Re: Relative stability of Testing vs Unstable

[Joe Pfeiffer]

My experience, solely as a user, has been that sometimes the unstable
distribution breaks and you're hosed.  I can't remember when I was last
burned by running testing.

Re: Spam on Debian lists

[Joe Pfeiffer]

Joel Rees  writes:

> On Fri, Apr 21, 2017 at 6:52 AM, Patrick Bartek  wrote:
>> On Thu, 20 Apr 2017 22:40:56 +0200 Jochen Spieker 
>> wrote:
>>
>>> fc:
>>> >
>>> > Actually -- does anyone monitor this list for this type of stuff?
>>>
>>> You have no idea *how much* spam is blocked by the work of the list
>>> masters. But it's not that anybody monitors all of the almost 300
>>> Debian lists¹ with thousands of posts each day.
>>>
>>> > I see these types of things come through periodically -- and 1
>>> > delete on the front end could prevent a lot of woe.
>>>
>>> Your help is appreciated:
>>> https://wiki.debian.org/Teams/ListMaster/ListArchiveSpam
>>>
>>> Obviously, this only affects the archive after all subscribers already
>>> received the spam message. Moderating all Debian lists is not a job
>>> that anybody wants to do (and it wouldn't even be appreciated).
>>>
>>> > *Even more so* -- it seems like unauthorized users can email this
>>> > list?
>>> >
>>> > Why not just restrict it to people who have subscribed?
>>>
>>> Because this excludes use cases that are deemed valid by the list
>>> masters.
>>
>> Like what?
>>
>> Why not this:  To post or reply to the list, you must be
>> a subscriber; but to read/browse (even search archives, etc.), you
>> do not. This is the way most of the lists I've been involved with have
>> been set up.  Works quite well controlling spurious posting by 'bots.
>> One list I used required annual renewal..
>
> How do you limit posts to subscribers?
>
> Login?
> Subscriber list?

Yes, that's easily done -- check the From: against the subscriber list.

That isn't perfect; a list I administer gets occasional spam from the
forged email address of a subscriber; the list is so incredibly low
volume that it's OK for me to just moderate it.  That wouldn't be the
case here.

> What happens when you need an answer, but you don't have access to a
> functional machine that you can trust?
>
> Also, I think there is a web forum that functions more or less as you 
> describe:
>
> http://forums.debian.net/

Re: Spam on Debian lists

[Joe Pfeiffer]

Patrick Bartek  writes:

> On Thu, 20 Apr 2017 22:40:56 +0200 Jochen Spieker 
> wrote:
>
>> fc:
>> >
>> > Actually -- does anyone monitor this list for this type of stuff?
>> 
>> You have no idea *how much* spam is blocked by the work of the list
>> masters. But it's not that anybody monitors all of the almost 300
>> Debian lists¹ with thousands of posts each day.
>> 
>> > I see these types of things come through periodically -- and 1
>> > delete on the front end could prevent a lot of woe.
>> 
>> Your help is appreciated:
>> https://wiki.debian.org/Teams/ListMaster/ListArchiveSpam
>> 
>> Obviously, this only affects the archive after all subscribers already
>> received the spam message. Moderating all Debian lists is not a job
>> that anybody wants to do (and it wouldn't even be appreciated).
>> 
>> > *Even more so* -- it seems like unauthorized users can email this
>> > list?
>> > 
>> > Why not just restrict it to people who have subscribed?
>> 
>> Because this excludes use cases that are deemed valid by the list
>> masters.
>  
> Like what?
>
> Why not this:  To post or reply to the list, you must be
> a subscriber; but to read/browse (even search archives, etc.), you
> do not. This is the way most of the lists I've been involved with have
> been set up.  Works quite well controlling spurious posting by 'bots.
> One list I used required annual renewal..

That would preclude those of us who still read it on usenet.

debugging TLS alert

[Joe Pfeiffer]

I'm trying to use exim4 to send email to another site.  My host
connects, negotiates a TLS connection and sends what seems to be a
reasonable amount of application data.

I then get an encrypted alert from the other host, the connection shuts
down, and the email doesn't get delivered.  I don't get any errors in my
exim4 log.

To the best of my knowledge, the other site is the only one I try to
send email to that is having the problem, and mine is the only host
having trouble sending to that site.

So...  any words of wisdom on how to debug this?  I've trying turning
all logging on in exim4, and didn't get anything helpful.  Is there a
way to turn logging on in gnutls?

Thanks,

Re: aptitude is dangerous - any replacement?

[Joe Pfeiffer]

Vincent Lefevre  writes:

> On 2017-03-21 21:39:40 +0100, Sven Joachim wrote:
>> On 2017-03-21 21:19 +0100, Vincent Lefevre wrote:
>> > aptitude ignores the apt preferences.
>> 
>> Huh?  At least on my systems, it obeys them.
>
> Perhaps with your configuration. And this is probably also true when
> the full resolver is not involved. But this is not the general case.
>
> By default, the apt preferences are such that unstable is preferred
> over experimental. But as explained in
>
>   https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=795228
>
> this is not the case with aptitude's resolver:
>
> | With the SolutionCost of "removals", aptitude doesn't take into account
> | installing by priorities or non-default releases, it just tries to
> | minimise the removals, so seing that it could solve the problem by
> | upgrading to 2.7-1~exp1, it just did that.

That appears (from my experience and reading for a few minutes, at any
rate) to only come into play when there are conflicts to be resolved.
In the normal case my experience is that the pinnings I've selected
appear to be honored.

All the same, thanks for alerting me to the SolutionCost setting.  I've
also got it set to "removals" which, after reading a bit, doesn't look
like a good idea.

Re: aptitude is dangerous - any replacement?

[Joe Pfeiffer]

Vincent Lefevre  writes:

> I've just noticed that aptitude upgraded packages from unstable to
> experimental versions (just with 'U' from the UI) without any warning!!!
> Again.
>
> Is there any replacement? Or a way to make aptitude ignore
> experimental packages?
>
> Note: I still want to keep experimental in my sources.list for the
> cases where I *explicitly* request experimental packages.

See the apt_preferences man page for information on prioritizing
distributions.  My own preferences file contains

Package: *
Pin: release a=testing
Pin-Priority: 700

Package: *
Pin: release a=stable
Pin-Priority: 650

Package: *
Pin: release a=unstable
Pin-Priority: 600

Package: *
Pin: release a=experimental
Pin-Priority: 550

So I see packages from all of these distributions, but when
automatically selecting versions to install it prefers testing, second
choice stable, third unstable and last experimental.

Re: A minimal relational database in Debian

[Joe Pfeiffer]

Curt  writes:

> On 2017-03-03, GiaThnYgeia  wrote:
>
>> I will top-post as it is meaningless to comment specifically.
>> I do have 5 and have not used base since I don't know when, on this
>> installation it was firsts.  I am made fun of having large fonts as I
>> refuse to wear glasses and my eyes get irritated trying to read small
>
> Is it because of the well-known optician conspiracy (and its attempt to
> get us to see the world rose-colored) or is it purely out of vanity that
> you refuse to make a spectacles out of yourself?
>
>> print.  I adjusted the menus, the cells, and everything else I can find
>> straight out of the LibreB menu for fonts.  And that 130% was
>> ridiculously exceeded on purpose to HUMONGOUS.
>
>
>> So, I am clueless to what you are saying!  Maybe it is some desktop
>> environment that is blocking the adjustment of fonts?
>
> I also hit (and mentioned in a previous post) the 130% limit. I got
> there by going to TOOLS/OPTIONS/VIEW/SCALING. What effect did your font
> manipulation have on the HELP files (because that's what we're talking
> about here, actually).

Hmmm, I just tried it (my version of LibreOffice is 1:5.2.5-1) and was
able to scale well past that; I took it up to 200%.  Just a warning for
others, the position of the spin button appears to be fixed; at 200%
scaling the label covered almost all of it, and I was just barely able
to click the "down" button to recover my usual 100% scale.

Re: How to fix I/O errors?

[Joe Pfeiffer]

David Christensen  writes:

> On 02/04/17 07:18, Ric Moore wrote:
>> I'm looking at a Seagate 750 gig drive that went south on me with a pile
>> of errors. Good luck getting Seagate to give a good gosh darn. In the
>> past I have had mixed results replacing the drive motherboard. I saved
>> two out of three. I doubt I will buy anything Seagate makes in the
>> future.
>
> Everything electrical and mechanical fails.  It's just a question of
> when, followed by whether or not you're prepared.
>
>
> I've found (and heard) that the worst thing I can do to a HDD is put
> it on the shelf and let it rot.  I've had more than a few that failed
> shortly after being put into a computer.

I hadn't heard this...  I've got a drive I've been keeping as a cold
spare.  Am I better off (in the sense of "is it more likely to actually
be useable when I need it") installing it and adding it to one of my
RAID1 arrays?  Can you point me to an article about it?

python time module

[Joe Pfeiffer]

I'm starting to try to learn pythonOCC (a wrapper on the OpenCascade 3d
modeling library) as an excuse to learn python itself, and am running
into some problems right off the bat.

(1) pythonOCC wants to be installed using a package manager called
conda (see http://www.pythonocc.org/download/ and
https://www.continuum.io/).  Is there a reasonably standard way to
integrate conda with apt (in the same sense that alien will let me
install an RPM), or do I just sort of have to expect to use a
different system for it?  Yes, I can download from source instead,
and may end up taking that approach, but would rather not if I can
avoid it.

(2) More seriously, I'm getting the following error and traceback trying
to run my very first script:

nowball:13$ python box1.py 
Traceback (most recent call last):
  File "box1.py", line 1, in 
from OCC.Display.SimpleGui import init_display
  File 
"/home/joseph/python/lib/python2.7/site-packages/OCC/Display/SimpleGui.py", 
line 20, in 
import logging
  File "/usr/lib/python2.7/logging/__init__.py", line 26, in 
import sys, os, time, cStringIO, traceback, warnings, weakref, collections
ImportError: No module named time

>From everything I'm finding, it looks like both the logging and time
modules should be installed by default with python; what do I need to
install that is missing at present to get the time module?

Re: Why I can not install software on debian easily?

[Joe Pfeiffer]

john cusey  writes:

> Why I can not download a .deb file, click it and it installs?
>
> I tried to install opera on Debian. 

Did you look at https://wiki.debian.org/Opera ?

Re: Python Alternatives?

[Joe Pfeiffer]

Ben Finney  writes:

> Michael Milliman  writes:
>
>> I currently have both Python 2.7 and Python 3.4 installed on my debain
>> 8.5 (jessie) system.  The default Python interpreter on the system is
>> Python 2.7 (as linked by /usr/bin/python).
>
> The policy for Python in Debian requires that “/usr/bin/python’ is the
> default Python 2 interpreter, and ‘/usr/bin/python3’ is the default
> Python 3 interpreter.
>
> There is no “default Python interpreter” in Debian. Python 2 and Python
> 3 are incompatible run-time systems.

snowball:518$ ls -l /usr/bin/python
lrwxrwxrwx 1 root root 9 Jun  3  2016 /usr/bin/python -> python2.7*

python2.7 is the default python on my system for any reasonable value of
"default".  I didn't set that link, the installation process did.

>> I would prefer [the ‘/usr/bin/python’ interpreter] to be Python 3.4.
>
> That would violate Debian Python policy. You are free to do it on your
> own system, but it will likely break many Python packages on Debian, and
> you get to keep all the pieces :-)

Re: parted is ALMOST suitable

[Joe Pfeiffer]

Brian  writes:

> On Wed 09 Nov 2016 at 11:27:11 +0100, to...@tuxteam.de wrote:
>
>> On Wed, Nov 09, 2016 at 10:12:13AM +, Brian wrote:
>> > 
>> > That gives "-bash: /dev/sda2: Permission denied" for me with a fixed
>> > disk. It's the same for a removable disk. The system came like that.
>> 
>> Hopefully. But that's not because bash checks that (as parted is).
>> It's because the permissions on the device file are set right!
>
> udev doesn't come into the picture for removable disks? It did on
> pre-Jessie.

What is the relevance of udev here?  Yes, udev sets the permissions, but
the issue is whether they're right not who sets them.

Re: parted is ALMOST suitable

[Joe Pfeiffer]

Felipe Salvador  writes:

> On Mon, Nov 07, 2016 at 06:37:53AM -0600, Richard Owlett wrote:
>> On 11/7/2016 6:20 AM, Felipe Salvador wrote:
>> > On Mon, Nov 07, 2016 at 06:11:50AM -0600, Richard Owlett wrote:
>> > > *HOWEVER* parted requires root privileges. That is not acceptable.
>> > > Suggestions?
>> > > TIA
>> > 
>> > lsblk -fr ?
>> > 
>> 
>> Debian is perverse ;{
>> man page suggested good things.
>> However when run as other than root, there is a column heading "FSTYPE".
>
>> It is blank for all partitions.
>> They are present when run as root.
>> Thanks for trying.
>
> I don't see this behaviour
>
> ~$ lsblk -fr
> NAME FSTYPE LABEL UUID MOUNTPOINT
> sda
> sda1 ext2  ... /boot
> sda2 ext4  ... /
> sda3 ext2  ... /tmp
> etc etc etc
>
> or
>
> file -s /dev/sda{1..5} | awk '{print $5}'

I was just about to post a very similar followup when I discovered a
gaping security hole (really, about as big as it gets) on my machine:

snowball:404$ ls -l /dev/sda2
brw-rw-rw- 1 root root 8, 2 Nov  7 07:54 /dev/sda2

You might want to check your permissions as well

Re: w3m stack smashing?

[Joe Pfeiffer]

Reco <recovery...@gmail.com> writes:


>   Hi.
>
> On Sat, 05 Nov 2016 19:22:02 -0600
> Joe Pfeiffer <pfeif...@cs.nmsu.edu> wrote:
>
>> Recently, I've been getting email messages that look like this:
>> 
>> *** stack smashing detected ***: /usr/bin/w3m terminated
>
> This is recently fixed bug #842623 in w3m.
>
>
>> This is with w3m 0.5.3-31, which appears to be the most current version
>> in Debian testing.
>
> Try 0.5.3-32 from the unstable.

Thanks -- upgrading now.

w3m stack smashing?

[Joe Pfeiffer]

Recently, I've been getting email messages that look like this:

*** stack smashing detected ***: /usr/bin/w3m terminated
=== Backtrace: =
/lib/x86_64-linux-gnu/libc.so.6(+0x70bcb)[0x7f03ee523bcb]
/lib/x86_64-linux-gnu/libc.so.6(__fortify_fail+0x37)[0x7f03ee5ac0e7]
/lib/x86_64-linux-gnu/libc.so.6(__fortify_fail+0x0)[0x7f03ee5ac0b0]
/usr/bin/w3m(+0x54574)[0x560e882b2574]
/usr/bin/w3m(+0x55974)[0x560e882b3974]
/usr/bin/w3m(+0x3dd18)[0x560e8829bd18]
/usr/bin/w3m(+0x3ec68)[0x560e8829cc68]
/usr/bin/w3m(+0x3f076)[0x560e8829d076]
/usr/bin/w3m(+0x42416)[0x560e882a0416]
/usr/bin/w3m(+0x20e26)[0x560e8827ee26]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf1)[0x7f03ee4d32b1]
/usr/bin/w3m(+0x2192a)[0x560e8827f92a]
=== Memory map: 
560e8825e000-560e8830f000 r-xp  09:01 5511742
/usr/bin/w3m
(lots more core dump lines deleted)

Am I seeing messages that are being sent from a buggy sender, or is my
local /usr/bin/w3m blowing up and spewing the detritus into the mail
message?

This is with w3m 0.5.3-31, which appears to be the most current version
in Debian testing.

Re: kernel header for 4.6.0-1-686

[Joe Pfeiffer]

Harry Putnam <rea...@newsguy.com> writes:

> Joe Pfeiffer <pfeif...@cs.nmsu.edu> writes:
>>
>> Any particular reason you need that particular version?  Could you
>> upgrade your virtualbox VM to a different kernel and use the headers
>> for that kernel (or if I'm misremembering which kernel requires the
>> headers, upgrade the host machine kernel)?
>
> I'm not sure about what you say there.  I'm not particularly
> knowledgable about this but far as I can tell The guest addtions
> require the guest OS's kernel headers to compile certain
> modules... without them... no guest additions.

Just tried it, so I could be sure I was giving good advice...

I'm suggesting you go into your package manager in the guest machine (I
use aptitude, but that shouldn't matter) and do an upgrade to the
current versions of everything (this will include the current kernel).
Once you've done that, you should be able to install
virtualbox-guest-utils (also inside the package manager); this should
pull in everything needed to run the guest additions including the
headers for the now-current kernel.

> I already tried to upgrade the kernel on the guest OS to one with
> headers available with apt-get but got into some trouble ending
> with an non-bootable mess...

You don't want to do it like that -- simply upgrading everything should
upgrade to the current version.

> I'd rather leave the kernel building etc to someone who better knows
> what they are doing.  It's said to be easily done(Updating a
> kernel) but my experience is a little different.  Or I'm just a little
> dimmer than the average bear.

No kernel building is needed for this.

Re: kernel header for 4.6.0-1-686

[Joe Pfeiffer]

Harry Putnam  writes:

> Juanjo Benages  writes:
>
>> El 25/10/16 a las 19:48, Harry Putnam escribió:
>>> Where can I get the kernel headers for my kernel 4.6.0-1-686?
>>>
>>> apt-get does not show that version.
>>>
>>> Googling for awhile here and not finding it either
>>>
>>> Trying to install vbox guest additions and it needs those headers.
>>>
>>>
>>>
>> Have you tried on snapshot.debian.org? It should be there.
>
> Using the binary tool search I finally found linux-headers-4.6.0-1-686
> But when I click on it, it listes available versions all of which
> start at 4.6.1 or better.
>
> I guess that means its not there eh?
>
> Any other suggestions?

Any particular reason you need that particular version?  Could you
upgrade your virtualbox VM to a different kernel and use the headers
for that kernel (or if I'm misremembering which kernel requires the
headers, upgrade the host machine kernel)?

Re: url redirected in chrome/chromium, but working fine, according to ping/traceroute, lynx, w3m, iceweasel.

[Joe Pfeiffer]

Richard Hector  writes:
>
> It appears that Montenegro only came into existence (most recently) in
> 2006 - it was part of Yugoslavia, then 'Serbia and Montenegro'. So all
> the 'good' codes were presumably taken.

I'd imagine .me would, like .tv (Tuvalo) be one that a small country
could use to bolster their income quite a bit...

> I'm looking here:
> http://www.domainsherpa.com/country-code-top-level-domains/
>
> But we're getting a bit off-topic :-)
>
> Richard

Re: A psgmlx that plays nice with emacs24?

[Joe Pfeiffer]

Johann Spies  writes:

> On 27 September 2016 at 23:34, Tony Baldwin 
> wrote:
>
> 
> 
> 
> Emacs?! People still use the crusty old thing?
> Perhaps he consider dumping that monstrosity and joining the rest
> of us in the 21st Century, and upgrade to a modern OS with a
> proper editor, such as Debian 8/Jessie with Vim!"
> 
> 
>
> Why so narrow-minded to choose only one of the two? I use both: Emacs
> for programming, org-mode and LaTeX and vim for editing config-files
> or very large text files.
>
> Sometimes I even use vim to edit ~/.emacs :) 

I suspect the post was intended humorously.

Re: Network issue........

[Joe Pfeiffer]

Charlie  writes:

> Hello Debian Users,
>
> I have a network issue that I find perplexing:
>
> When I do: # netstat -r -n
>
> or 
>
> # route
>
> Kernel IP routing table
> Destination Gateway  Genmask Flags MSS Window irtt Iface
> 0.0.0.0   10.80.2.85  0.0.0.0 UG 00  0 eth0
> 10.80.2.84  0.0.0.0  255.255.255.252 U  00  0 eth0
>
> But when I check it through my windows box it comes up as it should,
> according to my ISP, with the gateway being 10.80.2.86
>
> What is happening? Is it allowed to be that slack, one or the other?

Do you know whether your gateway is doing any sort of network address
translation?  It seems odd to me that you're getting an address range
that matches the external address of your modem, and that your external
address is a 10.x.x.x (since those are all non-routable private IPs).
I'd expect either the former if no translation is being done, or the
latter if your external address weren't private.  But seeing both at the
same time surprises me.

On my home system, for instance, my comcast cable modem is at 10.1.10.1
internally, but 173.163.240.62 externally.

I have seen DHCP assign different addresses to Windows than to Linux,
but in your case the gateway box should be grabbing its address for
itself, and giving your computer the other available address.

Re: laptop power management

[Joe Pfeiffer]

Nicolas George <geo...@nsup.org> writes:

> Le quintidi 25 fructidor, an CCXXIV, Joe Pfeiffer a écrit :
>> I'm using an old 32 bit laptop (Samsung N120) running Debian testing; up
>> until recently I've been able to configure it so when I close the lid it
>> turns off the screen, but leaves the laptop running.  With a recent
>> update (possibly this morning, but I couldn't swear to it) I'm not able
>> to do this any more -- when I close the lid, the laptop suspends.
>
> Systemd has taken upon itself to handle the power buttons and lid sensors.
> It has been quite some time now, but I suppose it might be the cause of your
> issue. See /etc/systemd/logind.conf and the corresponding man page.

That was it!  Thank you,

laptop power management

[Joe Pfeiffer]

I'm using an old 32 bit laptop (Samsung N120) running Debian testing; up
until recently I've been able to configure it so when I close the lid it
turns off the screen, but leaves the laptop running.  With a recent
update (possibly this morning, but I couldn't swear to it) I'm not able
to do this any more -- when I close the lid, the laptop suspends.

I'm using xfce, with power management through xfce4-power-management.
At this point, I've got it configured so that the power, sleep, and
hibernate buttons are all set to "do nothing" and the laptop lid is set
to "switch off display" for both battery and AC power.  The power
manager seems to be active, as when I turn on "handle brightness keys" I
can adjust brightness, while I if I turn that off I can not.

The current version of xfce-power-management is 1.4.4-4.  I've tried
both going back to 1.4.1-1 and forward to 1.6.0-1 with no difference.

To the best of my knowledge I have no screensavers or screen lockers
installed.

Any thoughts where I should look?

Re: mount MTP phone

[Joe Pfeiffer]

Ivan Petrov <ip...@yandex.ru> writes:

> 06.09.2016 22:19, Joe Pfeiffer пишет:
>> Curt Howland <howl...@priss.com> writes:
>>
>>> I've never been able to mount any MTP phone to Linux. Different
>>> phones, different Android versions, different Linux installations,
>>> nada.
>>
>> I've only been able to do it using jmtpfs (I don't have it set up to
>> automount using that).
>>
>
> Why I have this messages, when I'm using it?
>
> $ jmtpfs /media/exfat
> Device 0 (VID=19d2 and PID=ffce) is UNKNOWN.
> Please report this VID/PID and the device model to the libmtp
> development team
> Android device detected, assigning default bug flags

First question:  when you plugged the phone in, did you turn on USB for
file transfers?  Without that I get almost exactly the same behavior you
see:

snowball:501$ jmtpfs /home/joseph/mnt
Device 0 (VID=18d1 and PID=4ee1) is a Google Inc (for LG Electronics/Samsung) 
Nexus 4/5/7/10 (MTP).
Android device detected, assigning default bug flags
snowball:502$ ls mnt
/bin/ls: cannot access 'mnt': Input/output error

If I click the notification on my phone and then select "File transfers"
I get

snowball:507$ jmtpfs /home/joseph/mnt
Device 0 (VID=18d1 and PID=4ee1) is a Google Inc (for LG Electronics/Samsung) 
Nexus 4/5/7/10 (MTP).
Android device detected, assigning default bug flags
snowball:508$ ls mnt
Internal storage/

which is of course what I want.

I'm not sure whether the actual error message you're getting

> Device 0 (VID=19d2 and PID=ffce) is UNKNOWN.
> Please report this VID/PID and the device model to the libmtp
> development team

is relevant; that's telling you that the vendor ID and product ID your
phone is using to identify itself aren't in libmtp's database.  If I
search for them at http://www.the-sz.com/ it tells me the vendor is ZTE
(which looking at the table appears to be a chipset manufacturer, not
the manufacturer of your phone) but it can't find your device.  On the
other hand, jmtpfs is seeing that your device is an Android device, so
it seems to have gotten around that somehow.

Re: mount MTP phone

[Joe Pfeiffer]

Tony Baldwin  writes:

> On 09/04/2016 09:22 PM, Carl Fink wrote:
>> On 09/04/2016 09:13 PM, Tony Baldwin wrote:
>>> I'm at a loss, friends:
>>> I have a phone (Motorola Droid Turbo), which functions as an MTP
>>> device, and auto-mounts on Win7, but not on Debian 8.
>>> AFAIK, I have all the requisite MTP packages installed .
>>> When I plug it in, lsusb doesn't show it at all, however.
>>> I don't know what else to do to try and diagnose or resolve the matter.
>>> Anyone?
>>> I'm running Jessie on AMD64
>> Have you looked at /var/log/syslog for the relevant period?
>
> Sorry Carl, you're getting this twice now (it's been a while since I
> was active on lists, I've forgotten how to behave), but
>
> I confess, I had not, until now.
>
> Seems I get a bunch of this:
>
> Sep  4 21:23:47 deathstar kernel: [26533.020450] hub 10-0:1.0: unable
> to enumerate USB device on port 1

That's your problem...  you should be seeing something more like

Sep  6 09:13:02 snowball kernel: [1030652.496395] usb 5-2.1.4: new high-speed 
USB device number 13 using xhci_hcd
Sep  6 09:13:02 snowball kernel: [1030652.596926] usb 5-2.1.4: New USB device 
found, idVendor=18d1, idProduct=4ee1
Sep  6 09:13:02 snowball kernel: [1030652.596930] usb 5-2.1.4: New USB device 
strings: Mfr=1, Product=2, SerialNumber=3
Sep  6 09:13:02 snowball kernel: [1030652.596931] usb 5-2.1.4: Product: Nexus 5
Sep  6 09:13:02 snowball kernel: [1030652.596933] usb 5-2.1.4: Manufacturer: LGE
Sep  6 09:13:02 snowball kernel: [1030652.596934] usb 5-2.1.4: SerialNumber: 
039d5ef3437cc5b0

Of course, the strings I'm showing are for a different phone!   Most
commonly when I've seen a device fail to enumerate, the problem has been
the cable (I know, you've already said you've tried multiple cables).
But you need to be enumerating before absolutely anything else will
happen and before anything else is relevant...  you aren't having
trouble with other USB devices?  No chance there's a USB enable/disable
on the phone?  (on mine, things enumerate properly like I showed, but
nothing else happens until I select "USB for file transfers" in a
dropdown on the phone).
-- 
"Erwin, have you seen the cat?" -- Mrs. Shrödinger

Re: mount MTP phone

[Joe Pfeiffer]

Curt Howland  writes:

> I've never been able to mount any MTP phone to Linux. Different 
> phones, different Android versions, different Linux installations, 
> nada.

I've only been able to do it using jmtpfs (I don't have it set up to
automount using that).
-- 
"Erwin, have you seen the cat?" -- Mrs. Shrödinger

Re: Resolved: mount MTP phone

[Joe Pfeiffer]

Anthony Baldwin  writes:

> Sorry for top-posting.
> Iḿ no t even sure what it was that DID solve this but the phone is now
> mounting at mtp://[usb:005,012]/
> Must have something to do with installing the jmtpfs pkg,
> because that's the only thing I can think of that I did yesterday that
> could be making a difference today.
> I did briefly put the phone on ptp/camera mode, and it let me download
> the photos on it, but now it's letting me completely access the
> storage for whatever (move photos, music, etc. phone<->computer)

Magic fixes like this *really* sound like a bad cable.

> On 09/05/2016 12:47 PM, Tony Baldwin wrote:
>> On 09/05/2016 11:09 AM, Nicolas George wrote:
>>> Le decadi 20 fructidor, an CCXXIV, Curt Howland a écrit :
 I've never been able to mount any MTP phone to Linux. Different
 phones, different Android versions, different Linux installations,
 nada.
>>>
>>> I have. With FUSE and gvfs, something like (the computer where I did
>>> it and
>>> kept notes is current shut down):
>>>
>>> gvfs-mount 'mtp://[usb:001,002]'
>>>
>>> You need to have gvfs-fuse already running, then it appears under
>>> /run/users/$UID/gvfs/.
>>>
>>> libmtp has a few tools to manipulate MTP devices without mounting
>>> them, but
>>> they have 0 documentation. Furthermore, MTP it awfully slow at startup
>>> for
>>> some tasks, so individual commands without a process to keep the library
>>> open will give awful results. I have started working on a tool with a
>>> script-friendly FTP-like interface, but not got very far yet due to other
>>> more pressing projects.
>>>
>>> Regards,
>>>
>>
>> Something has changed since yesterday (IN which I installeed jmtpfs).
>> Now when I do ls usb, I see this:
>> Bus 005 Device 012: ID 22b8:2ea4 Motorola PCS
>> I'm also seeing possibly relevant (I don't know) stuff in dmesg, such as
>> this:
>> [quote][75390.506984] hub 5-0:1.0: port 5 disabled by hub (EMI?),
>> re-enabling...
>> [75390.507000] usb 5-5: USB disconnect, device number 8
>> [75406.588441] usb 5-5: new high-speed USB device number 9 using ehci_hcd
>> [75406.722345] usb 5-5: New USB device found, idVendor=22b8, idProduct=2ea6
>> [75406.722360] usb 5-5: New USB device strings: Mfr=1, Product=2,
>> SerialNumber=3
>> [75406.722368] usb 5-5: Product: XT1254
>> [75406.722374] usb 5-5: Manufacturer: motorola
>> [75406.722379] usb 5-5: SerialNumber: ZY2234TGKZ
>> [81479.602219] usb 5-5: USB disconnect, device number 9
>> [81479.876441] usb 5-5: new high-speed USB device number 10 using ehci_hcd
>> [81480.010291] usb 5-5: New USB device found, idVendor=22b8, idProduct=2ea6
>> [81480.010305] usb 5-5: New USB device strings: Mfr=1, Product=2,
>> SerialNumber=3
>> [81480.010314] usb 5-5: Product: XT1254
>> [81480.010319] usb 5-5: Manufacturer: motorola
>> [81480.010325] usb 5-5: SerialNumber: ZY2234TGKZ
>> [81483.946880] usb 5-5: USB disconnect, device number 10
>> [81484.220444] usb 5-5: new high-speed USB device number 11 using ehci_hcd
>> [81484.354763] usb 5-5: New USB device found, idVendor=22b8, idProduct=2ea6
>> [81484.354778] usb 5-5: New USB device strings: Mfr=1, Product=2,
>> SerialNumber=3
>> [81484.354786] usb 5-5: Product: XT1254
>> [81484.354792] usb 5-5: Manufacturer: motorola
>> [81484.354797] usb 5-5: SerialNumber: ZY2234TGKZ
>> [81484.399977] usb 5-5: USB disconnect, device number 11
>> [81508.968204] usb 5-5: new high-speed USB device number 12 using ehci_hcd
>> [81509.102613] usb 5-5: New USB device found, idVendor=22b8, idProduct=2ea4
>> [81509.102627] usb 5-5: New USB device strings: Mfr=1, Product=2,
>> SerialNumber=3
>> [81509.102635] usb 5-5: Product: XT1254
>> [81509.102641] usb 5-5: Manufacturer: motorola
>> [81509.102647] usb 5-5: SerialNumber: ZY2234TGKZ
>> [/quote]
>>
>>
>> is your work in a github repo, sourceforge project, or similar, where
>> others could have a look, and, perhaps, contribute, Nicolas?
>>

-- 
"Erwin, have you seen the cat?" -- Mrs. Shrödinger

Re: Advice on downloading software please

[Joe Pfeiffer]

Brian  writes:

> On Sat 27 Aug 2016 at 09:15:50 -0500, limpia wrote:
>
>> On 2016-08-27 08:55, Steve Greig wrote:
>> >I would like to download a programme (opencpn) onto my laptop which is
>> >running debian. It is so long since I have done this I can not
>> >remember how to start. Also I am not sure which version of debian I
>> >have which seems to be relevant according to the website. Instructions
>> >are given at x [1].
>  ^
>  |
>  |
> What happened here-| ? The link has disappeared, Censorship at work?

limpia appears to have edited the URL out of his reply, since he is
recommending not using it.  It is still there in the original post, as
one would expect.

>> >Any advice would be very  much appreciated. Steve
>> >
>> >
>> >Links:
>> >--
>> 
>> You can use 'uname -a ' to see what version of Debian
>> you have.
>
> Really? Positive?

More usefully, no, uname -a won't give the Debian version.
/etc/debian_version should contain the correct version, if it hasn't
been modified on your site.  As the documentation for base-files says,
however, the contents of your /etc/apt/sources.list and
/etc/apt/sources.list.d/* are a more reliable way to determine it.

>>  However, you also might want to read this:
>> https://wiki.debian.org/DontBreakDebian
>> 
>>  It is not a good idea to be using ubunto ppa's like
>> the instructions say on the link you posted.
>
> If it works, it works.

And if it doesn't, you're screwed.

>>  As far as installing Debian packages goes, there is
>> plenty of documentation on that:
>> https://wiki.debian.org/DebianPackageManagement
>> 
>> If you really need this program, and the source code
>> is available, you can follow the instructions here:
>> https://wiki.debian.org/CreatePackageFromPPA
>
> If it works, it works.

And if it doesn't, you're screwed.

>>  But it still something, personally I would not
>> recommend.
>
> Even if it works?

It is a really good idea to check the contents of the PPA, and make sure
it doesn't bring in nonstandard versions of system libraries.  I don't
remember ever having been burned by a PPA specifically, but I did once
have to spend a day cleaning stuff from debian-multimedia (long ago
enough that it wasn't deb-multimedia yet) off my system.

I frankly have to worry about somebody who is asking for help installing
from a PPA then being able to clean up if it goes sour.

Re: Any idea when CVE-2016-5696 is going to get fixed?

[Joe Pfeiffer]

"John T. Haggerty"  writes:

> On Fri, Aug 26, 2016 at 9:11 PM, Perry E. Metzger 
> wrote:
>
>On Fri, 26 Aug 2016 21:06:15 +0200 Frederic Marchal
> wrote:
>
> > The download must be long
> > enough (more than one minute) for the attacker to discover the
> set
> > of parameters that will make the attack successful.
> 
> You've forgotten how the modern web works. People have http:
> connections live for very long periods of time, with dynamic
> content
> flittering back and forth over the channel. It isn't like 1996 any
> more where someone downloaded some static HTML and closed the TCP
> connection until the next page was downloaded when they clicked
> again. It hasn't been like that in a very long time.
>
> So you are referring to the "netstat" output from the system itself?
> So physically redraw the page they are on even if they haven't
> refreshed the page?

I'm not sure how netstat is relevant here  but think of protocols
like AJAX where, indeed, content on a web page can be updated without
any user activity.  Do you have a facebook account?  I frequently have a
browser open to it for days at a time while it updates my feed (in
fairness, that's https: not http:, but the point about long-lived
connections remains valid).

Re: epub files and debian?

[Joe Pfeiffer]

Ben Finney  writes:

> Dan Ritter  writes:
>
>> evince cannot handle epub at all. Your statement about "many epub
>> files don't open correctly in evince" should read "no epub files are
>> opened by evince".
>
> Bah, you're right. This is a long-standing request (since 2008!) in the
> Gnome BTS .
>
> The sticking point appears to be that Evince fundamentally assumes the
> document naturally has pages, whereas EPub presents a continuous
> document. No-one has suggested a good resolution to that.

This surprises me.  I'd expect the sticking point to be that the actual
content in an EPUB is HTML, and evince doesn't do HTML.

Re: Terminal

[Joe Pfeiffer]

Темир Урокбаев  writes:

> Hello. Tell me, is there a
> comprehensive list of terminal
> commands, and where to find it
> or download.

Others have given good information; I'll just add that there can't be a
comprehensive list:  in addition to the built-in shell commands and the
common utilities, any package you install is likely to add more
available commands.

Re: ThinkPad fan

[Joe Pfeiffer]

Francesco Montanari  writes:

> Hi,
>
> I recently installed Jessie on a Lenovo ThinkPad T420. The fan usage
> looks reasonable. However, high temperatures (96 C) are reached when
> CPUs are running intensively for more than one minute or so. The fan
> speed at those temperatures is about 4500 rpm.
>
> Do you think it is ok, or do you suggest to force lower temperatures,
> e.g., with thinkfan [1]?
>
> Thanks,
> Francesco
>
> [1] https://packages.debian.org/jessie/thinkfan

That's unlikely to hurt anything, but from the package description it
sounds like will only slow a fan down if it doesn't need to run fast,
which isn't your problem.

>From my experience on overheating laptops, the first thing I'd do would
be to clean everything.  Laptops have much more constrained air paths
than desktops, and it's easy for air inlets (this is where I've seen
problems most frequently) or heat sinks to get clogged so you don't get
the flow you need.

Re: Big dummy at work again

[Joe Pfeiffer]

cbannis...@slingshot.co.nz writes:

> On Sat, Jun 11, 2016 at 10:04:27PM -0500, David Wright wrote:
>> On Sat 11 Jun 2016 at 20:49:27 (-0400), Gene Heskett wrote:
>> > On Saturday 11 June 2016 17:35:11 Lisi Reisz wrote:
>> 
>> > > No, Gene.  All created because you didn't trust the package manager. 
>> > > Not that IMHO Synaptic is that trustworthy. ;-)  I have just looked in
>> > > my /usr/bin.  I have firefox-esr and firefox-real.  So I have
>> > > experimented.  Firefox-esr is the one aptitude installed, and is
>> > > Firefox 45.2.0 and Firefox is 47.0, and is the one I installed.  I
>> > > have both in my menu, clearly labelled Firefox-esr and Firefox.  I let
>> > > aptitude do its thing.  It has left life simple.
>> > >
>> > Maybe, but that ncurses face on aptitube is a total turn-off, and will be 
>> > until aptitude figures out how to make ncurses redraw the whole screen 
>> > instead of leave a kilobyte of text covering 20% of the screen real 
>> > estate.  And it been that disaster since I first saw it 18 years ago.
>> 
>> As usual, your rant is in need of some explanation as to exactly
>> what your problem is.
>
> It seems obvious to me, he's talking about screen corruption.
> Gene, Is this when you close it down?
> I'd give apt-get a try, I prefer it to aptitude and esp. prefer it to 
> synaptic.

He might be, except I've never seen screen corruption when running
aptitude.

I like aptitude when I'm searching for packages or doing updates; I like
apt-get when I'm just installing a package whose name I know.

Re: pam_smbpass.so

[Joe Pfeiffer]

Sven Hartge <s...@svenhartge.de> writes:

> Joe Pfeiffer <pfeif...@cs.nmsu.edu> wrote:
>
>> I'm seeing a large number of entries in my /var/log/syslog that look
>> like this:
>
>> Feb 16 09:07:31 snowball auth: PAM unable to dlopen(pam_smbpass.so):
>> /lib/security/pam_smbpass.so: cannot open shared object file: No
>> such file or directory
>> Feb 16 09:07:31 snowball auth: PAM adding faulty module: pam_smbpass.so
>
>> So...  any idea what's going on here, and more importantly how to fix
>> it?  I also see consistently that this is a harmless message, but it
>> bugs me...
>
> Do a 
>
>   rgrep smbpass /etc/pam.d
>
> and see, if you have an old config file there which references
> pam_smbpass.so via absolute path. This can cause this error.

snowball:606$ rgrep smbpass /etc/pam.d
/etc/pam.d/common-password:password optional
pam_smbpass.so nullok use_authtok use_first_pass
/etc/pam.d/common-auth:auth optionalpam_smbpass.so 
migrate

> Odds are, you can just remove the old config file and purge the package
> you installed, because you don't need it.

Re: pam_smbpass.so

[Joe Pfeiffer]

Christian Seiler <christ...@iwakd.de> writes:

> Hi,
>
> On 02/17/2016 05:11 PM, Joe Pfeiffer wrote:
>> Christian Seiler writes:
>>> [Suggesting journalctl -o verbose to debug this]
>> I'm running a current Debian testing installation, and journal is
>> enabled.
>> 
>> It turns out it's only coming from /usr/lib/dovecot/auth.  What's
>> weird is in /etc/pam.d/, the only files using the module are
>> common-auth and common-password, so I'd expect to see the error coming
>> either every time someone authenticates through anything, or any time
>> someone changes their password, and I'm not seeing either of those
>> cases -- just dovecot.
>
> Just a hunch: do you run dovecot chroot'ed? If so, then it is most
> likely the case that the specific PAM module is not available within
> the chroot and that's why it produces that message.

No, it isn't chrooted -- if it were, I'd expect the other pam modules to
give the same issues (for that matter, I'd expect it to not be able to
find pam.d!).

> If that's not the case: what's the contents of /etc/pam.d/dovecot?
> And /etc/pam.d/common-auth?

/etc/pam.d/dovecot:
#%PAM-1.0

@include common-auth
@include common-account
@include common-session

/etc/pam.d/common-auth:
#
# /etc/pam.d/common-auth - authentication settings common to all services
#
# This file is included from other service-specific PAM config files,
# and should contain a list of the authentication modules that define
# the central authentication scheme for use on the system
# (e.g., /etc/shadow, LDAP, Kerberos, etc.).  The default is to use the
# traditional Unix authentication mechanisms.
#
# As of pam 1.0.1-6, this file is managed by pam-auth-update by default.
# To take advantage of this, it is recommended that you configure any
# local modules either before or after the default block, and use
# pam-auth-update to manage selection of other modules.  See
# pam-auth-update(8) for details.

# here are the per-package modules (the "Primary" block)
authrequiredpam_abl.so 
config=/etc/security/pam_abl.conf
auth[success=1 default=ignore]  pam_unix.so nullok_secure try_first_pass
# here's the fallback if no module succeeds
authrequisite   pam_deny.so
# prime the stack with a positive return value if there isn't one already;
# this avoids us returning an error just because nothing sets a success code
# since the modules above will each just jump around
authrequiredpam_permit.so
# and here are more per-package modules (the "Additional" block)
authoptionalpam_mount.so
authoptionalpam_smbpass.so migrate
authoptionalpam_cap.so
# end of pam-auth-update config

(note the line
authoptionalpam_smbpass.so migrate
is added when libpam-smbpass is installed)

pam_smbpass.so

[Joe Pfeiffer]

I'm seeing a large number of entries in my /var/log/syslog that look
like this:

Feb 16 09:07:31 snowball auth: PAM unable to dlopen(pam_smbpass.so): 
/lib/security/pam_smbpass.so: cannot open shared object file: No such file or 
directory
Feb 16 09:07:31 snowball auth: PAM adding faulty module: pam_smbpass.so

Googling the error mesage, I see lots of very old bug reports (mainly
2008 timeframe, though a few are as recent as 2012 -- still pretty old)
that look like this ubuntu forum thread:

http://ubuntuforums.org/showthread.php?t=770724

but nothing recent.

The fix in almost all those cases seems to have been to install
libpam-smbpass (which makes sense, since that's where pam_smbpass.so
lives).  I have that package installed, so I have
/lib/x86_64-linux-gnu/security/pam_smbpass.so

So... this is, of course, not the directory that's being reported as the
location of the missing file, however PAM is supposed to search
/lib/security/ and /lib/x86_64-linux-gnu/security for modules.  I've
tried putting a symbolic link from the real file to
/lib/security/pam_smbpass.so, which only gives me a raft of different
error messages having to do with failed page mappings.

I don't have any i386 PAM modules (nor libpam0g:i386) installed, and
(checking) have no /lib/i386/security/ directory at all.

/lib/security has very few files in it (only pam_dbus.so
pam_gnome_keyring.so  pam_mount.so  pam_unix2.so), and they're all
x86-64.  All my other PAM modules are (as far as I know) in
/lib/x86_64-linux-gnu/security/, but pam_smbpass.so is the only one
throwing the "no such file" error.

So...  any idea what's going on here, and more importantly how to fix
it?  I also see consistently that this is a harmless message, but it
bugs me...

Re: Meta key for 'emacs -nw'

[Joe Pfeiffer]

Bob Bernstein  writes:

> On Tue, 2 Feb 2016, Lisi Reisz wrote:
>
>> :-)  "There are no dumb questions.  Only dumb answers."
>
> Okay. Here's one -- I was going to post it in gnu.emacs.help, but you
> changed my mind! Emacs running in X honors Alt as its Meta key. But if
> I launch 'emacs -nw' to avoid running in X that understanding (Meta ==
> Alt) evaporates. Perhaps I need to set something specific in my .emacs
> for the minibuffer?
>
> I've got amd64 Jessie running without systemd and icewm for X.

I've always used  for  -- you have to press-and-release, not
hold like for  and , but it works for me in a window, with
-nw, and in console mode.  I'd actually been using it for a couple of
decades before I learned about  for , and have never made the
switch.

Re: RIP and Thank You.

[Joe Pfeiffer]

Martinx - ジェームズ  writes:

> Rest in Peace Ian!
>
> I wanna know exactly what happened with him.
>
> On 30 December 2015 at 20:44, Lisi Reisz  wrote:
>
> http://www.theregister.co.uk/2015/12/30/ian_murdock_debian_founder/

Yes.  The story as reported so far is really bizarre.

Re: grub2 security problem

[Joe Pfeiffer]

Michael Fothergill  writes:

> Dear Folks,
>
> I noticed some articles suggesting that there is a security problem in
> grub2.
>
> E.g.
>
> http://thehackernews.com/2015/12/hack-linux-grub-password.html
>
> ​Is there any substance to this?

Yes, for the microscopic proportion of people who put a password on
their bootloader in the first place, and for whom it is possible for
someone to to access their console to defeat it but not possible for
them to simply take the disk out and steal it.

Re: command not found

[Joe Pfeiffer]

Bob Holtzman  writes:

> Running Wheezy (7.9) on a reinstall after launching Jessie thru a wall.
> Reinstalled all my s/w including msmtp and fetchmail. I brought
> .fetchmailrc over from my backup as well as .msmtprc. Both had been
> working flawlessly on the previous install. Now when I run "fetchmail" I
> get "command not found". Perms on .fetchmailrc are 700. A search on
> various variations of Wheezy fetchmail "command not found" came up with 
> nothing.
>
> Any ideas, pointers or flames appreciated.

I assume when you say you get "command not found" you mean you get
"fetchmail: command not found".  If you mean you get anything else you
can stop reading right now, because everything else in what I'm writing
is assuming that.

Are you *sure* you installed fetchmail on the new system?  I know you
said you did, but that would be the most obvious problem.  You can run

dpkg -l fetchmail

and if it's there, you should get a line like

ii  fetchmail  6.3.26-2   amd64  SSL 
enabled POP3, APOP, IMAP mail gatherer/forwarder

in your output.  Assuming it's there, run

dpkg -L fetchmail | less

and see where the fetchmail program itself is installed.  On my system, I get

/usr/bin/fetchmail

Now, try

echo $PATH

to see if /usr/bin is in your path.  It sort of has to be; I can't
imagine how a system could be useable without it.

Anyway, those are some thoughts...

Re: Slow Display of Graphics in Chrome

[Joe Pfeiffer]

moxalt  writes:

> On Fri, 27 Nov 2015 19:32:25 +, Alan Chandler 
> wrote:
>
>> Just recently, I notice a sudden slowing down of the display of areas of 
>> the screen in Chrome when it fully maximised ( but still with toolbar 
>> etc on display)
>
> If you want a Chrome-like browser, what's wrong with Chromium? Is it lacking 
> in
> some way which necessitates the use of Chrome?
>
>> I am running a dual monitor gnome 3 setup with intel display driver.
>> 
>> Quite frequently, but not predictably so (say once every few minutes) 
>> Chrome, when displaying a web page whilst fully maximized gets very slow 
>> and displays blocks of the screen but then hangs for a few seconds 
>> before continuing with some more and hanging again.
>> 
>> You can stop it doing this by dragging the toolbar down from the fully 
>> maximized position.  I don't think I have seen it when not fully maximized.
>> 
>> Anyone else experiencing the same?
>
> Although I don't use Chrome or Chromium anymore, I experience this with
> *Iceweasel*, of all web browsers. In my experience Chromium has always been a
> lot more responsive and snappier- so I experience and associate exactly
> this problem with the Firefox-based browsers; Abrowser, Iceweasel, IceCat, 
> etc.
>
> Strange.
>
> Does anyone else experience these strange hangs with the Firefoxes?
>
> For me they are almost identical to the problem Alan has with Chrome.

What I find even more interesting is that I've pretty much given up on
Chromium (the only program, let alone browser, which is showing
painfully slow response on my system) and switched back to iceweasel...

Deleting i386 packages

[Joe Pfeiffer]

For historical reasons, my x86-64 architecture computers have a large
number of i386 packages on them that I'd just as soon be rid of.  is
there a good way to simply tell a package manager that I want everything
involving that architecture deleted?  The best answer I've found on my
own has been to use dpkg and grep to find everything with :i386, and
then construct a huge dpkg --purge command to get rid of them all.
Hoping for something a little simpler...

Re: Coder friendly font Anonymous Pro (ttf-anonymous-pro)

[Joe Pfeiffer]

David Wright <deb...@lionunicorn.co.uk> writes:

> Quoting Joe Pfeiffer (pfeif...@cs.nmsu.edu):

>> Looks interesting -- I've been using Terminus for quite a while -- it's
>> another fixed-width programmer-friendly font, Comparing it with
>> Anonymous Pro, it seems a bit narrows and doesn't seem to have as much
 ^^^narrower
>> variation in apparent weight (Anonymous Pro's W is so much darker than
>> the other characters on a line I'm looking at that it looks like it's in
>> Bold!).
>
> So you've installed it? Are you using it in a VC or an xterm?
>
> I'm not sure how you would use it: the package contains four TTF files
> and that's it.

I appear to have installed at some point; I tend to just install the
fonts that come up in the repository without thinking about it.

I tried it in an xterm (more specifically, xfce4-terminal).  I just
went to the preference's editor, saw that it was one of the font
options, and switched to it.  All the text in all my terminal windows
was suddenly in Anonymous Pro.

Re: Coder friendly font Anonymous Pro (ttf-anonymous-pro)

[Joe Pfeiffer]

Cindy-Sue Causey  writes:

> One more then I hear my bird feeders calling. Couple days ago I was
> trying to find a pirate friendly font via an "apt-cache search"
> inquiry. No pirates (that weren't part of a *2GB* package, yar!),
> but stumbled on a font called "Anonymous Pro" that is billed as a
> "fixed width sans serif font designed for coders".
>
> Further description is: "Anonymous Pro (2009) is a family of four
> fixed-width fonts designed  especially with coding in mind. Characters
> that could be mistaken for  one another (O, 0, I, l, 1, etc.) have
> distinct shapes to make them  easier to tell apart in the context of
> source code."
>
> Since I had just like the day before installed "devscripts", it
> sounded like a potential win worth pursuing. It looks very similar to
> Monospace, but my brain still keeps actively noticing that there is
> definitely a user-friendly difference..
>
> Sharing because it might just help someone else who spends a lot of
> time using terminals. As I write that, for some reason it comes to
> mind that it may be standard with large installs. If not, the package
> name again is ttf-anonymous-pro.

Looks interesting -- I've been using Terminus for quite a while -- it's
another fixed-width programmer-friendly font, Comparing it with
Anonymous Pro, it seems a bit narrows and doesn't seem to have as much
variation in apparent weight (Anonymous Pro's W is so much darker than
the other characters on a line I'm looking at that it looks like it's in
Bold!).

Re: Coder friendly font

[Joe Pfeiffer]

rlhar...@oplink.net writes:

> On Mon, September 21, 2015 11:33 am, Cindy-Sue Causey wrote:
>> a font called "Anonymous Pro" that is billed as a "fixed width sans serif
>> font designed for coders".
>> Further description is: "Anonymous Pro (2009) is a family of four
>> fixed-width fonts designed  especially with coding in mind. Characters
>> that
>> could be mistaken for  one another (O, 0, I, l, 1, etc.) have distinct
>> shapes to make them  easier to tell apart in the context of source code."
>
> Regarding disambiguity, Courier is one of the best fonts; likewise Times
> Roman.  The only problem with Times Roman with respect to coding is that
> it is not fixed-width.

And courier is downright ugly.

> Sans-serif is not a desirable attribute, except for certain applications
> such as newspaper headlines.  Studies have shown that serif fonts are more
> readable and less fatiguing, because the serifs of a letter or numeral
> convey much information.

It may be billed as sans serif, but there are a number of characters in
the font that do indeed have serifs.

> But Americans crave novelty, so there always is a market for the
> "different" when it is claimed to be "new" and "improved".

Re: pulseaudio module to mute other sources?

[Joe Pfeiffer]

rlhar...@oplink.net writes:

 On Wed, August 12, 2015 9:46 pm, Joe Pfeiffer wrote:
 Is there a pulseaudio module that will detect when an audio source isn't
 silent, and mute other sources in that cases?

 It does not directly address your need, but I think that
 pulseaudiovolumecontrol (pavucontrol) is almost essential, and really
 should be installed by default with the pulseaudio system.

Yep, I've pretty much always got that window up!

 I'd like to be listening to music, but when a call comes in on
 the radio have it mute any other audio.

 In the realm of broadcasting and commercial sound, the function you are
 describing is of the general category ducking (as in duck your head so
 you do not bump it on the obstruction ahead).  But ducking typically
 reduces the level of the primary audio stream, rather than completely
 muting it.

 You may find an inexpensive hardware solution; check broadcast suppliers
 such as Markertek, bswUSA, and FullCompass.

 Because dead air is anathema to a broadcaster, there are silence
 monitors which automatically switch to an alternate audio source if the
 primary source goes silent.  One of these might serve your needs.

 Utility modules of this sort for audio are widely used and tend to be
 surprisingly inexpensive.

Something like the
http://www.markertek.com/product/st-vp2/rdl-st-vp2-automatic-ducking-module
seems like it might meet my needs... it seems like this ought to be the
sort of thing that pulseaudio can do with software, though.

Re: pulseaudio module to mute other sources?

[Joe Pfeiffer]

rlhar...@oplink.net writes:

 On Wed, August 12, 2015 10:22 pm, rlhar...@oplink.net wrote:
 On Wed, August 12, 2015 9:46 pm, Joe Pfeiffer wrote:
 ...
 Is there a pulseaudio module that will detect when an audio source
 isn't silent, and mute other sources in that cases?
 ...
 In the realm of broadcasting and commercial sound, the function you are
 describing is of the general category ducking...
 ...

 Here is a URL I turned up with a search for audio ducking with pulseaudio:

 http://unix.stackexchange.com/questions/118512/how-to-reduce-the-volume-of-a-background-music-stream-when-a-different-audio-sou

 The URL provides other sources on the subject.

If I understand the ducking module, it doesn't seem to do what I want.
As I understand that one, it'll mute other streams when you start a new
one.  That's valuable, but my use case involves the radio stream always
being present, just silent 90% of the time.

Re: pulseaudio module to mute other sources?

[Joe Pfeiffer]

doug dmcgarr...@optonline.net writes:

 On 08/12/2015 10:46 PM, Joe Pfeiffer wrote:
 I'm reasonably confident there must be a better place to post this than
 this newsgroup, so answers telling me where to look will be just as
 welcome as any that address the question!  I'm a Debian user, so this is
 where I'm trying first

 Is there a pulseaudio module that will detect when an audio source isn't
 silent, and mute other sources in that cases?

 It may be more clear if I describe my particular desire use case.  I
 have a ham radio, whose audio output I can plug into my sound card
 input.  I'd like to be listening to music, but when a call comes in on
 the radio have it mute any other audio.

 Any thoughts?


 I suspect that your ham radio is VHF--2 meters, or possibly 440MHz,
 since you say when a call comes in.

Yes, it's a 2m/70cm HT (a cheap Baofeng, I confess).

 Therefore, the radio certainly has a squelch circuit. As a ham, you
 should be familiar enough with electronics
 to find a voltage in the radio that switches level when the squelch is
 activated or deactivated. You can
 buffer this voltage with a transistor or a digital comparator, and use
 it to control the music via a relay.

 Hint: squelch circuits usually (but not always) are controlled by the AGC.

Well, yes...  the squelch is actually software-controlled in the radio,
but there is a line I could use to detect when it comes off.

I guess I wasn't clear enough about the environment:  I want to route
the audio out from the HT through my computer, and I'm listening to
music on that same computer.  An all-software solution is certainly
possible; the question is whether it exists already.

Re: pulseaudio module to mute other sources?

[Joe Pfeiffer]

Ralph Katz ralph.k...@rcn.com writes:

 On 08/12/2015 10:46 PM, Joe Pfeiffer wrote:
 I'm reasonably confident there must be a better place to post this than
 this newsgroup, so answers telling me where to look will be just as
 welcome as any that address the question!  I'm a Debian user, so this is
 where I'm trying first
 
 Is there a pulseaudio module that will detect when an audio source isn't
 silent, and mute other sources in that cases?
 
 It may be more clear if I describe my particular desire use case.  I
 have a ham radio, whose audio output I can plug into my sound card
 input.  I'd like to be listening to music, but when a call comes in on
 the radio have it mute any other audio.
 
 Any thoughts?

 Joe -- Back in May we discussed pausing vlc with skype.  Your situation
 is not identical, but take a look at  /etc/pulse/default.pa  for ideas.
  Wish I knew more about pulseaudio...

 https://lists.debian.org/debian-user/2015/05/msg00643.html

It looks like that discussion is about whether or not skype should pause
multimedia players when a call comes in -- the usual behavior is it
does, and the poster was asking for a way to keep it from happening.

This is using the cork module, which will stop other streams when a new
stream starts; if I'm reading it correctly, skype creates a new stream
when a call comes.

Hmmm, this may be the start of an answer for me -- start a stream when
the input audio level gets above some threshold, end it after it dips
below...

(note the squelch discussion at
http://www.windytan.com/2013/07/squelch-it-out.html
doesn't start and stop the stream, it just replaces samples in the
stream with 0s, so that's not going to do it...)

pulseaudio module to mute other sources?

[Joe Pfeiffer]

I'm reasonably confident there must be a better place to post this than
this newsgroup, so answers telling me where to look will be just as
welcome as any that address the question!  I'm a Debian user, so this is
where I'm trying first

Is there a pulseaudio module that will detect when an audio source isn't
silent, and mute other sources in that cases?

It may be more clear if I describe my particular desire use case.  I
have a ham radio, whose audio output I can plug into my sound card
input.  I'd like to be listening to music, but when a call comes in on
the radio have it mute any other audio.

Any thoughts?

Re: CPU overheating then shutting down while encoding video with ffmpeg

[Joe Pfeiffer]

Chris Bannister cbannis...@slingshot.co.nz writes:

 On Sun, Jul 19, 2015 at 10:17:31AM -1000, Joel Roth wrote:
 
 In addition, you could put your laptop up on some blocks, so
 the airflow is better, or use an external fan or
 vaccuum cleaner to help. 

 I'd be a bit hesitant to recommend to use a vacuum cleaner, 
 wouldn't static electricity be a risk?

I've heard that concern, but never seen any evidence there is any basis
for it.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1boaj3r02k@pfeifferfamily.net

Re: CPU overheating then shutting down while encoding video with ffmpeg

[Joe Pfeiffer]

I'll echo the advice on cleaning the fan(s), air vents, and any ducts.
I've seen laptops which have been used in enviroments like sitting on a
blanket which whose vents have become completely blocked.  Also, of
course, make sure the fans are actually turning, and haven't failed!

I'll also echo the advice to put it on blocks, and also note that you
should make sure the vents aren't obstructed by anything.
-- 
Erwin, have you seen the cat? -- Mrs. Shrödinger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bvbdf95on@pfeifferfamily.net

Re: IP address

[Joe Pfeiffer]

John J. Boyer john.bo...@abilitiessoft.org writes:

 None of these solutions work. ifconfig is not available on Jessie. ip 
 seems to be inapropriate. dig produces nothing. I have used ifconfig on 
 other distros.

 John

 On Fri, Jul 17, 2015 at 05:41:27PM +0200, Elimar Riesebieter wrote:
 * John J. Boyer john.bo...@abilitiessoft.org [2015-07-17 08:32 -0500]:
 
  I have Jessie set up for CLI only. The machine is on a local network 
  using dhcp. What command will tell me what ip address it is using?
 
 $ dig  +short `hostname -f`

I'd be really surprised if there were no ifconfig on your system.  Maybe
a path issue?  Try /sbin/ifconfig

What do you mean when you say inappropriate?
-- 
Erwin, have you seen the cat? -- Mrs. Shrödinger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bd1zpa4or@pfeifferfamily.net

Re: IP address

[Joe Pfeiffer]

John J. Boyer john.bo...@abilitiessoft.org writes:

 Why isn't ifconfig available on Jessie? There id no package. The 
 command produces an error message that it has not been found.

 John

 On Fri, Jul 17, 2015 at 01:44:35PM -0400, Miles Fidelman wrote:
 John J. Boyer wrote:
 I have Jessie set up for CLI only. The machine is on a local network
 using dhcp. What command will tell me what ip address it is using?
 
 ifconfig -a
 is always a good one

ifconfig is in the net-tools package.
-- 
Erwin, have you seen the cat? -- Mrs. Shrödinger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1b8uada4l3@pfeifferfamily.net

Re: IP address

[Joe Pfeiffer]

Patrick Wiseman pwise...@gmail.com writes:

 On Fri, Jul 17, 2015 at 2:25 PM, John J. Boyer
 john.bo...@abilitiessoft.org wrote:
 I have net-tools. ifconfig works only for root. WHY? On other distros
 ordinary users can use it.

 You haven't been listening to what others have been telling you.
 ifconfig resides in sbin, which is in root's but not the ordinary
 user's path. So either use '/sbin/ifconfig' or 'sudo ifconfig' as an
 ordinary user, or su to root and run ifconfig.

While suggestions of running it as root will work, it strikes me as a
bad habit to get into (running things as root that don't have to be).
Using the full path will work without running the command as root, so it
seems like a better solution to me.
-- 
Erwin, have you seen the cat? -- Mrs. Shrödinger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1b4ml1a4fz@pfeifferfamily.net

Re: YAGF is a seriously screwed package

[Joe Pfeiffer]

Renaud (Ron) OLGIATI ren...@olgiati-in-paraguay.org writes:

 On Fri, 10 Jul 2015 11:15:17 -0400
 The Wanderer wande...@fastmail.fm wrote:

 If you substitute in madams for mesdames (since, AFAIK, mesdames
 is just the French equivalent of the same word), it makes more sense.

 remember that madam is usually apposite for the manageress of a brothel...

Depends on context.  Dear Sirs/Madams: no more implies that the women
are brothel managers than it does the men are knights.

The funny thing to me is what caused this whole mini-tempest in a
micro-teapot was an attempt to be polite.  I can't remember when I last
used a salutation on usenet, if I ever have.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1btwtbvpqf@pfeifferfamily.net

Re: Foolproof disk device name in fstab

[Joe Pfeiffer]

andmalc andm...@gmail.com writes:

 I have a Jessie VPS with external disks attached.  The disks are specified
 in /etc/fstab with traditional /dev/sdXX naming.  I recently made
 changes to the disks that made a device name invalid but didn't
 notice. When I rebooted, the disk couldn't be found and boot halted in
 rescue mode.

 My question is: how can I specify devices in fstab so if they can't be
 found boot proceeds proceeds normally instead of halting?  Would
 mounting with systemd with the 'device-timeout' option as described
 here be a good way?

 https://wiki.archlinux.org/index.php/fstab#Automount_with_systemd

Use UUIDs.  A typical line in my /etc/fstab looks like

UUID=edef6980-c414-4718-952e-a6b5e41204c3 /   ext4
errors=remount-ro 0   1

Here's an introduction to them:

https://liquidat.wordpress.com/2013/03/13/uuids-and-linux-everything-you-ever-need-to-know/


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1br3urmm2p@pfeifferfamily.net

Re: remove me from this list

[Joe Pfeiffer]

Talitha Thalya ravencoun...@gmail.com writes:

 My name was never meant to show up on a google search like this linked
 to Debian. and
 dated back in 2001 Not Ok it was meant to go to the cause. this is a
 misuse of trust. please remove me. name stated in this email address.
 Thanks

 Taliban woman 2001
 https://lists.debian.org/debian-devel/2001/04/msg00110.html

Let me get this straight -- you put your name on a petition that was
requested to get forwarded everywhere possible, and thirteen years later
you're surprised to find that it's been saved and indexed by google?
Good luck with that.

I'd say Debian has a lot more cause to have a beef with Mr. Jackson than
you do, since he seemed to feel that a political petition was a valid
use of a software developers' mailiing list...

Do you understand that your post today is also archived, so you've put
your name out to be searched again?


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1by4pdoxos@pfeifferfamily.net

Re: FW: Time for compassion and the Init GR

[Joe Pfeiffer]

Jerry Stuckle jstuc...@attglobal.net writes:

 The tone is subtle, and not necessarily something a native English
 speaker would see.  But I see it there.

Are you seriously claiming that a non-native speaker would be likelier
to pick up on subtle, and quite possibly subconcious, cues than a native
speaker?


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1btx27prvn@pfeifferfamily.net

Re: Joey Hess is out?

[Joe Pfeiffer]

Carl ca...@panix.com writes:

 Really? I may have unthinkingly assumed everyone reading was a native
 speaker is American English. In my dialect, out means openly
 homosexual far more often than quitting. The joke had nothing to do
 with Mr. Hess and everything to do with mocking my own first reading
 of the subject line. 

FWIW, I recognized your joke as a joke, and it got a chuckle from me.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bppcvprd0@pfeifferfamily.net

Re: umask has no man page?

[Joe Pfeiffer]

Carl Fink c...@finknetwork.com writes:

 When I wanted the options for umask, I typed 'man umask' and got the man
 page for it as a C header diretive? (I'm not a C programmer, but it seemed
 to be for C header files and came from section 2.)

 This is darn confusing for a new user. I have been around long enough
 (slink) that I quickly realized it must be a Bash builtin and found that man
 page, but how would a beginner know that? Surely a symbolic link could be
 set up for umask as well as the others (bg, eval, fg, read, etc.)?

 Should I file this as a bug against Sid? I know there's no chance it will
 make it into Wheezy.

The underlying problem is that umask isn't a standalone command, it's a
shell builtin.  So if you look at the bash manpage you can find the
(very terse) documention; of course, there's no hint anywhere that you
should do that.  Just as for (looking at some other builtins) ulimit,
unalias, unset


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1b38a1dkq6@pfeifferfamily.net

Re: apache2 what is the standard way to enable modules?

[Joe Pfeiffer]

Harry Putnam rea...@newsguy.com writes:

 googling to learn how to enable modules in apache2-2.4.10-1+b1

 I'm getting a little too much input to really see what to do.

 what is the name of cgi module?  That would be very useful for the
 `a2enmod' cmd.  And for something real simple like making sure it is
 installed. 

The modules you have available are in /etc/apache2/mods-available/

To enable one, create a symbolic link from one the module and its
configuration in /etc/apache2/mods-enabled.  I've got cgi enabled on my
system with

babs:511$ pwd
/etc/apache2
babs:512$ ls -l mods-enabled/*cgi*
lrwxrwxrwx 1 root root 26 Sep 26  2013 mods-enabled/cgi.load - 
../mods-available/cgi.load
babs:513$ 


 I see several files in [...]/mods-available with the string `cgi' in
 them.  None of those show up in [...]/mods-enabled.

Right -- the way you enable them is to create the symbolic link.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bmwaggiy3@pfeifferfamily.net

Re: since demise of encfs what to use for encrypting dir

[Joe Pfeiffer]

B lazyvi...@gmx.com writes:

 On Sat, 02 Aug 2014 18:10:10 -0600
 Joe Pfeiffer pfeif...@cs.nmsu.edu wrote:

  Do you have more information on encfs being declared a security
 hazard? Your post is the first I've heard of it.

 http://defuse.ca/audits/encfs.htm
 (You'll note that dangerous attack vectors are quite low).

Ah, OK -- I'd seen that (I'm on the encfs list), but there's a ways from
what that audit turned up to saying it's a security hazard and making
reference to its demise.  I'm still not sure why it isn't in testing at
the moment, but as long as it's in both stable and unstable I'm not
going to worry too much.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1b38dct1js@snowball.wb.pfeifferfamily.net

Re: since demise of encfs what to use for encrypting dir

[Joe Pfeiffer]

 Harry Putnam wrote:

 I guess encfs and its companion on windows of truecrypt have been
 declared serious security hazards... encfs is not even available .. at
 least in jessie repos.
 
 What are people using as a replacement?  Hopefully something as easy
 to use and encfs was.

Do you have more information on encfs being declared a security hazard?
Your post is the first I've heard of it.

Checking...  while it does appear to be missing from jessie, it is in
both wheezy and sid.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1by4v6saal@snowball.wb.pfeifferfamily.net

Re: Pin package to any version, don't remove?

[Joe Pfeiffer]

Andrei POPESCU andreimpope...@gmail.com writes:

 Note: I'd probably have the same problem with nvidia-driver, but my 
 aptitude is not suggesting removals as first option anymore due to:

 // tweak Aptitude to not suggest removals as first option
 Aptitude::ProblemResolver::SolutionCost removals;

Whether it helps the OP or not, it will certainly help me!  Aptitude
wanting to delete packages as its first option annoys me a lot...


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bvbqs2iy3@snowball.wb.pfeifferfamily.net

Re: android connection

[Joe Pfeiffer]

Esther Carillo ecarillo1...@gmail.com writes:

 On 2014-07-12, François Patte francois.pa...@mi.parisdescartes.fr wrote:
 I don't know anything to android but I have to connect an android
 device to a computer.

 I run a smbd server on my desktop and ES File Explorer app on my phone.
 Scanning my network with the app detects the Samba share and I can
 up/download files wirelessly through the router.

Another option is you can plug the phone in, and it should be recognized
as an MTP device.  The FUSE mtpfs filesystem can now be used to talk to
it.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bk37c3tbw@snowball.wb.pfeifferfamily.net

Re: simple database solution without root access

[Joe Pfeiffer]

Joel Rees joel.r...@gmail.com writes:

 2014/07/07 5:08 Nuno Magalhães nunomagalh...@eu.ipp.pt:

 On Sun, Jul 6, 2014 at 9:03 PM, kamaraju kusumanchi
 raju.mailingli...@gmail.com wrote:
  I am still exploring all the suggestions given by others. But SQLite looks
  very promising. There is a Perl DBI Interface to SQLite which might be what
  I am after.

   2) I want the data to be in text format.

 SQLite keeps data in binary files.

 What do you mean by that?

Presumably, as opposed to human-readable text files.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1btx6um0da@snowball.wb.pfeifferfamily.net

Re: simple database solution without root access

[Joe Pfeiffer]

Joel Rees joel.r...@gmail.com writes:

 2014/07/07 10:39 Joe Pfeiffer pfeif...@cs.nmsu.edu:

 Joel Rees joel.r...@gmail.com writes:

  2014/07/07 5:08 Nuno Magalhães nunomagalh...@eu.ipp.pt:
 
  On Sun, Jul 6, 2014 at 9:03 PM, kamaraju kusumanchi
  raju.mailingli...@gmail.com wrote:
   I am still exploring all the suggestions given by others. But SQLite 
   looks
   very promising. There is a Perl DBI Interface to SQLite which might be 
   what
   I am after.
 
2) I want the data to be in text format.
 
  SQLite keeps data in binary files.
 
  What do you mean by that?

 Presumably, as opposed to human-readable text files.

 Uhm, is text not a subset of binary (when talking about the contents of the 
 files that implement a database)?
 Does SQLite encode  text fields in some non-human-readable manner?

 Okay, thinking about it a bit, the lack of delimiters, and the puzzling 
 nature of binary zero when trying to
 read it as text, might be what Nuno was referring to. Comma delimited files 
 provide visible, understandable
 delimiters, 

 Oh, and the INTEGER PRIMARY KEY is never readable as TEXT.

 For some people seeking to keep data in text format, that might disqualify 
 SQLite. Apparently not  the OP?

My typical experience is that when people distinguish text vs
binary files, they mean the whole file can reasonably be made sense
of in a text editor (that's not a precise definition, of course, but I
think it serves the purpose).  When I open an SQLite database I have
handy with emacs, it is rife with nulls and other non-printing
characters.  Similarly, when I try to run 'less' on it, the response
is

babs:506$ less house.db
house.db may be a binary file.  See it anyway?

Arguably, as people typically use the distinction, it's not a text
file.  Yes, I can extract the text fields as human-readable ASCII,
but that does not make it a text file.

Okay, thinking about it a bit, the lack of delimiters, and the puzzling nature 
of binary zero when trying to read it as text,
might be what Nuno was referring to. Comma delimited files provide visible, 
understandable delimiters,

And what just about anybody else would mean by a text file, as well.

Oh, and the INTEGER PRIMARY KEY is never readable as TEXT.

For some people seeking to keep data in text format, that might
disqualify SQLite. Apparently not the OP? 

My impression (I'd have to go back and recheck) is that it
disqualified it for the OP, as well.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bmwcln9ul@snowball.wb.pfeifferfamily.net

Re: simple database solution without root access

[Joe Pfeiffer]

Joel Rees writes:
2014/07/07 10:39 Joe Pfeiffer pfeif...@cs.nmsu.edu:

 Joel Rees joel.r...@gmail.com writes:

  2014/07/07 5:08 Nuno Magalhães nunomagalh...@eu.ipp.pt:
 
  On Sun, Jul 6, 2014 at 9:03 PM, kamaraju kusumanchi
  raju.mailingli...@gmail.com wrote:
   I am still exploring all the suggestions given by others. But SQLite 
   looks
   very promising. There is a Perl DBI Interface to SQLite which might be 
   what
   I am after.
 
2) I want the data to be in text format.
 
  SQLite keeps data in binary files.
 
  What do you mean by that?

 Presumably, as opposed to human-readable text files.

Uhm, is text not a subset of binary (when talking about the contents of the 
files that implement a database)? Does SQLite encode 
text fields in some non-human-readable manner?

My typical experience is that when people distinguish text vs
binary files, they mean the whole file can reasonably be made sense
of in a text editor (that's not a precise definition, of course, but I
think it serves the purpose).  When I open an SQLite database I have
handy with emacs, it is rife with nulls and other non-printing
characters.  Similarly, when I try to run 'less' on it, the response
is

babs:506$ less house.db 
house.db may be a binary file.  See it anyway? 

Arguably, as people typically use the distinction, it's not a text
file.  Yes, I can extract the text fields, as human-readable ASCII,
but that does not make it a text file.

Okay, thinking about it a bit, the lack of delimiters, and the puzzling nature 
of binary zero when trying to read it as text,
might be what Nuno was referring to. Comma delimited files provide visible, 
understandable delimiters, 

And what just about anybody else would mean by a text file, as well.

Oh, and the INTEGER PRIMARY KEY is never readable as TEXT.

For some people seeking to keep data in text format, that might disqualify 
SQLite. Apparently not  the OP?

My impression (I'd have to go back and recheck) is that it
disqualified it for the OP, as well.


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/21434.3686.317980.890...@snowball.wb.pfeifferfamily.net

Re: preferred overlay/union filesystem?

[Joe Pfeiffer]

Chris Angelico ros...@gmail.com writes:

 On Wed, May 28, 2014 at 3:21 PM, Kushal Kumaran
 kushal.kumaran+deb...@gmail.com wrote:
 Chris Angelico ros...@gmail.com writes:

 On Wed, May 28, 2014 at 5:47 AM, Joe Pfeiffer pfeif...@cs.nmsu.edu wrote:
 Well, I don't want to keep two separate files (that's what I'm trying to
 get away from).  It seems like the overlay filesystem would be a bit
 cleaner if it can work, but symbolic links elsewhere would be my second
 choice.


 As a variant of Kushal's suggestion of two symlinks to the same file,
 you could have the real file in one place (preferably where it's
 written to) and a symlink to it from the other place.


 If I understand the situation correctly, only one of the locations will
 be accessible at any time.  When the user logs in, the original files
 will be hidden under the files provided by the encrypted filesystem.
 So, both files are actually ~/.bogofilter (say), just at different
 times.  So you cannot have a symlink from one location to the other.

 Ah! Gotcha. Then, yes; symlinks from both to the same destination. Not
 sure what a suitable destination is, though.

 ChrisA

For the moment, I have /home/acct/file, /home.unenc/acct/file, and
/home.enc/acct/file.

The real file is in /home.unenc/acct/file, and there are symbolic links
to it from both home/acct/file and /home.enc/acct/file.  When I'm not
logged in, the daemon sees /home/acct/file (and hence really sees
/home.unenc/acct/file); when I log in, /home.enc/acct is
unencrypted and laid on top of /home/acct, so the daemon still sees
/home.unenc/acct/file.

Inelegant (so I'd really like to find a way for a union or overlay
filesystem to do it!) but it seems to work...


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bfvju8234@snowball.wb.pfeifferfamily.net

Re: preferred overlay/union filesystem?

[Joe Pfeiffer]

Steve Litt sl...@troubleshooters.com writes:

 On Tue, 27 May 2014 11:27:56 +0530
 Kushal Kumaran kushal.kumaran+deb...@gmail.com wrote:

 Joe Pfeiffer pfeif...@cs.nmsu.edu writes:
 
  First, here's what I'm trying to do:
 
  I'm using encfs to give myself an encrypted home directory, and I'm
  successfully mounting it automatically using pam_mount when I log
  in.
 
  My email is processed by a .procmailrc file in my home directory,
  and I'm passing the email through bogofilter.  So, at present I
  have a nearly empty home directory containing .procmailrc
  and .bogofilter/ which is accessed when I'm not logged in, and my
  real home directory which is available when I am logged in.  This
  is, of course, less than optimal since I have to keep two versions
  of .procmailrc and .bogofilter/ in sync.
 
 
 You could keep the two files outside your home, and create symlinks to
 them in both the unencrypted and encrypted homes.

 That sounds good to me.

 If you want to keep two separate files, it's as simple as running a
 shellscript to make sure they are the same. For instance,
 my .procmailrc got so big and complicated that I split it into parts.
 For instance, all the trolls and irritants go into something called
 badguys.rc. Mailing lists go in maillists.rc. So to filter yet another
 person, I'd edit badguys.rc, run my shellscript, and it would create a
 new ~/.procmailrc with everything in the right order (bad guys get
 filtered out before mailing list decisions are made, etc). There's one
 file called vitals.rc that rises to the top of the file, and is
 reserved for anything from a person or organization I consider
 absolutely essential, and I want to see it even if it would normally be
 moved by other filters. My shellscript could just as easily write a
 copy in two different places.

Well, I don't want to keep two separate files (that's what I'm trying to
get away from).  It seems like the overlay filesystem would be a bit
cleaner if it can work, but symbolic links elsewhere would be my second
choice.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bbnujq9zu@snowball.wb.pfeifferfamily.net

preferred overlay/union filesystem?

[Joe Pfeiffer]

First, here's what I'm trying to do:

I'm using encfs to give myself an encrypted home directory, and I'm
successfully mounting it automatically using pam_mount when I log in.

My email is processed by a .procmailrc file in my home directory, and
I'm passing the email through bogofilter.  So, at present I have a
nearly empty home directory containing .procmailrc and .bogofilter/
which is accessed when I'm not logged in, and my real home directory
which is available when I am logged in.  This is, of course, less than
optimal since I have to keep two versions of .procmailrc and
.bogofilter/ in sync.

I'd like to have an overlay filesystem, so the unencrypted .procmailrc
and .bogofilter are still visible through a hole in my encrypted
filesystem.

I've been looking through the various union and overlay filesystems
(unionfs, unionfs-fuse, aufs, overlayfs) and (1) I'm not sure which
of them are currently being developed and supported, and (2) it isn't
clear any of them actually do what I want (either I can't just overlay a
new filesystem on top of an existing one, or else the existing one has
to be read-only).

Any guidance would be appreciated!


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bppiznazx@snowball.wb.pfeifferfamily.net

Re: 'box' as noun, was: wireless can DHCP but not DNS?

[Joe Pfeiffer]

Tom Roche tom_ro...@pobox.com writes:

 Lisi Reisz Fri, 23 May 2014 17:10:49 +0100
 box is a verb, so I found it confusing.

 You are indeed confused. As a native speaker of English, I can assure
 you, 'box' is both noun and verb.
 Also, having been in computing in the US for decades, I can assure
 you, 'box' as a noun is widely used to refer to informatic devices
 generically.

While you are correct, the sentence structure leads one to expect 'box'
as a verb and 'ethernets' as a noun.  Speaking as another native
speaker, it took me two passes to read what was intended there.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1b38g02pmd@snowball.wb.pfeifferfamily.net

Re: Systemd

[Joe Pfeiffer]

Ralf Mardorf info.mard...@rocketmail.com writes:

 On Fri, 2014-04-25 at 21:49 +0200, Ralf Mardorf wrote:
 On Fri, 2014-04-25 at 15:12 -0400, Steve Litt wrote:
  Kinda seems like the (de) evolution of cars, doesn't it?  As a kid,
  I could tune up my beater flat head 6 1959 Plymouth in 20 minutes with
  a 10 inch adjustable and a gapping tool. With today's cars, I'm lucky
  to be able to change the air filter.
 
 That's a nice analogy :). We could repair our old cars using a hammer
 and screwdriver, today we need to be computer experts to repair a car.

 For example, to correct idling mixture there under the hood was a screw
 on the carburetor, nowadays there's a connector for a computer inside
 the driver's cab. Assumed your car strikes in the middle of the
 wilderness, what kind of car do you prefer?

I've never had a car stop dead due to the idle mixture going out of
spec.  When something has gone seriously enough wrong with one of my
cars to stop it, the fact that it had a carburetor (as my daily driver,
a '78 Chrysler Newport, does) or a computer has been totally
irrelevant.

I maintain that anybody who is afraid of modern fuel injection has never
rebuilt a ThermoQuad.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bwqebu4uu@snowball.wb.pfeifferfamily.net

Re: Systemd

[Joe Pfeiffer]

staticsafe m...@staticsafe.ca writes:

 On 4/24/2014 19:15, latin...@vcn.bc.ca wrote:
 Hello list:
 Have you read this stupid things?
 
 http://boycottsystemd.org/
 https://igurublog.wordpress.com/2014/04/03/tso-and-linus-and-the-impotent-rage-against-systemd/

 What a waste of bits indeed.

But there's something just special about

1. systemd flies in the face of the Unix philosophy: do one thing
and do it well, representing a complex collection of dozens of
binaries. snip

since lots of little binaries is *exactly* how one has each binary do
one thing and do it well.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bppk6ging@snowball.wb.pfeifferfamily.net

Re: Source.list URL

[Joe Pfeiffer]

ray r...@aarden.us writes:

 Marc,

 Thank you for your efforts.  I tried it again and reached 
 http://debian.uchicago.edu/debian/ which seems like the URL in source.list is 
 just a relay.

 ray

One small note -- it would be helpful to include a little bit of context
in your messages (like I did here) to help people see what you're
responding to.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1b8urnpwz5@snowball.wb.pfeifferfamily.net

Re: Source.list URL

[Joe Pfeiffer]

ray r...@aarden.us writes:

 The URL in the default source.list doesn't seem correct:

 /etc/apt/sources.list  - 
 deb http://http.debian.net/debian/ wheezy main contrib non-free

 The above path is to a directory with other folders.  But Wheezy is a 
 subdirectory of dists, not debian.  So it seems the URL should be:
 deb http://http.debian.net/debian/dists/ wheezy main contrib non-free

 (I added contrib non-free)

 I have just started with Linux and have not had much luck with apt-get so I 
 am wondering if this might be a problem.  

The syntax for an entry in the sources.list file isn't what I would call
obvious.  What you're seeing is in fact correct, even though the dists
directory doesn't appear in it.

(if it weren't, you wouldn't be able to do an update nor get packages!)


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bppl0plxg@snowball.wb.pfeifferfamily.net

Re: clean upgrade 32 - 64?

[Joe Pfeiffer]

Darac Marjal mailingl...@darac.org.uk writes:

 On Mon, Mar 03, 2014 at 05:08:37PM -0700, Joe Pfeiffer wrote:
 I'm about to replace one of my old 32-bit x86 Debian boxes with a
 64-bit; I'll actually just be moving the disk drives out of the old box
 into the new one and doing any minor configuration changes that'll be
 neede (which will be very minor).  So, while I'm at it, I'm curious --
 is there any clean way to do an update that will simply replace all the
 32 bit packages with 64 (and do all the other necessary multiarch
 stuff)?

 It depends.

 Arguably the cleanest solution is to re-install using a 64-bit
 installer. However, this will blow away your configuration, choice of
 packages and so on.

 If you want to cross-grade, the accepted procedure is detailed here:
 https://wiki.debian.org/CrossGrading

Thanks!  Finally got the new machine set up to try the crossgrading, and
everything worked just fine.  There's just one issue remaining:  a huge
number of i386 packages on the machine show dependencies on either apt
or dpkg, so any attempt to upgrade reinstalls the i386 apt and dpkg.  Is
there an easy way around this, or do I need to plan on replacing all of
these packages specifying amd64?

Here's an example from the end of the list of conflicts I got:

 texlive-pictures : Depends: dpkg (= 1.14.18) but it is not going to be 
installed.
 module-init-tools : PreDepends: dpkg (= 1.15.7.2) but it is not going to be 
installed.
The following actions will resolve these dependencies:

 Install the following packages: 
1) apt [0.9.16.1 (testing, unstable)]
2) dpkg [1.17.6 (testing, unstable)] 



Accept this solution? [Y/n/q/?] q
Abandoning all efforts to resolve these dependencies.
Abort.
babs:46$ dpkg -l dpkg
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name Version   Architecture  Description
+++--=-=-=
ii  dpkg 1.17.6amd64 Debian package 
management system


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1bzjk8q561@snowball.wb.pfeifferfamily.net

clean upgrade 32 - 64?

[Joe Pfeiffer]

I'm about to replace one of my old 32-bit x86 Debian boxes with a
64-bit; I'll actually just be moving the disk drives out of the old box
into the new one and doing any minor configuration changes that'll be
neede (which will be very minor).  So, while I'm at it, I'm curious --
is there any clean way to do an update that will simply replace all the
32 bit packages with 64 (and do all the other necessary multiarch
stuff)?

I'm not actually doing anything with it that will require 64 bits; if
the answer is no I'll simply continue to run it as a 32 bit machine.

So, is it possible?

Thanks,


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1btxbe3jwa@snowball.wb.pfeifferfamily.net

Re: Netflix on Sid, no wine.

[Joe Pfeiffer]

Hugo Vanwoerkom hvw59...@care2.com writes:

 Hi,

 A few days ago Google News carried this:
 http://www.makeuseof.com/tag/easily-enable-silverlight-watch-netflix-linux/

 I tried it and it works as advertized, an easy installation and
 Netflix works.

 IMO the latter is overrated: mostly old hat hu hum movies.

 Hugo

Thanks!  It's very nice to have netflix available at last!  You're
right, it is overrated, but it's what's available; it seems to me like
it was better a few years ago.  It looks like a pay-per-play model is
winning this particular battle.  It's really unfortunate that
non-DRM-encumbered files seem to be losing (when they pretty much won
the audio content battle, I had high hopes).


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1b7g8c4mbi@snowball.wb.pfeifferfamily.net

Re: Determine IP Adrress from MAC Address

[Joe Pfeiffer]

Muntasim-Ul-Haque tranjees...@inventati.org writes:

 Hi,
 How can I determine the IP address if I already have the MAC address
 or Hardware Address? What is the most convenient way?

 -Muntasim Ul Haque

The fact that you already have the MAC address doesn't matter in finding
out your IP address.

ifconfig will give you both the MAC address and IP address of all your
network interfaces:

snowball:593$ ifconfig
eth0  Link encap:Ethernet  HWaddr d4:3d:7e:01:8c:8f  
  inet addr:192.168.13.2  Bcast:192.168.13.255  Mask:255.255.255.0
  inet6 addr: fe80::d63d:7eff:fe01:8c8f/64 Scope:Link
  UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
  RX packets:75979320 errors:0 dropped:0 overruns:0 frame:0
  TX packets:79860948 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:1000 
  RX bytes:26691502625 (24.8 GiB)  TX bytes:34703381554 (32.3 GiB)

loLink encap:Local Loopback  
  inet addr:127.0.0.1  Mask:255.0.0.0
  inet6 addr: ::1/128 Scope:Host
  UP LOOPBACK RUNNING  MTU:65536  Metric:1
  RX packets:590910 errors:0 dropped:0 overruns:0 frame:0
  TX packets:590910 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:0 
  RX bytes:38390044 (36.6 MiB)  TX bytes:38390044 (36.6 MiB)

So eth0 (my ethernet card) has IP address 192.168.13.2, and lo (the
loopback device, used for programs on my machine to talk to other
programs on the same machine) has 127.0.0.1 -- it's actually always
127.0.0.1

If you have network devices that aren't configured at the moment,
ifconfig -a will tell you about them too (but they won't have IP
addresses).


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1bk3ctpk5n@snowball.wb.pfeifferfamily.net

Re: Determine IP Adrress from MAC Address

[Joe Pfeiffer]

Joe Pfeiffer pfeif...@cs.nmsu.edu writes:

 Muntasim-Ul-Haque tranjees...@inventati.org writes:

 Hi,
 How can I determine the IP address if I already have the MAC address
 or Hardware Address? What is the most convenient way?

 -Muntasim Ul Haque

 The fact that you already have the MAC address doesn't matter in finding
 out your IP address.

 ifconfig will give you both the MAC address and IP address of all your
 network interfaces:

 snowball:593$ ifconfig
 eth0  Link encap:Ethernet  HWaddr d4:3d:7e:01:8c:8f  
   inet addr:192.168.13.2  Bcast:192.168.13.255  Mask:255.255.255.0
   inet6 addr: fe80::d63d:7eff:fe01:8c8f/64 Scope:Link
   UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
   RX packets:75979320 errors:0 dropped:0 overruns:0 frame:0
   TX packets:79860948 errors:0 dropped:0 overruns:0 carrier:0
   collisions:0 txqueuelen:1000 
   RX bytes:26691502625 (24.8 GiB)  TX bytes:34703381554 (32.3 GiB)

 loLink encap:Local Loopback  
   inet addr:127.0.0.1  Mask:255.0.0.0
   inet6 addr: ::1/128 Scope:Host
   UP LOOPBACK RUNNING  MTU:65536  Metric:1
   RX packets:590910 errors:0 dropped:0 overruns:0 frame:0
   TX packets:590910 errors:0 dropped:0 overruns:0 carrier:0
   collisions:0 txqueuelen:0 
   RX bytes:38390044 (36.6 MiB)  TX bytes:38390044 (36.6 MiB)

 So eth0 (my ethernet card) has IP address 192.168.13.2, and lo (the
 loopback device, used for programs on my machine to talk to other
 programs on the same machine) has 127.0.0.1 -- it's actually always
 127.0.0.1

 If you have network devices that aren't configured at the moment,
 ifconfig -a will tell you about them too (but they won't have IP
 addresses).

Ah, re-reading:  when you said the MAC address, my previous response
was based on the assumption that you meant your IP address.  If you
need the IP address of some other machine, arping should give it to you.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1bfvnhphb8@snowball.wb.pfeifferfamily.net

Re: GPT + RAID + boot

[Joe Pfeiffer]

Sven Hartge s...@svenhartge.de writes:
PaulNM deb...@paulscrap.com writes:

snip

Thank you both for your help -- your suggestions were exactly what I
needed (I delayed responding until I was confident I had everything
working).

I'm puzzled as to why parted refers to these partition types as
flags -- seeing that when using the program, and when reading the
documentation, coupled with the fact that bios_grub doesn't turn up in
the man page (but only in info), and that disks using MBR actually can
have a boot flag on a partition, certainly helped send me off in the
wrong direction!

But again, thank you both for helping me navigate through this one.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1b8uuh5aas@snowball.wb.pfeifferfamily.net

GPT + RAID + boot

[Joe Pfeiffer]

My goal here is to be able to have a bootable, running system in the
event of a disk failure.  I've been running two disks in a RAID-1
configuration, with grub installed on both disks, for some time.  My
/etc/mdadm/mdadm.conf is essentially empty, as mdadm has been
successfully finding my RAID partitions and assembling my arrays at boot
time without it.

I've gotten a new 3TB disk (to replacing an old, failing disk), so I'm
setting up my first GPT partition table.

On my old (MBR) disk, parted shows my first partition as

1  32.3kB  1500GB  1500GB  primary  boot, raid

and I'm able to boot successfully.

On my new (GPT) disk, I am only able to install grub if I've set the
bios_grub flag (note that this flag doesn't appear in the man page,
though it does appear in the documentaiton at
http://www.gnu.org/software/parted/manual/parted.html#set) on my
partition.  If I set the boot or legacy_boot flags, I get

snowball:518$ sudo grub-install /dev/sdb
/usr/sbin/grub-setup: warn: This GPT partition label has no BIOS Boot 
Partition; embedding won't be possible!.
/usr/sbin/grub-setup: error: embedding is not possible, but this is required 
for cross-disk install.

Unfortunately, I don't seem to be able to set both the bios_grub and the
raid flags at the same time.  If I set the bios_grub flag, printing the
partition table shows no raid flag; if I set the raid flag, printing the
partition table shows no bios_grub flag.

Booting with the bios_grub flag set, my raid array isn't assembled
properly:  the partition with bios_grub set isn't added into the array
(fortunately, my other disk is good!).

So:  how can I go about setting up my new disk so I will have a two-disk
RAID array if both disks are good, and be able to boot with a degraded
array in the event of either disk failing?
-- 
Erwin, have you seen the cat? -- Mrs. Shroedinger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1bfvoz35op@snowball.wb.pfeifferfamily.net

Re: MIT discovered issue with gcc

[Joe Pfeiffer]

Robert Baron robertbartlettba...@gmail.com writes:

 Aren't many of the  constructs used as examples in the paper are commonly used
 in c programming.  For example it is very common to see a function that has a
 pointer as a parameter defined as:

 int func(void *ptr)
     {
     if(!ptr) return SOME_ERROR;
     /* rest of function*/
     return 1;
     }

 Isn't it interesting that their one example will potentially dereference the
 null pointer even before compiler optimizations (from the paper):

 struct tun_struct *tun=;
 struct sock *sk = tun-sk;
 if(*tun) return POLLERR; 

 The check to see that tun is non-null should occur before use, as in - quite
 frankly it is useless to check after as tun cannot be the null pointer (the
 program hasn't crashed):

 struct tun_struct *tun=;
 if(*tun) return POLLERR; 
 struct sock *sk = tun-sk;

The paper points out that the code contains a bug; the claim in the
paper is that it is a minor bug as written (it only gets past the
tun-sk dereference if page 0 has somehow been made readable), but
becomes a possible privilege escalation after the check has been
optimized away.

 I am under the impression that these problems are rather widely known among c
 programmers (perhaps not the kids fresh out of college).  But this is why
 teams need to have experienced people. 

 Furthermore, it is very common to find code that works before optimization,
 and fails at certain optimization levels.  Recently, I was compiling a library
 that failed its own tests under the optimization level set in the makefile but
 passed its own test at a lower level of optimization.

Isn't that, and an analysis of when this can happen, the main point of
the paper?


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1bvbzgtj13@snowball.wb.pfeifferfamily.net

Re: MIT discovered issue with gcc

[Joe Pfeiffer]

Robert Baron robertbartlettba...@gmail.com writes:

 Second question:

 Doesn't memcpy allow for overlapping memory, but strcpy does not?  Isn't this
 why memcpy is preferred over strcpy?

According to the man page for memcpy, The memory areas must not
overlap.  Use memmove(3)  if  the memory areas do overlap.

strcpy will stop copying at the first null byte.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1br4a4tiy3@snowball.wb.pfeifferfamily.net

Re: sudo and UNIXes

[Joe Pfeiffer]

Curt cu...@free.fr writes:

 On 2013-11-02, Joe Pfeiffer pfeif...@cs.nmsu.edu wrote:
 
 Again -- isn't basically equivalent to giving everyone uid=0.  Permits
 someone who *has* sudo access to avoid retyping a password.

 Not only that. Permits someone who already has sudo access to continue
 having such access indefinitely, ignoring being excluded from sudoers
 altogether.

 You made a specific claim, that sudo without patches is basically
 equivalent to giving everyone uid=0.  You have yet to say anything that
 even begins to substantiate that claim.


 How about this bug:

 http://www.sudo.ws/sudo/alerts/sudo_debug.html
  
  Impact: Successful exploitation of the bug will allow a user to run arbitrary
  commands as root.

  Exploitation of the bug does not require that the attacker be listed in the
  sudoers file. As such, we strongly suggest that affected sites upgrade from
  affected sudo versions as soon as possible. 

OK, there has been a bug that will cause the claimed behavior if the
sysadmin updated his system between February and November 2011 but not
since, and you've got a seriously malicious user.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1bvc09paki@snowball.wb.pfeifferfamily.net

Re: sudo and UNIXes

[Joe Pfeiffer]

Reco recovery...@gmail.com writes:

  Hi.

 On Sat, 2 Nov 2013 11:46:48 -0500
 Cybe R. Wizard cybe_r_wiz...@earthlink.net wrote:
  How about this bug:
  
  http://www.sudo.ws/sudo/alerts/sudo_debug.html
   
   Impact: Successful exploitation of the bug will allow a user to run
  arbitrary commands as root.
  
   Exploitation of the bug does not require that the attacker be listed
  in the sudoers file. As such, we strongly suggest that affected sites
  upgrade from affected sudo versions as soon as possible. 
  
 How valid is that considering that Wheezy is using sudo
 version 1.8.5p2-1+nmu1 ?

 Perfectly valid, considering that this part of thread is about using
 sudo in the UNIX environment, not Linux one.


 May I assume that there are still a lot of non-upgraded machines out there?

 Depends. For example, AIX 5, 6 and 7 all have sudo-1.6.7p5-3 (the only
 version built officially by IBM). Unless you build sudo from the source
 - no upgrades for you.
 Solaris 11.1 has sudo-1.8.6.7 out of the box.

Note that neither of these is subject to vulnerability in the bug
report.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1br4axpaik@snowball.wb.pfeifferfamily.net

Re: sudo and UNIXes

[Joe Pfeiffer]

Reco recovery...@gmail.com writes:

 On Mon, Oct 28, 2013 at 10:19:43AM -0600, Joe Pfeiffer wrote:
 Reco recovery...@gmail.com writes:
  You also have to add to the picture such a vulnerability, and I haven't
  noticed any.
 
  If we're speaking of public vulnerabilities:
 
  CVE-2010-0427.
 
 Does not permit users outside of those in the sudoers file (or with the
 root password) to escalate privileges.

 Lessens attack surface, but doesn't void the existence of vulnerability.

 
  CVE-2013-1775 (allows bypass sudoders modification to retain root
  privileges).
 
 Again -- isn't basically equivalent to giving everyone uid=0.  Permits
 someone who *has* sudo access to avoid retyping a password.

 Not only that. Permits someone who already has sudo access to continue
 having such access indefinitely, ignoring being excluded from sudoers
 altogether.

You made a specific claim, that sudo without patches is basically
equivalent to giving everyone uid=0.  You have yet to say anything that
even begins to substantiate that claim.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1b4n7vik0q@snowball.wb.pfeifferfamily.net

Re: sudo and UNIXes

[Joe Pfeiffer]

Reco recovery...@gmail.com writes:

 On Sun, Oct 27, 2013 at 09:28:51PM -0600, Joe Pfeiffer wrote:
 Reco recovery...@gmail.com writes:
  True, you need to add to the picture that curious user who just read on
  Bugtraq or Full Disclosure about fresh vulnerability in sudo. Or that
  disgruntled user who needs /etc/system changed right here and now. Or
  that developer who needs to do this 'small change, nobody will notice'
  on a production server.
  And if you don't have such people there - good for you, as here we can
  always find such person here.
 
 You also have to add to the picture such a vulnerability, and I haven't
 noticed any.

 If we're speaking of public vulnerabilities:

 CVE-2010-0427.

Does not permit users outside of those in the sudoers file (or with the
root password) to escalate privileges.

 CVE-2013-1775 (allows bypass sudoders modification to retain root
 privileges).

Again -- isn't basically equivalent to giving everyone uid=0.  Permits
someone who *has* sudo access to avoid retyping a password.

 I have no knowledge about private 0days.

 Reco


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1bvc0hcqqo@snowball.wb.pfeifferfamily.net

Re: sudo and UNIXes

[Joe Pfeiffer]

Reco recovery...@gmail.com writes:
 Tom H tomh0...@gmail.com wrote:
 On Fri, Oct 25, 2013 at 9:16 PM, Reco recovery...@gmail.com wrote:

  Considering that primary usage of sudo is to provide controlled
  privilege escalation to uid=0, using unsupported (therefore - not
  updated unless local sysadmins care about security) sudo on these OSes
  is basically equivalent to giving everyone uid=0.
 
  Somewhat exaggerated :)
 
  No offense meant, but probably you're living in a some kind of IT
  paradise ;) 'Nobody does no evil, nobody does any mistakes' kind of
  paradise.
 
 Not updating/patching sudo isn't equivalent to giving everyone root
 access! It's a BIG leap!

 True, you need to add to the picture that curious user who just read on
 Bugtraq or Full Disclosure about fresh vulnerability in sudo. Or that
 disgruntled user who needs /etc/system changed right here and now. Or
 that developer who needs to do this 'small change, nobody will notice'
 on a production server.
 And if you don't have such people there - good for you, as here we can
 always find such person here.

You also have to add to the picture such a vulnerability, and I haven't
noticed any.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1b38nmdqfg@snowball.wb.pfeifferfamily.net

Re: endianness (was Re: sysadmin qualifications (Re: apt-get vs. aptitude))

[Joe Pfeiffer]

Jonathan Dowland j...@debian.org writes:

 On Thu, Oct 17, 2013 at 05:29:33PM +0200, berenger.mo...@neutralite.org wrote:
 Speaking about endianness, it really is hard to manage:
 
 void myfunction( ... )
 {
 #ifdef BIG_ENDIAN
 move_bytes_in_a_specific_order
 #else
 move_bytes_in_the_other_specific_order
 #endif
 }

 Bad way to manage endian in C. Better to have branching based on C
 itself (rather than preprocessor), otherwise you run the risk of never
 testing code outside the branch your dev machine(s) match. E.g. use

 char is_little_endian( … ) {
   int i = 1;
   int *p = i;
   return 1 == *(char*)p;
 }

 Or similar. The test will likely be compiled out as a no-op anyway with
 decent compilers (GCC: yes; Sun Workshop: no.)

What's wrong with htonl and other similar functions/macroes?


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1bfvrz6v1o@snowball.wb.pfeifferfamily.net

Re: Printer brand recommendations

[Joe Pfeiffer]

Doug dmcgarr...@optonline.net writes:

 On 09/09/2013 06:16 PM, David Christensen wrote:
 On 09/09/13 14:42, ken wrote:
 I've used Epson with success, but won't another one. The cost of the
 cartridges is so high, it's like I'm buying the printer over and over
 again every year.
 
 HP cartridges are also very expensive -- $100+.  So:
 
 1.  I bought a black-and-white printer (1 cartridge), rather than color 
 (4 cartridges).
 
 2.  I buy high capacity generic cartridges (~$30).
 
 
 HTH,
 
 David
 
 

 Another thread that's been running on one of the lists, where someone
 in Switzerland was looking for a printer recommendation, pointed out
 that HP printers are no longer the Sherman tanks they used to be, and
 that they may no longer be built to last; also that HP does not support
 any product after they stopped making it for five years. Not at all--
 you may not even find a manual. (OTOH, I can still get replacement toner
 units for a 12 year old LaserJet 2200dn, even tho I can't get any info
 from HP about it.) I also have an Epson all-in-one, a WP-4530.
 It's quick, it prints, copies, faxes, scans, and has usb, Ethernet and
 wireless interfaces. (And a phone connection for fax.) Does duplex, of
 course. I do a fair amount of printing, and I don't need new cartridges
 very often. It uses four cartridges, CMYK, so if you mostly print black,
 that will be the only one you need to replace. The color is very good,
 both from the net and scanned in. For BW images, it's about on a par
 with the LaserJet. And Epson has Linux drivers on their website for all
 their printers and scanners, in both deb and rpm formats. Someone on the
 other list found the WP-4530 for $189 on Amazon.
 I have no financial or other interest in Epson.

I was a strong HP fan for years, to the point that when a printer died I
would only look at HP replacements.  After a couple of short-lived HPs
in a row, I broadened my horizons to take at least a token look at other
brands -- and brought home an Epson Workforce 645 all-in-one which is
behaving like HPs used to.  It's reliable, the cartridges last a long
time, the tray holds a *lot* of paper...  just a good piece of
equipment.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1bob812w6f@snowball.wb.pfeifferfamily.net

Re: please read i am not getting satisfaction from mytablet l got it in april i dont download anything on it and it is slow take long to load so what should i do please tell me thank you.

[Joe Pfeiffer]

phillip johnson jamaicabarne...@gmail.com writes:
silence
When your subject line is three lines long (on my display, anyway) maybe
you should move it to the body of your post.

When the body of your post is empty, you should *definitely* move
something in there.

Does your table run Debian Linux?  If not, why are you asking here?


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1bk3iw7ytq@snowball.wb.pfeifferfamily.net

Re: strange bash behavior

[Joe Pfeiffer]

William Hopkins we.hopk...@gmail.com writes:

 On 09/03/13 at 03:45pm, Joe Pfeiffer wrote:
 Stephen Powell zlinux...@wowway.com writes:
 
  Interesting.  If break appears out of context, you should get
  an error message something like:
 
 bash: break: only meaningful in a 'for', 'while', or 'until' loop
 
  You didn't get an error message, so part of bash thinks it is in context.
  Yet it did not exit the loop.  It seems to me that you should get one
  behavior or the other.  Either you should get an error message or it
  should exit the loop.
 
 Good point -- it is odd that it isn't giving the error message.

 The loop context is inherited by the subshell, so break thinks it is fine. It
 is only that it is totally meaningless to break there, since that signal 
 cannot
 be captured by parent shell environment. 

How is the context passed to the subshell?  And what signal?

 This seems to be expected behavior..


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1bd2oo7yqh@snowball.wb.pfeifferfamily.net

Re: strange bash behavior

[Joe Pfeiffer]

David Guntner da...@guntner.com writes:

 Darac Marjal grabbed a keyboard and wrote:
 On Mon, Sep 02, 2013 at 08:06:17AM -0700, David Guntner wrote:
 Matej Kosik grabbed a keyboard and wrote:
 Hello,

 This morning I have been puzzled by bash.
 After typing the following command:

for i in `seq 1 5`;do echo $i; test $i = 3  break; done

 I see:

1
2
3

 Which is OK.

 However, if the break command appears in a subshell:

for i in `seq 1 5`;do echo $i; test $i = 3  (break); done

 then the break command does not seem to have any effect

1
2
3
4
5

 I am curious, is this something to be expected?

 What do you mean by appears in a subshell?
 
 From man 1 bash:
 (list) list  is  executed in a subshell environment (see COMMAND EXECU‐
TION ENVIRONMENT below).  Variable assignments and builtin  com‐
mands  that  affect  the  shell's  environment  do not remain in
effect after the command completes.  The return  status  is  the
exit status of list.

 Ok, I'm still not following you.

 What, exactly, is it that you are doing at your keyboard, in order to
 run it in this subshell?  I'm assuming that in your main one you're
 just typing the expression and hitting enter.  So what are you doing
 when the second example fails?

Based on his post, he is putting 'break' in parentheses in the second
case.  This causes a new shell (the subshell) to be created, and the
break command to be executed in that subshell.  The subshell then exits.

This is exactly the behavior I would expect:  the break command isn't
being executed by the same shell as your for-loop.  The for-loop has no
way of knowing that the subshell executed a break command (note that the
break command is a shell built-in command -- it is interpreted and
executed by the shell itself, no new process is created for it).


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1br4d5mvmi@snowball.wb.pfeifferfamily.net

Re: strange bash behavior

[Joe Pfeiffer]

Stephen Powell zlinux...@wowway.com writes:

 Interesting.  If break appears out of context, you should get
 an error message something like:

bash: break: only meaningful in a 'for', 'while', or 'until' loop

 You didn't get an error message, so part of bash thinks it is in context.
 Yet it did not exit the loop.  It seems to me that you should get one
 behavior or the other.  Either you should get an error message or it
 should exit the loop.

Good point -- it is odd that it isn't giving the error message.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1bmwntmvfc@snowball.wb.pfeifferfamily.net

Re: Thanks

[Joe Pfeiffer]

Thod Motte tmo...@mail.com writes:

 Thanks to debian and Gnome 3 for making my desktop as buggy and unstable as 
 Windows 95 was in 1997 and less
 customizable.

 I'm just going to revert to squeeze, that desktop actually worked.

Another vote for xfce.  I switched to it quite a while ago, and have
been happy since.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1bob8djn8v@snowball.wb.pfeifferfamily.net

Re: How to get rid of an entry in grub?

[Joe Pfeiffer]

David Guntner da...@guntner.com writes:

 Hugo Vanwoerkom grabbed a keyboard and wrote:
 David Guntner wrote:
 Hmmm.  I wonder if the MBR for the drive sill has a loader on it,
 even though I removed all partitions and repartitioned it?  Is there a
 utility out there that can wipe the MBR of a drive without touching the
 rest of the contents?
 
 There is http://bootinfoscript.sourceforge.net/ also.

 Ok, I got that, thanks.  Ran it and it does look like there's a remnant
 of the old 6.0 system that lived on that drive at one time.

 So, how to *wipe* the MBR for that drive without losing everything on
 it? :-)  Is there a way?

 'Cause face it, if I get rid of the boot record on the drive, grub won't
 see it anymore even without my disabling the probe script. :-)

That is the cleanest way to do it!

Looking around a bit, I found the following:

https://bbs.archlinux.org/viewtopic.php?id=119702


The command to do it was

$sudo dd if=/dev/zero of=/dev/sdx bs=446 count=1

Something the poster didn't emphasize strongly enough (IMHO) is that
he's only wiping the first 446 bytes of the MBR -- this leaves the
partition table alone.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1bli3iyczp@snowball.wb.pfeifferfamily.net

Re: How to get rid of an entry in grub?

[Joe Pfeiffer]

David Guntner da...@guntner.com writes:

 Joe Pfeiffer grabbed a keyboard and wrote:
 David Guntner da...@guntner.com writes:
 
 Hugo Vanwoerkom grabbed a keyboard and wrote:
 David Guntner wrote:
 Hmmm.  I wonder if the MBR for the drive sill has a loader on it,
 even though I removed all partitions and repartitioned it?  Is there a
 utility out there that can wipe the MBR of a drive without touching the
 rest of the contents?

 There is http://bootinfoscript.sourceforge.net/ also.

 Ok, I got that, thanks.  Ran it and it does look like there's a remnant
 of the old 6.0 system that lived on that drive at one time.

 So, how to *wipe* the MBR for that drive without losing everything on
 it? :-)  Is there a way?

 'Cause face it, if I get rid of the boot record on the drive, grub won't
 see it anymore even without my disabling the probe script. :-)
 
 That is the cleanest way to do it!
 
 Looking around a bit, I found the following:
 
 https://bbs.archlinux.org/viewtopic.php?id=119702
 
 
 The command to do it was
 
 $sudo dd if=/dev/zero of=/dev/sdx bs=446 count=1
 
 Something the poster didn't emphasize strongly enough (IMHO) is that
 he's only wiping the first 446 bytes of the MBR -- this leaves the
 partition table alone.

 Seems like a good idea.  I tried it, but it doesn't look like it worked:

 # dd if=/dev/zero of=/dev/sdb bs=446 count=1
 1+0 records in
 1+0 records out
 446 bytes (446 B) copied, 0.000867678 s, 514 kB/s
 # update-grub
 Generating grub.cfg ...
 Found background image: /usr/share/images/desktop-base/desktop-grub.png
 Found linux image: /boot/vmlinuz-3.2.0-4-amd64
 Found initrd image: /boot/initrd.img-3.2.0-4-amd64
 Found memtest86+ image: /memtest86+.bin
 Found memtest86+ multiboot image: /memtest86+_multiboot.bin
 Found Debian GNU/Linux (6.0.7) on /dev/sdb1
 done

 Grub still seems to think there's Linux on /dev/sdb1.  That's
 aggravating..  I guess I'll just try moving the stuff off of the one
 and only partition on that drive to somewhere else temporarily, and then
 delete the partition outright and recreate it.  Maybe *THAT* will
 finally do it!

   --Dave

Ah, I missed that it was on a partition (and not the MBR).  What
*should* work is

$sudo dd if=/dev/zero of=/dev/sdb1 bs=512 count=1

More information:
http://www.novell.com/documentation/suse91/suselinux-adminguide/html/ch07.html


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1b4na6lhwy@snowball.wb.pfeifferfamily.net

Re: Moved MAC addresses

[Joe Pfeiffer]

M.Atıf CEYLAN meh...@atifceylan.com writes:

 On 07/25/2013 03:14 AM, Joe Pfeiffer wrote:
 You haven't said a lot about your topology; are both interfaces visible
 to the firewall through whatever series of switches you might have?
 Yes
 Actually, that pretty much has to be the case, otherwise your firewall
 couldn't see both MAC addresses.
 The both subnets are defined on the fw.

I guess I'm trying to think of a topology in which both cards are
visible to the firewall in which you're better off than just having a
single card.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1btxjibplu@snowball.wb.pfeifferfamily.net