apt-get dist-upgrade.

  covered by the man page.


  songbird

Re: Questions after doing update and upgrade on Stretch

[Richard Owlett]

On 05/21/2017 09:23 AM, Pascal Hambourg wrote:

Le 21/05/2017 à 15:34, Richard Owlett a écrit :


1. In the first run, I don't understand:
 Processing triggers for initramfs-tools (0.130) ...
 update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
 I: The initramfs will attempt to resume from /dev/sda5
 I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
 I: Set the RESUME variable to override this.
 As /dev/sda5 is my SWAP.


Check with blkid that sda5 has this UUID.


Gparted reports that UUID for /dev/sda5





2. I don't understand any implications of:
 The following packages have been kept back:
   linux-image-686-pae xorg xserver-xorg


These packages cannot be upgraded due to dependency changes.
This is usually solved with dist-upgrade.


I'll hold off on that until I can double-check on packages that were 
reported as "automatically installed" but no longer needed. I thought 
that one was installed manually. Won't be able to follow up until this 
afternoon.


Thank you.

Re: Questions after doing update and upgrade on Stretch

[Dejan Jocic]

On 21-05-17, Richard Owlett wrote:
> I had done:
>   apt-get update
>   apt-get upgrade
> The tail end of the output was:
> ...
> Setting up libkde3support4 (4:4.14.26-2) ...
> Setting up libktexteditor4 (4:4.14.26-2) ...
> Setting up libkdewebkit5 (4:4.14.26-2) ...
> Setting up libkhtml5 (4:4.14.26-2) ...
> Setting up libplasma3 (4:4.14.26-2) ...
> Setting up kdelibs5-plugins (4:4.14.26-2) ...
> Processing triggers for initramfs-tools (0.130) ...
> update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
> I: The initramfs will attempt to resume from /dev/sda5
> I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
> I: Set the RESUME variable to override this.
> Processing triggers for libc-bin (2.24-10) ...
> root@stretch-2nd:/home/richard#
> 
> Before asking for confirmation to do the "upgrade" it said 3 packages would
> not be upgraded.
> If it said which packages, I didn't spot it.
> I then reran with following result.
> 
> root@stretch-2nd:/home/richard# apt-get upgrade
> Reading package lists... Done
> Building dependency tree
> Reading state information... Done
> Calculating upgrade... Done
> The following packages were automatically installed and are no longer
> required:
>   dconf-cli gir1.2-notify-0.7 libconfig9 libgtkspell3-3-0 libindicator3-7
> mate-indicator-applet
>   mate-indicator-applet-common python3-psutil python3-setproctitle
> Use 'apt autoremove' to remove them.
> The following packages have been kept back:
>   linux-image-686-pae xorg xserver-xorg
> 0 upgraded, 0 newly installed, 0 to remove and 3 not upgraded.
> root@stretch-2nd:/home/richard#
> 
> My questions:
> 
> 1. In the first run, I don't understand:
>  Processing triggers for initramfs-tools (0.130) ...
>  update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
>  I: The initramfs will attempt to resume from /dev/sda5
>  I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
>  I: Set the RESUME variable to override this.
>  As /dev/sda5 is my SWAP.
> 
> 2. I don't understand any implications of:
>  The following packages have been kept back:
>linux-image-686-pae xorg xserver-xorg
> 
> TIA
> 
> 
> 
> 
As for number 1 can't say much about it, I do not get it either. But 2
happens because you've used apt-get upgrade instead of apt-get
dist-upgrade. Packages that will uninstall some packages already
installed on your system and that will change some dependencies 
require dist-upgrade. It happens always in case of linux-image packages. 
It will leave your previous working linux-image on though, but will 
uninstall one older than that, so you will always end up with chance to 
boot in working kernel, if new one messes up some things.

Re: Questions after doing update and upgrade on Stretch

[Pascal Hambourg]

Le 21/05/2017 à 15:34, Richard Owlett a écrit :


1. In the first run, I don't understand:
 Processing triggers for initramfs-tools (0.130) ...
 update-initramfs: Generating /boot/initrd.img-4.9.0-2-686-pae
 I: The initramfs will attempt to resume from /dev/sda5
 I: (UUID=5d0c821b-26b2-4d38-b7fe-dc7db1b72576)
 I: Set the RESUME variable to override this.
 As /dev/sda5 is my SWAP.


Check with blkid that sda5 has this UUID.


2. I don't understand any implications of:
 The following packages have been kept back:
   linux-image-686-pae xorg xserver-xorg


These packages cannot be upgraded due to dependency changes.
This is usually solved with dist-upgrade.

Re: Questions de novice sur -dbgsym

[François TOURDE]

Le 17237ième jour après Epoch,
Vincent Lefevre écrivait:

> On 2017-03-08 11:44:49 +0100, François TOURDE wrote:
>> Le 17233ième jour après Epoch,
>> Olivier écrivait:
>> > 1. À quoi sert exactement un paquet -dbgsym ? Doit-on installer
>> >  et -dbgsym ou -dbgsym à la place de 
>> > ?
>> 
>> C'est un paquet qui va contenir des infos de débug, des symboles
>> donc. Le paquet s'installe à la place de l'autre, et contient des
>> binaires un peu (voire beaucoup) plus gros, à cause des infos de debug.
>
> Non, les deux paquets coexistent. On peut même installer le -dbgsym
> après le core dump, et il est automatiquement utilisé par gdb.

Oui, tu as raison, désolé d'avoir répondu trop vite sans vérifier...

Re: Questions de novice sur -dbgsym

[Vincent Lefevre]

On 2017-03-08 11:44:49 +0100, François TOURDE wrote:
> Le 17233ième jour après Epoch,
> Olivier écrivait:
> > 1. À quoi sert exactement un paquet -dbgsym ? Doit-on installer
> >  et -dbgsym ou -dbgsym à la place de 
> > ?
> 
> C'est un paquet qui va contenir des infos de débug, des symboles
> donc. Le paquet s'installe à la place de l'autre, et contient des
> binaires un peu (voire beaucoup) plus gros, à cause des infos de debug.

Non, les deux paquets coexistent. On peut même installer le -dbgsym
après le core dump, et il est automatiquement utilisé par gdb.

-- 
Vincent Lefèvre  - Web: 
100% accessible validated (X)HTML - Blog: 
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)

Re: Questions de novice sur -dbgsym

[Haricophile]

Le Wed, 08 Mar 2017 11:44:49 +0100,
fra-duf-no-s...@tourde.org (François TOURDE) a écrit :

> J'espère avoir éclairé un peu ta lanterne.

Pour compléter l'éclairage, strace peut avoir son utilité, non ?

Re: Questions de novice sur -dbgsym [RESOLU]

[Olivier]

Le 8 mars 2017 à 11:44, François TOURDE  a
écrit :

> Bonjour,
>
> Le 17233ième jour après Epoch,
> Olivier écrivait:
>
> > Pardonnez-moi si mes questions semblent idiotes mais:
>
> Pas de questions idiotes, seules les réponsent peuvent l'être :-p
>
> Pardonne moi, donc, si mes réponses sont idiotes...
>
> > 1. À quoi sert exactement un paquet -dbgsym ? Doit-on installer
> >  et -dbgsym ou -dbgsym à la place de 
> > ?
>
> C'est un paquet qui va contenir des infos de débug, des symboles
> donc. Le paquet s'installe à la place de l'autre, et contient des
> binaires un peu (voire beaucoup) plus gros, à cause des infos de debug.
>
> En général, les programmes sont compilés avec et sans les infos de
> débug, ou alors juste avec puis sont "strippés" (man strip). Cela donne
> deux versions des objets, l'un avec et l'autre sans les infos de debug.
>
> > 2. Quel rapport entre -dbgsym et la production de fichier
> > coredump ?
>
> Avec la production du coredump, rien, mais par contre pour l'analyse de
> ce coredump, avoir les infos de débug est utile. Tu peux voir le nom des
> variables en clair plutôt que les adresses hexa, tu vois la pile sous
> forme de nom de fonctions et non sous forme d'adresses...
>
> > 3. Plus généralement, dans mon imaginaire "un fichier coredump est la
> > condition nécessaire et suffisante pour qu'un développeur (ie pas un
> > mainteneur) puisse analyser un plantage". Est-ce plutôt exact sinon
> qu'est
> > ce qui le serait d'avantage ?
>
> Ce n'est ni nécessaire, ni suffisant, mais ça peut être super pratique.
>
> Le coredump (illustré des infos de debug) est un instantané pris au
> moment du plantage, avec lequel tu peux voir les variables, la pile,
> etc... Mais tu ne vois pas par exemple les conditions initiales, le
> chemin du programme pour en arriver là, etc...
>
> J'espère avoir éclairé un peu ta lanterne.
>

Parfaitement !




>
> --
> Do not sleep in a eucalyptus tree tonight.
>
>

Re: Questions de novice sur -dbgsym

[François TOURDE]

Bonjour,

Le 17233ième jour après Epoch,
Olivier écrivait:

> Pardonnez-moi si mes questions semblent idiotes mais:

Pas de questions idiotes, seules les réponsent peuvent l'être :-p

Pardonne moi, donc, si mes réponses sont idiotes...

> 1. À quoi sert exactement un paquet -dbgsym ? Doit-on installer
>  et -dbgsym ou -dbgsym à la place de 
> ?

C'est un paquet qui va contenir des infos de débug, des symboles
donc. Le paquet s'installe à la place de l'autre, et contient des
binaires un peu (voire beaucoup) plus gros, à cause des infos de debug.

En général, les programmes sont compilés avec et sans les infos de
débug, ou alors juste avec puis sont "strippés" (man strip). Cela donne
deux versions des objets, l'un avec et l'autre sans les infos de debug.

> 2. Quel rapport entre -dbgsym et la production de fichier
> coredump ?

Avec la production du coredump, rien, mais par contre pour l'analyse de
ce coredump, avoir les infos de débug est utile. Tu peux voir le nom des
variables en clair plutôt que les adresses hexa, tu vois la pile sous
forme de nom de fonctions et non sous forme d'adresses...

> 3. Plus généralement, dans mon imaginaire "un fichier coredump est la
> condition nécessaire et suffisante pour qu'un développeur (ie pas un
> mainteneur) puisse analyser un plantage". Est-ce plutôt exact sinon qu'est
> ce qui le serait d'avantage ?

Ce n'est ni nécessaire, ni suffisant, mais ça peut être super pratique.

Le coredump (illustré des infos de debug) est un instantané pris au
moment du plantage, avec lequel tu peux voir les variables, la pile,
etc... Mais tu ne vois pas par exemple les conditions initiales, le
chemin du programme pour en arriver là, etc...

J'espère avoir éclairé un peu ta lanterne.

-- 
Do not sleep in a eucalyptus tree tonight.