Re: TTL expired in transit to qemu virtual machine.

[Igor Cicimov]

Hi Mimiko,

On Fri, Mar 17, 2017 at 9:58 PM, Mimiko  wrote:

> Hello.
>
> I've setup qemu/kvm and installed several virtual machines. Access and
> ping to some virtuals are ok, but one have a stable problem not receiving
> correctly packets. First, this is the environment:
>
> >uname -a
> Linux 3.2.0-4-amd64 #1 SMP Debian 3.2.84-1 x86_64 GNU/Linux
>

That's an really old kernel, I don't start anything virtual these days
without at least 3.13.x kernel.


>
> >libvirtd --version
> libvirtd (libvirt) 0.9.12.3
>
> >cat /etc/network/interfaces
> auto lo
> iface lo inet loopback
>
> auto eth0
> iface eth0 inet manual
>
> auto eth1
> iface eth1 inet manual
>
> auto bond0
> iface bond0 inet manual
> bond-slaves eth0 eth1
> bond-mode balance-alb
> bond-miimon 100
> bond-downdelay 200
> bond-updelay 200
>
> auto br0
> iface br0 inet static
> address 10.10.10.10
> netmask 255.255.0.0
> vlan-raw-device bond0
> bridge_ports bond0
> bridge_stp off
> bridge_fd 0
> bridge_ageing 0
> bridge_maxwait 0
>

Hmmm, this doesn't make much sense to me, more specifically this part:

vlan-raw-device bond0
bridge_ports bond0

Whats the purpose exactly of the vlan? Usually, and that is how I do it,
you would split the VLAN's coming from the switch trunk port over the bond
and attach them to separate bridges lets say:

# VLAN51
auto br-vlan51
iface br-vlan51 inet manual
vlan-raw-device bond0
bridge_ports *bond0.51*
bridge_stp off
bridge_maxwait 0
bridge_fd 0

# VLAN52
auto br-vlan52
iface br-vlan52 inet manual
vlan-raw-device bond0
bridge_ports *bond0.52*
bridge_stp off
bridge_maxwait 0
bridge_fd 0

If the intention was to pass through the tagged traffic to the VM's then
vlan-raw-device part is not needed at all.


> Virtual machines connects to LAN using bridge:
> >virt-install  --network=bridge:br0,model=virtio 
>
> All virtuals has networking configuret like that. Also in iptables is an
> entry to allow access to virtuals:
>
> >iptables -L FORWARD -nv
> Chain FORWARD (policy DROP 0 packets, 0 bytes)
>  pkts bytes target prot opt in out source
>  destination
>  XX ACCEPT all  --  br0br0 0.0.0.0/0
> 0.0.0.0/0
>
> Most virtuals does not have networking problems, but some times they can't
> be reached. For now only one virtual machines have this problem:
> From windows machine ping virtual machine:
>
> >ping 10.10.10.3
>
> Reply from 10.10.10.10: TTL expired in transit.
> Reply from 10.10.10.10: TTL expired in transit.
> Reply from 10.10.10.10: TTL expired in transit.
> Reply from 10.10.10.10: TTL expired in transit.
> Request timed out.
> Request timed out.
> Request timed out.
> Request timed out.
> Request timed out.
> Request timed out.
> Reply from 10.10.10.10: TTL expired in transit.
> Reply from 10.10.10.10: TTL expired in transit.
> Reply from 10.10.10.10: TTL expired in transit.
> Reply from 10.10.10.10: TTL expired in transit.
>
> >tracert -d 10.10.10.3
>
> Tracing route to 10.10.10.3 over a maximum of 30 hops
>
>   1<1 ms<1 ms<1 ms  10.10.10.10
>   2<1 ms<1 ms<1 ms  10.10.10.10
>   3<1 ms<1 ms * 10.10.10.10
>   4 1 ms<1 ms<1 ms  10.10.10.10
>   5<1 ms<1 ms * 10.10.10.10
>
> So packet goes round on interfaces of server hosting virtuals.
>

Yep typical routing loop.


>
> Virtuals are linux different flavour and one windows. This problem may
> occur on any of this virtuals.
>
> I've observed that for this particular virtual, which have problem, the
> arp table of host assigned self mac to the virtual's IP, not the mac
> configured for virtual machine.
>

That's strange indeed, except if br0 is used by something else like libvrit
network that sets up the interface for proxy-arp. What's the output of:

# brctl showmacs br0
# ip route show
# arp -n

on the host, and:

# ip link show
# ip route show
# arp -n

on the problematic vm and on one of the good vm's?

To find the loop I would start by doing ping between good and bad vm (both
directions in turns) and check the traffic on the host interface:

# tcpdump -ennqt -i br0 \( arp or icmp \)

and corresponding network devices on both vm's too.


>
> What could be the problem?
>
>
Any sysctl settings you might have changed on the host?


> --
> Mimiko desu.
>
>

Re: Guide(s?) to backup philosophies

[David Christensen]

On 03/17/2017 03:31 AM, Dan Purgert wrote:

David Christensen wrote:

On 03/13/2017 05:38 AM, Dan Purgert wrote:

Currently, the system here is

 - every PC has a cronjob backing up $HOME to a central "server" (read -
   repurposed PC with decent WD drives), just an rsync script that runs
   daily.


Don't forget security:

1.  With a "push" arrangement (e.g. each workstation backs up itself to
the server) -- if a workstation gets compromised, the backups are at risk.

2.  With a "pull" arrangement (e.g. the server backs up all the
workstations) -- if a workstation gets compromised, the backups should
be safe (and might have clues about the intrusion).  Additionally, the
backup server can be completely firewalled (e.g. no open ports).


I should clarify that:

"The backup server can be firewalled with no incoming ports and
outgoing ports limited to SSH and other required ports".


I still need to figure out the "other required outgoing ports". 
Suggestions and comments are welcome.




Since the PCs are laptops, they're not always here, so I was never able
to figure out how to get pull to work with the condition that we were on
vacation (or the laptops were otherwise "not home").

Though, yeah, the stuff that's statically here (desktop, server, etc.)
are rsync-by-pull.


I haven't dealt with the "roaming laptop on the Internet" use-case yet, 
but I do have a desire to solve it.  My idea has been, and remains, for 
the backup server to poll for a "job file" on the laptop, and to execute 
it when found (once; idempotent).  This implies a network connection 
between the backup server and the laptop.  OpenVPN is a technology that 
might be able to facilitate this.



David

Re: Where is data stored when Synaptic scans DVDs?

[David Wright]

On Fri 17 Mar 2017 at 15:29:58 (-0500), Richard Owlett wrote:
> On 03/17/2017 11:15 AM, songbird wrote:
> >Richard Owlett wrote:
> >>As I have limited bandwidth, I install from DVDs.
> >>Also I frequently experiment with various options by doing a fresh
> >>install to a different partition.
> >>It is annoying to keep re-scanning a set of 13 DVDs.
> >>Where is the information stored?
> >>/etc/apt/sources.list is obvious, but where is the data about contents
> >>of each DVD stored?
> >
> >  man apt-get
> 
> It implies *BUT DOES _NOT_ STATE* the all the requested data is in
> the /var/lib/apt directory.

(Three levels of emphasis?)

IIRC the command to scan is apt-cdrom. Are DVDs handled any
differently? Anyway,   man apt-cdrom   says:

"APT uses a CD-ROM ID to track which disc is currently in the drive
 and maintains a database of these IDs in /var/lib/apt/cdroms.list"

I don't think the apt suite has much choice about where to put this
information; the FHS mandates it in §5.8.1.

But I don't think /var/lib is designed to be shared between systems,
and so I think you may need to delve into the intricacies of how
the information is stored if you're going to share it in any way.

I assume you can find out which files are involved just by scanning a
disc and looking at the file timestamps. I would also assume that one
disc≡one file rather like the Packages files, but have nothing to
check this out with. If it's that simple, it then just comes down to
how the files link with the contents of cdroms.list. Looking at the
latter, I would imagine they use the GUID in there for the
lists/. I'm not sure about the -2 suffix I see in there.

> I hadn't thought of looking at apt-get when I posted, even though I
> had just chased down some differences on how synaptic and apt-get
> differed in how they logged some trans action. mea culpa ;>
> 
> >  and as i've previously said, you can copy the
> >dvds to spinning rust or a SSD and avoid all
> >future needs to reload from them again...
> 
> Yes but 
> in my peculiar situation it is better to know on which DVD something
> resides than to spend ~1000 times more storage space to duplicate
> the data.

I thought you had copied your DVDs back in 2015. I remember only
because you incorrectly shouted that you'd used the past tense when
you wrote "I have copied…". No, it was also because I objected to
the word "VIOLATES" (also shouted).

> >  and futhermore, if you are doing partitions
> >you can do some minimal install to get the base
> >system going, then do a partition copy of that
> >and you'll not have to redo the previous steps
> >again to recover if you want to try something
> >else.

I presume it's those "previous steps" that interest you,
in continuance of your research into preseeding installations.

Cheers,
David.

Re: programmatically determining the desktop environment of a system

[Dan Hitt]

Thanks Dominik, Steve, and Tomas for your mail.

For reference, Steve Litt responded to me privately to consider first
running 'ps axjf' by hand, in which case the environment should be
obvious, presumably doing this in a few different environments, and
then attempting to write some parsing code which goes through the same
mental steps.   Certainly sounds reasonable absent an api, so thanks
Steve for the suggestion.

Thanks Dominik for the stack exchange link, which is basically the
same question (except that i would not be programming it in bash),
with some interesting ideas about environment variables to inspect.
(Part of the SE discussion in fact suggests using ps.)

Finally thanks for your response Tomas, with which i think most people
would agree.

That is, i think usually the best practice is to do capability
detection (like the autoconf apparatus or jquery does) and then use
whatever capabilities you detect.

I'm not sure this is always possible, though, because if it is
something about the using the hardware in a certain way you may not be
able to detect this (but rather rely on what you know about the
environment).

I was sort of hoping that somebody would answer saying "Use the foo()
function in the XYZ api".  Then at least i'd know to go find the XYZ
api and study it, because it would probably have a lot of other
goodies in it as well.  And i'm still interested in that, but given
Dominick's reference i'm kind of doubtful i'll find it.

Anyhow, thanks again everybody for your help!  :)

dan



On Fri, Mar 17, 2017 at 1:02 AM,   wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> On Thu, Mar 16, 2017 at 10:17:37PM -0700, Dan Hitt wrote:
>> Is there a programmatic way that a piece of software can learn what
>> desktop environment it is executing in?
>
> Just out of curiosity: what is (roughly) your use case?
>
> (Personally I think it's a bad idea, but I'd like to learn)
>
> regards
> - -- t
> -BEGIN PGP SIGNATURE-
> Version: GnuPG v1.4.12 (GNU/Linux)
>
> iEYEARECAAYFAljLmC4ACgkQBcgs9XrR2kYBoACZAVOgMl1I2DyFdJSp1FMMy5eA
> /rMAn0BTT1VByF5rhECqo47jAexuAFhG
> =ucsZ
> -END PGP SIGNATURE-
>

Re: Netinstall of Debian 8 on DELL laptop fails

[Lisi Reisz]

On Friday 17 March 2017 17:55:12 Klaus Jantzen wrote:
> Hello,
>
> I am trying to net-install Debian 8.7. (Graphical expert install)
>
> After loading a couple of modules for the installation of the base
> system the installer claims that
>
> beginning with 'groff-base' all modules are corrupted and cannot be loaded.
>
> This is wrong as I used the same CD to install Debian on my PC without
> any problems.

CDs can get damaged.  Or sometimes in practice a CD can be read in one drive 
but not in another.
>
> Before that I had installed Wheezy on the laptop without any problems.
>
>
> Does anyone have an idea on how to solve the problem?

Download the file again and burn a new CD, preferably with a different 
make/type of CD on a different drive - if you can manage it.  Do check sums 
carefully.  Try again.  If you still have the problem, tell us.

Lisi

Re: Where is data stored when Synaptic scans DVDs?

[Richard Owlett]

On 03/17/2017 03:57 PM, Brian wrote:

On Fri 17 Mar 2017 at 15:29:58 -0500, Richard Owlett wrote:


On 03/17/2017 11:15 AM, songbird wrote:

Richard Owlett wrote:

As I have limited bandwidth, I install from DVDs.
Also I frequently experiment with various options by doing a fresh
install to a different partition.
It is annoying to keep re-scanning a set of 13 DVDs.
Where is the information stored?
/etc/apt/sources.list is obvious, but where is the data about contents
of each DVD stored?


 man apt-get


It implies *BUT DOES _NOT_ STATE* the all the requested data is in the
/var/lib/apt directory.


Stop pissing about. Is the data in /var/lib/apt or not?




Who knows. I don't!
I has some, but is it all???

Re: Where is data stored when Synaptic scans DVDs?

[Brian]

On Fri 17 Mar 2017 at 15:29:58 -0500, Richard Owlett wrote:

> On 03/17/2017 11:15 AM, songbird wrote:
> >Richard Owlett wrote:
> >>As I have limited bandwidth, I install from DVDs.
> >>Also I frequently experiment with various options by doing a fresh
> >>install to a different partition.
> >>It is annoying to keep re-scanning a set of 13 DVDs.
> >>Where is the information stored?
> >>/etc/apt/sources.list is obvious, but where is the data about contents
> >>of each DVD stored?
> >
> >  man apt-get
> 
> It implies *BUT DOES _NOT_ STATE* the all the requested data is in the
> /var/lib/apt directory.

Stop pissing about. Is the data in /var/lib/apt or not?

Re: Comment s'opère la mise à jour du noyau dans Stretch ? [RESOLU][Etait: Pourquoi linux-headers-`uname r` échoue sur Stretch et pas sur Jessie ?]

[Jean-Marc]

Fri, 17 Mar 2017 15:16:15 +0100
Olivier  écrivait :

> Dois-je retenir qu'un dist-upgrade est nécessaire à chaque changement de
> version du noyau ?
> Ai-je raté une alerte me signalant qu'un changement de noyau est possible
> ou recommandé ?
> Que risque-ton avec un dist-upgrade (si on n'a pas changé les données de
> /etc/apt) ?


Pas d'alerte en Stretch.  C'est la version de test.

Comme je l'ai dit, un (apt upgrade> n'installe pas de nouveau paquets.
Or dans ton cas, tu voulais installer un nouveau paquet avec un noyau 4.9.

Seul un  staisfait ta demande.

Plus généralement, je n'utilise que .

Et aucun cas, un  dans une version de test.

Ce qui me permet de vérifier ce qui sera mis à jour, enlever ou ajouter avant 
de le faire.

> 
> En tout cas, merci pour le tuyau !

De rien.

Et j'ai une question : pourquoi le noyau linux-image-686-pae ?

Tu as une machine particulière qui le demande ?

Jean-Marc 


pgpqFCAJQU3ID.pgp
Description: PGP signature

Re: Finding firmware (and SHA sums etc), was Re: Linux Mint, Ubuntu, Arch Linux recognize my Acer Aspire S wifi, debian does not

[Brian]

On Fri 17 Mar 2017 at 17:02:04 +, Lisi Reisz wrote:

> On Thursday 16 March 2017 20:08:16 Brian wrote:
> > and we are told
> > it is not the purpose of this list to answer practical problems by
> > by guiding people to it. :)
> 
> Oh, come Brian.  We are also told that it is not the purpose of Debian to 
> pander to freaks and Geeks.  We are told that the moon is made of green 
> cheese and the earth is flat.  Oh, and young ladies don't do that.  We are 
> not going to worry about what we are told, are we?

When it comes to the desirability of reading documentation on such a
basic and important procedure as installing Debian I'm inclined not to
be ambivalant. That doesn't mean I think every single word of it needs
to be understood but preparation underpins an install which is
successful. Otherwise, why put so much effort into writing the Manual?

You cut some of the paragraph I wrote. Its main intended purpose was to
point the OP to a solution to his issue, If he is still reading the
thread he started he might say whether it does. It could help others.

-- 
Brian.

Re: Where is data stored when Synaptic scans DVDs?

[Richard Owlett]

On 03/17/2017 11:15 AM, songbird wrote:

Richard Owlett wrote:

As I have limited bandwidth, I install from DVDs.
Also I frequently experiment with various options by doing a fresh
install to a different partition.
It is annoying to keep re-scanning a set of 13 DVDs.
Where is the information stored?
/etc/apt/sources.list is obvious, but where is the data about contents
of each DVD stored?


  man apt-get


It implies *BUT DOES _NOT_ STATE* the all the requested data is in the 
/var/lib/apt directory.


I hadn't thought of looking at apt-get when I posted, even though I had 
just chased down some differences on how synaptic and apt-get differed 
in how they logged some trans action. mea culpa ;>





  and as i've previously said, you can copy the
dvds to spinning rust or a SSD and avoid all
future needs to reload from them again...


Yes but 
in my peculiar situation it is better to know on which DVD something 
resides than to spend ~1000 times more storage space to duplicate the data.




  and futhermore, if you are doing partitions
you can do some minimal install to get the base
system going, then do a partition copy of that
and you'll not have to redo the previous steps
again to recover if you want to try something
else.


  songbird

Re: tor -- way OT

[Reco]

On Fri, 17 Mar 2017 21:48:02 +0200
Teemu Likonen  wrote:

> Reco [2017-03-17 22:05:45+03] wrote:
> 
> > Sure. Deinstall/purge current Tor and privoxy. Install Debian's
> > version of Tor. It should start itself. Point your Firefox to SOCKS5
> > proxy located at localhost port 9050. That all you (ever) need,
> > judging from your timezone.
> >
> > Do not mix HTTP proxy into your setup. You won't gain anything short
> > of complicating your setup.
> 
> In practice the Tor network, especially hidden services, work badly
> without a http proxy. It has been a while since I last tried but I
> always ended up setting up polipo http proxy.

I'm using Tor just right now via SOCK5, and that includes several kinds
of L7 including HTTP. Care to elaborate what exactly should 'work
badly' for me? 

Reco

Re: tor -- way OT

[Teemu Likonen]

Reco [2017-03-17 22:05:45+03] wrote:

> Sure. Deinstall/purge current Tor and privoxy. Install Debian's
> version of Tor. It should start itself. Point your Firefox to SOCKS5
> proxy located at localhost port 9050. That all you (ever) need,
> judging from your timezone.
>
> Do not mix HTTP proxy into your setup. You won't gain anything short
> of complicating your setup.

In practice the Tor network, especially hidden services, work badly
without a http proxy. It has been a while since I last tried but I
always ended up setting up polipo http proxy.

-- 
/// Teemu Likonen   - .-..    //
// PGP: 4E10 55DC 84E9 DFF6 13D7 8557 719D 69D3 2453 9450 ///


signature.asc
Description: PGP signature

Re: Suitable text editor [NOT word processor] or workaround?

[davidson]

On Fri, 17 Mar 2017, david...@freevolt.org wrote:


In case OP decides to install emacs, given his previously expressed
appreciation of fine documentation, I recommend installing also (for
the chosen version NN of emacs) the emacsNN-common-non-dfsg package
from the non-free repository.

This is the simplest way of installing the info pages for emacs.


Hrm. Installing XXX is the simplest way of installing XXX. Very
insightful.

The second instance of "installing" should probably have been
"obtaining".

Re: Suitable text editor [NOT word processor] or workaround?

[davidson]

On Thu, 16 Mar 2017, John Hasler wrote:


Richard Owlett writes:

I require two things:
1. a search and replace which can include a "newline" in new string.
2. display/edit 2 files simultaneously *side by side*


Emacs.


In case OP decides to install emacs, given his previously expressed
appreciation of fine documentation, I recommend installing also (for
the chosen version NN of emacs) the emacsNN-common-non-dfsg package
from the non-free repository.

This is the simplest way of installing the info pages for emacs.

That they happen to reside in non-free is conceivably unexpected, so
it bears special mention.

Re: Finding firmware (and SHA sums etc), was Re: Linux Mint, Ubuntu, Arch Linux recognize my Acer Aspire S wifi, debian does not

[Brian]

On Fri 17 Mar 2017 at 10:46:57 -0500, David Wright wrote:

> On Thu 16 Mar 2017 at 20:08:16 (+), Brian wrote:
> > 
> > There was never had any reason to disagree with the analysis. The
> > queries starting this subthread
> > 
> >  > Is it just me, or are the "unofficial" images hard to get to?
> > 
> > have the answers "no" and "yes". However, the routes to the images are
> > varied and guessing which one a user takes to get there isn't fruitful.
> 
> The routes exist and don't need to be guessed. The reason that a
> particular group of users follows a route is what has to be guessed.

Apropos nothing. Columbus knew exactly where he was going and why. He
just didn't get there. The exact route he took didn't exist (or maybe
philosohically it did but hadn't been travelled). But it was planned
using the Google documentation of the time. Perhaps prospective
Debian users could plan their journey into the unknown in the same way.

Point taken, though. Web page design requires the author to put herself
(or try to) in the position of a typical user.

A good place to mention the firmware netinst ISO would be the Debian
front page instead of 4 or more layers down. Depends on how much of a
problem it is seen to be. Don't hold your breath waiting for a change;
there are other factors.

> For every selection made/link clicked etc, the user had an intention.
> That intention has to be guessed by the web designer/programmer etc.
> Few designers actually experience a stream of tyros sitting down to
> use their design, and give them reasons.
> 
> Similarly, documentation authors must have in-depth knowledge of
> their subject, but also have the ability to read the mind of the
> ignorant reader.

A decent technique which can be fruitful with documentation writing is
to read what you have written a year later. If you still understand it
and can easily follow any instructions it could stand the test of time.

-- 
Brian.

Re: tor -- way OT

[Reco]

Hi.

On Fri, 17 Mar 2017 12:28:33 -0600
Glenn English  wrote:

> I'm trying to use the Tor Browser. They don't seem to have any support
> (beyond an FAQ) on their site, so I'm asking here.
> 
> Jessie and XFCE on a Supermicro workstation connected through a T1.
> 
> I installed Tor a few days ago and it was working fine -- Gmail said
> it was having authentication problems and that I was using Firefox on
> Winders in Paris. Just what I'd hoped for.
> 
> Then I installed privoxy and set the regular browser (Firefox) to use
> it. Tor started saying it couldn't find the proxy it'd been configured
> to use. Tor hadn't been configured to do any such thing, but I looked
> at the config to make sure.

A big red "you're doing it wrong" sign flashes in my head as I read
this.

First things first. Tor provides you with SOCKS5 proxy server already.
Why bother chaining it with conventional HTTP proxy?

Second thought. While Tor has some (limited) ability to work via proxy
[1], it is by no means required to force Tor to use one.

 
> Firefox is the same -- no proxy. It used to use privoxy, but I turned it off 
> I removed Tor (Aptitude purge) and reinstalled (with Aptitude), and it
> connected to DuckDuckGo, Amazon, and Newegg (maybe others, I haven't
> tried). 

It's not a valid test as it does not prove anything. Try [2]. If you
see security.debian.org - you're set.


> But it couldn't find the proxy when I pointed at local sites
> and the virtuals on my server (again, maybe others).
> 
> So I removed privoxy, and now Tor can't even connect to its own network.
> 
> Anybody seen this, have an idea, know a fix?? Is there some kind of
> (defective) interaction between the Tor and Firefox configs?

Sure. Deinstall/purge current Tor and privoxy.
Install Debian's version of Tor. It should start itself.
Point your Firefox to SOCKS5 proxy located at localhost port 9050.
That all you (ever) need, judging from your timezone.

Do not mix HTTP proxy into your setup. You won't gain anything short of
complicating your setup.

Reco

[1] https://www.torproject.org/docs/faq.html.en#NeedToUseAProxy
[2] http://sgvtcaew4bxjd7ln.onion/

Re: Guide(s?) to backup philosophies

[Glenn English]

On Fri, Mar 17, 2017 at 4:31 AM, Dan Purgert  wrote:
> David Christensen wrote:
>> On 03/13/2017 05:38 AM, Dan Purgert wrote:
>>> Currently, the system here is
>>>
>>>  - every PC has a cronjob backing up $HOME to a central "server" (read -
>>>repurposed PC with decent WD drives), just an rsync script that runs
>>>daily.
>>
>> Don't forget security:
>>
>> 1.  With a "push" arrangement (e.g. each workstation backs up itself to
>> the server) -- if a workstation gets compromised, the backups are at risk.
>>
>> 2.  With a "pull" arrangement (e.g. the server backs up all the
>> workstations) -- if a workstation gets compromised, the backups should
>> be safe (and might have clues about the intrusion).  Additionally, the
>> backup server can be completely firewalled (e.g. no open ports).
>
> Since the PCs are laptops, they're not always here, so I was never able
> to figure out how to get pull to work with the condition that we were on
> vacation (or the laptops were otherwise "not home").

Amanda and tape, but hopefully useful with other programs

Amanda's hard core pull -- there's a server on the LAN that does all
the LAN and DMZ backing up. Part of Amanda's configuration is a list
of things around the network to back up. To deal with wandering
laptops, I have a shell script that looks around to find out what's
there and creates a modified version of that list just before Amanda
runs. Anything that doesn't respond to a ping doesn't get backed up.

--
Glenn English

tor -- way OT

[Glenn English]

I'm trying to use the Tor Browser. They don't seem to have any support
(beyond an FAQ) on their site, so I'm asking here.

Jessie and XFCE on a Supermicro workstation connected through a T1.

I installed Tor a few days ago and it was working fine -- Gmail said
it was having authentication problems and that I was using Firefox on
Winders in Paris. Just what I'd hoped for.

Then I installed privoxy and set the regular browser (Firefox) to use
it. Tor started saying it couldn't find the proxy it'd been configured
to use. Tor hadn't been configured to do any such thing, but I looked
at the config to make sure.

Firefox is the same -- no proxy. It used to use privoxy, but I turned it off

I removed Tor (Aptitude purge) and reinstalled (with Aptitude), and it
connected to DuckDuckGo, Amazon, and Newegg (maybe others, I haven't
tried). But it couldn't find the proxy when I pointed at local sites
and the virtuals on my server (again, maybe others).

So I removed privoxy, and now Tor can't even connect to its own network.

Anybody seen this, have an idea, know a fix?? Is there some kind of
(defective) interaction between the Tor and Firefox configs?

--
Glenn English

Re: Suggested edit

[Lisi Reisz]

On Friday 17 March 2017 17:24:53 Catherine Gramze wrote:
> You used Libranet, too? Are you still in touch with Tal? That was my
> favorite distro EVER. The only real help I ever got was in the Libranet
> beta testing group, for that last release that never made it because John
> was so ill.

I still mourn it and no, though I was in touch with Daniel for some time.

Oh it did make it, briefly - I bought it!!  John died almost immediately after 
it was released, and Tal withdrew it while he decided what to do.  I always 
felt that he chose to kill it, but could have given it to the community if he 
didn't want it.

But, I have a PAID FOR copy of Libranet 3.  Somewhere.  I loved, loved, loved 
it.  Nothing else can touch it - not even Debian itself!!  And it had a logo 
to die for. ;-)

Lisi

Netinstall of Debian 8 on DELL laptop fails

[Klaus Jantzen]

Hello,

I am trying to net-install Debian 8.7. (Graphical expert install)

After loading a couple of modules for the installation of the base 
system the installer claims that


beginning with 'groff-base' all modules are corrupted and cannot be loaded.

This is wrong as I used the same CD to install Debian on my PC without 
any problems.


Before that I had installed Wheezy on the laptop without any problems.


Does anyone have an idea on how to solve the problem?
Thanks in advance.
--

K.D.J.

Re: Suggested edit

[Joe]

On Fri, 17 Mar 2017 11:18:57 -0400
Catherine Gramze  wrote:

> > On Mar 17, 2017, at 10:28 AM, Jonathan Dowland 
> > wrote:
> > 
> > That would presume that the majority of users of the installer
> > wanted to install a desktop environment. This is not necessarily
> > true even for beginners, on say, server machines.  
> 
> Beginners are installing server systems as their first experience
> with Linux? I think not. Most people don't ever touch a server in
> their entire life.

You might be surprised. I was given a Red Hat 5.2 disc to play with,
and I bought a Linux magazine. It contained an article about compiling
kernels, so I tried it. It was much later that I found this was
considered an advanced task.

I tried a Debian, possibly Potato, I'm not sure. I certainly didn't get
it working, and I can remember an entire screenful of mouse
configuration options, none of which turned out to be correct.

I went to one of the early Mandrakes, did a couple of LFS builds, and
didn't go back to Debian until Sarge, which became my home server... my
current server is a direct descendent of this installation. My LFS was
pure command-line, I was bored enough by the compilation business that I
never did try a GUI, and in those days the Linux desktop was more a
curiosity than a tool.

-- 
Joe

Re: Suggested edit

[Catherine Gramze]

Sent from my iPad

> On Mar 17, 2017, at 12:57 PM, Lisi Reisz  wrote:
> 
> Not, it isn't.  Debian is for experts, or would-be experts, or those who will 
> never be experts but like to pretend.
> 

I strongly disagree. The point of a genuinely free and open operating system is 
accessibility at all levels. Debian is more difficult than some other systems 
not because it is meant for experts, but because of the dedication to the core 
concepts of FOSS. Or, why the non-free repository is not included in a standard 
installation source list. 
> 
> Yes, we were.  And we either managed, had help (I had a lot AND used Libranet 
> (an easier true derivative - used Debian sources, but pinned) or used a 
> derivative.  
> 

You used Libranet, too? Are you still in touch with Tal? That was my favorite 
distro EVER. The only real help I ever got was in the Libranet beta testing 
group, for that last release that never made it because John was so ill. 

> We neither expect nor demand that Windows be installable by all and sundry, 
> from scratch, on an empty machine.

We don't? It's a lot more idiot proof than Debian.


Cathy

Re: Suggested edit

[Joe]

On Fri, 17 Mar 2017 16:57:19 +
Lisi Reisz  wrote:

> On Friday 17 March 2017 16:45:31 Catherine Gramze wrote:
> > It is more important to not intimidate the beginner.   
> 
> Not, it isn't.  Debian is for experts, or would-be experts, or those
> who will never be experts but like to pretend.
> 
> I am so sad that Debian is becoming more and more Ubuntu-ised.  For
> those who want and like Ubuntu, Ubuntu exists.
> 
> > We were all a beginner 
> > once.  
> 
> Yes, we were.  And we either managed, had help (I had a lot AND used
> Libranet (an easier true derivative - used Debian sources, but
> pinned) or used a derivative.  
> 
> We neither expect nor demand that Windows be installable by all and
> sundry, from scratch, on an empty machine.
> 

Actually, these days, it's a pretty trivial job. You basically give it
a user name, a computer name and a timezone. I've done a few 7 Pro
installations in the last year or two.

I don't know how easy a Debian non-expert CD/DVD installation is, the
last time I tried that, I ended up without networking. Not just without
it configured, without an eth0 at all. I did report it as a bug, but
apparently it was a feature.

-- 
Joe

Re: Suggested edit

[Catherine Gramze]

Sent from my iPad
> On Mar 17, 2017, at 12:04 PM, Lisi Reisz  wrote:
> 
> There are plenty of dumbed down Debian derivatives for those that want them.  
> Please don't let's dumb Debian down any further.  Base install means base 
> install.  The net install disk by default installs Gnome (I have on occasion 
> gone to sleep and failed to prevent it).  Some WANT a base install and should 
> be able to have it easily.  You are describing Ubuntu or Mint. Ubuntu and 
> Mint exist.

No, Lisi, I don't mean like Mint or Ubuntu. My point was only that beginners 
can accidentally do a base install only, not knowing what a base install is, 
and that knowing what they did to cause it and fixing it is not something a 
beginner can easily do. 

My very first Linux installation was in 1993, a Slackware disc that came free 
in Wired or PCMagazine. It installed successfully, and there I was looking at a 
blank screen. I knew 2 things instantly: that the installation was a success, 
and that I had no idea what command might invoke a graphic interface, if one 
existed. Or what any Linux command might be. 

I gave up on Linux for a few years, until 1999 and my horrific Red Hat 6.0 
experience. I moved on to Debian in mere months, switching to the derivatives 
when Debian did not support my hardware. (Which was regularly) But I clearly 
remember that first Linux experience and that disheartening blank screen. 

I agree it is important to be able to get a base install if you want one, but 
you can do that manually simply by backing out of the installation script after 
the reboot. It is more important to not intimidate the beginner. We were all a 
beginner once.

Re: Finding firmware (and SHA sums etc), was Re: Linux Mint, Ubuntu, Arch Linux recognize my Acer Aspire S wifi, debian does not

[Lisi Reisz]

On Thursday 16 March 2017 20:08:16 Brian wrote:
> and we are told
> it is not the purpose of this list to answer practical problems by
> by guiding people to it. :)

Oh, come Brian.  We are also told that it is not the purpose of Debian to 
pander to freaks and Geeks.  We are told that the moon is made of green 
cheese and the earth is flat.  Oh, and young ladies don't do that.  We are 
not going to worry about what we are told, are we?

Lisi

Re: Suggested edit

[Lisi Reisz]

On Friday 17 March 2017 16:45:31 Catherine Gramze wrote:
> It is more important to not intimidate the beginner. 

Not, it isn't.  Debian is for experts, or would-be experts, or those who will 
never be experts but like to pretend.

I am so sad that Debian is becoming more and more Ubuntu-ised.  For those who 
want and like Ubuntu, Ubuntu exists.

> We were all a beginner 
> once.

Yes, we were.  And we either managed, had help (I had a lot AND used Libranet 
(an easier true derivative - used Debian sources, but pinned) or used a 
derivative.  

We neither expect nor demand that Windows be installable by all and sundry, 
from scratch, on an empty machine.

Lisi

Re: Where is data stored when Synaptic scans DVDs?

[songbird]

Richard Owlett wrote:
> As I have limited bandwidth, I install from DVDs.
> Also I frequently experiment with various options by doing a fresh 
> install to a different partition.
> It is annoying to keep re-scanning a set of 13 DVDs.
> Where is the information stored?
> /etc/apt/sources.list is obvious, but where is the data about contents 
> of each DVD stored?

  man apt-get

  and as i've previously said, you can copy the
dvds to spinning rust or a SSD and avoid all
future needs to reload from them again...

  and futhermore, if you are doing partitions
you can do some minimal install to get the base
system going, then do a partition copy of that
and you'll not have to redo the previous steps
again to recover if you want to try something
else.


  songbird

Re: Suggested edit

[Lisi Reisz]

On Friday 17 March 2017 15:18:57 Catherine Gramze wrote:
> > On Mar 17, 2017, at 10:28 AM, Jonathan Dowland  wrote:
> >
> > That would presume that the majority of users of the installer wanted to
> > install a desktop environment. This is not necessarily true even for
> > beginners, on say, server machines.
>
> Beginners are installing server systems as their first experience with
> Linux? I think not. Most people don't ever touch a server in their entire
> life. The closest they get to a server is a thin client. Any sane server
> administrator is going to set up a test system first, and probably on a
> desktop so they can easily mess around with it while they learn about the
> tools it offers.

In general, Debian is not recommended for total newbies who are unassisted for 
precisely this sort of reason.

There are plenty of dumbed down Debian derivatives for those that want them.  
Please don't let's dumb Debian down any further.  Base install means base 
install.  The net install disk by default installs Gnome (I have on occasion 
gone to sleep and failed to prevent it).  Some WANT a base install and should 
be able to have it easily.  You are describing Ubuntu or Mint. Ubuntu and 
Mint exist.

The mentioned problem in finding things definitely exists, especially when it 
comes to check sums, especially for those of us who are partially sighted.  
The developers, bless them, are aware and are working on it.  That is great, 
and good enough for me.

But if you want a dumbed down newbie distro, use a dumbed down newbie distro.  
If Debian is too Geekish, leave it to the Geeks.

Meanwhile, Brian, not all of us are blessed with 35 year old brains and 20/20 
vision.

Lisi

Re: Finding firmware (and SHA sums etc), was Re: Linux Mint, Ubuntu, Arch Linux recognize my Acer Aspire S wifi, debian does not

[David Wright]

On Thu 16 Mar 2017 at 20:08:16 (+), Brian wrote:
> On Thu 16 Mar 2017 at 12:04:14 -0500, David Wright wrote:
> 
> > On Thu 16 Mar 2017 at 14:05:17 (+0100), to...@tuxteam.de wrote:
> > > On Thu, Mar 16, 2017 at 12:52:00PM +, GiaThnYgeia wrote:
> > > > Brian:
> > > > 
> > > > > Documentation dispels ignorance. Dispelling ignorance requires the
> > > > > co-operation of the user.
> > > 
> > > :-)
> > > 
> > > > To recapitulate, the purpose of the list as I understand it is not to
> > > > answer practical problems with "read the manual".  It defeats the 
> > > > purpose.
> > > 
> > > I think you're being a bit unfair here. Brian has surely done a bit
> > > beyond saying "read the manual". He has provided leads to the relevant
> > > spots in there.
> > 
> > He showed a much better path to the important web page
> > http://cdimage.debian.org/debian-cd/current/i386/iso-cd/
> > He hasn't disagreed with my analysis of why people can get
> > stuck on another pathway to the ISO files on that page,
> > particularly the people desiring a network installation.
> 
> There was never had any reason to disagree with the analysis. The
> queries starting this subthread
> 
>  > Is it just me, or are the "unofficial" images hard to get to?
> 
> have the answers "no" and "yes". However, the routes to the images are
> varied and guessing which one a user takes to get there isn't fruitful.

The routes exist and don't need to be guessed. The reason that a
particular group of users follows a route is what has to be guessed.
For every selection made/link clicked etc, the user had an intention.
That intention has to be guessed by the web designer/programmer etc.
Few designers actually experience a stream of tyros sitting down to
use their design, and give them reasons.

Similarly, documentation authors must have in-depth knowledge of
their subject, but also have the ability to read the mind of the
ignorant reader.

> > But the Installation Manual comment, particularly the rhetorical
> > question, was a cheap shot IMO. Navigating these pages should be
> > straightforward. Any indication that the Installation Manual
> > is going to help that navigation, or should help, is laughable
> > and contradicts the very raison d'être of the web.
> 
> Note that the OP knew he might require firmware for his install. The
> Manual doesn't remove the need for web page navigation to be improved
> but it does give a destination for firmware images, so any port in a
> storm etc. I'd use the text version for searching.

Interesting; that might make the headings less influential when
reading it.

> Fortunately, the OP was helpfully directed to the correct netinst ISO.
> Unfortunately, it is quite possible it does not contain the firmware for
> what looks like the Atheros-based chipset he has. Off to backports and
> firmware-atheros, then. I'd link to pages on the wiki but documentation
> does not appear to be in good standing at the moment and we are told
> it is not the purpose of this list to answer practical problems by
> by guiding people to it. :)

Cheers,
David.

Re: Suggested edit

[Catherine Gramze]

> On Mar 17, 2017, at 10:28 AM, Jonathan Dowland  wrote:
> 
> That would presume that the majority of users of the installer wanted to
> install a desktop environment. This is not necessarily true even for 
> beginners,
> on say, server machines.

Beginners are installing server systems as their first experience with Linux? I 
think not. Most people don't ever touch a server in their entire life. The 
closest they get to a server is a thin client. Any sane server administrator is 
going to set up a test system first, and probably on a desktop so they can 
easily mess around with it while they learn about the tools it offers. 

Re: Plus de son pour regarder des vidéos avec Firefox 52

[MERLIN Philippe]

Le vendredi 17 mars 2017, 14:24:07 CET MERLIN Philippe a écrit :
> Le vendredi 17 mars 2017, 13:56:48 CET heho a écrit :
> > MERLIN Philippe a écrit le 17/03/2017 à 13:21 :
> > > Je visionne souvent une vidéo youtube depuis aujourd'hui impossible
> > > d'entendre les commentaires, cette vidéo est devenue muette. En
> > > Googolisant j'ai vu qu'à partir de la Firefox 52  des plugins sont
> > > abandonnés. Savez vous comment faire pour remédier à ce problème.?
> > 
> > Bonjour Philippe,
> > 
> > 
> > je ne sais pas si cela pourrait résoudre votre problème,
> > mais j'utilise parfois ce "hack" (quelque soit le browser)
> > pour forcer la visualisation en flash ( pas taper ;)
> > quand c'est parfois nécessaire:
> > 
> > et donc, il suffit de rajouter
> > 
> > "=1" (sans les guillemets) à l'url youtube
> > 
> > Cordialement,
> > hého
> 
> En effet cela résout le problème , merci.
> Maintenant j'aimerais savoir pourquoi regarder une page web en Html5 il n'y
> a plus de possibilité de la visionner avec le son .
> Philippe MERLIN
J'ai trouvé la réponse à ma question, ce n'est pas très encourageant,
http://www.omgubuntu.co.uk/2017/03/firefox-52-no-sound-pulseaudio-alsa-linux
Firefox  qui m'était très sympathique a là une attitude un peu surprenante et 
très désagréable.
Circulez il n'y a rien à voir. Utilisez Pulseaudio si vous voulez avoir du son 
pour vos vidéos autrement Basta!!!
Philippe Merlin

Re: Plus de son pour regarder des vidéos avec Firefox 52

[bernard . schoenacker]

- Mail original -
De: "MERLIN Philippe" 
À: debian-user-french@lists.debian.org
Envoyé: Vendredi 17 Mars 2017 14:24:07
Objet: Re: Plus de son pour regarder des vidéos avec Firefox 52

Le vendredi 17 mars 2017, 13:56:48 CET heho a écrit :
> MERLIN Philippe a écrit le 17/03/2017 à 13:21 :
> > Je visionne souvent une vidéo youtube depuis aujourd'hui impossible
> > d'entendre les commentaires, cette vidéo est devenue muette. En
> > Googolisant j'ai vu qu'à partir de la Firefox 52  des plugins sont
> > abandonnés. Savez vous comment faire pour remédier à ce problème.?
> 
> Bonjour Philippe,
> 

> je ne sais pas si cela pourrait résoudre votre problème,
> mais j'utilise parfois ce "hack" (quelque soit le browser)
> pour forcer la visualisation en flash ( pas taper ;)
> quand c'est parfois nécessaire:
> 
> et donc, il suffit de rajouter
> 
> "=1" (sans les guillemets) à l'url youtube
> 
> Cordialement,
> hého

En effet cela résout le problème , merci.
Maintenant j'aimerais savoir pourquoi regarder une page web en Html5 il n'y a 
plus de possibilité de la visionner avec le son .
Philippe MERLIN


bonjour,


je conseille d'installer midori qui accepte les vidéos en html5

http://midori-browser.org/download/debian/


slt
bernard

Re: Suitable text editor [NOT word processor] or workaround?

[Gene Heskett]

On Friday 17 March 2017 10:26:05 Jonathan Dowland wrote:

> On Fri, Mar 17, 2017 at 07:37:11AM -0400, Gene Heskett wrote:
> > I do, I am a very longtime user of amanda, since 1998 TBE.
>
> OK - I misread your post as suggesting you had lost 6 months of work.
>
> > And yet for me, "life" is long, I'm 82 and counting.
>
> No point wasting any of it, no matter how much you end up getting.

My point exactly. I've chased electrons to make them do useful work for  
close to 70 of those 82 years, so I can't wean me at this late stage. :)

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

Re: Suggested edit

[Jonathan Dowland]

On Thu, Mar 16, 2017 at 11:43:05AM -0400, Catherine Gramze wrote:
> That "feature" is not limited to the cd version. It is also in the netinst
> version and the dvd versions. It would be nice if people intuitively
> understood that the "base system" was not going to meet their expectations of
> a minimal install, which includes a desktop environment, but noobs don't
> understand that. Equally nice would be an "advanced installation" option to
> only do a base system installation.  Making the selection of a mirror and
> desktop environment an unskippable part of the installation, unless you chose
> that advanced "base system" would be very user friendly, particularly for
> noobs.

That would presume that the majority of users of the installer wanted to
install a desktop environment. This is not necessarily true even for beginners,
on say, server machines.

-- 
Jonathan Dowland
Please do not CC me, I am subscribed to the list.


signature.asc
Description: Digital signature

Re: Where is data stored when Synaptic scans DVDs?

[Jonathan Dowland]

On Fri, Mar 17, 2017 at 08:47:01AM -0500, Richard Owlett wrote:
> /etc/apt/sources.list is obvious, but where is the data about contents of
> each DVD stored?

/var/cache/apt/* I believe

-- 
Jonathan Dowland
Please do not CC me, I am subscribed to the list.


signature.asc
Description: Digital signature

Re: Suitable text editor [NOT word processor] or workaround?

[Jonathan Dowland]

On Fri, Mar 17, 2017 at 07:37:11AM -0400, Gene Heskett wrote:
> I do, I am a very longtime user of amanda, since 1998 TBE.

OK - I misread your post as suggesting you had lost 6 months of work.

> And yet for me, "life" is long, I'm 82 and counting.

No point wasting any of it, no matter how much you end up getting.

-- 
Jonathan Dowland
Please do not CC me, I am subscribed to the list.


signature.asc
Description: Digital signature

Re: Comment s'opère la mise à jour du noyau dans Stretch ? [RESOLU][Etait: Pourquoi linux-headers-`uname r` échoue sur Stretch et pas sur Jessie ?]

[Olivier]

Le 17 mars 2017 à 10:34, Jean-Marc  a écrit :

> Fri, 17 Mar 2017 10:26:49 +0100
> Olivier  écrivait :
>
> salut,
>
> > $ sudo apt-get -s dist-upgrade
> > Lecture des listes de paquets... Fait
> > Construction de l'arbre des dépendances
> > Lecture des informations d'état... Fait
> > Calcul de la mise à jour... Fait
> > Les paquets suivants ont été installés automatiquement et ne sont plus
> > nécessaires :
> >   bijiben libsofia-sip-ua-glib3 libsofia-sip-ua0 telepathy-rakia
> > Veuillez utiliser « sudo apt autoremove » pour les supprimer.
> > Les NOUVEAUX paquets suivants seront installés :
> >   firmware-linux-free gnome-calendar irqbalance libopenmpt0 libpgm-5.2-0
> > libproxy1-plugin-gsettings
> >   libproxy1-plugin-networkmanager libsaxonhe-java
> > linux-image-4.9.0-2-686-pae
> > Les paquets suivants seront mis à jour :
> >   gnome gnome-core gstreamer1.0-libav libavdevice57 libavfilter6
> > libavformat57 libgegl-0.3-0 libopencv-calib3d2.4v5
> >   libopencv-core2.4v5 libopencv-features2d2.4v5 libopencv-flann2.4v5
> > libopencv-highgui2.4-deb0 libopencv-imgproc2.4v5
> >   libopencv-objdetect2.4v5 libopencv-video2.4v5 libxmlbeans-java libzmq5
> > linux-image-686-pae
> > 18 mis à jour, 9 nouvellement installés, 0 à enlever et 0 non mis à jour.
> >
> >
> > Si j'en crois ce qui précède, linux-image n'est pas mis à jour par
> > dist-upgrade.
>
> Si tu regardes attentivement, tu verras linux-image-686-pae dans la liste
> des paquets qui seront mis à jour et linux-image-4.9.0-2-686-pae dans la
> liste des nouveaux paquets.
>
> Je pensais que c'est ce que tu voulais.
>
>
> Jean-Marc 
>

En effet, je suis complètement miro !
J'arrête de boire de l'eau ;-))


Dois-je retenir qu'un dist-upgrade est nécessaire à chaque changement de
version du noyau ?
Ai-je raté une alerte me signalant qu'un changement de noyau est possible
ou recommandé ?
Que risque-ton avec un dist-upgrade (si on n'a pas changé les données de
/etc/apt) ?

En tout cas, merci pour le tuyau !

Where is data stored when Synaptic scans DVDs?

[Richard Owlett]

As I have limited bandwidth, I install from DVDs.
Also I frequently experiment with various options by doing a fresh 
install to a different partition.

It is annoying to keep re-scanning a set of 13 DVDs.
Where is the information stored?
/etc/apt/sources.list is obvious, but where is the data about contents 
of each DVD stored?

TIA

Re: Plus de son pour regarder des vidéos avec Firefox 52

[MERLIN Philippe]

Le vendredi 17 mars 2017, 13:56:48 CET heho a écrit :
> MERLIN Philippe a écrit le 17/03/2017 à 13:21 :
> > Je visionne souvent une vidéo youtube depuis aujourd'hui impossible
> > d'entendre les commentaires, cette vidéo est devenue muette. En
> > Googolisant j'ai vu qu'à partir de la Firefox 52  des plugins sont
> > abandonnés. Savez vous comment faire pour remédier à ce problème.?
> 
> Bonjour Philippe,
> 

> je ne sais pas si cela pourrait résoudre votre problème,
> mais j'utilise parfois ce "hack" (quelque soit le browser)
> pour forcer la visualisation en flash ( pas taper ;)
> quand c'est parfois nécessaire:
> 
> et donc, il suffit de rajouter
> 
> "=1" (sans les guillemets) à l'url youtube
> 
> Cordialement,
> hého

En effet cela résout le problème , merci.
Maintenant j'aimerais savoir pourquoi regarder une page web en Html5 il n'y a 
plus de possibilité de la visionner avec le son .
Philippe MERLIN

Plus de son pour regarder des vidéos avec Firefox 52

[MERLIN Philippe]

Bonjour,
Je visionne souvent une vidéo youtube depuis aujourd'hui impossible d'entendre 
les commentaires, cette vidéo est devenue muette. En Googolisant j'ai vu qu'à 
partir de la Firefox 52  des plugins sont abandonnés. Savez vous comment faire 
pour remédier à ce problème.?
Philippe Merlin

Re: XFCE - move items in notification area

[Celejar]

On Fri, 17 Mar 2017 00:55:48 -0400
Alex  wrote:

> How does one move items within the notification area?

Not sure one can:

https://www.reddit.com/r/linux4noobs/comments/33vu14/is_there_any_way_to_order_the_indicators_on/
https://forum.xfce.org/viewtopic.php?id=7034

Celejar

Re: Comment s'opère la mise à jour du noyau dans Stretch ? [Etait: Pourquoi linux-headers-`uname r` échoue sur Stretch et pas sur Jessie ?]

[Jean-Marc]

Fri, 17 Mar 2017 10:26:49 +0100
Olivier  écrivait :

salut,

> $ sudo apt-get -s dist-upgrade
> Lecture des listes de paquets... Fait
> Construction de l'arbre des dépendances
> Lecture des informations d'état... Fait
> Calcul de la mise à jour... Fait
> Les paquets suivants ont été installés automatiquement et ne sont plus
> nécessaires :
>   bijiben libsofia-sip-ua-glib3 libsofia-sip-ua0 telepathy-rakia
> Veuillez utiliser « sudo apt autoremove » pour les supprimer.
> Les NOUVEAUX paquets suivants seront installés :
>   firmware-linux-free gnome-calendar irqbalance libopenmpt0 libpgm-5.2-0
> libproxy1-plugin-gsettings
>   libproxy1-plugin-networkmanager libsaxonhe-java
> linux-image-4.9.0-2-686-pae
> Les paquets suivants seront mis à jour :
>   gnome gnome-core gstreamer1.0-libav libavdevice57 libavfilter6
> libavformat57 libgegl-0.3-0 libopencv-calib3d2.4v5
>   libopencv-core2.4v5 libopencv-features2d2.4v5 libopencv-flann2.4v5
> libopencv-highgui2.4-deb0 libopencv-imgproc2.4v5
>   libopencv-objdetect2.4v5 libopencv-video2.4v5 libxmlbeans-java libzmq5
> linux-image-686-pae
> 18 mis à jour, 9 nouvellement installés, 0 à enlever et 0 non mis à jour.
> 
> 
> Si j'en crois ce qui précède, linux-image n'est pas mis à jour par
> dist-upgrade.

Si tu regardes attentivement, tu verras linux-image-686-pae dans la liste des 
paquets qui seront mis à jour et linux-image-4.9.0-2-686-pae dans la liste des 
nouveaux paquets.

Je pensais que c'est ce que tu voulais.


Jean-Marc 


pgphvUbXuTHry.pgp
Description: PGP signature

Re: Suitable text editor [NOT word processor] or workaround?

[Gene Heskett]

On Friday 17 March 2017 05:49:30 Jonathan Dowland wrote:

> On Thu, Mar 16, 2017 at 09:54:42AM -0400, Gene Heskett wrote:
> > gedit has caused me to have to start over again, 3 times, on a 600+
> > line configuration file for machine control on several occasions. As
> > the file it trashed was 6 months worth of adding new features to
> > that machines control, that was not appreciated. 1st time is a wtf?,
> > 2nd time sends me a message, 3rd time I shut it down w/o the save
> > and never ran it again.
>
> Whilst this is very frustrating, one other thing it tells me is you
> need to set up a backup system (although you are quite right to
> abandon gedit based on your experiences)

I do, I am a very longtime user of amanda, since 1998 TBE. I've never 
contaminated my thinking with a windows install, going from a coco3 with 
nitros9, to amigados, to linux at RH-5.0. amanda does all 5 of these 
machines every night at about 1:20 AM and it got used all 3 times. The 
problem with that is that the status of the backup was 15 to 20 hours 
old, so I lost that days work and had to re-invent that particular batch 
of work.  I keep a rotating vtape setup of 30 days here, on a terabyte 
drive, and amanda can go back for recovery as much as 26 days if 
required. Since this is not a real tape setup where access needs to read 
the whole tape to get to the last block of data, but random access on 
rotating media, recovery's are as much as 100x faster than from tape. 
And so far its been 100% dependable, which is a heck of a lot better 
than losing 4 or 5 tapes a month, and a drive a year because I never had 
an affordable drive that didn't have to spend the thanksgiving to new 
years holidays in Oklahoma City being refurbished. Disk drives have been 
100x more dependable. That drive now has in excess of 62 thousand 
spinning hours on it, has had 25 reallocated sectors since the first 
time I looked at it with smartctl at about 5 thousand hours. It will 
die, but smartctl has so far given me enough notice that I can run up to 
Staples and get another drive and rsync to it, no data lost.

> > gedit is like asking if freshly poured concrete will crack.  Wrong
> > question because it will, the right question is "when", its just a
> > matter of time. Life is too short for putting up with that.
>
> I like this comparison :)

And yet for me, "life" is long, I'm 82 and counting.

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

Re: TTL expired in transit to qemu virtual machine.

[Dominik George]

Hi,

> >iptables -L FORWARD -nv
>Chain FORWARD (policy DROP 0 packets, 0 bytes)
>pkts bytes target prot opt in out source  
>destination
>XX ACCEPT all  --  br0br0 0.0.0.0/0   
>0.0.0.0/0

What is that supposed to do?

Forwarding on the IP layer, by definition, happens between different interfaces.

Although this rule is most likely a no-op, you should remove it to reduce 
confusion.

-nik

Re: mother nueva

[walter salazar]

Hola javier. Gracias por tu respuesta. Incluso tu comentario y el anterior
en cierta forma coinciden.
Inmediatamente la pongo en practica. Luego les comento. Gracias

El 16 de marzo de 2017, 8:21, JAP  escribió:

> El 15/03/17 a las 23:12, walter salazar escribió:
>
> Hola. Compre una mother gigabyte z170x con un i7 6ta generacion y 16mg
>> de RAM compatibles. Funciona, pero cuando uso firefox o chrome se
>> clava(cuelga o freeze) o uso android studio lo mismo,etcc.
>>   Con windows 7 no se clava, ni le pasa nada, sera que la mother es
>> nueva y todavia no esta muy  desarrollado Debian para este hardware?
>> Como puedo saber esto?
>>  Alguna respuesta o consejo seran muy bienvenidos!!!
>> Gracias
>>
>
> Tengo exactamente el mismo equipo.
> El la tarjeta de video.
> Eliminá todo lo que sea nvidia
> # atp-get purge nvidia*
>
> JAP
>
>

TTL expired in transit to qemu virtual machine.

[Mimiko]

Hello.

I've setup qemu/kvm and installed several virtual machines. Access and ping to some virtuals are ok, but one have a stable problem not receiving 
correctly packets. First, this is the environment:


>uname -a
Linux 3.2.0-4-amd64 #1 SMP Debian 3.2.84-1 x86_64 GNU/Linux

>libvirtd --version
libvirtd (libvirt) 0.9.12.3

>cat /etc/network/interfaces
auto lo
iface lo inet loopback

auto eth0
iface eth0 inet manual

auto eth1
iface eth1 inet manual

auto bond0
iface bond0 inet manual
bond-slaves eth0 eth1
bond-mode balance-alb
bond-miimon 100
bond-downdelay 200
bond-updelay 200

auto br0
iface br0 inet static
address 10.10.10.10
netmask 255.255.0.0
vlan-raw-device bond0
bridge_ports bond0
bridge_stp off
bridge_fd 0
bridge_ageing 0
bridge_maxwait 0

Virtual machines connects to LAN using bridge:
>virt-install  --network=bridge:br0,model=virtio 

All virtuals has networking configuret like that. Also in iptables is an entry 
to allow access to virtuals:

>iptables -L FORWARD -nv
Chain FORWARD (policy DROP 0 packets, 0 bytes)
 pkts bytes target prot opt in out source   destination
 XX ACCEPT all  --  br0br0 0.0.0.0/00.0.0.0/0

Most virtuals does not have networking problems, but some times they can't be 
reached. For now only one virtual machines have this problem:
From windows machine ping virtual machine:

>ping 10.10.10.3

Reply from 10.10.10.10: TTL expired in transit.
Reply from 10.10.10.10: TTL expired in transit.
Reply from 10.10.10.10: TTL expired in transit.
Reply from 10.10.10.10: TTL expired in transit.
Request timed out.
Request timed out.
Request timed out.
Request timed out.
Request timed out.
Request timed out.
Reply from 10.10.10.10: TTL expired in transit.
Reply from 10.10.10.10: TTL expired in transit.
Reply from 10.10.10.10: TTL expired in transit.
Reply from 10.10.10.10: TTL expired in transit.

>tracert -d 10.10.10.3

Tracing route to 10.10.10.3 over a maximum of 30 hops

  1<1 ms<1 ms<1 ms  10.10.10.10
  2<1 ms<1 ms<1 ms  10.10.10.10
  3<1 ms<1 ms * 10.10.10.10
  4 1 ms<1 ms<1 ms  10.10.10.10
  5<1 ms<1 ms * 10.10.10.10

So packet goes round on interfaces of server hosting virtuals.

Virtuals are linux different flavour and one windows. This problem may occur on 
any of this virtuals.

I've observed that for this particular virtual, which have problem, the arp table of host assigned self mac to the virtual's IP, not the mac 
configured for virtual machine.


What could be the problem?

--
Mimiko desu.

Re: Chrome pour Debian 32 bits ?

[baal]

re bonjour, un peut orienté y'a sa 
http://assiste.com/Chrome_VS_Chromium.html



Le 17/03/2017 à 11:35, erwin a écrit :

Le Fri, 17 Mar 2017 11:22:37 +0100
andre_deb...@numericable.fr écrivait:



Quelle différence entre Chromium et Chrome ?


bonjour,
j'ai trouvé cela :
http://www.commentcamarche.net/faq/27735-introduction-a-chromium-google-chrome
bonne journée
Erwin

Re: Guide(s?) to backup philosophies

[Dan Purgert]

David Christensen wrote:
> On 03/13/2017 05:38 AM, Dan Purgert wrote:
>> Currently, the system here is
>>
>>  - every PC has a cronjob backing up $HOME to a central "server" (read -
>>repurposed PC with decent WD drives), just an rsync script that runs
>>daily.
>
> Don't forget security:
>
> 1.  With a "push" arrangement (e.g. each workstation backs up itself to 
> the server) -- if a workstation gets compromised, the backups are at risk.
>
> 2.  With a "pull" arrangement (e.g. the server backs up all the 
> workstations) -- if a workstation gets compromised, the backups should 
> be safe (and might have clues about the intrusion).  Additionally, the 
> backup server can be completely firewalled (e.g. no open ports).

Since the PCs are laptops, they're not always here, so I was never able
to figure out how to get pull to work with the condition that we were on
vacation (or the laptops were otherwise "not home").

Though, yeah, the stuff that's statically here (desktop, server, etc.)
are rsync-by-pull.

-- 
|_|O|_| Registered Linux user #585947
|_|_|O| Github: https://github.com/dpurgert
|O|O|O| PGP: 05CA 9A50 3F2E 1335 4DC5  4AEE 8E11 DDF3 1279 A281

Re: Chrome pour Debian 32 bits ?

[erwin]

Le Fri, 17 Mar 2017 11:22:37 +0100
andre_deb...@numericable.fr écrivait:


> Quelle différence entre Chromium et Chrome ?
> 

bonjour,
j'ai trouvé cela :
http://www.commentcamarche.net/faq/27735-introduction-a-chromium-google-chrome
bonne journée
Erwin

Re: Chrome pour Debian 32 bits ?

[andre_debian]

On Friday 17 March 2017 08:48:25 baal wrote:
> bonjour, essais sa 
> 
http://frankyleundeu.over-blog.com/article-installer-google-chrome-sur-debian-ubuntu-86207444.html

Merci,

J'ai fait ce qui est demandé au lien ci-dessus :
"Impossible de récupérer 
http://dl.google.com/linux/chrome/deb/dists/stable/Release  
Impossible de trouver l'entrée « main/binary-i386/Packages » attendue dans le 
fichier Release : ligne non valable dans sources.list ou fichier corrompu
Le téléchargement de quelques fichiers d'index a échoué, ils ont été ignorés, 
ou les anciens ont été utilisés à la place"

Sans modifier le sources.list, le paquet :
google-chrome-stable - The web browser from Google
est proposé que j'ai installé.

Dommage que Google ne propose pas Chrome en 32 bits.

Il parait que la dernière version de Google Chrome
n'accepte plus les certificats StartSSL,
et j'aurais voulu tester.
(la version installée l'accepte)

Quelle différence entre Chromium et Chrome ?

Bonne journée,  

André.


> > Le 16/03/17 à 23:46, andre_deb...@numericable.fr a écrit :
> > AF> Y a t-il une version du navigateur Chrome 32 bits ?
> > Non, mais y'a chromium https://packages.debian.org/jessie/chromium

Re: Suitable text editor [NOT word processor] or workaround?

[Jonathan Dowland]

On Thu, Mar 16, 2017 at 09:54:42AM -0400, Gene Heskett wrote:
> gedit has caused me to have to start over again, 3 times, on a 600+ line 
> configuration file for machine control on several occasions. As the file 
> it trashed was 6 months worth of adding new features to that machines 
> control, that was not appreciated. 1st time is a wtf?, 2nd time sends me 
> a message, 3rd time I shut it down w/o the save and never ran it again.

Whilst this is very frustrating, one other thing it tells me is you need
to set up a backup system (although you are quite right to abandon gedit
based on your experiences)

> gedit is like asking if freshly poured concrete will crack.  Wrong 
> question because it will, the right question is "when", its just a 
> matter of time. Life is too short for putting up with that.

I like this comparison :)


signature.asc
Description: Digital signature

Re: Comment s'opère la mise à jour du noyau dans Stretch ? [Etait: Pourquoi linux-headers-`uname r` échoue sur Stretch et pas sur Jessie ?]

[Olivier]

Le 16 mars 2017 à 19:13, Francois Lafont  a écrit :

> On 03/16/2017 05:21 PM, Olivier wrote:
>
> > Ce que j'ai du mal à comprendre, c'est pourquoi le passage de 4.8 en 4.9
> ne
> > s'opère pas automatiquement : je n'ai jamais spécifié la version 4.8,
> j'ai
> > à l'époque spécifié la version courante qui était la 4.8.
> > Si la version courante change, je m'attends à ce que mon noyau suive, ni
> > plus ni moins, ne serait-ce que justement, pour garder la possibilité des
> > en-têtes si j'en ai envie..
>
> Que donne « dpkg -l | grep linux-image » sur ta machine ?
>
> Normalement, c'est le principe du métapackage linux-image-amd64 de faire
> ce que tu dis.
>
> --
> François Lafont
>
>

$ dpkg -l | grep linux-image
ii  linux-image-4.8.0-2-686-pae
4.8.15-2 i386 Linux 4.8 for modern PCs
(signed)
ii  linux-image-686-pae
4.8+77   i386 Linux for modern PCs
(meta-package)


cf aussi mon post précédent avec apt policy.

Existe-t-il une commande qui permet à oeil exercé de comprendre pourquoi un
paquet figure explicitement dans la liste des paquets conservés et pourquoi
un autre figure dans la liste de ceux mis à jour ?

Re: Comment s'opère la mise à jour du noyau dans Stretch ? [Etait: Pourquoi linux-headers-`uname r` échoue sur Stretch et pas sur Jessie ?]

[Eric Degenetais]

Le 17 mars 2017 10:27, "Olivier"  a écrit :



Le 16 mars 2017 à 17:42, Jean-Marc  a écrit :

> Thu, 16 Mar 2017 17:21:45 +0100
> Olivier  écrivait :
>
> > En d'autres termes, que font les utilisateurs de Stretch, qui comme moi
> > veulent simplement rester synchrone avec la distribution ?
>
> Tu as essayé :
>  ?
>

$ sudo apt-get -s dist-upgrade

Lecture des listes de paquets... Fait
Construction de l'arbre des dépendances
Lecture des informations d'état... Fait
Calcul de la mise à jour... Fait
Les paquets suivants ont été installés automatiquement et ne sont plus
nécessaires :
  bijiben libsofia-sip-ua-glib3 libsofia-sip-ua0 telepathy-rakia

Veuillez utiliser « sudo apt autoremove » pour les supprimer.
Les NOUVEAUX paquets suivants seront installés :
  firmware-linux-free gnome-calendar irqbalance libopenmpt0 libpgm-5.2-0
libproxy1-plugin-gsettings
  libproxy1-plugin-networkmanager libsaxonhe-java
linux-image-4.9.0-2-686-pae
Les paquets suivants seront mis à jour :

  gnome gnome-core gstreamer1.0-libav libavdevice57 libavfilter6
libavformat57 libgegl-0.3-0 libopencv-calib3d2.4v5
  libopencv-core2.4v5 libopencv-features2d2.4v5 libopencv-flann2.4v5
libopencv-highgui2.4-deb0 libopencv-imgproc2.4v5
  libopencv-objdetect2.4v5 libopencv-video2.4v5 libxmlbeans-java libzmq5
linux-image-686-pae
18 mis à jour, 9 nouvellement installés, 0 à enlever et 0 non mis à jour.


Si j'en crois ce qui précède, linux-image n'est pas mis à jour par
dist-upgrade


Bonjour, de ce que j'en comprends, ce qui porte la version du noyau et est
effectivement mis à jour est le paquet spécifique de la plate-forme (ici
linux-image-686-pae, qui figure bel et bien dans la liste des paquets mis à
jour). Ce paquet de plate-forme a avec linux-image une relation de type
"provide"



Jean-Marc 
>

Re: Comment s'opère la mise à jour du noyau dans Stretch ? [Etait: Pourquoi linux-headers-`uname r` échoue sur Stretch et pas sur Jessie ?]

[Olivier]

Le 16 mars 2017 à 17:42, Jean-Marc  a écrit :

> Thu, 16 Mar 2017 17:21:45 +0100
> Olivier  écrivait :
>
> > En d'autres termes, que font les utilisateurs de Stretch, qui comme moi
> > veulent simplement rester synchrone avec la distribution ?
>
> Tu as essayé :
>  ?
>

$ sudo apt-get -s dist-upgrade
Lecture des listes de paquets... Fait
Construction de l'arbre des dépendances
Lecture des informations d'état... Fait
Calcul de la mise à jour... Fait
Les paquets suivants ont été installés automatiquement et ne sont plus
nécessaires :
  bijiben libsofia-sip-ua-glib3 libsofia-sip-ua0 telepathy-rakia
Veuillez utiliser « sudo apt autoremove » pour les supprimer.
Les NOUVEAUX paquets suivants seront installés :
  firmware-linux-free gnome-calendar irqbalance libopenmpt0 libpgm-5.2-0
libproxy1-plugin-gsettings
  libproxy1-plugin-networkmanager libsaxonhe-java
linux-image-4.9.0-2-686-pae
Les paquets suivants seront mis à jour :
  gnome gnome-core gstreamer1.0-libav libavdevice57 libavfilter6
libavformat57 libgegl-0.3-0 libopencv-calib3d2.4v5
  libopencv-core2.4v5 libopencv-features2d2.4v5 libopencv-flann2.4v5
libopencv-highgui2.4-deb0 libopencv-imgproc2.4v5
  libopencv-objdetect2.4v5 libopencv-video2.4v5 libxmlbeans-java libzmq5
linux-image-686-pae
18 mis à jour, 9 nouvellement installés, 0 à enlever et 0 non mis à jour.


Si j'en crois ce qui précède, linux-image n'est pas mis à jour par
dist-upgrade.


>
> Jean-Marc 
>

Re: Chrome pour Debian 32 bits ?

[Frédéric MASSOT]

Le 16/03/2017 à 23:46, andre_deb...@numericable.fr a écrit :
> Bonsoir,
> 
> Y a t-il une version du navigateur Chrome 32 bits ?
> je ne vois que pour Debian (deb) et Fedora (rpm) 64 bits.

En décembre 2015 Google a annoncé l'arrêt de Chrome 32 bits pour Linux :

-
https://groups.google.com/a/chromium.org/forum/#!topic/chromium-dev/FoE6sL-p6oU

- http://www.webupd8.org/2016/03/fix-failed-to-fetch-google-chrome_3.html

Pour le 32 bits, il faut utiliser Chromium, Google a maintenu les
options de compilation.


-- 
==
|  FRÉDÉRIC MASSOT   |
| http://www.juliana-multimedia.com  |
|   mailto:frede...@juliana-multimedia.com   |
| +33.(0)2.97.54.77.94  +33.(0)6.67.19.95.69 |
===Debian=GNU/Linux===

Re: programmatically determining the desktop environment of a system

[tomas]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Thu, Mar 16, 2017 at 10:17:37PM -0700, Dan Hitt wrote:
> Is there a programmatic way that a piece of software can learn what
> desktop environment it is executing in?

Just out of curiosity: what is (roughly) your use case?

(Personally I think it's a bad idea, but I'd like to learn)

regards
- -- t
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAljLmC4ACgkQBcgs9XrR2kYBoACZAVOgMl1I2DyFdJSp1FMMy5eA
/rMAn0BTT1VByF5rhECqo47jAexuAFhG
=ucsZ
-END PGP SIGNATURE-

Re: Chrome pour Debian 32 bits ?

[baal]

bonjour, essais sa 
http://frankyleundeu.over-blog.com/article-installer-google-chrome-sur-debian-ubuntu-86207444.html





Le 17/03/2017 à 08:44, Daniel Caillibaud a écrit :

Le 16/03/17 à 23:46, andre_deb...@numericable.fr a écrit :
AF> Y a t-il une version du navigateur Chrome 32 bits ?

Non, mais y'a chromium https://packages.debian.org/jessie/chromium

Re: Chrome pour Debian 32 bits ?

[Daniel Caillibaud]

Le 16/03/17 à 23:46, andre_deb...@numericable.fr a écrit :
AF> Y a t-il une version du navigateur Chrome 32 bits ?

Non, mais y'a chromium https://packages.debian.org/jessie/chromium

-- 
Daniel

Quand on est trop bonne pâte, on risque de finir
dans le pétrin.
Pierre Dac

Re: Outil d'audit de charge

[Daniel Caillibaud]

Le 08/03/17 à 18:10, fra-duf-no-s...@tourde.org (François TOURDE) a écrit :

FT> Le 17233ième jour après Epoch,
FT> hams...@suna.fdn.fr écrivait:
FT> 
FT> > Le 08/03/2017 17:15, Pierre Malard a écrit :
FT> >> Mais je voudrait pouvoir écrire ces valeurs dans un fichier exploitable
FT> >> dans la durée. Auriez-vous une idée ?
FT> >
FT> > Ce dont tu parle ressemble fort a munin ou nagios. En plus ils se
FT> > chargent de faire les graphiques a ta place.
FT> 
FT> +1 pour munin, qui est bien plus léger à mettre en place que nagios dans
FT> ce genre de cas.

Je l'utilise aussi et c'est un bon outil permettant d'écrire facilement ses 
propres sondes
dans le langage que l'on veut. 

Mais pour ton besoin, si munin ne produit pas directement la sortie qui 
t'intéresse
(exploitable dans un tableur), regarde atop dont les résultats seront 
probablement plus facile
à exploiter (à coup de sed|awk pour produire du csv).

atop affiche un mix de ce que l'on peut avoir avec top|htop|iotop|iftop et 
l'enregistre dans un
log binaire, que tu peux ensuite lire rétrospectivement avec atop -r pour avoir 
une sortie
texte, dont les infos sont faciles à extraire avec un pipe vers awk (cf man 
atop au § PARSEABLE
OUTPUT).

-- 
Daniel

Internet permet à ceux qui n'ont rien à dire de le dire quand même.
Toorop

Re: Serveur bloqué par de multiples CRON -f ?

[Daniel Caillibaud]

Le 16/03/17 à 11:27, Charles Plessy  a écrit :
CP> (Résumé des épisodes précédents, j'ai une machine virtuelle sur laquelle il
CP> devient parfois impossible de se connecter.  Les sessions existantes 
continuent
CP> de fonctionner normalement, une grande partie des logs ne sont plus écrits, 
et
CP> un processus cron par heure se lance, semble bloquer et s'accumule.)


CP> Il a fallu attendre, mais le plantage nouveau est arrivé.
CP> 
CP> `killall cron` enlève toutes les tâches cron bloquées, mais il est toujours
CP> impossible de se connecter.  C'est donc un symptôme et pas une cause.
CP> 
CP> Dans kern.log, je note:
CP> 
CP> Mar 14 11:51:56 dgt-med vmunix: [3190021.268633] rsession invoked 
oom-killer:
[…]
CP> Mar 14 13:17:30 dgt-med vmunix: [3195156.740951] Task in /lxc/dgt-med 
killed as a result of
CP> limit of /lxc/dgt-med

À partir du moment où tu as de l'oom kill, ça sent mauvais et c'est pas si 
étonnant que ça parte
en vrille ensuite.

J'ai eu des pbs d'oomkill avec lxc + btrfs + kernel 4.9, les tâches btrfs qui 
suivent des modifs
de snapshots déclenchaient de l'oomkill (sur le host) et aboutissaient au 
plantage total du
host (repasser en 4.8 a réglé le pb pour le moment).

Tu peux installer atop sur le host, et le régler avec une mesure par minute (10 
par défaut,
dans /etc/default/atop mettre `INTERVAL=60`), ça devrait te permettre après 
coup de voir à
chaque minute l'état complet du host, par ex atop -r 
/var/log/atop/atop_mmdd -b hh:mm
pour avoir un top amélioré de cette minute là, que tu peux trier par conso RAM, 
CPU, disque,
etc. (man atop pour les détails).

La suggestion de François sur le rsyslog distant est aussi une bonne idée pour 
pister l'origine
du pb.

Mais regarde quand même la mémoire dispo dans cette VM et les besoins attendus 
de rstudio, si
c'est trop juste y'a p'tet pas vraiment d'autre solution que d'allouer plus de 
RAM à cette VM.

-- 
Daniel

- Aujourd'hui, c'est la chasse à l'ours.
  Où cours-tu le lapin? Tu ne risques rien!
- Eh, t'es con! J'ai pas mes papiers!
Coluche