Latest kernel update fixed crashes?

[Neal Murphy]

The latest kernel update seems to have fixed the panics and GPFs I was 
experiencing.

I was experiencing nearly predictable crashes whenever RAM was filled with 
cached disk blocks. At that point, it seemed that anything that addressed the 
cache would cause a crash: use a program that needed RAM or compile something. 
Even 'sync' would cause a crash. I think I even reinstalled the linux-image 
pkg, thinking something might've been corrupted.

I'm running Wheezy 64-bit. I had 8GiB of GSkill 9-9-9-24 RAM in my quad 
PhenomII, Gigabyte MA790FXT-UD5P. Because a memtest showed there might be a 
bad bit somewhere, I bought new RAM: 16GiB GSkill 7-7-7-21. It made no 
difference. As soon as there was pressure on RAM, the system would crash. 
Sometimes with triple faults. I was really beginning to doubt the CPU (memory 
controller) and other hardware.

Then the latest kernel update was released (3.2.51, I believe). I installed it 
and the problems ceased. I can build my firewall without crashes now; the 
complete build fills 16GiB RAM (cached disk blocks).

I perused the kernel changelogs from .46 to .51; nothing stood out to my 
apprentice's eyes.

So it comes down to two questions. Was there a kernel change that would 
account for this improvement in stability? Or is it more likely that some 
kernel/module was corrupted on disk and the new kernel erased that error, 
transparently fixing the 'problem'?

Thanks,
N


-- 
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/201310281709.53389.neal.p.mur...@alum.wpi.edu

Re: Init system deba{te|cle}

[Neal Murphy]

On Tuesday, October 29, 2013 05:48:20 PM Jonathan Dowland wrote:
> On Tue, Oct 29, 2013 at 04:55:44PM -0400, John wrote:
> > Could someone who has been following the giant fuss on -devel over
> > init systems explain why there's such a sense of dire urgency?
> 
> I think it's largely driven by frustration over how bipartisan the
> discussion is and how long it has been going on (it has been repeating
> over and over for years), combined with a desire on the part of most
> folks for Debian to move to *some* modern init system (the debate
> being, which).

I imagine part of the debate includes the fact that systemd integrates all 
kinds of systems and subsystems in an attempt to become the do-all and end-all 
of services control. This effort moves far away from the old UNIX concept of 
'do one thing and do it well'. Were I to be unkind, I would opine that systemd 
is an attempt to make Linux more like Windows, where everything has tentacles 
everywhere.

I think a 'next-gen' sysvinit could be developed--from sysvinit--that would 
satisfy most requirements of a services monitor, and continue to do what 
sysvinit was intended to do in the first place: start daemons and keep them 
running as best as possible without creating all manner of interdependencies.

Were one to look, on would see that init is really only used for two purposes 
today: (1) to restart gettys and (2) to inherit processes and reap zombies. It 
isn't used to start, maintain and stop daemons; when a daemon dies, it can be 
days before its absence is noticed.

So who is up to extend sysvinit to 16-32 run levels with an 'init tree' (or 
trees) instead of the current 'init table'? Extend it to follow its original 
purpose: start, stop and maintain daemons and services. Allow daemons to be 
started in parallel (up to the count of available CPUs), but add checkpoints 
to the tree that pause until pending parallel startups are complete before 
continuing to spawn more daemons (i.e., handle startup dependencies). And 
design it so that the tree is human-readable, so that it doesn't require a 
whole new suite of incomprehensible programs to (re)start and stop daemons. 
Design this 'next-gen' init to use the existing start/stop scripts in 
/etc/init.d. It's possible. It's doable. It would require learning a 'new way' 
but would be far less invasive than current alternatives.

N


-- 
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/201310291825.02817.neal.p.mur...@alum.wpi.edu

Re: Choosing default OS in multi boot system

[Neal Murphy]

On Wednesday, October 30, 2013 10:48:55 AM Richard Owlett wrote:
> Don't know if I'd call Grub2 bloated, but Grub-legacy was friendlier.

Maybe not bloated, but grub2 was certainly broken the last time I tried it. 
Specifically, (1) when I installed a system using grub2, it would install on 
the first drive it found grub on, never mind that I told it to install to a 
different drive. (2) I could never get it to work on an ISO, a thumb drive 
with the contents of the ISO (but not an ISO FS) and a hard drive. 3-4 weeks I 
spent building, testing, debugging. I gave up and went back to grub legacy. 
Four hours later I had it working as desired. If grub2 has improved in the 
past couple years (sounds like it hasn't), I'll eventually look at it again; 
if it hasn't, my system will remain with grub legacy (built with the huge 
RedHat patchset and a few of my own bug fixes).

N


-- 
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/201310301115.47020.neal.p.mur...@alum.wpi.edu

Re: Only in America! ? (was ... Re: sudo and UNIXes (was: audacity export wma format[1 more question]))

[Neal Murphy]

On Thursday, October 31, 2013 02:22:40 PM Chris Bannister wrote:
> On Mon, Oct 28, 2013 at 03:38:12PM -0600, Bob Proulx wrote:
> > Case 1: I find that someone in my family who lives in my house has
> > rumaged through my underwear drawer.  A violation of trust has
> > occurred.  I am unhappy and will talk with them and give them a harsh
> > lecture.  This is not appropriate behavior.
> > 
> > Case 2: I find someone who is not a member of my family and who does
> > not live in my house and who I don't know has rummaged through my
> > underwear drawer.  A very serious crime has been committed.  I live in
> > a state where I am fully legally protected if I shoot them dead.
> :
> :(
> 
> And yet you can't shoot the family member?  I think most murders are
> actually committed by family members against other family members, if my
> memory serves me correctly.
> 
> So you could shoot kids in halloween costumes for illegally being on
> your property?


An uninvited stranger in someone's home rifling through the occupants' 
belongings should expect to have a short life expectancy.

There is a certain amount of responsibility involved before one exercises the 
power to use arms (with rights come responsibilities). But we begin to digress 
from this list's purpose.


-- 
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/201310311514.05426.neal.p.mur...@alum.wpi.edu

Re: ANDROID (back to the OQ)

[Neal Murphy]

On Thursday, October 31, 2013 02:56:21 PM ken wrote:
> On 10/31/2013 02:02 PM Beco wrote:
> > On 31 October 2013 13:12, ken  > 
> > > wrote:
> > Alex,
> > 
> > As you can see (from this long conversation), there are a variety of
> > interpretations of what "free" means and its value to the end user.
> > Getting back to your original concerns, here are some observations
> > I've made about android which indicate to me that it's not free.
> > 
> > When you buy a phone with android on it, you don't have root access
> > to the system.  You're just a regular user.  Yes, you can root the
> > phone, but then you invalidate the warrantee, from what I
> > understand, both the software and hardware warrantees.  So if
> > something fails on your phone, the company whom you bought it from
> > won't provide support.  If something breaks (whether it's software
> > or hardware), you're on your own.  There are some android-specific
> > lists which could be helpful.
> > 
> > [cut]
> > 
> > 
> > Hello Ken,
> > 
> > I agree with you in all the topics bellow (the [cut]) but this one above.
> > 
> > The fact that you cant be root doesn't add to Android not being FOSS.
> > 
> > Lets say, for example, that you create an "enterprise" that makes
> > software (and hardware, to be more close to the example. Suppose you
> > build a small computer using go'old Z80 processor. The motherboard isn't
> > that big. You call it Z80-Alive! )
> > 
> > Now, you sell this machines in your community (school, church, whatever)
> > with a support contract, and state: You'll be THE only sysadmin, you'll
> > have root access and buyers will be a regular users. As long as buyers
> > don't try to gain root access, you'll give support to software and
> > hardware.
> > 
> > In some enterprises, if you try to get root access, you may be fired! :)
> > But Z80-Alive!, as someone buy the piece of hardware and you are just
> > helping out, the buyer can't (won't) be fired, just lose warranty.
> > 
> > Well, for me, this enterprise can't be called "not free" based only on
> > that.
> > 
> > I agree with the other topics in your email: closed softwares installed
> > without your agreement, and other stuffs (closed hardware, drivers,
> > etc). But to isolate the "feature" -->become root<--, suppose this
> > enterprise will only install FOSS, will only use public domain hardware,
> > and ask you if you are ready for an update before pushing it to your Z80
> > machine.
> > 
> > Avoiding users to become root is just a policy matter of an
> > organization, in which you are part.
> > 
> > Of course you can become root anyway and void warranty. That is not bad.
> > That is just an weighted conscious option.
> > 
> > My best,
> > Beco.
> 
> Beco,
> 
> This could get us into another abstract ontological discussion about
> what constitutes FOSS and how to define it... a sort of discussion I
> don't really care to engage in right now.  I'll just say that, in your
> example, perhaps the machine is free for you, but not free to those you
> sell it to.  And at work I might have root access to a FOSS system
> running a webserver, but visitors to that website don't.  True, this
> doesn't mean that it's not FOSS.  But I own the system, the visitors
> don't.  If someone else at work has root access to a machine and I'm
> just the DBA and don't have root access, true, it's still FOSS; if
> something's not right with the system, the sysadmin can change it
> (because that's his job); but he doesn't have to beg the whim of the
> owner or vendor of proprietary software.  So the distinction between
> FOSS and proprietary remains.
> 
> In the case of android, I've paid for the hardware and for someone to
> install and support the software and provide updates.  Vendors don't
> advertise the fact at all that you don't get root access and that,
> actually, other unseen people are controlling your phone.  And that's
> what it really comes down to-- who has control.  And this is a prime
> condition for FOSS, that *you* have full control of something you
> bought, not someone else.
> 
> It's also true that I could root my phone and accept that I've voided
> the warrantee.  But part of the purchase price I paid for the phone
> includes support and the reasonable guarantee that the hardware won't
> fail in the first year (or whatever the term paid for).  So by rooting
> my android, by simply taking control of something I paid for, I'm losing
> something else I paid for.  With FOSS, I think we could agree, this sort
> of conundrum doesn't arise.

In the US, the Magnusson-Moss act *may* act to invalidate such warrantee 
denials; 15-USC2302(c) is the specific section:

-
Prohibition on conditions for written or implied warranty; waiver by 
Commission 
No warrantor of a consumer product may condition his written or implied 
warranty of such product on the consumer’s using, in connection with such 
product,

Re: Hosting advice

[Neal Murphy]

On Friday, November 01, 2013 08:49:28 AM Craig L. wrote:
> Good points. The reason for going for hosting at the moment is it will
> give us a quick and easy solution. The reason for the Linux requirement
> is that we will be looking into a dedicated or virtual solution in the
> future. If I am going to manage it, it will be Linux.

If Q-n-E and cost are the primary requirement, you'd be hard-pressed to do 
worse than AdvantageCom. See http://simplywebhosting.com/plans-vps/index-
economy-hwvps.shtml: $60/year for a reasonably capable Xen VM running 64-bit 
Centos.

AdvantageCom also have a MailFoundry-based email filtration service for 
$60/year; this is very good at removing spam and has very few false positives.

So, for $10/mo., you can have spam- and virus-filtered email up and running in 
a day. But read their terms and conditions; there are some restrictions on 
email. In essence, they don't like spammers. If you are reasonably 
conscientious about keeping your VPS systems fairly secure, they'll inform you 
when they detect and disable spammers who've hacked into your system. If you 
don't care about security, they'll invite (and encourage) you to go elsewhere.

Check forum.schmolie.com to review their reliability.

I've been using them for years with almost no trouble.

N


-- 
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/201311011212.33040.neal.p.mur...@alum.wpi.edu

Re: Hosting advice

[Neal Murphy]

On Friday, November 01, 2013 01:46:26 PM Celejar wrote:
> I'm curious: how important is getting updates fast in the context of a
> server? I understand that for desktops, some want the latest features,
> or support for new hardware, etc. but servers? Doesn't it make more
> sense to just run something like Debian *stable*?

You're right; a server generally needs to be stable more than it needs to have 
the latest and greatest versions of software. Because Debian stable uses 
somewhat older versions of software, it tends to be quite stable (more 
reliable, requires fewer updates); security and bug fixes are the order of the 
week.

I run Debian stable on my desktop. My computer is a tool that must work. I 
almost never need the latest versions of software. Granted, my platform had 
problems with kernel panics and GPFs with the previous Wheezy kernel (it would 
crash when RAM--8GiB or 16GiB--was filled and cached disk blocks needed to be 
dropped). But the latest kernel eliminated the problem. I don't know if it was 
really a kernel problem or if some kernel files were corrupted on disk; if no 
one else ever had the problem, I must assume the latter.


-- 
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/201311011408.57211.neal.p.mur...@alum.wpi.edu

Re: sd card not detected

[Neal Murphy]

On Saturday, November 02, 2013 06:55:39 AM Alex Mestiashvili wrote:
> I use xfce4 4.10, but in this case I think it has nothing to do with the
> problem.
> The problem is that the device is not detected by the kernel .
> 
> it is not visible in the dmesg output and not visible in the output of
> fdisk -l.
> But other usb devices work fine - a usb stick is detected and mounted
> automatically.
> 
> ls -l /dev/sd*
> 
> brw-rw 1 root floppy 8, 112 Nov  2 11:32 /dev/sdh
> 
> /dev/sdh -is the cardreader device.
> 
> after inserting a sd card nothing happens - /dev/sdh1 doesn't appear,
> but it appears if I run fdisk -l /dev/sdh for example.

It appears as though udev is working correctly. If sdh1 appears when you plug 
in the reader with the card already inesrted, udev is definitely working.

What appears not to be working is the daemon that periodically opens and 
closes USB devices in the event that a medium was plugged in (used to be hald, 
I think). Remember that USB1 and USB2 are master/slave; the host end must 
initiate and perform all communication. If the daemon is malfunctioning, the 
system will not detect cards inserted into readers: the USB device cannot 
autonomously notify the host of the change in status.

The daemon works for me (64-bit Wheezy). (Verifying ... yes, it does work.)

You could probably simulate the daemon by running:

while true; do
  # Read a block; ignore failures
  dd if=/dev/sdh of=/dev/null bs=512 count=1
  sleep 3
done >/dev/null 2>&1


or even:

while true; do
  # Open the device; close it on success
  exec 3/dev/null 2>&1


Neither is optimal. Such a script should really look for non-partition USB 
additions to /dev/disk/by-id. When found, it should periodically open/close 
the device until success or until the device is deleted. Then start over.


-- 
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/201311021222.26166.neal.p.mur...@alum.wpi.edu

Re: Init system deba{te|cle}

[Neal Murphy]

On Saturday, November 02, 2013 08:23:45 AM Joel Rees wrote:

> I'm repeating myself, but good engineers don't do that.

No, they don't. They prepare new footings and pour a new foundation before 
moving the house to the new location.

It's nice to know I haven't misperceived the situation.


-- 
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/201311021228.39675.neal.p.mur...@alum.wpi.edu

Re: Installing same packages in a Squeeze installation in a new Wheezy installation

[Neal Murphy]

On Monday, November 04, 2013 01:20:01 PM Tony van der Hoff wrote:
> No, it didn't work for me. I would have much preferred to automate this.
> However, AFAICT the tools aren't available, and the time comes where you
> feel you're beating your head against a wall. That's when I gave up.
> 
> I'm not suggesting anyone should follow my path, it's dismal. I was
> merely relating my experience. If someone would contribute a WIKI on
> exactly how to do this, they'd win my eternal gratitude.

[This is an attempt to summarize various suggestions in a single msg.]

The theory is that --get-selections will list all installed pkgs in a form 
that can be used by --set-selections to (re)install them.

So, on the existing system:
  # Mount a thumb drive; change 'sdg' as needed
  mount /dev/sdg1 /mnt

  # Save the list of pkgs
  dpkg --get-selections > /mnt/current_installed_pkgs.txt

  # Umount, then unplug the drive
  umount /mnt

And on the new system, first netinstall a basic system. Then
  # Mount the thumb drive; change 'sdd' as needed
  mount /dev/sdd1 /mnt

  # Set the list of pkgs to install
  dpkg --set-selections < /mnt/current_installed_pkgs.txt

  # Unmount and remove the thumb drive
  umount /mnt

  # Start the 'upgrade'
  apt-get dselect-upgrade
When done, the new system should have the same pkgs as the old system. It 
won't be identical, but it'll be close.

This is almost straight out of dpkg's man page. If it doesn't work, perhaps a 
bug report should be opened.


To make the new system nearly identical to the old system, plug the new 
system's drive into the old system (hot-plug SATA or SAS is good; the new 
drive needs to be at least as large as the old drive).

  # Copy the current drive to the new drive; change sda and sdc as needed.
  dd if=/dev/sda of=/dev/sdc bs=1024k

Mount the new system's filesystems and adjust the hostname, IP address, etc. 
Then unmount and unplug. You may need to let it replay journals and fsck the 
FSen.

Plug the new drive into the new system and power it on. It should be a replica 
of the old, except for the ID bits you changed.


If you want to start with Squeeze, for example, and end up with Wheezy, it'd 
probably be better to get Squeeze's pkgs, install Squeeze on the new system, 
upgrade to Wheezy, then use the resulting list from --get-selections to 
install more Wheezy systems.

Note that this addresses pkgs, *not* configs. Configuring the new systems is a 
separate step.


-- 
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/201311041556.26034.neal.p.mur...@alum.wpi.edu

Re: No space left on device (28) but device is NOT full!

[Neal Murphy]

On Tuesday, November 05, 2013 02:21:36 AM Richard Hector wrote:
> On 05/11/13 16:51, Stan Hoeppner wrote:
> > Second, you have a serious problem here because it is your root
> > filesystem that has run out of inodes.  You need to ask yourself why you
> > have 1.7M files in your rootfs.  That's very dumb.
> 
> Or perhaps "That's not generally advisable." or similar.

'Contra-indicated' is a suitable military term. :)


-- 
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/201311050247.36507.neal.p.mur...@alum.wpi.edu

(64-bit) linux 3.10 and nouveau oddity

[Neal Murphy]

Due to GPF problems I've been having with wheezy's linux 3.2, I decided to try 
testing's 3.10 kernel (only installed the newer initramfs and 3.10 pkgs). 
Yesterday it ran fine all day, even built my firewall system without a lick of 
the troubles I've been having with 3.2. (So the GPF problem might be a linux 
3.2 bug. But that's fodder for another thread.)

[The log I present below is a contiguous section of syslog. There was nothing 
useful in messages.]

Today, I booted, read a forum thread about someone wanting to buy a new system 
and disabling CPUs. "Huh," thought I. "I wonder if disabling CPUs in the 
running system changes power consumption." I've toyed with disabling and 
enabling CPUs on prior versions of Debian with no trouble, so I figured 
there'd be no harm. Hah!

So I disabled CPUs 3-1 on my quad Phenom II. The immediate result was the 
message:


Nov  5 12:37:45 playground kernel: [ 2314.586965] do_IRQ: 3.65 No irq handler 
for vector (irq -1)
Nov  5 12:37:45 playground kernel: [ 2314.587039] smpboot: CPU 3 is now 
offline
Nov  5 12:38:11 playground kernel: [ 2341.205069] smpboot: CPU 2 is now 
offline


Immediately xterm started (re)displaying quite slowly. Otherwise, the system 
seemed to be OK.

But after a while (especially after I tried to start iceweasel), it seemed to 
hang. Then my X session disappeared and some nouveau messages appeared. So I 
switched to a text console; this took a while. It, too, kept hanging and 
unhanging. And the display would switch to show nouveau messages, then the X 
display would reappear a few seconds later, then the text console would appear 
in a couple seconds. And I'd have maybe a half-second to type stuff in the 
console (but it did seem to buffer keystrokes). The mouse was inoperative. I 
eventually rebooted to clear the problem.

Is it fair to say this is a bug? In *something*? Or just an incompatibility 
between Wheezy and the newer kernel?

I'm not complaining, mind you. Like the doctor said (after the patient 
groused, "Doctor, it hurts when I do dis"), "Well, don't do dat." Or, as the 
character Spock was known to intone, "Curious."

The rest of the syslog selection:


Nov  5 12:39:01 playground /USR/SBIN/CRON[5868]: (root) CMD (  [ -x 
/usr/lib/php5/maxlifetime ] && [ -d /var/lib/php5 ] && find /var/lib/php5/ -
depth -mindepth 1 -maxdepth 1 -type f -ignore_readdir_race -cmin 
+$(/usr/lib/php5/maxlifetime) ! -execdir fuser -s {} 2>/dev/null \; -delete)
Nov  5 12:40:14 playground kernel: [ 2463.527521] smpboot: Booting Node 0 
Processor 3 APIC 0x3
Nov  5 12:40:14 playground kernel: [ 2463.541412] LVT offset 1 assigned for 
vector 0x400
Nov  5 12:40:19 playground kernel: [ 2468.651015] smpboot: Booting Node 0 
Processor 2 APIC 0x2
Nov  5 12:52:30 playground ntpd[3327]: 0 out of 1 peers valid
Nov  5 12:52:30 playground ntpd[3327]: bad peer 10.20.30.1 (10.20.30.1)
Nov  5 13:09:01 playground /USR/SBIN/CRON[5958]: (root) CMD (  [ -x 
/usr/lib/php5/maxlifetime ] && [ -d /var/lib/php5 ] && find /var/lib/php5/ -
depth -mindepth 1 -maxdepth 1 -type f -ignore_readdir_race -cmin 
+$(/usr/lib/php5/maxlifetime) ! -execdir fuser -s {} 2>/dev/null \; -delete)
Nov  5 13:12:11 playground dbus[3133]: [system] Activating service 
name='org.freedesktop.NetworkManager' (using servicehelper)
Nov  5 13:12:11 playground dbus[3133]: [system] Activated service 
'org.freedesktop.NetworkManager' failed: Launch helper exited with unknown 
return code 1
Nov  5 13:17:01 playground /USR/SBIN/CRON[6106]: (root) CMD (   cd / && run-
parts --report /etc/cron.hourly)
Nov  5 13:27:43 playground kernel: [ 5315.329604] nouveau E[ DRM] GPU 
lockup - switching to software fbcon
Nov  5 13:27:43 playground acpid: client 4135[0:0] has disconnected
Nov  5 13:27:46 playground kernel: [ 5318.346648] nouveau E[Xorg[4135]] failed 
to idle channel 0x0001 [Xorg[4135]]
Nov  5 13:27:49 playground kernel: [ 5321.348690] nouveau E[Xorg[4135]] failed 
to idle channel 0x0001 [Xorg[4135]]
Nov  5 13:27:51 playground kernel: [ 5323.350749] nouveau E[   PFIFO]
[:01:00.0] channel 3 [Xorg[4135]] unload timeout
Nov  5 13:27:54 playground kernel: [ 5326.352083] nouveau E[Xorg[4135]] failed 
to idle channel 0x [Xorg[4135]]
Nov  5 13:27:57 playground kernel: [ 5329.354118] nouveau E[Xorg[4135]] failed 
to idle channel 0x [Xorg[4135]]
Nov  5 13:27:59 playground kernel: [ 5331.355750] nouveau E[   PFIFO]
[:01:00.0] channel 2 [Xorg[4135]] unload timeout
Nov  5 13:27:59 playground kdm[4115]: X server for display :0 terminated 
unexpectedly
Nov  5 13:28:00 playground acpid: client connected from 6283[0:0]
Nov  5 13:28:00 playground acpid: 1 client rule loaded
Nov  5 13:28:03 playground kernel: [ 5334.465574] nouveau E[iceweasel[6250]] 
failed to idle channel 0x [iceweasel[6250]]
Nov  5 13:28:06 playground kernel: [ 5337.467618] nouveau E[iceweasel[6250]] 
failed to idle channel 0x [iceweasel[6250]]
Nov  5 13:28:08 playground kernel: [ 5339.469093] nou

Re: No space left on device (28) but device is NOT full!

[Neal Murphy]

On Wednesday, November 06, 2013 07:44:18 AM Wawrzek Niewodniczanski wrote:
> This is a bit off main topic, but definitely 'on' for this list. Lets
> imagine a scenario there is nothing to delete on the troublesome
> partition, but there is another disk. What would be the best tool to
> move data to another partition having the same size, but higher number
> of inodes?

Assuming the problem is /var/log is part of the root filesystem and is crammed 
with millions of files. Assume other drive is /dev/sdb. The general process is 
as follows.

1. Reboot to single-user
2. Add partition #1 to /dev/sdb
3. 'mkreiserfs /dev/sdb1'  # to avoid the whole issue of inodes
4. 'mount /dev/sdb1 /mnt'
5. 'cd /var/log; find . -depth | cpio -pdv /mnt'
6. 'if [ $? -eq 0 ]; then cd ..; mv log log-; rm -rf log-; fi&'
7. 'mkdir log; chmod 755 log
8. 'echo "/dev/sdb1 /var/log reiserfs defaults,notail 0 1" >> /etc/fstab'
9. 'wait'
10. 'umount /mnt; init 6'


-- 
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/201311061143.09913.neal.p.mur...@alum.wpi.edu

Re: No space left on device (28) but device is NOT full!

[Neal Murphy]

On Wednesday, November 06, 2013 12:11:33 PM Beco wrote:
> On 6 November 2013 13:43, Neal Murphy  wrote:
> > Assuming the problem is /var/log is part of the root filesystem and is
> > crammed with millions of files. Assume other drive is /dev/sdb. The
> > general process is as follows.
> > 
> > 1. Reboot to single-user
> > 2. Add partition #1 to /dev/sdb
> > 3. 'mkreiserfs /dev/sdb1'  # to avoid the whole issue of inodes
> > 4. 'mount /dev/sdb1 /mnt'
> > 5. 'cd /var/log; find . -depth | cpio -pdv /mnt'
> > 6. 'if [ $? -eq 0 ]; then cd ..; mv log log-; rm -rf log-; fi&'
> > 7. 'mkdir log; chmod 755 log'
> > 8. 'echo "/dev/sdb1 /var/log reiserfs defaults,notail 0 1" >> /etc/fstab'
> > 9. 'wait'
> > 10. 'umount /mnt; init 6'
> 
> Hi Neal,
> 
> I think I'm going to ask about the easier part:
> 
> What is "9. wait" for?

You want the background delete to complete before rebooting. If it already 
finished, wait returns immediately.

N


-- 
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/201311061246.43209.neal.p.mur...@alum.wpi.edu

Re: Problem understanding/using dpkg-scanpackages

[Neal Murphy]

On Wednesday, November 06, 2013 07:11:24 PM Zenaan Harkness wrote:
> On 11/7/13, Richard Owlett  wrote:
> > I purchase complete DVD sets.
> > I am doing multiple clean installs to determine my "optimum"
> > solution.
> > Shuffling DVDs became a pain. I set aside a partition for myown
> > repository.
> > I copied the /dists and /pool directories from all DVDs in the
> > distribution set to that partition - it is mounted as
> > /media/repo6.

I've been following this for a while. How hard would it be to create a local 
web site that mirrors the essential parts of, say, ftp.us.debian.org? Then 
tell net install to load from that host. Would the net install then function 
as expected (albeit somewhat quicker)?


-- 
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/201311062010.00554.neal.p.mur...@alum.wpi.edu

Re: SSD as Cache?

[Neal Murphy]

On Friday, November 08, 2013 09:12:08 AM basti wrote:
> Hello,
> 
> on my Webservers, I have 1x 128GB SSD and a Raid 1 (1TB).
> 
> Now I plan to improve the performance of my Webapplication.
> The Cache is about 10.0 GB in 200 files.
> 
> Can this cache moved to SSD?
> Months ago I read articles about SSD and Flash Memory like:
> - "Disable logging on SSD"
> - "Disable cache on SSD"
> - "Don't swap on SSD" ...
> 
> And today?
> How long will the SSD work without data loss?

Using the -noatime mount option will extend the lifetime of the SSD. The final 
lifetime depends on how heavily the servers are used. If they're used 24x7 and 
the cache is always changing, you might get a year or two out of the SSD.

RAM is fairly cheap these days. Instead, if you can, increase RAM by 16GiB, 
leave the cache on rotating media, and let Linux cache the files in RAM. After 
that, performance improvements will come from fixing inefficient code.


-- 
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/201311081239.30381.neal.p.mur...@alum.wpi.edu

Re: Why Debian

[Neal Murphy]

On Sunday, November 10, 2013 03:54:31 PM Ralf Mardorf wrote:
> On Sun, 2013-11-10 at 14:27 -0500, Miles Fidelman wrote:
> > linuxfromscratch.org
> 
> IMO there's no need for a user to know all the details, however,
> something FreeBSD port like, e.g. Arch Linux IMO is more pleasant than
> Linux from scratch. It's also possible to learn about Linux, by using
> FreeBSD, e.g. Debian's FreeBSD instead of Linux. OTOH a user is free to
> start with a minimal Debian install, so Linux from scratch, Arch,
> FreeBSD etc. aren't needed to learn the details.

If one only wants to learn to use the system, Debian with XFCE may be one of 
the easiest.

But if one wants to learn how to *build* a working GNU/Linux system from 
source (that is, learning the steps involved), LFS may be hard to beat.


-- 
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/201311101603.14358.neal.p.mur...@alum.wpi.edu

Re: Questions about new hardware & Debian (or Linux in general)

[Neal Murphy]

On Monday, November 11, 2013 05:25:18 PM Karl E. Jorgensen wrote:
> Hi
> 
> On Mon, Nov 11, 2013 at 05:03:44PM -0500, Jon N wrote:
> > Hi,
> > 
> > I have been a Debian user for I'd guess 7 or 8 years now.  I would
> > like to thank all of you that helped create a system that is so
> > useful.  Despite using it for so long I am far from a power users, I
> > basically rely on the fact that it works well (or if it doesn't it
> > will soon :-)).
> > 
> > My current computer is also about that old and I am finally planning
> > on buying some new hardware.  A lot of stuff has changed in that time
> > that's new to me, so I would like to ask some general questions before
> > I purchase to save some potential headaches later.
> 
> Rule of thumb: stick with off-the-shelf stuff :-) You don't want to be
> the first guy with some esoteric hardware.

To expand on this point a little, buy mainboards, NICs (even if built-in) and 
video cards that were manufactured not much later than early 2012. Debian 
stable doesn't necessarily support all newer hardware.


-- 
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/20131822.24108.neal.p.mur...@alum.wpi.edu

Re: Questions about new hardware & Debian (or Linux in general)

[Neal Murphy]

On Monday, November 11, 2013 09:28:26 PM Jon N wrote:
> Neal,
> 
> Well, I think you've found my weak point.  I have been looking at
> fairly new stuff.  When shopping for processors I found there was
> little difference in price between Intel's Haswell verses Ivy Bridge
> (at least for Pentiums, which what I'm thinking of getting).  So the
> mainboards have new chipsets (either H81 or H87).  Is the chipset an
> issue?

It can be. If the chipset isn't supported, the kernel mayn't know to or know 
how to walk the device tree, or simply know how, to initialize all the buses, 
bridges, and devices. Alas, I was unable to find any document anywhere that 
states which chipsets (even just the 20 newest) that a kernel supports when I 
looked 2-4 years ago.


-- 
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/20132209.37485.neal.p.mur...@alum.wpi.edu

Re: who uses dual boot? [was: How to start using a free OS]

[Neal Murphy]

On Tuesday, November 12, 2013 03:06:23 PM Nemeth Gyorgy wrote:
> 2013-11-12 14:32 keltezéssel, Miles Fidelman írta:
> > That's a very interesting point, but I wonder if it's true.  There are
> > real-world reasons to run both windows on linux on the same machine
> > (personal example: running Linux on my laptop for development and
> > demonstrations; running Windows for office applications).
> > 
> > But, having said that, when one really uses two operating systems on the
> > same machine, I expect it's more common to run one under virtualization,
> > so you can run both at the same time - dual booting is a real pain if
> > one is really USING both operating systems.
> 
> There can be a lot of reasons to use natively two operating systems on
> the same computer. One can have hardware which is handled only by
> Windows for example. Virtualization is a solution sometimes but there
> are always overheads and drawbacks. Sometimes it is not a real problem,
> in other cases it is.

With working hardware virtualization support and adequate RAM, the difference 
should be nearly undetectable. This is true running Debian in vbox on WinXp on 
my 2006 HP laptop w/2GiB RAM, and is true running 6 KVMs on Wheezy on my quad 
Phenom II desktop w/16GiB RAM and 3TB of storage. The speed difference may be 
measurable, but it is imperceptible for all intents and purposes.

If you do not have hardware virt. support, then VMs will run much slower.


--
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/201311121530.48877.neal.p.mur...@alum.wpi.edu

Re: (64-bit) linux 3.10 and nouveau oddity

[Neal Murphy]

On Friday, November 08, 2013 06:13:44 AM berenger.mo...@neutralite.org wrote:
> Le 05.11.2013 21:26, Neal Murphy a écrit :
> > Is it fair to say this is a bug? In *something*? Or just an
> > incompatibility
> > between Wheezy and the newer kernel?
> 
> It is indeed a bug, if it was known to work previously.
> I would have thought about a nouveau bug, but by reading the logs, I do
> not know... lot of non graphic-related stuff failing too.
> 
> I think to ensure that it's not nouveau's bug, you can try to use
> another driver, like, say, fbdev or vesa. Things will be slower of
> course, since there won't be as efficient hardware acceleration, but if
> it is nouveau's bug, the behavior will disappear. Otherwise, it could be
> xorg, but I strongly doubt it, or the kernel itself, which is more
> 
> probable since even non graphical stuff are buggy according to the log:
> > Nov  5 13:12:11 playground dbus[3133]: [system] Activating service
> > name='org.freedesktop.NetworkManager' (using servicehelper)
> > Nov  5 13:12:11 playground dbus[3133]: [system] Activated service
> > 'org.freedesktop.NetworkManager' failed: Launch helper exited with
> > unknown
> > return code 1
> 
> Or maybe the org.freedesktop.NetworkManager service is linked with
> xorg-dbus? With a lot of modern applications comes strange
> dependencies... and so wrong designs, but that's another problem that
> can't be fixed :)

Hmmm. It ain't doin' it now. On either 3.2 or 3.10 kernels. The only system 
change: I removed two sticks of RAM. Which seems to have stabilized my system. 
I guess my mainboard just can't handle having all four slots filled and was 
giving random errors.

But I did uncover a potential bug in top(). Start it and hit '1' to show all 
the CPUs. Then stop all but CPU-0 (which you can't stop anyway). top() 
dutifully removes the extra CPUs. Now re-enable the other CPUs and see that 
top doesn't resume displaying the other CPUs.


--
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/201311130153.00062.neal.p.mur...@alum.wpi.edu

Re: eth0 "disabled" after upgrade Squeeze to Wheezy

[Neal Murphy]

On Friday, November 15, 2013 02:18:26 PM Glenn English wrote:
> On Nov 15, 2013, at 12:01 PM, Lisi Reisz  wrote:
> >> Re-adding the two etho lines should fix it all for you - unless a
> >> reboot deletes them, somehow!
> > 
> > it did. :-(
> 
> Noob here. How about fixing /etc/network/interfaces, removing all write
> permissions from the file, rebooting, and looking for write errors? In
> hopes of finding what's writing over the file...

Close, Glenn. Set the file's 'immutable' bit so it *can't* be changed or 
deleted. That should surely raise the hackles of the quisling.
--
mkdir /run/network
# edit /etc/network/interfaces to restore eth0

chattr +i /run/network /etc/network/interfaces
--


-- 
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/201311151512.44290.neal.p.mur...@alum.wpi.edu

Re: eth0 "disabled" after upgrade Squeeze to Wheezy

[Neal Murphy]

On Friday, November 15, 2013 11:38:16 AM Lisi Reisz wrote:
> Thanks, Andrei,
> 
> On Friday 15 November 2013 16:15:20 Andrei POPESCU wrote:
> > On Vi, 15 nov 13, 15:06:59, Lisi Reisz wrote:
> > > I have just upgraded a client's computer to Wheezy.  It appeared
> > > to go well and there was certainly an internet connection: it
> > > would not have been able to upgrade otherwise!
> > > 
> > > Now there is  none.  I have checked /etc/network/interfaces and
> > > changed "allow-hotplug" to "auto", just for something to try. :-(
> > > 
> > >  It made no difference.  I pinged the gateway, largely so that I
> > > 
> > > could report that I had done so.  I got the error message
> > > "Network is unreachable".
> > > 
> > > KControl tells me that eth0 is "disabled".  How and why is it
> > > disabled?  More importantly, **how do I enable it?**
> > 
> > How is the network configuration handled? Network Manager,
> > ifupdown, etc.?
> 
> /etc/network/interfaces file and ifupdown, etc..  (Network Manager and
> I are not on speaking terms.)
> 
> Lisi

'ls /sys/class/net' to see all known network devices.
'ip link' to see their states: up/down, (no)carrier, PtP/broadast, et alia

If eth0 is there and appears to be a broadcast medium with carrier present, 
you can manually bring it up and verify connectivity:
-
ip link set dev eth0 up
ip addr add 192.168.0.2/24 dev eth0
ip route add default via 192.168.0.1
ping 8.8.8.8
-

If it works, then the immutable bit in my previous message should aim your 
20mm cannon on the guilty culprit.


-- 
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/201311151522.42892.neal.p.mur...@alum.wpi.edu

Re: Wheezy/XFCE: remote desktop service difficulties with VNC server

[Neal Murphy]

On Sunday, November 17, 2013 03:53:14 PM Reco wrote:
> On Sun, 17 Nov 2013 20:15:02 +
> 
> Ron Leach  wrote:
> > X could not detect the attached screen because its cable is switched
> > across a KVM which seems to destroy the EDID information; I'd already
> > manually configured a Modeline for 1440x900.  (This works fine on the
> > attached screen even through the KVM.)  The remote desktop appears to
> > be 4x3 shape, and something closer to 800x600, I would guess.
> 
> Vnc4server doesn't (and should not) take into account any EDID. The
> entire point of VNC is to be able run even if video card(s) is
> physically absent at the host.
> Try experimenting with '-screen' and '-dpi' VNC options.

To expand, vncserver creates an X11 server session that uses a virtual video 
card (in effect). Hence the 'virtual' in 'virtual network computing'.

'vncserver -geometry 1024x300' will create a 1024 wide x 300 high virtual 
display available via :1; 'vncviewer :1' will dutifully display it in a 
1024x300 window (plus trimmings).

VNC isn't rdesktop. As far as I know, the viewer can only display the virtual 
X session.

If you had a local X server configured to use a 2x2 matrix of 1920x1080 
displays as one logical display, you could have it run vncviewer to display a 
remote 3840x2160 VNC session fullscreen. (In theory, at least.)


-- 
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/201311171629.08146.neal.p.mur...@alum.wpi.edu

Re: Automatic installs

[Neal Murphy]

On Sunday, November 17, 2013 04:20:25 PM Bob Proulx wrote:

> P.S. Yes I know mixing awk and grep is silly since awk can do it all.
> 
>   dpkg --purge $(dpkg -l | grep ^rc | awk '{print$2}' | grep ^lib)
> 
> I normally would have said this and done it all with awk.
> 
>   dpkg --purge $(dpkg -l | awk '/^rc/&&$2~/^lib/{print$2}')
> 
> But I didn't think that was a clear of intent as the former in an
> email.  So I went with the more obvious logic even though it mixed
> commands in a silly way.

If the command was intended to be run once every second, the latter might be 
preferred. But since absolute efficiency isn't required, the former is clearer 
and is far more easily understood by non-experts.

Actually, if efficiency was important, the --purge option would accept a 
regex. Or there'd be a --purgex option.

Enough pedantry for now. :)


-- 
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/201311171639.10897.neal.p.mur...@alum.wpi.edu

Re: debian-user-digest Digest V2013 #1401

[Neal Murphy]

> help

Could you be a little more specific? The dearth of data prevents us from 
offering meaningful, targetted assistance.

This has be the least informative request ever seen in a community support 
forum.

. . .

Oh, wait. Mayhap the poster was trying to get the mailing list's 
command/operation help sheet.

Never mind.

:) (for the humor-impaired)


-- 
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/201311171708.04838.neal.p.mur...@alum.wpi.edu

Re: Finding pertinent inforation on WEB

[Neal Murphy]

On Tuesday, November 19, 2013 12:45:53 PM Richard Owlett wrote:
> Linux-Fan wrote:
> > Also, you can google "GNU Info Pages online" without quotes. I just
> > found http://linux.about.com/od/lts_guide/a/gdelts69t02.htm by entering
> > that query which also suggests the first page I have mentioned.
> > 
> > HTH
> > Linux-Fan
> 
> You are right on so many counts ;)
> 
> I was looking, this time, for information on GRUB. I had done a
> half dozen Google searches, but with poor choice of keywords.

I've found that searching for 'man topic' takes me right to many pages with 
topic's manuals. linux.die.net, tldp.net, and even man.cx now, are a few other 
typical sites.

I've also found that info pages on debian often have nothing more than the 
respective man page, and man pages often contain cursory information. 
Sometimes I think it should be a bug to have a useless man page in place.


-- 
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/201311191306.05623.neal.p.mur...@alum.wpi.edu

Re: Extending fluxbox menu with a script.

[Neal Murphy]

On Thursday, November 21, 2013 01:15:34 PM Sharon Kimble wrote:
> I am trying to write a bash script which has this line in it -
> 
> mv ~/.fluxbox/menu ~/.fluxbox/menu-$(/bin/date +%Y%m%d-%R); mmaker
> fluxbox -f;;
> 
> This line is creating a fluxbox menu which ends with these lines -
> ' [restart] (Restart)
>   [separator]
>   [exit] (Exit)
>   [end]
> [end]'
> 
> What I want to do is to add this script onto the end of the first
> command, and it is -
> #!/bin/bash
> # title - submenu
> 
> echo  [submenu] (My Menu)
> echo  [include] (~/.fluxbox/usermenu)
> echo  [end]
> echo  [separator]
> 
> So the final running command will look like this -
> 'mv ~/.fluxbox/menu ~/.fluxbox/menu-$(/bin/date +%Y%m%d-%R); mmaker
> fluxbox -f; submenu;;'
> 
> and the main menu will look like this at the end -
>   [restart] (Restart)
>   [separator]
>   [exit] (Exit)
>   [end]
>   [submenu] (My Menu)
>   [include] (~/.fluxbox/usermenu)
>   [end]
>   [separator]
> 
> But how do I do it please?

sed is your friend. If I've followed your message (you want menu-`date` to 
contain the submenu), this should do the trick--*provided* the last [end] is 
the last line in the file:

mv ~/.fluxbox/menu ~/.fluxbox/menu-$(/bin/date +%Y%m%d-%R); \
sed -i -e '$ i \
\t\t\[submenu] (My Menu)\
\t\t\[include] (~/.fluxbox/usermenu)\
\t\t\[end]\
\t\t\[separator] ~/.fluxbox/menu-$(/bin/date +%Y%m%d-%R); \
  mmakerfluxbox -f; submenu;;

If the last [end] is the only one in the file at the beginning of the line, 
change the first line of the sed chant to:

sed -e '/^\[end]/ i \

(Instead of addressing the last line of the file, address the line with [end] 
at B.o.L.)

N


-- 
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/201311211735.38957.neal.p.mur...@alum.wpi.edu

Re: MIT discovered issue with gcc

[Neal Murphy]

On Saturday, November 23, 2013 04:23:05 PM Stan Hoeppner wrote:

> I didn't read the full paper yet, but I'm wondering how/if the
> optimization flag plays a part in this.  I.e. does "O2" produce these
> bugs but "OO" (default) or "Og" (debugging) does not?

Or -O3...


-- 
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/201311231636.18718.neal.p.mur...@alum.wpi.edu

Re: A rookie's query: Want to about Debian and the related

[Neal Murphy]

On Tuesday, November 26, 2013 11:27:03 AM Lisi Reisz wrote:
> On Tuesday 26 November 2013 16:03:54 AP wrote:
> > On Mon, Nov 25, 2013 at 12:05 AM, Andrew Wood
> > 
> >  wrote:
> > > After many years of using Linux on servers and my primary desktop
> > > I would only recommend Debian. Its solid and reliable, other
> > > distros ive found to be very buggy
> 
> [snip]
> 
> > > My primary machine is only a Core 2 Duo but its running Debian 7
> > > with Gnome 3 without any issues and is more than adequate for
> > > LibreOffice and web browsing speedily
> > 
> > Oh well. By saying solid and reliable, you mean it is worth
> > installing.
> 
> It is certainly worth installing, but what he means, I think, is that
> it will rarely, if ever, crash, be almost bug-free (nothing is
> totally bug-free in all possible situations), has good security
> updates and can be relied on.  It Just Works.
> 
> My husband, who likes a trouble-free life, is quite clear that he
> wants the stable version of Debian.   (He gets Debian anyway.)
> 
> If you want the latest, greatest, newest hardware and software, then
> Debian Stable is not for you.  If you want reliability, an ability to
> run on almost anything, and a peaceful life, then Wheezy - 7.2 - (the
> current Stable) is the one for you.
> 
> Lisi

Well met good lady! You've stated it about as succinctly as could be done.


-- 
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/201311261136.31510.neal.p.mur...@alum.wpi.edu

Re: A rookie's query: Want to about Debian and the related

[Neal Murphy]

On Tuesday, November 26, 2013 11:33:18 AM Lisi Reisz wrote:
> On Tuesday 26 November 2013 16:25:08 AP wrote:
> > On Tue, Nov 26, 2013 at 7:55 PM, Lisi Reisz 
> 
> wrote:
> > > Much depends on whether you (AP) are ready to get your feet wet
> > > and use the CLI (command line interface).  By all means stick to
> > > the GUI (graphical user interface) for installing things if that
> > > is your preference.
> > 
> > Because I guess CLI is a bit more typical and for advanced users
> > but ok...slowly-slowlyI would try to
> 
> Some of us are old enough for the command line to be more familiar.
> DOS anyone? ;-)

Never liked it. Going back a little further, VMS, AOS/VS, AOS, RT-11, TOPS-20, 
TOPS-10. PDP-11 where two people could play a space wars game using the 
console switches, where entire programs could be toggled in using the switches 
(CLI not needed).

GUIs have their uses; but they limit what one can do. If one wants to learn 
how a computer works, CLI is nearly the only way to go.

N


-- 
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/201311261144.52550.neal.p.mur...@alum.wpi.edu

Re: split(): syntax error near unexpected token `(' ... [OT]

[Neal Murphy]

Bash does not have a split function. To split in bash, one has to do it the 
old-fashioned way:

declare -i i
declare -a myArray

OIFS=$IFS
IFS=":"
set `egrep root /etc/passwd`
IFS=$OIFS

# Store the new positional parameters in indexed array myArray.
i=0
while [ ! -z $1 ]; do
  myArray[$i]=$1
  shift
  i=i+1
done

# Display the elements of myArray
i=0
while [ $i -lt ${#myArray[@]} ]; do
  echo ${myArray[$i]}
  i=i+1
done


-- 
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/201311282331.22018.neal.p.mur...@alum.wpi.edu

Re: Won't complete bootup (gdm3 problem?)

[Neal Murphy]

On Friday, November 29, 2013 02:44:58 PM Bob Proulx wrote:

> Now some people might claim that GNOME *is* the operating system.  But
> don't believe it.  It is not.  This is proven by the number of people
> that use Debian every day but do not have GNOME on the system at all.

I would say they don't know the difference between an operating system and a 
graphical user interface. Generally speaking, an operating system provides 
controlled access to system (hardware) resources. A grahical user interface is 
supposed to provide a pleasant and efficient intermediate layer between the 
user and the OS; but not all succeed.

User interfaces include front panel switches (ala PDP-10/PDP-11), card 
punches, console (command line interface), and graphical. There are even 
hybrids. But they certainly are not operating systems, as you said.


-- 
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/201311291526.17623.neal.p.mur...@alum.wpi.edu

Re: Won't complete bootup (gdm3 problem?)

[Neal Murphy]

On Saturday, November 30, 2013 02:57:13 PM Ralf Mardorf wrote:
> On Sat, 2013-11-30 at 20:43 +0100, Ralf Mardorf wrote:
> > On Sat, 2013-11-30 at 00:16 +, Joe wrote:
> > > I haven't tried KDE for a while, and I'm not really interested in a big
> > > beast. I do keep a Knoppix DVD handy for emergencies, but it has used
> > > LXDE for some time now.
> > 
> > Yesterday I installed Razor-Qt using KWin to one of my Linux installs.
> > There are issues with the mouse theme and while the install should be in
> > English only, some Window titles and bottoms are in German, but the
> 
>   ^^^ buttons
> 
> > Window content is in English. FWIW, the mouse cursor theme is ok when
> > the cursor is above the desktop, but when it's above a Window, the
> > cursor theme changes to an unwanted theme. If there wouldn't bee those
> > issues, it likely would become my new DE. JFTR, it's a Kubuntu install,
> > I'll try Razor-Qt with Debian and/or Arch Linux, to see if the issues
> > are *buntu related.
> > 
> > Regards,
> > Ralf

That's the only mistake? Gotta say, Debian's voice recognition is much better 
than most smart phones. I'd hate to see how a smart phone would've garbled 
Ralf's text. :D


-- 
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/201311301518.19159.neal.p.mur...@alum.wpi.edu

Re: A rookie's query: Want to about Debian and the related

[Neal Murphy]

On Monday, December 02, 2013 11:38:47 PM Celejar wrote:
> On Mon, 02 Dec 2013 14:17:20 +0100
> 
> Ralf Mardorf  wrote:
> > On Mon, 2013-12-02 at 08:09 -0500, Celejar wrote:
> > > Sorry, don't know rocketmail.
> > 
> > Nowadays occupied by Yahoo, POP and SMPT settings seems to be equal.
> 
> Don't understand this statement.

Yahoo bought rocketmail?


-- 
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/201312022347.35899.neal.p.mur...@alum.wpi.edu

Re: A rookie's query: Want to about Debian and the related

[Neal Murphy]

On Tuesday, December 03, 2013 02:04:18 AM Ralf Mardorf wrote:

> Sorry for the extra broken English, I've got a cold :S and better don't
> correct or try to rephrase my mails

I quite understand. My fingers get stuffy. What I type seems oddly, mmm, 
distant, blurred. And my messages are often garbled when my hands have 
coughing spasms. And that's when they're awake; most of the time they fall 
asleep. :)

Hope it passes quickly.

(Your extra-broken English is still better than that of many native writers of 
it.)

N


-- 
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/201312030234.44280.neal.p.mur...@alum.wpi.edu

kernels 3.0, 3.2, 3.4 won't boot in KVM

[Neal Murphy]

I've encountered a problem with Debian Squeeze. Wheezy doesn't seem to have a 
bigmem kernel, so I can't test it there.

I have a custom firewall based on Smoothwall. I've been testing it (kernel 
2.6.35 with PAE and SMP) on Squeeze in KVM without trouble for a couple years.

When I upgraded my firewall's kernel to 3.0, it would no longer boot in KVM. 
Instead, it hangs right after displaying the first three lines on the console, 
and the host CPU goes to 100%. Trying other kernels, I found that none of 3.0, 
3.2 or 3.4 will boot.

Investigating further, I found that those kernels *will* boot if I use 
Debian's x86 or x86_64 kernel. But they *won't* boot when using Squeeze's 
bigmem (PAE) kernel.

I tried both Squeeze's kvm as well as a build of qemu-kvm 1.1.x.

In a sentence, my 3.x kernels with PAE and SMP will not boot in KVM on Squeeze 
when using a bigmem (PAE) kernel, but they *do* boot when using Squeeze's 686 
or x86_64 kernel, as well as Wheezy's x86_64 kernel.

Can anyone verify this, or is it just me? Is it possible I missed a kernel 
config parameter required to allow the PAE-enabled kernel to boot in a KVM 
session? Could it be a Debian build problem? An upstream kernel problem? Or a 
qemu-kvm problem?

Here's an example ISO (29MiB, linux 3.4.9):
  http://roadster.agcl.us/release/agcl-0.0.0-offroad.iso
Select the 'Explore' option from the boot menu.

Thanks!


-- 
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/201209121704.41004.neal.p.mur...@alum.wpi.edu

Re: dpkg/apt voodoo to ask "what version of is installed, if any?"

[Neal Murphy]

On Wednesday, September 12, 2012 04:54:27 PM Andrei POPESCU wrote:
> On Mi, 12 sep 12, 15:44:59, Kris Deugau wrote:
> > Is there a single command that can do this for both virtual and real
> > packages, a la "rpm -q --whatprovides"?
> 
> I have no idea what that command does, can you provide an example?

Try 'dpkg-query -s name' and 'dpkg-query -S name'.


-- 
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/201209121716.17249.neal.p.mur...@alum.wpi.edu

Re: deprecated (Re: What Version To Install On iMac?)

[Neal Murphy]

On Wednesday, September 12, 2012 07:44:03 PM Chris Bannister wrote:
> > How do you call it when software or an administrator is being
> > deprecated, i. e. the process of deprecating something/making something
> 
> I can only think of the word "redundant" (surplus to requirements) at
> the moment. e.g. The administrator was made redundant because of
> restructuring.

I believe the idiom you are looking for is 'put out to pasture': it is no 
longer needed in any way and serves no function, but remains until death and 
the knacker has carted it off.


-- 
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/201209122115.48848.neal.p.mur...@alum.wpi.edu

Re: Installation

[Neal Murphy]

On Tuesday, September 18, 2012 01:06:33 PM Ross Boylan wrote:
> On Tue, 2012-09-18 at 10:03 -0400, The Wanderer wrote:
> > At its root, my objection in this subthread isn't necessarily to the
> > information
> > and functionality available in the current installer, but to the
> > implied
> > statement that "you shouldn't install if you don't have a second,
> > already-working computer" - or at least that, if you try to install
> > without
> > having one, you're doing it wrong.
> 
> While having a second computer is both helpful and likely, my somewhat
> dated experience with the installer doesn't suggest it is essential.  I
> remember someone saying "a pigeon could install Debian" because if you
> just click to accept the default responses you're likely to end up with
> a reasonable--not necessarily optimal, but reasonable--system.
> 
> The pigeon thing is probably a bit strong, since you'll need a machine
> name and possibly networking info, but then you would with MS Windows
> installs too.
> 
> Finally, there are two other routes, though probably neither is natural
> for a newbie.  You could install a system, use it a bit (to read
> installer docs), and then install again, over the first.  Or you could
> make an initial install into a small partition and then do the "real"
> install.

Or to best learn: install, find the things you missed and did wrong and 
install again, then find more things you missed and done did wrong and install 
a third time. Along about now, you'll've a rather firm grasp of the install 
essentials. It's a given that you'll need a local install CD/DVD if you don't 
have a fast internet connection. Or a decent local debian mirror.


-- 
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/201209181335.13639.neal.p.mur...@alum.wpi.edu

Re: networking with virtual machine

[Neal Murphy]

On Tuesday, September 18, 2012 03:25:45 PM lee wrote:
> Chris Davies  writes:
> > lee  wrote:
> >> Yes and when I replace the interface I have now (eth1) with a bridge
> >> device (br1), then how do I tell shorewall that the guest is in the dmz
> >> (for example)?
> > 
> > You need "bridge" and "routeback" set in your shorewall interfaces file.
> 
> Ok, all the examples in the shorewall documentation I'm seeing say that
> I need the "routeback" option with bridge devices.  I'm fine with that.
> 
> This option doesn't tell me how to treat the bridge device as two
> different interfaces which seem to be needed for shorewall to work.
> 
> All the examples in the shorewall documentation I'm seeing assume that I
> would have several interfaces rather than only one.
> 
> > Take a look at http://www.shorewall.net/SimpleBridge.html and
> 
> This example uses two interfaces while I would have only one.
> 
> > http://www.shorewall.net/KVM.html.
> 
> This example isn't really explained.  It refers you to [1], which also
> requires two interfaces.  There is other information linked to it which
> brings tunneling/tapping stuff into the setup and doesn't explain
> anything about that.  The example script it refers to is probably
> deprecated: It's 4 years old, and there are already start-scripts for
> qemu/kvm in effect in Debian.
> 
> Do I need tunneling/tapping?
> 
> 
> [1]: http://www.shorewall.net/two-interface.htm
> 
> > I think that the second reference will be particularly useful for you
> > - ignore the references to wlan0, and replace "eth0" and "br0" with
> > "eth1" and "br1" respectively.
> 
> Well, [2] even says clearly:
> 
> 
> 1.) "IP addresses are properties of systems, not of interfaces."
> 
> 2.) "All IP addresses configured on firewall interfaces are in the $FW
> (fw) zone."
> 
> 
> Number 2.) is definitely *not* what I want.
> 
> Would I need to create a tunneling/tapping interface for the host and
> one for each guest to circumvent 2.)?  Would that be safe to do?  Would
> that be better than using the currently unused physical interface eth0
> instead of the currently used eth1 to make a(n independent) bridge
> device? --- I'm probably not going to have more than two guests running
> at the same time.
> 
> 
> [2]: http://www.shorewall.net/two-interface.htm

A bridge device merely connects two or more devices--be they real NICs, TAPs, 
virtual NICs or some such--at layer 2.

On my desktop, I have 3 NICs assigned to 3 bridges, and a fourth bridge 
without a NIC. I can start KVM sessions tapped into any combination of bridges 
for testing multi-zone firewalls and other systems. The host creates the 
bridges and assigns the NICs to them. (I turned off NetworkManager because it 
always interferes.) I named the bridges in the Smoothwall fashion: RED, GREEN, 
ORANGE and PURPLE. I anally named the taps similarly (and included the KVM 
instance #) so I can see what's connected to what. I generate explicit MAC 
addresses for each virt. NIC so I know which bridge they're connected to; the 
addr includes the KVM instance # and the virt. NIC number. The script deals 
with booting from HD, ISO, CD/DVD, USB/flash, etc. Only took about 300 lines 
of bash script. The only thing I haven't yet figured out how to do remotely is 
attach and detach USB devices to/from VMs (to perform plug-n-play backups, 
installs and restores).

The result is that every VM is addressable from any LAN and can even have 
ports forwarded to it from the perimeter F/W. I can daisy-chain firewalls and 
still forward ports from the perimeter to the innermost system. I can simulate 
systems connected through single-, double-, and even triple-NATting. I can 
simulate the public internet by using real addresses on the bridge that has no 
NIC.

So yes, if you want 'real' networking, you'll need bridges and taps. The only 
drawback is that bridges seem to have limited throughput (I'd expect my quad 
PhenomII to do better than 90Mb/s); I haven't yet seen GigE speeds from the 
virtual GigE NICs through to any of the wired GigE LANs, or even across the 
bridge.


-- 
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/201209181613.38416.neal.p.mur...@alum.wpi.edu

Re: networking with virtual machine

[Neal Murphy]

On Tuesday, September 18, 2012 05:59:47 PM lee wrote:
> Neal Murphy  writes:
> > So yes, if you want 'real' networking, you'll need bridges and taps.
> 
> Thank you, I'll have to look into taps then.
> 
> Do you think it's a good idea to just create a bridge device with the
> unused eth0 for this?  I could leave eth1 as is and would basically only
> have to add a zone and appropriate policy and rules in the shorewall
> configuration.

If that is the only firewall method you have then yes, enable forwarding, add 
the bridge to a second shorewall zone, and add iptables rules that drop, 
reject, allow and deny traffic as you desire. All of your VMs can easily be 
tapped into the bridge.


-- 
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/201209191045.04927.neal.p.mur...@alum.wpi.edu

Re: networking with virtual machine

[Neal Murphy]

On Wednesday, September 19, 2012 08:42:57 PM lee wrote:
> Neal Murphy  writes:
> > On Tuesday, September 18, 2012 05:59:47 PM lee wrote:
> >> Neal Murphy  writes:
> >> > So yes, if you want 'real' networking, you'll need bridges and taps.
> >> 
> >> Thank you, I'll have to look into taps then.
> >> 
> >> Do you think it's a good idea to just create a bridge device with the
> >> unused eth0 for this?  I could leave eth1 as is and would basically only
> >> have to add a zone and appropriate policy and rules in the shorewall
> >> configuration.
> > 
> > If that is the only firewall method you have then yes, enable forwarding,
> > add the bridge to a second shorewall zone, and add iptables rules that
> > drop, reject, allow and deny traffic as you desire. All of your VMs can
> > easily be tapped into the bridge.
> 
> The router has a firewall and I'm running shorewall on the host behind
> that.  It should be save enough, and it gives me some things like
> traffic shaping which the router doesn't do.  I'm not doing firewall
> testing and like to keep things simple.
> 
> So now I know which way to go and what to read about, thanks :)

  o Remember to put the bridge and VMs in a LAN different from your NIC.
And be sure there's a route to the bridge LAN on your router/FW so
it knows where to send reply packets; if you wanted to be fancy, you
could NAT the bridge from ethX, but that's a lot more work (port
forwarding, SNAT, DNAT, et alia.)
  o The bridge's IP addr (on the host) is the default gateway for all VMs
on that bridge.


-- 
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/201209192159.27665.neal.p.mur...@alum.wpi.edu

Re: Display hurtful on LCD screen with Wheezy

[Neal Murphy]

On Thursday, September 20, 2012 03:55:48 PM Lionel Trésaugues wrote:
> > And the sub-pixel order is equal too?
> 
> The sub-pixel order is set to rgb (like I do in Ubuntu). I can
> definitely see a (worse) change if I use a different mode.
> Disabling the sub-pixel smoothing didn't improve anything relatively to
> the pain generated by the exposure to the screen.
> 
> I'll come back to all of you in three days with some pictures to allow
> you to properly compare the rendering in Ubuntu-based distro (currently
> Maya13) and Debian Wheezy with the same graphic drivers, on the same
> computer.
> 
> Nevertheless, to me, it seems that one of the best way to properly
> identify and describe in a more objective manner my issue should be to
> get on of these color calibrator such as the ColorHug.
> 
> Until them, I'll keep on reading the thread in silence, but please, stay
> calm and respectuous with each other : my issue is not worth turning
> people mad at each other ;)

Come now. It just isn't a proper internet discussion without a flamewar thrown 
in. :)

I forget if you said you are using an LCD or a CRT. I believe you said you've 
tried different monitors and different video cards. How about cables? Can you 
switch between VGA and DVI cables?

Other things that come to mind.
  - have you turned the brightness/contrast down
  - if you don't need correct colors, can you reduce R, G and/or B until
whites are neutral or slightly on the red/yellow side?
  - how about adjusting gamma?
  - if a CRT, can you (or anyone with very good ears) hear a very high pitched
whine, nearly in the ultrasonic spectrum? A bad cap in a CRT will give
me a headache in mere minutes.
  - if a CRT, does the display waver or shimmy at all?

That one GNU/Linux distro is OK and Debian causes trouble somewhat befuddles 
me. Does Debian now stand for DEranges Brains In A Nanosecond? Should it be 
nicknamed Torquemada? :)


--
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/201209201637.43866.neal.p.mur...@alum.wpi.edu

Re: IA64 or AMD64?

[Neal Murphy]

On Friday, September 21, 2012 04:53:21 PM Stan Hoeppner wrote:
> It's not writing style but attitude.  My attitude is that people should
> be self reliant.  Only when they search and can't find an answer should
> they ask on a mailing list.  Especially in this case, when the answer is
> so damn easy to find, literally clicking one link on the Debian home
> page and reading the subsequent page.

I wholly disagree. First, when someone lowers himself to ask a question in a 
public forum, it is usually because she has been unable to find the answer; it 
is the duty of others in the forum to respond politely and civilly. Second, 
mayhap you want others to treat you like a fetid, steaming dog turd; most of 
us don't want anyone treated like that. We don't like being bullied; rather, 
we wish to be treated with civility and respect, and we try to treat others in 
the same way.

If one cannot respond with civility and respect, one shouldn't respond at all.

Attitudes like yours drive people away from OSS.


-- 
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/201209211716.39796.neal.p.mur...@alum.wpi.edu

Re: hard drive clean / microsoft hidden partition

[Neal Murphy]

On Saturday, September 22, 2012 03:33:47 PM Scarletdown wrote:
> On 9/22/2012 12:20 PM, Andrei POPESCU wrote:
> > On Sb, 22 sep 12, 14:31:06, d d wrote:
> >> I need advice on removal of the microsoft hidden partition and other
> >> software from a hard drive
> >> leaving a completely clean hard drive.
> > 
> > dd if=/dev/zero of=/dev/yourdrive bs=1M
> 
> But does that remove the partitions themselves?  I thought the OP was
> wanting to actually delete the MS partitions, which are used to restore
> a computer to its factory default with all the overbloated crappeware
> that gets put on them by the OEM.

Yes. The command writes zeroes to every block on the drive. Recovering the 
data would require dismantling the drive and using specialized hardware to 
read the edges of each track (to find residual data).

It's not a cryptographically secure erase, but for the OP's purpose, it should 
suffice.


-- 
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/201209221556.27066.neal.p.mur...@alum.wpi.edu

Re: ntpd crashes.

[Neal Murphy]

On Sunday, September 23, 2012 05:08:19 AM Rick Thomas wrote:
> On Sep 22, 2012, at 6:51 AM, Camaleón wrote:
> > Anyway, no NTP daemon should crash because of skewed time; one thing
> > is
> > that it refushes to sync (which can be fine, and should log this fact
> > so the admin can make the proper measures) but a different thing is
> > completely killing the service.
> 
> Hi Camaleón!  A bit of NTP history:
> 
> That issue has been argued on the NTP developer mailing lists.
> Crashing the daemon is Dave Mills' way of telling the admin that
> something is badly broken here and needs to be fixed.  Several
> developers (myself included) disagreed with him at the time, but he
> was adamant on the subject.  So that's the way it is.

I must agree that that was a somewhat braindead decision. 'Failure by design' 
is never an valid option.

If a problem is important enough to require an admin's attention, a daemon 
should demand it via the console and the system logs. Nag repeatedly and 
incessantly until the problem is corrected. Interrupt booting until the admin 
has acknowledged the problem. But never fail silently.


--
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/201209230728.13002.neal.p.mur...@alum.wpi.edu

Re: IA64 or AMD64?

[Neal Murphy]

On Tuesday, September 25, 2012 02:12:10 AM Ross Boylan wrote:

> I wish you would be more civil instead of more clever about being
> incivil.

Never mind that Miss Manners would be utterly aghast that someone would try to 
claim the right to behave rudely, impolitely and/or barbarously simply because 
some notorious person behaves in such a manner.


-- 
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/201209250242.52406.neal.p.mur...@alum.wpi.edu

Re: Installation

[Neal Murphy]

On Tuesday, September 25, 2012 02:26:11 AM Chris Bannister wrote:
> On Tue, Sep 18, 2012 at 04:12:31PM +, Camaleón wrote:
> > Okay, I agree the user does not need to hold a MS in Computer Science as
> > a previous requirement for installing an OS and managing a computer.
> 
> But an MS in Astronomy (Astrophysics?) won't help you build a telescope
> either. Remember, a computer is to Computer Science as a telescope is to
> Astronomy.

Thinking on this, a computer is to Computer Science as a computer is to 
Astrophysics. In fact, computers have become the basic tools of many 
disciplines: entertainment, automotive design and operation, mechanical 
engineering, civil engineering, electronic engineering, astrophysics, several 
aspects of medicine, law enforcement, civil governance, navigation, and 
subterranean resource exploration to name a few. CS professionals are the 
'tool & die' makers of the virtual world.

System installation should be as automated as is possible. In fact, the 
installer should ask as few questions as possible. One such question might be, 
"Where are you?" Suitable answers might be the person's 
city/village/state/county/country, zip/postal code, lat/lon, or something like 
"near Lyons, France". Combining the answer and the answer's language with the 
automatically-obtained IP address should allow the installer to deduce correct 
settings for many of the questions currently asked. But that's a lot to ask, I 
guess.


--
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/201209250306.06098.neal.p.mur...@alum.wpi.edu

Re: ls file exist, but can't access.

[Neal Murphy]

On Wednesday, September 26, 2012 12:33:42 PM lina wrote:
> $ ls -lrt
> total 8
> -rw-r--r-- 1 lina lina  367 Sep 27 00:15  RET
> drwx-- 2 lina lina 4096 Sep 27 00:16 auto-save-list

Look closer: there's a space you are overlooking.
  ls -ls " RET"
might work, and
  ls -ls *RET
will definitely work (in case that character is not a space).

That's the main reason I always use a fixed-pitch font for CLI stuff and 
email.

Re: ls file exist, but can't access.

[Neal Murphy]

On Wednesday, September 26, 2012 12:48:38 PM Tony van der Hoff wrote:
> > That's the main reason I always use a fixed-pitch font for CLI stuff and
> > email.
> 
> So why was your post HTML?

Hoist on me own petard!

I thought I had all that turned off long ago. It keeps sneaking back, though.


-- 
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/201209261330.48301.neal.p.mur...@alum.wpi.edu

Re: Display hurtful on LCD screen with Wheezy

[Neal Murphy]

On Thursday, September 27, 2012 02:53:19 AM lee wrote:
> Lionel Trésaugues  writes:
> > http://i1058.photobucket.com/albums/t405/Alkalyzer/Ubuntu_Terminal_zps831
> > 03ab7.png
> > http://i1058.photobucket.com/albums/t405/Alkalyzer/Debian_Terminal_zps18
> > 67e859.png
> 
> Now I see what you mean, things look better in Ubuntu.  There doesn't
> seem to be much difference in the magnified versions, and in the
> terminal and writer screen shots I can see that the line spacing in
> Ubuntu seems larger, i. e. leaving more blank space between each line.
> Perhaps that's what makes the crucial difference?

Other than the fonts (they *are* different), I see little difference between 
the two. The sub-pixel rendering seems to behave the same on each.

Discern which font is used on Ubuntu and use that font on Debian. That should 
fix the inter-line spacing and the often indecipherable glyphs.

N


--
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/201209270905.36554.neal.p.mur...@alum.wpi.edu

Re: why would fdisk -l take so long?

[Neal Murphy]

On Thursday, September 27, 2012 10:52:26 PM Albretch Mueller wrote:
> $ date; fdisk -l; date
> Thu Sep 27 22:48:21 UTC 2012
> ...
> Thu Sep 27 22:48:59 UTC 2012

Failing boot sector? Some other sector it has to read is failing? Check the 
logs. Try (from smartmontools):
  smartctl -A /dev/sda | egrep -i "sector|realloc"


-- 
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/201209272347.31024.neal.p.mur...@alum.wpi.edu

Re: IA64 or AMD64?

[Neal Murphy]

On Friday, September 28, 2012 02:35:49 AM Stan Hoeppner wrote:
> The only permanent solution to this confusion is for Debian to rename
> the IA64 port to "Itanium" and rename the AMD64 port to something like
> "AMDINTL64".

Something wrong with 'x86_64'?


-- 
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/201209280252.47782.neal.p.mur...@alum.wpi.edu

Re: why would fdisk -l take so long?

[Neal Murphy]

On Friday, September 28, 2012 08:23:59 AM Dom wrote:
> >1 Raw_Read_Error_Rate 0x000f   115   082   006Pre-fail
> > 
> > Always   -   96695847
> 
> Ok, your disk is dying. The Raw_Read_Error_Rate should be zero, or very
> low.


Not necessarily. At least one disk mfr (Seagate?) puts large values in these 
fields. Cause me a few moments' consternation the first time I saw it on my 
own drives


-- 
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/201209281448.32568.neal.p.mur...@alum.wpi.edu

Re: command isn't working in crontab

[Neal Murphy]

On Friday, September 28, 2012 07:35:04 PM Tomas Hulata wrote:
> Hello, below command works in command line but not as a cronjob can
> someone explain me why?
> 
> 23 58  * * *   rootcd /some_path/;mkdir "CAM1-$(date +%d.%m.%Y)";mv
> ./CAM1/*.* ./"CAM1-$(date +%d.%m.%Y)"/;mkdir "CAM2-$(date +%d.%m.%Y)";mv
> ./CAM2/*.* ./"CAM2-$(date +%d.%m.%Y)"/
> 
> I want to move files from /some_path/CAM1 directory to directory called
> CAM1-'today date' and the  same for /some_path/CAM2 directory

Perhaps cron doesn't know how to run those commands. Try:
23 58  * * *   root/bin/bash -c "cd /some_path/;mkdir \"CAM1-$(date +%d.
%m.%Y)\";mv ./CAM1/*.* ./\"CAM1-$(date +%d.%m.%Y)\"/;mkdir \"CAM2-$(date +%d.
%m.%Y)\";mv ./CAM2/*.* ./\"CAM2-$(date +%d.%m.%Y)\"/"


-- 
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/201209282005.23020.neal.p.mur...@alum.wpi.edu

Re: why would fdisk -l take so long?

[Neal Murphy]

On Friday, September 28, 2012 08:58:47 PM Albretch Mueller wrote:
> > > >1 Raw_Read_Error_Rate 0x000f   115   082   006Pre-fail
> > > > 
> > > > Always   -   96695847
> > > 
> > > Ok, your disk is dying. The Raw_Read_Error_Rate should be zero, or very
> > > low.
> > 
> > Not necessarily. At least one disk mfr (Seagate?) puts large values in
> > these fields. Cause me a few moments' consternation the first time I saw
> > it on my own drives
> 
> ~
>  Indeed! Something "spooky" may be going on. After taking the drive
> out in order to back it up, I have run "fdisk -l" with no disk and
> sometimes with a pen drive inserted and these bellow are the results I
> got.
> ~
>  Don't you find all of this downright weird?

Not yet.

> ~
>  I don't believe in ghosts ;-) What do you think I could do to
> troubleshoot this further?

Have you tried "fdisk -l /dev/sda"? I think you did, to no avail. It might be 
searching some other drive or type of drive first that is slow to respond.

Check the drive's power state:
  hdparm -C /dev/sda

How about:
  tail -f /var/log/messages  # In a separate window
  time dd if=/dev/sda of=/dev/null bs=1024k count=1
  time fdisk -l /dev/sda
to see if the drive is dozing.

Or (from hdparm's man page:  Disable  the  automatic power-saving
function of certain Seagate drives...):
  hdparm -Z /dev/sda


-- 
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/201209282134.29827.neal.p.mur...@alum.wpi.edu

[NIT] Re: Debian Small CD install "netinst.iso"

[Neal Murphy]

On Sunday, October 07, 2012 03:18:25 PM Paul E Condon wrote:
> Windows and Debian use different file systems on disk. I think Windows
> is incapable of modifying data on extN formatted disks that Debian
> uses.


A minor NIT to pick.

There is an EXT3 driver for Winders that enables it to read and write certain 
Linux file systems. It worked quite well, though it's been some years since I 
used 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/201210071528.45778.neal.p.mur...@alum.wpi.edu

Re: OT: man in the middle attack ?

[Neal Murphy]

On Wednesday, October 10, 2012 03:14:23 PM Joe wrote:
> 
> And finally, there are a few people who are just plain prickly... but
> one of the most important of all freedoms is the freedom to offend.

As is the freedom to choose to brush off offenses--be they real or perceived-- 
or to take them personally. I'm sure Miss Manners has waxed eloquently on the 
topic more than once.


-- 
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/201210101528.27699.neal.p.mur...@alum.wpi.edu

Re: newbie question on port forwarding(and ssh, netcat)

[Neal Murphy]

On Wednesday, October 10, 2012 08:19:25 PM houkensjtu wrote:
> Thanks for great reply!!
> I have to apologize for sth... I forgot to say that all these experiments
> were done in home on my laptop...omg So, now I solved the problem with
> echo "1">/proc/sys/net/ipv4/ip_forward
> 
> What is this file? Is there any other way to check or configure my laptop
> with out writing directly to this file?

That is exactly how you tell linux to forward traffic between NICs.


-- 
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/201210102046.03522.neal.p.mur...@alum.wpi.edu

Re: arp cache incomplete entries

[Neal Murphy]

On Thursday, October 11, 2012 08:47:37 PM Celejar wrote:
> 
> I'm no expert, but my impression is that any machine which is asked to
> connect to some other host by IP address will issue such an ARP
> request, so if I do 'ping x.x.x.x', and x.x.x.x has not been recently
> in contact with my machine, my machine will issue an ARP request.

Recall that all ethernet communication (layer 2) is done using MAC addresses. 
Also recall that point-to-point links do not use ARP.

A little more specifically, if the IP->MAC translation entry is not in the 
cache, the host will issue the ARP request. If the address is not on the local 
LAN, the host will issue an ARP request for the appropriate gateway.


-- 
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/201210112108.31906.neal.p.mur...@alum.wpi.edu

Re: arp cache incomplete entries

[Neal Murphy]

On Saturday, October 13, 2012 12:23:31 AM Panayiotis Karabassis wrote:
> 
> In a pattern that is becoming all too familiar, the problematic machine
> sends an ARP request, to which the nameserver replies. But the reply is
> never received by the asking machine. So says wireshark.
> 
> Could this be a fault NIC, or is the packet lost somewhere else?

It's not unknown. First, verify that the ARP cache is correct (host names/adds 
match the MAC addrs) on both hosts; on GNU/linux, use 'arp'. Second, directly 
connec the requester to the target (use a crossover cable if neither NIC is 
GigE). If the ARP tables are correct and direct connection works, then look in 
between them; specifically, look for a bad switch (some switches are known to 
corrupt their ARP tables and, thus, no longer function correctly as L2 
switches). If direct connection doesn't work (with and without crossover 
cable), then start looking at the NICs.

If you cannot directly observe the NICs (with tcpdump or equiv.), then you'll 
have to deduce the fault by observing whatever traffic that *does* traverse 
the LAN.


-- 
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/201210130053.02343.neal.p.mur...@alum.wpi.edu

Re: GRUB location on Dual-Boot with TWO hard drives

[Neal Murphy]

On Saturday, October 13, 2012 12:40:40 AM Wally Lepore wrote:
> Hi Debain Users,
> 
> I'm at the final stages of Installing NOT Ubuntu but  Debian 'Squeeze'
> on my dual-boot system. Windows is installed on the 1st hard drive
> (/dev/sda) and Debian will be installed on the 2nd hard drive
> (/dev/sdb).
> 
> The installer is asking me where I want Grub installed. It says:
> 
> -begin-
> 
> The following other operating systems have been detected on this
> computer: Microsoft Windows 2000 Professional
> 
> *begin*
> If all of your operating systems are listed above, then it should be
> safe to install the boot loader to the master boot record of your
> first hard drive. When your computer boots, you will be able to choose
> to load one of these operating systems or your new system. Install the
> GRUB boot loader to the master boot record? No  or  Yes ?
> *end*
> 
> This is the make it or break it point! Debian is installed on my 2nd
> drive (/dev/sdb) NOT the 1st drive (/dev/sda). I also created a
> partition on the Debian drive (/dev/sdb) called  "/boot". GRUB was to
> be installed at this /boot location and then I would go into BIOS
> after install and switch the boot order to boot the Debian drive
> (/dev/sdb). This would then present the menu for which OS I would like
> to boot (Windows or Debian).
> 
> If I choose NO to the installer's question as to placing GRUB in the
> MBR of the 1st drive. What are my choices as to where to install it? I
> don't want to answer "NO" to the question only to advance the
> installer to a dead end. I have no idea what may happen next if I
> answer NO. Any ideas or suggestions please?
> 
> Thank you

You may have no choice. I had lenny installed on sda. I tried to install 
squeeze on sdb so I could play with KVM and Xen. Grub *insisted* on installing 
itself to sda regardless of what I told it to do. I had to boot lenny's rescue 
disk several times to fix grub before I gave up.

You may have to pull the first drive during the install. And, if you're lucky, 
your BIOS will set the second drive as the 'first' when you tell it to boot 
from it.

If you find no joy, try grub legacy if it's an option. I *know* legacy works; 
I can make a bootable ISO that uses grub, copy the ISO contents to a flash 
drive and make it bootable with a trivial change, and legacy installs where I 
tell it to, not where it decides to because it knows better.

To be clear, my grub2 problems were with v1.97-1.99; I've not tried it since. 
And probably won't until legacy truly dies (i.e., RH's patch set falls into 
disrepair).

In a short phrase, be prepared to stub your toes some more.


-- 
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/201210130110.48090.neal.p.mur...@alum.wpi.edu

Re: Another OT: GRUB location on Dual-Boot with TWO hard drives

[Neal Murphy]

On Monday, October 15, 2012 05:19:29 PM Ralf Mardorf wrote:
> Assumed that you are not blind, perhaps a YouTube video will help you to
> learn faste, resp. it might better explain how to e.g. become root in a
> terminal emulation.

Or, perhaps, a simple list of ways to become root without any clutter.

Log in as root on a console, and return to the GUI:
  Type  through  to get to a console
  Log in as root
  Log out (typically using  or exit)
  Type  to return to your GUI. (It usually runs on virt.
cons. 7, but sometimes on VC6 or VC8).

Become root in the current directory:
  su
  su root
  sudo su
  sudo su root
  sudo /bin/bash

Execute a command as root:
  su -c "/sbin/sbin-command with options and args"
  su root -c "/sbin/sbin-command with options and args"
  sudo /sbin/sbin-command with options and args
  sudo -u root /sbin/sbin-command with options and args

Become root as though root had logged in on the terminal (or text console);
this give you root's PATH and other ENV settings and puts you in root's home 
dir:
  su -
  su - root
  sudo su -
  sudo su - root
  sudo /bin/bash -l   # Except this one leaves you in the current dir

Execute a command in root's environment:
  su - -c "/sbin/sbin-command with options and args"
  su - root -c "/sbin/sbin-command with options and args"
  sudo su - -c "/sbin/sbin-command with options and args"
  sudo su - root -c "/sbin/sbin-command with options and args"

Notes:
  - Su will always ask for the target user's password unless you are
already root.
  - Sudo (on most modern GNU/Linux dostros) will ask for the user's
password instead of root's password.
  - Having su run a command (or having sudo run a command other than su)
can be insecure (but it is not necessarily so)
  - Some distros will install only sudo by default.

For security purposes (partly because the X11 protocols can be insecure), most 
GNU/Linux distros do not allow root to log in to the GUI.

To learn more, 'man su', 'man sudo', and/or 'man bash'.

Re: aptitude interface help

[Neal Murphy]

On Tuesday, October 16, 2012 12:34:24 AM james gray wrote:
> Question:
> 
> where in the file system is the individual file
> 
> for help in the aptitude interface environment.

The man pages are typically in /usr/share/man. The man page for aptitude is 
found in section 8. So, try:
  zcat /usr/share/man/man8/aptitude.8.gz | \
groff -man -T ps | \
ps2pdf - aptitude.pdf

Then use the normal method(s) to print a PDF. You may have to install gzip, 
groff and ghostscript for this to work (if they aren't already installed).

While there may be a 'proper' way to do it, it works for me.

Re: Wally Lepore

[Neal Murphy]

On Friday, October 19, 2012 07:00:35 PM John Hasler wrote:
> > ...and requires emacs to be installed.
> 
> So what?

I used emacs back when it was written in TECO and have used a few flavors 
since. I'd *never* advise a new user to use emacs. They have enough to learn; 
they don't need to double or triple the load, or steepen the learning curve.

Re: Wally Lepore

[Neal Murphy]

On Friday, October 19, 2012 08:27:25 PM Chris Bannister wrote:
> ...
> > I will add that, if anyone does take the emacs+gnus route they will have
> a powerful and versatile system which a lot of developers/users swear by.
> 
> On the other hand, a lot of developers/users swear by the vim+mutt route.

Alas, not everyone who uses a computer is a developer.

Re: Kernel documentation

[Neal Murphy]

On Sunday, October 21, 2012 04:14:17 AM Lisi wrote:
> 
> We advise studying the README files in this root directory of the kernel
> source, and Documentation/Changes or the documentation index of the kernel
> in Documentation/00-INDEX.
> 
> 
> Presumably I have to download a kernel source to get at the README, but
> where can I find the Documentation/00-INDEX?
> 
> If I download an older kernel than the one that I am using, in oredr to
> read the README file,  will GRUB continue to use the present newer kernel?
>  I am still not very confident about editing GRUB 2.
> 
> And yes, I have googled - and I'm afraid that I found the answers as clear
> as mud. :-(
> 
> Thanks,
> Lisi

When you unpack the kernel *source*, it will create a tree (such as 
linux-2.6.35.14). In there, you will find README and Documentation/, among 
other things. Under Documentation/, you will find 00-INDEX and many squirrels.

Debian's kernel source package may be unpacked in /usr/src (or somewhere under 
there). I don't know; it's been a long time since I unpacked a Debian kernel 
source deb. (All my build work these past few years has involved vanilla 
kernels from kernel.org.) It's fair to assume that Debian don't change the 
kernel source tree structure.

Downloading any Debian source package should not alter the running system's 
behaviour. And DLing/unpacking a kernel source definitely won't change the 
system's boot; you have to take deliberate steps to make grub boot a new 
custom-built kernel. (Source packages are quite different from regular 'built' 
packages.)

Re: Screenshots

[Neal Murphy]

On Monday, October 22, 2012 07:23:58 PM lee wrote:
> Frank McCormick  writes:
> > ImageMagicks import commmand seems to work better to select a portion
> > of the page...but it needs a new file name for each shot..which makes
> > it a little awkward.
> > 
> > Does anyone have suggestions...perhaps a batch file to rename the
> > files after they've been taken ?
> 
> , [ ~/.fvwm/.fvwm2rc ]
> 
> | Key Print   A   A   Screenshot
> | 
> | 
> | # Screenshot - using import from ImageMagick
> | # taken from fvwm-crystal
> | 
> | DestroyFunc Screenshot
> | AddToFunc Screenshot
> | + I Exec exec mkdir -p $HOME/fvwm/screenshots
> | + I Exec exec import -window root -quality 100
> | $HOME/fvwm/screenshots/screenshot-`date +%Y%m%d_%H%M`.png
> | 
> | 
> | DestroyFunc Screenshot-Delay
> | AddToFunc Screenshot-Delay
> | + I Exec exec mkdir -p $HOME/fvwm/screenshots
> | + I Exec exec import -pause 10 -window root -quality 100
> | $HOME/fvwm/screenshots/screenshot-`date +%Y%m%d_%H%M`.png
> | 
> | DestroyFunc Screenshot-Frame
> | AddToFunc Screenshot-Frame
> | + I Exec exec mkdir -p $HOME/fvwm/screenshots
> | + I Exec exec import -frame -quality 100
> | $HOME/fvwm/screenshots/screenshot-`date +%Y%m%d_%H%M`.png
> | 
> | 
> | # A little menu...
> | DestroyMenu Screenshot
> | AddToMenu Screenshot
> | + "Screenshot Menu" Title
> | + "Fullscreen"  Screenshot
> | + "Delayed fullscreen"  Screenshot-Delay
> | + "Frame"   Screenshot-Frame
> 
> `
> 
> 
> You can adjust so you need to select an area, and you can do similar
> with i3 with something like "bindsym $mod+Print exec --no-startup-id
> import ...".

Or, in the traditional UNIX way using pipes and filters, grab a pic of the 
window, then adjust it:
===
#! /bin/bash

if [ $# -ne 1 ]; then
  echo "Usage: $0 filename"
  echo "where 'filename' is the name of the destination PNG filename; '.png'"
  echo "will be appended to 'filename'."
  echo
  echo "Run this script, bring the window you want to save to the front,"
  echo "wait for the cursor to change to a big +, and click inside the"
  echo "target window. Then use your favorite image editor to crop and"
  echo "adjust the image as desired."
  exit 1
fi

echo "You have three seconds to bring the target window to the top."
echo "When the cursor changes to a big +, xwd will copy the window you click."

# Wait a few seconds so the target window can be brought to the top
sleep 3

# Now get the sceenshot.
xwd | xwdtopnm | pnmtopng > $1.png
===

Note that this only works with the visible portion of a window, and is limited 
to the resolution of your display.

Re: OT: A question about bash scripting

[Neal Murphy]

On Monday, October 29, 2012 03:26:20 PM Ralf Mardorf wrote:
> On Mon, 2012-10-29 at 15:00 -0400, Wolf Halton wrote:
> > On Mon, Oct 29, 2012 at 6:57 AM, Ralf Mardorf
> > 
> >  wrote:
> > > Hi :)
> > > 
> > > how can I get rid of the variable "seconds"?
> > > 
> > > ((seconds=(done-started)-(((done-started)/60)*60)+100))
> > > min_sec=$(((done-started)/60))":"${seconds: -2}
> > > 
> > >^^^ the math should replace the
> > > 
> > > variable.
> > > 
> > > ### Killall and Restore session
> > > 
> > >  started=$(date +%s)
> > >  bash "$song_path/session/start-session-$startversion"
> > >  
> > >  ### Time
> > >  month=$(date +%B)
> > >  mon=$(date +%b)
> > >  d_y_t=$(date '+/%d/%Y %T')
> > >  done=$(date +%s)
> > >  ((seconds=(done-started)-(((done-started)/60)*60)+100))
> > >  min_sec=$(((done-started)/60))":"${seconds: -2}
> > >  echo
> > >  echo"Attended time to restore session: $min_sec"
> > >  echo -n "Session restored at " ; printf %9.9s $month ; echo $d_y_t
> > >  echo
> > > 
> > > TIA,
> > > Ralf
> > 
> > Ralf,
> > I don't understand.
> > What are you wanting to do with the script?
> > If you don't like the name of the variable, make up another one.
> > 
> > Wolf
> 
> This part of a larger script is a stopwatch and I want the math
> ((done-started)-(((done-started)/60)*60)+100))   inside the formatted
> string   ${seconds: -2}   , instead of a variable.
> 
> Something similar to
> ${((done-started)-(((done-started)/60)*60)+100)): -2}
> 
> Similar, because I'm missing something.

You missed the $((...)) syntax as exemplified by the '(done-started)/60' just 
before it:
  min_sec=$(((done-started)/60))":"
  min_sec=${min_sec}$(((done-started)-(((done-started)/60)*60)+100))

Re: OT: A question about bash scripting

[Neal Murphy]

On Monday, October 29, 2012 04:31:03 PM Ralf Mardorf wrote:

> FOR YOUR EXAMPLE, IIUC IT SHOULD BE? ...
> 
>  ### Killall and Restore session
>  started=$(date +%s)
>  sleep 2
> 
>  ### Time
>  month=$(date +%B)
>  mon=$(date +%b)
>  d_y_t=$(date '+/%d/%Y %T')
>  done=$(date +%s)
>  #((seconds=(done-started)-(((done-started)/60)*60)+100))
>  #min_sec=$(((done-started)/60))":"${seconds: -2}
>  min_sec=$(((done-started)/60))":"$(((done-started)-(((done-started)/60)*60
> )+100)) echo
>  echo"Attended time to restore session: $min_sec"
>  echo -n "Session restored at " ; printf %9.9s $month ; echo $d_y_t
>  echo
> 
> ... RESULT ...
> 
> Attended time to restore session: 0:102
> Session restored at   October/29/2012 21:11:43
> 
> ... RESP. ...
> 
>  ### Killall and Restore session
>  started=$(date +%s)
>  sleep 2
> 
>  ### Time
>  month=$(date +%B)
>  mon=$(date +%b)
>  d_y_t=$(date '+/%d/%Y %T')
>  done=$(date +%s)
>  #((seconds=(done-started)-(((done-started)/60)*60)+100))
>  #min_sec=$(((done-started)/60))":"${seconds: -2}
>  min_sec=$(((done-started)/60))":"$(((done-started)-(((done-started)/60)*60
> )+100)) min_sec=${min_sec: 2}
>  echo
>  echo"Attended time to restore session: $min_sec"
>  echo -n "Session restored at " ; printf %9.9s $month ; echo $d_y_t
>  echo
> 
> ... RESULT ...
> 
> Attended time to restore session: 102
> Session restored at   October/29/2012 21:17:26
> 
> BUT I NEED ...
> 
>  ### Killall and Restore session
>  started=$(date +%s)
>  sleep 2
> 
>  ### Time
>  month=$(date +%B)
>  mon=$(date +%b)
>  d_y_t=$(date '+/%d/%Y %T')
>  done=$(date +%s)
>  ((seconds=(done-started)-(((done-started)/60)*60)+100))
>  min_sec=$(((done-started)/60))":"${seconds: -2}
>  echo
>  echo"Attended time to restore session: $min_sec"
>  echo -n "Session restored at " ; printf %9.9s $month ; echo $d_y_t
>  echo
> 
> ... THIS RESULT ...
> 
> Attended time to restore session: 0:02
> Session restored at   October/29/2012 21:21:32
> 
> ... WHILE I WONT THIS 2 lines, AS ONE LINE, INCLUDING THE FORMATTING:
> 
>  ((seconds=(done-started)-(((done-started)/60)*60)+100))
>  min_sec=$(((done-started)/60))":"${seconds: -2}
> 
> Regards,
> Ralf

What's the '+100' supposed to do? Add 100 to the remaining seconds? Or 
subtract 100 from it? (That is, increase or decrease the number of seconds?) 
The way it is now, the number of seconds will never be less than 100 and your 
': -2' tweak will never trigger anyway.

What you are asking cannot be done. You cannot nest substitutions in the 
manner you wish, and getting the leading zero on the seconds is problematic 
using only bash.

I don't think you can put that many conditionals on a single line and have it 
remain comprehendable.

However, you might be able to do it using awk:

  ### Killall and Restore session
  started=$(date +%s)
  sleep 2
 
  ### Time
  month=$(date +%B)
  mon=$(date +%b)
  d_y_t=$(date '+/%d/%Y %T')
  done=$(date +%s)
  min_sec=$(((done-started)/60))":"$(echo $done $started | awk '{s=($1-$2)%60; 
if (s==0) {s=2;} printf ("%2.2d", s);}')
  echo
  echo"Attended time to restore session: $min_sec"
  echo -n "Session restored at " ; printf %9.9s $month ; echo $d_y_t
  echo

But regardless, the line is going to be rather long. Unless you use a shell 
function:

getSeconds () {
  echo $done $started | \
awk '{
  s=($1-$2)%60;
  if (s==0) {s=2;}
  printf ("%2.2d", s);
}'


Then use:
  min_sec=$(((done-started)/60))":"$(getSeconds)

But I still don't see what the '+100" is supposed to do.

Re: Advice on system purchase

[Neal Murphy]

I hate waiting for my computer to do things. Swapping and paging at all? Add 
more RAM. CPU-starved while running multiple processes? Add more CPUs.

By and large, for most desktop purchases, the most economical and reliable 
system will have a Gigabyte 790 or 970 mboard (I've never had a Gigabyte 
mboard fail), a quad-core AMD PhenomII-965 CPU, the fastest RAM one can 
reasonably afford (likely 1333 or 1666), enough RAM so that swap is rarely 
used, and SATA 3GB/s or 6GB/s IFs with one or more single-platter drives 
(certain Hitachi 7Kx000 drives are good). All the data and programs one 
normally works with should fit in RAM with room to spare. Clamav and squid 
each want about 500MB RAM (program and disk cache) to work efficiently. Gnome 
and KDE have become memory hogs. After using KDE for about 10 years, I've 
switched to XFCE; it's a lot faster and I don't need the UI wiping my butt and 
nose.

With 8GiB 1066 RAM, I can build my firewall system from scratch (source 
packages predownloaded), keep all four CPUs pegged during compiles (GCC and 
Linux will keep them pegged for about 5 minutes each, and IO waits still 
remain around 0), and rarely see IO wait states greater than 0.1. Linux's disk 
caching really is that good. The whole thing built uses about 6GiB (both 
cached in RAM and stored on disk). The build takes about 100 minutes building 
either on iron or in a KVM session with write-back disk access. Using write-
though caching raises IO wait states to (for me) unacceptable levels. I bought 
this system a few years ago and I'm *still* not sorry I did.

There is one very good reason to always have at least a dual-core CPU: X11. 
It's a pig; it slows down single-core system (has done for at least 15 years; 
my old dual PII-266 was noticeably faster/smoother than my PIII-800). A second 
core allows X11 to run independently of other apps.

Power-conscious? Use the on-demand cpu-freq governor and get idle CPUs down to 
the slowest frequency available (800MHz for the PhenomII-965). I don't know of 
a way to get it any better outside of actively controlling system voltages. 
Performance mode--all CPUs at max clock of 3.4GHz--uses 20-40W more power on 
my system. Turning CPUs off (disabling them via /proc) doesn't change power 
consumption. [I don't know why Intel and AMD don't use (the king of low power) 
Motorola's (Freescale's) trick of turning off the clock in the parts of the 
CPU that aren't being used at any given time.]

This is what I've done and will continue to do. It works for me®. Your mileage 
may vary®. Consult a doctor if you experience any side effects®.

Re: Advice on system purchase

[Neal Murphy]

On Saturday, November 03, 2012 01:47:40 AM Stan Hoeppner wrote:
> Motorola 680x0, DEC Alpha, SGI MIPS, HP PA-RISC, Motorola/IBM PowerPC,
> Sun SPARC, Cray Vector, Intel Itanium (irony here).

You missed Moto's 88K which vastly outperformed the 68K. The Moto/Freescale 
embedded PPC (though clearly not desktop) beats the pants off most of the 
competition. IBM's Power CPUs still have a strong niche. DEC did make some 
*very* good ICs. The Alpha was very good, and many of their 100Mb/s ethernet 
chips are still in use. I've no evidence, but I suspect Intel incorporated a 
good deal of DEC's technology in their ethernet offerings.

The x86 architecture may seem quick, but that's due more to lack of 
competition. There are other architectures that flat out smoke x86. But they 
gain no traction because Win doesn't run on them, so they cannot gain from the 
economies of scale.

Foresight, planning, execution and ethics combine to determine the winners and 
losers in the marketplace. Standards also play a role.

Re: Advice on system purchase

[Neal Murphy]

On Saturday, November 03, 2012 02:50:00 PM Stan Hoeppner wrote:
> I do support AMD, and I never said they're
> on a collision course with bankruptcy.  What I did say is competing head
> to head with Intel in the x86 CPU market is a tough game, and they have
> made many missteps along the way.
> 
> Someone would likely acquire them long before bankruptcy would loom.
> AMD has too many valuable assets, including CPU and GPU patents, people,
> etc.  A likely suitor, should things get that bad for AMD, would be
> nVidia.  They've wanted to enter the x86 market, and specifically the
> CPU/GPU market, for some time.  Acquiring a weakened AMD on the cheap
> would get them there quickly.  Damn, did I make another prediction?
> Shame on me. ;)

AMD's been around a looong time. They've always made good ICs. In particular, 
their 2903 bit slice processor was a fine piece of engineering in its day. I 
particularly liked the clockless ALU: apply the inputs (no clocking or 
latching needed) and read the outputs.

AMD's been weak before. And they've survived.

Re: [OT] debian package maintain tutorial

[Neal Murphy]

On Sunday, November 04, 2012 11:24:34 AM lina wrote:
> Please also understand that people's mind not programmed the same. We
> learned things by different ways. I really had difficulty reading
> manuals even I messed up lots of things by blindly try.
> I read manual, but just don't get it.

Don't feel badly. I have a hard time understanding many English-language 
technical manuals. And English is my native language!

And yes, everyone learns differently. Some by mimicing what others have done, 
some by striking out on their own and learning from all the mistakes they 
make, and most somewhere in between. *How* and *how fast* you learn don't 
matter. *That* you learn does.

Re: Mysterious packet

[Neal Murphy]

On Thursday, November 08, 2012 11:58:33 AM Darac Marjal wrote:
> On Thu, Nov 08, 2012 at 03:26:23PM +, Hendrik Boom wrote:
> > I've started getting messages like the following:
> > 
> > [12332.047451] IN=ppp0 OUT=ppp0 SRC=74.125.133.188 DST=25.46.128.71
> > LEN=40 TOS=0x00 PREC=0x00 TTL=50 ID=46353 PROTO=TCP SPT=5228 DPT=44380
> > WINDOW=0 RES=0x00 RST URGP=0 [79.489288] IN=ppp0 OUT=ppp0
> > SRC=74.125.133.188 DST=25.45.89.15 LEN=40 TOS=0x00 PREC=0x00 TTL=50
> > ID=25315 PROTO=TCP SPT=5228 DPT=43491 WINDOW=0 RES=0x00 RST URGP=0
> > 
> > Now these IP numbers are not on my LAN, which is masqueraded.  They also
> > bear no relationship to my external-world IP number.  If it's about a
> > packet being sent from 4.125.133.188 to either of the others, my ISP
> > shouldn't even be sending it to me.  Do I understand the message
> > correctly?
> 
> Yep. As I understand it 74.125.133.188:5228 is sending a RESET packet
> to 25.46.128.71:44380. By the looks of things, though, your kernel is
> responding as you'd expect it to and re-routing the packet back out your
> PPP connection (that is, it came in on ppp0, it's not for you, so you
> pass it back out on the default route which I imagine is ppp0).

Presented this way, it could be a DDoS attack on either the src or the dest.


-- 
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/201211081215.55984.neal.p.mur...@alum.wpi.edu

Re: Mysterious packet

[Neal Murphy]

On Friday, November 09, 2012 06:30:37 PM Tom Furie wrote:
> Not sure it helps any, but the 74.125.0.0/16 block belongs to Google and
> the 25.0.0.0/8 block belongs to the UK's MoD. Looks like some sort of
> attack attempt to me.

Were I a paranoid type, I might think that someone was inventing a new attack, 
where they try to induce two large sites to DoS each other, or try to clog a 
trans-oceanic cable.


-- 
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/201211091850.31046.neal.p.mur...@alum.wpi.edu

Re: Awk, filtering match through external command

[Neal Murphy]

On Friday, November 09, 2012 11:28:29 PM T o n g wrote:
> Any way to filter through external command to variable, somewhat like:
> 
>  head /etc/group | awk '{print $0 | "cut -d':' -f1" | getline result ;}'
> 
> Any way to make it works?

You'd *think* there'd be a way to do that, but I don't think awk works that 
way. I don't see a way to pipe both ends of an external command. I'm not sure 
even *perl* could do that.

Your best bet is to follow the ancient UNIX mantra of writing a program or 
script to do one thing well. Split the awk script into two parts, as in:
  awk -f part1 < file | external_filter | awk -f part2

Of course, if it's a simple filter like the one you illustrated, you can do it 
in awk:
  awk '{split($0, result, ":"); print result[1];}' http://lists.debian.org/201211100122.30219.neal.p.mur...@alum.wpi.edu

Re: Directory ownership

[Neal Murphy]

On Sunday, November 11, 2012 12:42:09 AM Gean Ceretta wrote:

>  *# chown -Rv gean:gean /home/gean/*

This is your trouble. You recursively changed the ownership of everything *in* 
/home/gean, but you did not change /home/gean itself. Try:
  chown -Rv gean:gean /home/gean
This will also change ownership of all the dot files/dirs (except for ..) in 
your home dir that your command also missed.


-- 
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/20120104.34270.neal.p.mur...@alum.wpi.edu

Re: Directory ownership

[Neal Murphy]

On Sunday, November 11, 2012 01:50:03 AM Gean Ceretta wrote:
> Thanks Neal and Charlie, I've tried:
>  *# chown -Rv gean:gean /home/gean* but the ownership stays the same root,
> maybe its important to say that the /home is an NTFS partition, mounted by
> /etc/fstab as:
> 
>  */dev/sda3 /home auto defaults 0 0*
> 
>  this is the correct way to mount this? the problem can be here? All the
> home dirs in this machine have root ownerships.

Knowing it's NTFS may make a big difference. You have to think how Winders 
handles access rights. You may have to do it using ACLs. This may be a dumb 
question: do any ACL tools work with NTFS?

Try:
  apt-get install acl
  getfacl /home/gean
and see what appears.

You could try:
  setfacl -m user:gean:rwx /home/gean
  setfacl -m group:gean:rwx /home/gean
It doesn't change ownership, but it may give you rwx access.

Getfacl/setfacl works on EXTn (and maybe ReiserFS); I don't know if it works 
with NTFS-3G. The ACLs work through samba on Win systems (and even allow Win 
systems to set access rights via Properties->Secuiryt). If it doesn't work, 
there may be other tools that work.


-- 
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/20120334.42933.neal.p.mur...@alum.wpi.edu

Re: ssh issue

[Neal Murphy]

On Sunday, November 11, 2012 02:39:14 PM Rainer Dorsch wrote:
> Hello,
> 
> I have on a Debian squeeze server an issue, that I can only login as user
> rd, not as user gpxrecorder, although there seems to be no difference in
> the accounts:

I'll bet some key files in .ssh/ are readable by other than the owner. Nothing 
in .ssh (and the dir itself) should have 'group' or 'other' perms set. Try:
  chmod -R g-rwx,o-rwx /home/gpxrecorder/.ssh
You should do that for /home/rd/.ssh as well.

If you have the file '.ssh/config', check it for oddities.


-- 
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/20121650.29545.neal.p.mur...@alum.wpi.edu

Re: Noob Question :-/ ....

[Neal Murphy]

On Sunday, November 11, 2012 04:08:47 PM David Christensen wrote:
> On 11/11/12 12:54, William A. Mahaffey III wrote:
> > Does Debian-kfreeBSD in fact have ext3fs support ?
> 
> I dunno -- perhaps that's the problem.  (I use Debian Squeeze i386 and
> Debian Wheezy amd64.)
> 
> 
> A console-only install of Debian Squeeze i386 can have a fairly small
> memory footprint (my CVS server uses 42 MB):
> 
> 2012-11-11 12:59:16 dpchrist@cvs ~
> $ top
> 
> top - 12:59:16 up 0 min,  1 user,  load average: 0.06, 0.01, 0.00
> Tasks:  62 total,   1 running,  61 sleeping,   0 stopped,   0 zombie
> Cpu(s):  1.7%us,  7.0%sy,  0.0%ni, 78.9%id, 12.3%wa,  0.1%hi,  0.0%si,
> 0.0%st
> Mem:514684k total,42284k used,   472400k free, 7544k buffers
> Swap:0k total,0k used,0k free,19004k cached
> 
> 
> Perhaps somebody with a console-only amd64 server can post data for that.

After clearing the caches (/sbin/sysctl vm.drop_caches=3), one of my amd64 
console-only systems uses about 94MiB. With apache2 and samba running, 
dropping the cache brings it down to about 105MiB.

I've noticed that 64-bit Wheezy likes to cache a lot more in RAM. Without 
doing much on my 64-bit Wheezy XFCE desktop, it has built up to 2.8GiB used 
(1.1GiB cached, 740MiB buffers); dropping the cache reduces mem usage to about 
620MiB. This is a departure from 32-bit Squeeze with PAE and KDE which never 
used that much memory 'doing nothing', though it might be that 32-bit+PAE 
performs cache-usage computations sees only 4GiB RAM (32-bit) rather than all 
8GiB.


-- 
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/20121727.17598.neal.p.mur...@alum.wpi.edu

Re: apt / aptitude question

[Neal Murphy]

On Sunday, November 11, 2012 04:59:34 PM David Guntner wrote:
> Hello,
> 
> Mandriva refugee here. :-)  New to Debian, but have been using some form
> of *NIX since 1986.  Have been a happy Mandriva user since the Mandrake
> 7 days, but that new company that purchased it, resulting in them losing
> most of their talent, has finally caused me to leave; the last update to
> the OS (which was the first under the new regime) left me with a system
> that barely runs.  So I looked around and decided that Debian might be a
> good distro for me to use (plus, my webhosting provider is using Debian
> on their servers; I decided it might be nice to not have to change
> mental gears when I ssh into my vps :-) ).
> 
> I've been working on figuring out the various quirks and differences
> between the two and Google Has Been My Friend in this a lot. :-)  Even
> managed to find a reference to using an Ubuntu source to install Freenx
> from the .deb packages that they provide since for some reason Debian
> doesn't (so that's something I won't have to do without - yay!).
> 
> However, I've had no luck finding any useful information on this one
> particular item; maybe I'm just not wording the search correctly.  I'm
> hoping someone here can give me a hand.
> 
> In Mandriva, they use a wrapper to rpm called urpm{x} (where "{x}"
> represents an action that you want to do "i" for install, "e" for erase,
> etc.).  There are options that allow you to see what files are related
> to packages, whether they are installed in the system or not.  There are
> two in particular that I'd like to have happen here.
> 
> In Mandriva if I enter:
> 
> urpmq -i {somepackage}
> 
> I will get an output that's pretty similar to what I get if I enter:
> 
> aptitude show {somepackage}
> 
> in Debian.  So far, so good.  However, on the Mandriva side, if I enter:
> 
> urpmq -fi {somepackage}
> 
> I'll get the package information *and* a listing of all files provided
> by that package.  I can't seem to figure out (or find out) how to do
> that same thing in Debian.  Is there a way?  And if so, how to do it?
> 
> As to the second variation on this theme, under Mandriva, if I enter:
> 
> urpmf {somefilename}
> 
> It will spit back a list of all packages that provide that filename or
> any part of it.  It's treated like a substring - for example if I type
> "urpmf kross" it will list the package that provides /usr/bin/kross, and
> will then go on to show other packages that have that string in it like
> /usr/lib64/libkokross.so.7 and so on.  This can be controlled by
> entering something like "urpmf /usr/bin/kcross", though other things in
> /usr/bin that start with "kcross" will also show up.
> 
> I haven't figured or found out yet how to get that kind of search
> functionality within Debian.  So as above, is there a way to do this,
> and if so, how?
> 
> Thanks for any help!
> 
>   --Dave

I believe you are looking for dpkg-query...

Which packages provides files with 'pattern' in the name?
  dpkg-query -S kross
  dpkg-query -S bin/kross

What files does kdelibs-bin provide?
  dpkg-query -L kdelibs-bin

What are the details for package kdelibs-bin?
  dpkg-query -p kdelibs-bin


-- 
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/20121746.06489.neal.p.mur...@alum.wpi.edu

Re: Grub and Wireless Keyboard

[Neal Murphy]

On Monday, November 12, 2012 06:32:22 PM Dr Beco wrote:
> Last week my keyboard broke, and I bought a wireless set
> keyboard+mouse that comes with a usb transmitter.
> 
> Yesterday I realize I can not chose an item from the grub menu during boot.
> ...
> But, still... Is there any solution? Maybe something to add grub a
> functionality?
> 
> Specially, the wireless keyboard, I think it should work. After all,
> it is just a USB keyboard from the system's perspective, isn't it?

Look for a BIOS setting involving USB and 'legacy' that would treat the USB 
kbd/mouse as though they were legacy wired units.


-- 
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/201211122010.30298.neal.p.mur...@alum.wpi.edu

Re: Help netbooting a diskless, headless system

[Neal Murphy]

Does your box have a serial port? Can it be configured to display the BIOS
screen on the serial port? Can Debian be installed using a serial port? That
is, connect a null-modem serial cable between the box to be installed and some
other computer and use minicom (Linux) or Hyperterm (Win).

Or put the hard drive into another regular computer, install there, then move
the drive back to the target box. And (1) enable a getty on /dev/ttyS0 and (2)
add 'console=ttys0,115200' to grub. The only oddity you should encounter here
is that the target box's NIC won't be eth0; this results from how udev works,
but it can be changed if desired.

Grub0 can be configured to poll both the VESA console and a serial port, then
use whichever gets the first keystroke (or time out and use the selected
default). I don't know how (or if) Grub2 handles this situation. So if the box
has no drive and no VESA console and you cannot redirect the BIOS to a serial
port, you'll be blind until grub starts.

Another possibility. Check if the system has a compact flash socket. If so,
you could effectively install a /boot to it with a custom initramfs that
contains enough command line tools (and libs) to run a fairly minimal 'live'
system in RAM; this might take a 100-300 MiB. As it boots, it can mount the
rest of what it needs over the net. My firewall has a smallish initramfs
(30MiB compressed CPIO, about 100MiB in RAM). It's a fairly fully usable
environment with a real init and a few real tools with busybox handling
others; I made it to ease debugging the install process. It works well on
standard computers and works well on headless systems like the Lanner 7530 and
7539 network appliances using a serial console.

You *can* do what you want, but it requires you to roll up your sleeves and
get up to your elbows in slimy bits. But then, Debian might have something for
this already. The hard part will be to redirect the install session to a 
serial port.


-- 
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/201211122157.39742.neal.p.mur...@alum.wpi.edu

Re: compiling (or rather, failing to compile) a kernel

[Neal Murphy]

On Tuesday, November 13, 2012 01:22:50 AM Tom Furie wrote:
> On Tue, Nov 13, 2012 at 12:10:55AM -0500, Tom H wrote:
> > That RHEL/Fedora dont' use "/usr/src" might, on its own, not make it
> > good practice, but since they're following kernel documentation
> > perhaps it does!
> 
> The kernel documentation does not say not to /usr/src, it says not to
> use /usr/src/linux. That doesn't mean you can't use
> /usr/src/linux-source-2.6.32-5, or whatever else you might like, it just
> means you shouldn't use /usr/src/linux.

For generic upstream packages including the kernel, one can build them in pert 
near any directory anywhere. The only caveat is that you have to specify that 
location when you build companion klibc, iptables, xtables-addons and openswan 
(to name a few packages).

I don't know the requirements of "The Debian Way", though I would expect 
Debian builds to be slightly restricted, like putting the unpacked source 
package dirs in a common parent dir.


-- 
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/201211130145.56022.neal.p.mur...@alum.wpi.edu

Re: Help netbooting a diskless, headless system

[Neal Murphy]

On Tuesday, November 13, 2012 12:19:34 PM Ross Boylan wrote:
> On Mon, 2012-11-12 at 21:57 -0500, Neal Murphy wrote:
> > Does your box have a serial port?
> 
> No.  USB and LAN ports.  I investigated SOL, serial over LAN, and IPMI,
> but can't get access to it; apparently it ordinarily must be enabled on
> the target machine as a first step (if it's there at all, which I
> suspect it is since an Intel mobo DH77DF).
> 
> > Can it be configured to display the BIOS
> > screen on the serial port? Can Debian be installed using a serial port?
> > That is, connect a null-modem serial cable between the box to be
> > installed and some other computer and use minicom (Linux) or Hyperterm
> > (Win).
> > 
> > Or put the hard drive into another regular computer,
> 
> The system is diskless.  Also, I have no other amd64 computers.
> 
> > install there, then move
> > the drive back to the target box. And (1) enable a getty on /dev/ttyS0
> > and (2) add 'console=ttys0,115200' to grub. The only oddity you should
> > encounter here is that the target box's NIC won't be eth0; this results
> > from how udev works, but it can be changed if desired.
> > 
> > Grub0 can be configured to poll both the VESA console and a serial port,
> > then use whichever gets the first keystroke (or time out and use the
> > selected default). I don't know how (or if) Grub2 handles this
> > situation. So if the box has no drive and no VESA console and you cannot
> > redirect the BIOS to a serial port, you'll be blind until grub starts.
> > 
> > Another possibility. Check if the system has a compact flash socket.
> 
> Since it has USB I believe I could boot from that.  But there again, it
> would expect me to control it from the local computer .
> 
> >  If so,
> > 
> > you could effectively install a /boot to it with a custom initramfs that
> > contains enough command line tools (and libs) to run a fairly minimal
> > 'live' system in RAM; this might take a 100-300 MiB. As it boots, it can
> > mount the rest of what it needs over the net. My firewall has a smallish
> > initramfs (30MiB compressed CPIO, about 100MiB in RAM). It's a fairly
> > fully usable environment with a real init and a few real tools with
> > busybox handling others; I made it to ease debugging the install
> > process. It works well on standard computers and works well on headless
> > systems like the Lanner 7530 and 7539 network appliances using a serial
> > console.
> > 
> > You *can* do what you want, but it requires you to roll up your sleeves
> > and get up to your elbows in slimy bits. But then, Debian might have
> > something for this already. The hard part will be to redirect the
> > install session to a serial port.
> 
> I've gotten a little further, though still no joy.
> I took the terminal-based testing netboot material for amd64 (only used
> the linux and pxelinux.0 files) and created a config file with the magic
> name including the UUID of the new client.  It has
> label install
>   menu label ^Install
>   menu default
>   kernel linux
>   append vga=788 root=/dev/nfs nfsroot=192.168.40.2:/mnt/amd64
>   timeout 50
>   totaltimeout 300
> This effectively bypasses most of the other files (and their screwy
> paths)
> 
> /mnt/amd64 has the results of running
> debootstrap --arch=amd64 --foreign --include openssh-server
> --variant=minbase testing amd64 http://debian.betterworld.us:\
> 3142/mainline
> from my testing chroot.  It's exported via NFS.
> 
> The idea was to create enough of a system that it would start and run
> the ssh server so I could ssh in.  On reflection, since none of the
> packages are fully installed, that probably  won't work even if I get
> further.
> 
> The the client shows the early linux kernel load but fails with
> No Filesystem could mount root, tried:
> Kernel panic - not syncing: VFS: Unable to mount root fs on
> unknown-block(0,255)
> 
> Possibilities:
> 1. The installer kernel does not support NFS.
> 2. I needed to load an NFS-related package as part of deboostrap. (Also,
> the man page says --include= is the syntax; I'm not sure if omitting the
> "=" is a problem).
> 3. Skipping the initrd suppliedd with the installer messed things up.
> (I was worried using it would take me into the installer).
> 4. The result of deboostrap --foreign is radically incomplete.  /boot is
> empty; several other key system files have stubs, including fstab.  I've
> been unable to discover exactly what one is supposed to do with the
> material created by 

Re: Gnome3 and Wheezy.

[Neal Murphy]

On Thursday, November 15, 2012 01:09:31 AM Ralf Mardorf wrote:
> Shortcuts to resize the view do
> work for Thunar, Xfce's file browser. Such shortcuts depend to the
> applications you're using, the view shortcuts are at least common for
> all web browsers I know.

I've found  works on a number of apps to scale the fonts 
and/or entire window content, in KDE, Gnome and (now) XFCE. If I can replace 
kmail, I can drop KDE completely (even though I might miss kmahjonng and 
kpatience, a tleast for a while).


-- 
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/201211150143.07962.neal.p.mur...@alum.wpi.edu

Re: CPU scaling problems

[Neal Murphy]

On Thursday, November 15, 2012 01:34:11 PM Mark Allums wrote:
> >  Francesco wrote:
> I have installed Debian testing on a X1 Carbon, processor i5-3427U.
> 
> My problem is the following: when the `ondemand' governor is active, the
> processor clock never scales up, it always stays at 800Mhz.  The situation
> changes when using the `conservative' or the `performance' governors.  More
> curiously, if I set `conservative'/`performance' as the governor for at
> least one core, the CPU will scale.  I am on the 3.6.6-trunk-amd64 kernel
> from experimental because the intel video driver 3.2 causes random freezes.
> To test if the scaling occurs I run `stress -c 4'.
> 
> Things I have tried:
> 
> * Enabling/disabling `cpufreqd', `laptop-mode', and `acpid'.  I currently
> run
>   `laptop-mode'.
> * The 3.2 kernel in testing
> * Enabling/disabling the `xfce4-power-manager', which I use to manage
>   suspends/hibernate.
> 
> There are some kind-of relevant BIOS options and I will try with those soon
> but I doubt that's the issue.
> 
> Does anybody have any idea on how to troubleshoot such a problem?  I
> suspect that the problem is either in the kernel or in some other software
> which is regulating the scaling.
> <
> 
> 
> At a guess, I would assume a configuration problem.  Or no problem.  Is
> your machine sluggish or unresponsive?  How do you know it stays at 800
> MHz? What tool do you use to monitor it?  Maybe your monitor is wrong.  I
> am not sure how to troubleshoot this, but if you don't need the governor,
> turn it off or use the 'performance' one.

In one shell, run 
  while :; do grep MHz /proc/cpuinfo; echo; sleep .25; done
and watch the clock(s); they should all be at minimum clock.

In another, run
  while :; do true; done
In the first shell, you should then see one CPU jump to full speed.  
and it should return to minimum.

If you run one for each CPU, they should all jump to full clock, the CPU temp 
should climb, and the CPU fan will speed up. Kill them all and everything 
should return to 'rest'.

For status and info, look in:
  ls /sys/devices/system/cpu/cpu*/cpufreq/


-- 
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/201211151358.03817.neal.p.mur...@alum.wpi.edu

Re: Thunar, USB-sticks and big files

[Neal Murphy]

On Sunday, November 18, 2012 07:57:09 AM Andreas Rönnquist wrote:
> > Hi,
> > 
> > On Sun, Nov 18, 2012 at 11:58:29AM +0100, Andreas Rönnquist wrote:
> > > Hi guys!
> > > 
> > > I have a problem copying big files to an USB-stick using thunar in
> > 
> > ...
> > 
> > > I do this to copy ripped DVD's to USB memory for viewing in a TV
> > > with
> > 
> > ...
> > 
> > > Nothing extremely serious, I can still copy the files using
> > > terminal as I said, but it would be nice to have it working using
> > > only Thunar too.
> > 
> > You see this list talking the same problem :-)
> > You are hitting 4GB limit of FAT.
> > 
> > Reformat USB stick with
> > 
> >  * ext2/3/4 if moving around Linux.
> >  * Darwin UFS, I think, for Mac
> >  * NTFS for Windows ...
> 
> (No need to CC the replies to me, I read the list)
> 
> But I am not nowhere near using 4GB... Files are around 700MB each, and
> one of the tried USB sticks is only 2GB big, so I would guess that
> doesn't apply? - and as I said, it works just fine when doing the same
> thing in a terminal.
> 
> "Copying" works just fine in both terminal and Thunar, but when
> unmounting the device (I guess its at this point the files are actually
> written to the device), the action times out (and the resulting
> files are not properly readable) when using Thunar, but not when using
> the terminal.
> 
> /Andreas

Try 'sync' after the write is supposedly complete; see what happens when the 
system actually tries to write to the device. Open a shell and 'tail -f 
/var/log/messages' to see if anything is griping about the device during 
writes.

Hmmm. You *are* waiting for the cached data to be flushed to the drive before 
unplugging it, right?


--
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/201211181231.01744.neal.p.mur...@alum.wpi.edu

Re: Thunar, USB-sticks and big files

[Neal Murphy]

On Sunday, November 18, 2012 01:43:54 PM Andreas Rönnquist wrote:
> > Try 'sync' after the write is supposedly complete; see what happens
> > when the system actually tries to write to the device. Open a shell
> > and 'tail -f /var/log/messages' to see if anything is griping about
> > the device during writes.
> > 
> > Hmmm. You *are* waiting for the cached data to be flushed to the
> > drive before unplugging it, right?
> 
> hmm, I now I have tried several different USB-sticks, and I still have
> the problem (It perhaps isn't quite as serious as I believed though)...
> 
> If having copied a "big" file, unmounting through Thunar pretty much
> 
> always results in the following error message:
> > Failed unmounting 'Devicename'
> > "Did not receive a reply. Possible causes include: the remote
> > application did not send a reply, the message bus security policy
> > blocked the reply, the reply timeout expired, or the network
> > connection was broken."
> 
> - after a short while (way before the actual writing is done) - But, if
> I keep waiting for some time _after_ this, the data is still written to
> the device as it should. However, it doesn't give any clue whatsoever
> to when it actually is finished writing the files to the device, so it
> is easy to (by mistake) physically take the USB-stick before the data is
> actually written.
> 
> However, I do have one USB-stick, that actually lights up a LED during
> copying, making it pretty easy to spot when it is safe to remove it and
> not, but I kind of wish that that sort of feature wouldn't be necessary
> to safely use an USB-stick...
> 
> If I don't copy any files, or just copy small files, I get that
> standard "It's now safe to remove the device" when unmounting - This
> is what I expect in the scenario with bigger files described above
> too after copying is done, but no...

The current version of teh command line umount does seem to wait until all 
data are flushed to the drive before exiting; older versions, IIRC, didn't 
necessarily wait.

Hmmm. You said 'timeout'. Could it be that Thunar doesn't wait long enough? 
See if there's a way to increase that timeout. 10 seconds generally ought to 
be long enough, unless you have a *very* slow flash drive.

Or, if there's a way to do it, tell Thunar to mount using "-o sync"; this 
should disable caching (that is, make the effective policy 'write-through'). 
Writes will be slower, but you won't have to wait for the flush to complete 
when unmounting.


--
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/201211181419.48442.neal.p.mur...@alum.wpi.edu

Black Screen of Death

[Neal Murphy]

Last night, I got two BSODs while building my firewall (actually building its 
toolchain). General Protection Fault. I snapped a pic of the second one.

This is on wheezy, 64-bit, updated, using the latest 3.2.0-4-amd64 kernel.

I last saw similar problems some years back when I had hardware troubles 
(failing on-board P/S). Thus I lean toward hardware trouble, but have a small 
suspicion it could be wheezy's new -4 kernel.

Has anyone else encountered such with wheezy?


-- 
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/201211191209.30492.neal.p.mur...@alum.wpi.edu

Re: Outrageous sexism. Was: Re: Mail client, threads, etc...

[Neal Murphy]

On Wednesday, November 21, 2012 12:54:33 PM Ralf Mardorf wrote:
> ... My humour is less good than Chaplin's ...

OK. Pointed, direct humor back at you. What was the shortest book ever 
published? "One Thousand Years of German Humor", which consisted of the 
frontispiece, the preface, and the introduction. :) :)


-- 
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/201211251755.11512.neal.p.mur...@alum.wpi.edu

Recent net stats oddity with Wheezy KVM sessions

[Neal Murphy]

In the past month or two, I have consistently encountered an odd problem with 
Wheezy on an AMD 8350 CPU. I never had this problem with my quad Phenom-II.

The problem
---
When I install recent builds of 32- and 64-bit Smoothwall Express 3.1 in a KVM 
on 64-bit Wheezy, everything seems to work well. Except for one thing. 
Smoothwall has a program that collects interface byte & packet peg counts from 
iptACCOUNT and possibly directly from netfilter every two seconds. The 
bandwidthbars and trafficmonitor UI pages are supposed to display the current 
bit rate. This problem makes the displayed rates climb very slowly and fall 
glacially. Also, the interfaces are weird. I've seen it display 170Mb/s 
throughput through a 100Mb/s NIC, and at least 1.9Gb/s through a physical gigE 
LAN. And no transmit/receive errors appear on any NIC's ifconfig output.

It happens on kernel 3.2, 3.12 and 3.14. It happens with Wheezy's kvm() and 
with qemu 1.7.x built from scratch. It happens with emulated e1000 NICs and 
with virtio.

Comparison
--
The exact same ISO installed on iron with a dual-core Athlon works as 
expected. And it works OK in that other computer in a KVM on Squeeze. (In 
fact, I NFS-mounted my KVM stuff from the desktop to this test system, so the 
method is identical.)



I haven't yet tried removing the HDs from my vishera desktop and installing 
the ISO directly. Or booting the Squeeze HD on the vishera. Or trying a 
different distro. Or even installing Wheezy from scratch on a spare drive.

Queries
---
Has anyone else seen behavior like this in KVM sessions on Wheezy? Or is it 
just my recent bad luck with computers?

Where/how should I dig to narrow down the scope of the problem?

Is it possible to install older kernels, qemu/kvm and other pkgs as needed to 
try to find a combination that works?

Thanks,
N


-- 
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/201407041214.30636.neal.p.mur...@alum.wpi.edu

Re: I'm not a huge fan of systemd

[Neal Murphy]

On Saturday, July 05, 2014 03:09:52 PM Erwan David wrote:
> I also think that the transition was far too fast : The testing fast,
> see wether it breaks anything should have been done *before* setting it
> the default. Not after.

A variant of the old corporate 'if it compiles, throw it over the wall at the 
customers and make them do the testing' trick?


-- 
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/201407051602.38733.neal.p.mur...@alum.wpi.edu

Re: simple database solution without root access

[Neal Murphy]

On Sunday, July 06, 2014 01:21:31 AM kamaraju kusumanchi wrote:
> I have some data in text format organized as follows
> 
> field_1,field_2,field_3,...,field_9
> val_1_1,val_1_2,val_1_3,...,val_1_9
> val_2_1,val_2_2,val_2_3,...,val_2_9
> ...
> val_100_1,val_100_2,val_100_3,...,val_100_9
> 
> 
> I want to do database (sql) like operations on this data. For example,
> ...
> Any ideas, pointers to existing code (perl scripts?) are much appreciated.

Google "perl dbi". DBI provides an interface that allows you to use SQL with 
just about *anything*, from RDBMSes, to CSV files, to Windows system files, to 
plain text files, to perl memory structures. It wouldn't surprise me if there 
was an interface (DBD) to packet capture logs.


-- 
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/201407060213.57707.neal.p.mur...@alum.wpi.edu

Re: Clone GPT partition table - with Lenny ?

[Neal Murphy]

Other than not being fully automated, what would be wrong with:
  - use dd to copy the first 10MiB of the old drive to the new,
  - use dd to skip all but the last 10MiB of the old drive and seek to the
same spot on the new drive
  - use dd if=/dev/zero to zero the first MiB of each partition.
That *should* guarantee that the partition structure is copied from the old 
drive to the new and that no partition contains RAID IDs. You have to think 
about what you're doing, but it should work and allow mdadm to equalize to the 
new drive.

Also, unless I'm mistaken, neither hard drives nor hard drive controllers have 
used CHS in many years, mostly because CHS has not been constant across the 
platter(s) in about as many years. Far as I know, only partitioning tools 
gripe about CHS alignment. It's why I stopped using fdisk, sfdisk and other 
ancient partitioners. And I came close to scrapping parted because it can't do 
simple arithmetic; but using 'unit s' works well enough, provided I do all the 
begin/end computations outside of parted.

N


-- 
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/201407061920.01540.neal.p.mur...@alum.wpi.edu

Re: what was broken with sysvinit? some thoughts on initialization "systems"

[Neal Murphy]

On Sunday, July 06, 2014 10:19:33 PM Joel Rees wrote:
> It took me several years (and playing with MSWindows's Wrong Way To Get It
> Right) to see that ad-hoc nature of sysvinit was, indeed, a feature, not a
> bug.

SysV init was a feature of the first computer I bought for my own use decades 
ago; I've been using it ever since. I do believe I'll dump Linux before I ever 
consider using an opaque, tentacled thing like systemd.

I still think SysV and its inittab could be enhanced to control hierarchies of 
daemons without introducing incomprehensible syntax and file structure that 
abstracts away all chance of human understanding via several layers of 
indirection. No one's done it because it would break compatibility with 
traditional SysV init (and I've other fish to fry).

N


-- 
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/201407062247.21502.neal.p.mur...@alum.wpi.edu

Re: I'm not a huge fan of systemd

[Neal Murphy]

On Monday, July 07, 2014 03:49:52 PM Michael Biebl wrote:
> Am 07.07.2014 21:29, schrieb Andrei POPESCU:
> > To prove my point (on a laptop with LXDE and just a few services):
> > $ grep sleep /etc/init.d/* | wc -l
> > 27
> > $ ls /etc/init.d/* | wc -l
> > 75
> 
> Yup, the boot speed improvements come from doing things correctly and
> event based. Socket activation doesn't necessarily mean things are
> delayed but simply that explicit orderings are unnecessary.
> 
> The numbers you have posted are depressing, but doing that over the
> complete archive is even more so.
> 
> The last time I did an archive wide check on this was early 2014, at
> that time we had 1235 SysV init scripts and 1124 occurences of sleep.

Whatever happened to semaphores (flags)? Seems to me that if a daemon is a 
dependency, the second-last startup thing it should do is connect to itself 
(since it may well be asynchronous); on success, it should run a flag up the 
pole (touch a file somewhere) to tell the world that it is up and ready to 
process requests. All of its dependents should wait for that flag to appear 
before they make their own services available. And later during operation, 
removal of the flag should cause dependent daemons to withdraw their services.


-- 
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/201407071625.41808.neal.p.mur...@alum.wpi.edu

Re: HTML5 => png or HTML5 => jpg.

[Neal Murphy]

On Tuesday, July 08, 2014 02:32:37 AM Reco wrote:
>  Hi.
> 
> On Mon, 07 Jul 2014 22:07:24 -0700
> 
> pe...@easthope.ca wrote:
> > Is there a software which can convert an HTML5 page to a
> > pixel map?  Conversion should apply on text, tables, images
> > and SVGs, all allowed in HTML5.
> > 
> > rsvg restricts the source to SVG.  What I want is similar
> > to a screen grab of iceweasel.  But invoked by a command
> > specifying the height and width of the image and not
> > including the menus and scroll bars of the browser.
> 
> This utility should do the job:
> 
> $ apt-cache show cutycapt | grep -A4 Description-en
> Description-en: utility to capture WebKit's rendering of a web page
>  CutyCapt is a small cross-platform command-line utility to capture
> WebKit's rendering of a web page into a variety of vector and bitmap
> formats, including SVG, PDF, PS, PNG, JPEG, TIFF, GIF, and BMP.
> Homepage: http://cutycapt.sourceforge.net/

Second this, but I don't know about HTML5. It does work very nicely to 
generate high-resolution images of an HTML UI that are suitable for 
publication in user manuals, though resolution selection may have room for 
improvement. It's *far* better than trying to make screenshots by hand.


-- 
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/201407080248.48488.neal.p.mur...@alum.wpi.edu