gmane.linux.debian.user

backing up

Frank McCormick — 2012-05-26T20:43:03

Can anyone suggest an easy way to back up a couple of profile 
directories (such as .thunderbird) when existing a window manager to 
return to GDM ?

Thanks

backing up

Frank McCormick — 2012-05-26T20:42:34

Can anyone suggest an easy way to back up a couple of profile 
directories (such as .thunderbird) when existing a window manager to 
return to GDM ?

Thanks

backing up

Frank McCormick — 2012-05-26T20:57:23

Is there an easy way to back up a couple of profile directories when
you exit a window manager.
I'd like to back up .thunderbird at least when I quit Icewm to GDM...

Thanks

changing sound parameter on startup

Ross Boylan — 2012-05-26T20:40:01

Simple version: what's the best way to get the effect of 
echo 1024 > /proc/asound/card0/pcm0p/sub0/prealloc
every time I start, preferably before KDE starts?

I've seen indications I should put something in rcS.d/, but also some
things saying stuff there gets executed at every run level change.
Also, I'm not sure where in the sequence it needs to be.

Longer version: The underlying issue is that I've been having problems
with sound under mythtv.  Hardware:
Audio device: Intel Corporation 82801G (ICH7 Family) High Definition
Audio Controller (rev 01) on a Pentium 4.

When myth starts it was complaining that I had a buffer too small and
that its attempts to reset it with the above command failed.  After
disabling sound on KDE I was able to resize the buffer.  I'm still
having sound problems, though I think not quite as frequently as before.

One solution would be to resolve the permission problem.

There are a lot of things I can think of that might be relevant to
either the permission issue or the buffer siz

Motion M1400 tablet upgrade woe

chris dunn — 2012-05-26T19:18:39

I'm trying to upgrade from Squeeze to Wheezy on a Motion M1400 tablet
and seem to have hit the wall.

Initially I had Squeeze installed, and I tried to upgrade via Synaptic
simply by changing the repositories from Squeeze-Updates to Wheezy,
reloading, then marking upgrades and applying.

This started ok but ran into a problem when I was told that the latest
686 kernel had bits that would not work on my single processor M1400.
I was informed that I needed to uninstall my 686 kernel and install a
486 kernel. Since there seemed to be no way of exiting the upgrade in
progress in order to achieve this, I shut the machine down after
deciding to reinstall from scratch. 

Reinstalled Squeeze from a cd using expert install, and selecting the
486 kernel to be installed. All went well and I now had a working M1400
with Squeeze on it.

Then attempted via Synaptic to again upgrade to Wheezy by changing the
repositories from Squeeze-Updates to Wheezy. All went well and
indications were that 1,646 packages would be install

Problems with flashplayer on a new installation of Squeeze.

Lisi — 2012-05-26T16:04:47

Is there a solution to the following problem?  (See below) I see that the bug 
was reported two years ago, so I am tempted to go for purge, but that may 
leave some detritus. :-(

And if I were to purge flashplugin-nonfree, what could I install that would 
actually work to play flash?

Would I do better to scrap, what is a very new installation of Squeeze - this 
was part of the installation process - and reinstall?

Thanks,
Lisi

root< at >Squeeze:/home/john# aptitude install flashplugin-nonfree
The following NEW packages will be installed:
  flashplugin-nonfree
0 packages upgraded, 1 newly installed, 0 to remove and 8 not upgraded.
Need to get 17.9 kB of archives. After unpacking 135 kB will be used.
Get:1 http://mirror.ox.ac.uk/debian/ squeeze/contrib
flashplugin-nonfree amd64 1:2.8.2 [17.9 kB]
Fetched 17.9 kB in 0s (50.4 kB/s)
Selecting previously deselected package flashplugin-nonfree.
(Reading database ... 136074 files and directories currently installed.)
Unpacking flashplugin-nonfree (from
.../flashplugin

Re Installing Squeeze with Firmware

2012-05-26T15:33:55





Hi all 
 
I today try install Squeeze on laptop Hp Compaq nc600, with stick 
write with cat fileiso.iso > /dev/sdb 
 
 
 
Regards 
 
 
Experimental2: syslog-unetbooting.txt > ask me choice from two netcard. 
I try another with stick created with unetbooting, only (initrd.gz-vmlinuz)  
StableNetinstall 
After 10 minutes, appears on screen request choice from! eth0:Broadcom  
Nextreme 
and Wlan: Atheros bla bla bla 

 
The log is this: 
 
Log for Experimental2: syslog-unetbooting.txt 
 
/* 
 
May 25 07:28:22 syslogd started: BusyBox v1.17.1 
 
May 25 07:28:22 kernel: klogd started: BusyBox v1.17.1 (Debian 1:1.17.1-8) 
 
May 25 07:28:22 kernel: [ 0.000000] Initializing cgroup subsys cpuset 
 
May 25 07:28:22 kernel: [ 0.000000] Initializing cgroup subsys cpu 
 
May 25 07:28:22 kernel: [ 0.000000] Linux version 2.6.32-5-486 (Debian  
2.6.32-45) (dannf< at >debian.org) (gcc version 4.3.5 (Debian 4.3.5-4) ) #1 Sun  
May 6 03:29:22 UTC 2012 
 
May 25 07:28:22 kernel: [ 0.000000] KERNEL supported

LVM, Resize failed (most mlikely user error)

2012-05-26T15:17:11

Dear Guys,

I had the (wonderful?!) idea to resize my LVM.  The result is, that I can not access my /home anymore :-(.

Environment: 
HP Mini, 160GB HDD
2 partitions (sda1, ext2, 254.99M, bootflag and sda2, 159784.29M)
LVM initial setup:
- root 300M
- home 136G
- var 2.8G
- usr 4.6G

During a dist upgrade, my /var went full and I started to do the following with the lvm commands:
reduce /home to 100G, 
increase /var by +5G, 
increase /usr by +6G, and
increase /root by 600M.

After that the whole thing went mad.

During boot it stops during fsck for /home with error 4 ("filesystem (according to superblock) is 36210688 blocks  The physical size of the device is 26249216" and "superblock or partition table is corrupt").  I can enter my root password and run a manual fsck with the same result.  If I do not abort I get "Error reading block 26279938 (invalid argument) while getting next inode from scan.  Ignore error?"

My idea is to use the partition table to "fix" the lvm (i. e. return to the initial state).  Is

whezzy + hostapd + ath9k = random fail

Gary Trotcko — 2012-05-26T13:59:52

Hi all. I got a problem with my soft AP.

I got an old PC running as router which has 3 NIC and 1 wifi PCI card
TP-Link TL-WN751ND. This box runs Debian Wheezy, hostapd, dhcpd etc.
The problem is that AP randomly gets unreachable for the clients while
hostapd is running and its interface is up. In fact, this router had
been working in 6 day w/o any problems since I set it up until the
yesterday.

What interfaces I have exactly:
--------------------------------------------
eth0 is down (3Com)
eth1 gets IP from ISP DHCP (Realtek)
eth2 takes IP for my LAN over DHCPD
ppp0 - it's an ISP VPN over PPTP

Configuration:
-------------------
$ cat /etc/debian_version
wheezy/sid

$ uname -r
3.2.0-2-686-pae

dpkg -l | grep linux-image
ii  linux-image-2.6-686                  3.2+44
Linux for modern PCs (dummy package)
ii  linux-image-3.2.0-2-686-pae          3.2.17-1
Linux 3.2 for modern PCs
ii  linux-image-686-pae                  3.2+44
Linux for modern PCs (meta-package)

$ hostapd -v
hostapd v1.0
User space daemon fo

wheezy: ext4 or ext3

Hans-J. Ullrich — 2012-05-26T13:27:12

Hi list,

I am now a proud user of a ssd drive. As I cloned my system (which was 
installed 5 years ago) to the drive, the filesystem on the source drive is 
ext3.

On my new drive I chose ext4 (with luks encryption) for as far as I read, most 
people are using ext4 instead of ext3 on ssd drives.
Is this really recommended? 

And if yes, should I change the filesystem from ext3 to ext4 on my old spin 
drive? I think, this can be done with tune2fs (when I remember correctly), 
without any loss of data. If I am wrong, please point me to the correct tool.

Most important thing for me is speed and security, and, of course, lifetime of 
the ssd is also important. I added noatime and discard to /etc/fstab, hope 
this is correct.

Just a last question: During installation, I forgot, how I formatted the "/" 
partition (ext3 or ext4?). Is there a way, to find out, how it is formatted? 
Strangely, the option "discard" in /etc/fstab seem not to work on this 
particular partition.

Thanks for any help!

Best regards

Ha

Less responsive input, 3-second freezes & blind-typing

John Magolske — 2012-05-26T00:07:50

Hello,

For a while now I've been experiencing a strange behavior where there
are frequent and regular "freezes" when typing or issuing commands
in the shell. All of a sudden I'll find myself "blind-typing" for 3
seconds or so before a backlog of characters burst onto the screen.

I seem to notice this a lot in Vim...maybe because that's what I'm
using most of the time. But this also happens for example when typing
a command into a shell and attempting to tab-complete a filename --
there will be a 3-second pause before the completion takes effect.
Trying another command immediately after, I see the tab-completion
is instant. Setting updatecount=0 in vim to avoid disk activity in
writing to a swap file made no difference.

One guess is that this may have to do with disk access, as when
experiencing this there's the telltale sound of the hard disk spinning
up during the pause. Not always -- at times the disk is already
spinning when I'm typing and then there's the annoying 3-second pause.
But never is there th

new tab or new terminal

BG — 2012-05-25T21:00:03

To monitor my system I usually open a terminal and run htop and then 
open another tab and run gkrellm. This is as a user without sudo 
privileges. I recently began using tcpspy and then tail syslog | grep 
tcpspy in yet another tab. However my user doesn't have sudo privileges 
so I have to su - to a user in the sudo group to do this.

Is it better from a security standpoint to su - in a tab as I have been 
doing or should I open another terminal for this user?

I am not overly paranoid and the programs used are irrelevant really and 
yes, I do realize I could get the same results as tcpspy without the 
privileged user using other commands such as ss or netstat. I am just 
curious what the implications of using mixed users with mixed privileges 
in a single terminal are if any.

Broken cups.

Ed Siefker — 2012-05-25T21:09:23

I am unable to print on my HP 1300n.   I have cups installed, along
with hpijs and hplip.
I will append an actual list of packages to this email.   My user is a
member of both
groups 'lp' and 'lpadmin'.   Dmesg identifies my printer as "usb 5-2",
and the device
entry looks like this:

crw-rw-r-T+ 1 root lp 189, 516 May 25 15:54 /dev/bus/usb/005/002

I have an HP2575 connected through the network.  I can print on this
printer perfectly.
My HP1300n is connected by USB.  I cannot print to this printer at all.

If I delete my 1300n from cups and physically restart the printer, I
get an entry
added automatically to my list.  Cups claims that the "make and model" is
"HP LaserJet 1300n pcl3, hpcups 3.12.2".

When I send a job to this printer, it claims to work.  It processes N pages
(where N is the number of pages in the print job) and returns to Idle
- "ready to print"
status.  This is a complete lie.  Nothing is printed and no error messages
are emitted.

If I delete this printer and physically restart the printe

Dual head Intel i7-2600S and DQ67SWB3

David Christensen — 2012-05-25T18:46:44

debian-user:

I'm attempting to get a dual head setup working using an Intel i7-2600S 
processor, Intel DQ67SWB3 motherboard, and Debian GNU/Linux.  (See end 
for console session.)


The DVI (Blue) Analog port is configured as primary in CMOS setup, and 
connects via an Intel DVI/ VGA adapter (furnished with motherboard) to a 
Nokia 445XiPlus 21" CRT monitor.


The DisplayPort port is configured as secondary in CMOS setup, and 
connects via a StarTech DP2VGA2 DP/ VGA adapter to a Nokia 445Xi 21" CRT 
monitor.


My OS is Debian GNU/Linux 6.0.5 amd64 (Squeeze).


My kernel is linux-image-3.2.0-0.bpo.2-amd64 from Debian Backports.


My X.org server is xserver-xorg-video-intel from Debian Backports.


I booted in recovery mode and ran "Xorg -configure" to create an 
xorg.conf.new file.


I temporarily connected the 445Xi to the DVI port and used "get-edit | 
parse-edid" to obtain an xorg.conf Monitor section.


"get-edid" doesn't work for the 445XiPlus, so I created an xorg.conf 
Monitor section using specificat

OT Seach for Elaborate Address Book Program

Thomas H. George — 2012-05-25T15:39:08

I am planning to put a vast array of handwritten notes into some kind of
order.  Notes include names and associated information - addresses,
phone numbers, email addresses, spouse's name, childrens' names, boat
names, birthdays, misc information.

An elaborate address book?  An apt-cache search addressbook found
several programs.  rubric seemed the closest to meeting my needs but
there was an installation problem:

Unpacking replacement rubrica ...
Processing triggers for hicolor-icon-theme ...
Processing triggers for desktop-file-utils ...
Processing triggers for gnome-menus ...
Processing triggers for man-db ...
Processing triggers for menu ...
Processing triggers for gconf2 ...
WARNING: failed to parse type name `boolean'
WARNING: invalid or missing type for schema (/schemas/apps/rubrica2/file/load_files)
WARNING: no <list_type> specified for schema of type list
WARNING: invalid or missing list_type for schema (/schemas/apps/rubrica2/file/files)
WARNING: failed to install schema `/schemas/apps/rubrica2/fi

MySQL upgrade from 5.1 to 5.5

Brad Alexander — 2012-05-25T15:06:01

I did an upgrade on my mediawiki box the other day to fix some
security issues. Well, in the mix, mysql-common was upgraded from
mysql-common_5.1.62-1 to mysql-common 5.5.23+dfsg-1. Nothing happened
until last night when we took a power hit due to a storm, and when it
rebooted, I got

May 25 09:49:06 sisko mysqld[31214]: 120525  9:49:06 [ERROR]
/usr/sbin/mysqld: unknown variable 'lc-messages-dir=/usr/share/mysql'
May 25 09:49:06 sisko mysqld_safe[31217]: ended
May 25 09:49:20 sisko /etc/init.d/mysql[31383]: 0 processes alive and
'/usr/bin/mysqladmin --defaults-file=/etc/mysql/debian.cnf ping'
resulted in
May 25 09:49:20 sisko /etc/init.d/mysql[31383]: /usr/bin/mysqladmin:
connect to server at 'localhost' failed
May 25 09:49:20 sisko /etc/init.d/mysql[31383]: error: 'Can't connect
to local MySQL server through socket '/var/run/mysqld/mysqld.sock'
(2)'
May 25 09:49:20 sisko /etc/init.d/mysql[31383]: Check that mysqld is
running and that the socket: '/var/run/mysqld/mysqld.sock' exists!
May 25 09:49:20 sisko /e

Help identifying an old wireless card

Michael Mehrazar — 2012-05-25T14:00:33

Hello everyone,

I apologize for perhaps this very simplistic question, but unfortunately
I have been unable to figure out what is the wireless card on an old
laptop. I've failed at the Google.

Here is the pastebin commands of lspci -v. http://paste.debian.net/171147/

Thank you, I appreciate any and all help. Also, as an optional
additional point, if anyone is familiar with this particular wireless
card, would you mind letting me know if it can support WPA/2? However,
that is not terribly important as I can (hopefully) figure that out on
my own.

Cheers and thanks again!

Michael Mehrazar

Wheezy printing problem: 2 identical (?) machines and 1 does notprint PDFs

Joost Kraaijeveld — 2012-05-25T12:39:01

Hi,

I have 2 identical Wheezy machines: they both have the same packages and
the same versions for those packages Wheezy, up to date as of this
morning). The configuration files in /etc/cups and /etc/foomatic are
exactly identical. Both machines should print to a network printer
(HP2200 DTN). I use Cups and Foomatic, and the recommended "HP LaserJet
2200 Foomatic/lj5gray" driver (the postscript driver has the same
behaviour).

Both machines print the Linux test page without problems. But one
machine cannot print PDFs while the other one prints them without
problem, be it a little slow (i.e. compared to the Windows machines that
use the same printer). The problematic machine prints a PDF file as an
almost empty pages with the following text:
"PCL CL error 
    Subsystem: IMAGE 
    Error:     ExtraData 
    Operator:  ReadImage 
    Position:  7"
I have re-installed/purged/rebooted etc. and nothing helped so far.

Does anyone recognises this problem? Which *machine-dependent* files
exist and in which directo

[OT] non-www to www redirect works with a little issue

J. Bakshi — 2012-05-25T11:35:44

Hello list,

I have place the following in my .htaccess to redirect non-www to www redirection 

`````````````
RewriteCond %{HTTP_HOST} ^example\.com$
RewriteRule ^(.*) http://www.example.com/$1 [R=301]

```````````````

The redirect work well but a little hitch.
Say If I visit http://example.com/page1.html it redirect to
http://www.example.com so at the home page. Where it should
append the www before the link, but not alter the link.

What modification should I add here to make it work ?

Thanks

how to update Debian OS properly

Muhammad Yousuf Khan — 2012-05-25T10:57:28

well, I have questions about upgrading Linux. since i have been using
Microsoft for years the concept of upgrade I think is different from
Linux upgrade.

in a testing environment I was trying to upgrade the security patches
and I run the command
apt-get upgrade.
It ran fine, but at the end it upgraded my whole OS, first my test
machine was showing Debian version 6.0.4 now after the upgrade it
shifted to Wheezy/SID which I think is unstable.

In Microsoft when we upgrade  the OS. it downloads only the security
and OS patches.

So the question are

1. how to upgrade only the security patches?
2. if the upgrade patch trigger any critical issue. Like any service
like hosting, filesharing or squid got effected by the update, how
come we know which patch cause this problem and how to remove that
specific patch because in Microsoft I have seen that security patches
and OS patches some time make problems when run along with ongoing
services.

3. how to revert back to old OS, for example, in my case i upgraded my
sy

Configure sudo

Denis Witt — 2012-05-25T08:13:05

Hi List,

we're running a server for a german bank. Of course we want to keep our services secure. A partner of us has to install a web based service (php, python and sql) on this machine. This partner will also be in charge in support and maintenance of this software.

So he needs access to the server, sftp isn't enough. There may be changes in the web server php.ini necessary from time to time. The web server needs some restarting, etc. Files must be edited and so on.

sudo might be a fine solution, but sudo is way too mighty in it's defaults. I know that you can allow and disallow certain commands only.

sudo su must be disabled of course, also /etc/sudoers must be write protected, even for root. This is no problem if you use chattr +i /etc/sudoers. 

But i think enable all commands and disallow some, line su and all known shells ;), isn't a good way to go. I would like to disallow all commands by default but allow some of them:

* restarting of web server
* editing of php.ini
* file transfer (ftp-ssl, sf