Re: [SLUG] ADSL modem recommendations (with bridging)

2008-02-28 Thread Tony Lissner
On 28/02/2008, Erik de Castro Lopo [EMAIL PROTECTED] wrote:
  I've been banging my head against my adsl modem and I wondering if it
  might not just be easier/cheaper to replace it with something better.
  Can anybody recommend an ADSL modem that does up to an including
  ADSL2+, is Linux friendly and easy to set up in bridging or half
  bridging mode? It would also be nice if the adminstrative functions
  were still accessible when it is in bridging mode.

I've been using a Thompson SpeedTouch 536 in straight bridge
mode in front of a WRT54GL (OpenWrt) with the WRT handling the pppoe dialing
but your modem should handle it in bridge mode OK. When you switch to bridge
mode you usually lose administrative access to the modem. I haven't played
with half bridging so not much help there. This is with Bigpong and it needs
the pppoe username and password.
You could try seeing if it works in gateway? mode before switching to bridge?
With this modem in gateway mode it has a fairly good CLI.

Tony
-- 
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html


Re: [SLUG] how to import weird gpg keys in mutt?

2007-01-24 Thread Tony Lissner
Sonia Hamilton wrote:
 After the keysigning at LCA, I'm receiving signings from people. Most of
 them I can import into my keyring in mutt using ^k, as the sender has
 used mime.
 
 But how to import the keys that don't use mime? I've tried gpg --import
 on the email, no joy. The mail headers look like this:
 
 ...
 ...
 Date: Sat, 20 Jan 2007 10:28:35 +1100
 X-Pgp-Agent: GPGMail 1.1.2 (Tiger)
 X-Mailer: Apple Mail (2.752.3)
 
 -BEGIN PGP MESSAGE-
 Version: GnuPG v1.4.5 (Darwin)
 
 hQIOAyLo+Up3tYIPEAf/Qg94dgzigZ7ZWBEIGeEum7mkycM5HVuDjnUixyirD6AX


I had trouble with a couple of keys also but I am using Icedove with
Enigmail. These ones were where gmail flagged them as spam and I had to
dig them out and forward them to my other account to decrypt. All the
ones that were forwarded automatically by gmail were OK.

This is probably not the right way to do it but it worked for me:)
Open the email and copy from BEGIN PGP... to END PGP... and save it to
somefilename.asc then email it to yourself as an attachment with nothing
in the body of the email and then it might decrypt and import OK???

#-BEGIN PGP MESSAGE-
Version: GnuPG v1.4.6 (GNU/Linux)

AFqU+IUbrwXpRMlAY/LQw8Ho7ikoVs7KtMkhGjHpfamDrSS/YaI8ZlKK
snip...
Bg==
#=F+uL
#-END PGP MESSAGE-

-- 
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html


Re: [SLUG] Extracting images from PDF file?

2006-11-25 Thread Tony Lissner
Robert Thorsby wrote:
 On 2006.11.24 09:23 Penedo wrote:
 Speaking of PDF processing - I have twice reserved rental
 cars through GINZ (ginz.com) and both times they sent me
 a scanned page of the reservation voucher inside a PDF file
 which won't print.

 The files display perfectly on my monitor and look like a
 scanned image.

 Is there some way to extract the image out of the PDF file
 so I can print it by some other means?
 
 Try pdf2ps which might give you a printable output.
 
 Try pdfimages.
 
 Try pdftk (PDF Toolkit)
 

If you only have a small number to do try the Gimp
open the pdf on the page number of the image and select it with the
dotted box thingy, crop and save as *.ps or whatever.

-- 
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html


Re: [SLUG] Making samba DWIM

2006-11-17 Thread Tony Lissner
On Fri, 2006-11-17 at 17:49 +1100, Adam Kennedy wrote:
 Hi gang
 
 I've been having some trouble with Samba I can't seem to work out 
 myself, and I was hoping someone can help.
 
 The setup I have for web work is a Windows 2000 desktop (for 
 Dreamweaver, Ultraedit, Firefox+IE+Opera, and so on) and a debian 
 dev/test server with Apache and Perl (just simple CGI in this case, but 
 it varies).
 
 To work on the files I mount a Samba share as a Windows drive, and just 
 edit everything directly on the Linux box (the apps are all set to use 
 Unix newlines etc). I can save/reload to see changes, and I generally 
 have half a dozens shell open on the box, and various other 
 infrastructure so that the who boxes work more like one hybrid machine 
 than two separate ones (in a previous incarnation they were even 
 duck-taped together).
 
 This workflow has worked really well for me for a number of years, but I 
 always have trouble fiddling the samba setup to the right place.
 
 For some reason, saving from the Windows box seems to reset the 
 permissions of the files. And in this particular case, that also means 
 644 (and thus not world-readable so nobody can do the CGI).
 
 Now, I assume that Windows is doing something funky and delete/write'ing 
 the file instead of overwriting the file.
 
 Could someone with more samba-fu than I explain if there's a way to make 
 these file writes behave themselves and maintain whatever permissions 
 they currently have?

Have you tried the
force create mode =
force directory mode =
man smb.conf


-- 
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html


Re: [SLUG] A Ubuntu/Debian and a grub question

2006-07-19 Thread Tony Lissner
[EMAIL PROTECTED] wrote:

 1) How will grub work out where to find menu.lst?  How will it 'know'
to use the updated /dev/hda7/boot/grub/menu.lst rather than the out
of date /dev/hda6/boot/grub/menu.lst version?
 
 2) I used apt to update the kernel image, and it thought I was still
using the / mounted on hda6 (it clobbered all the extra hda7 entries
I'd made and used to boot from).  How do I tell apt that I'm running
on hda7, not the hda6 which I was using when I first installed?
 
What made apt decide that hda6 is / - that disagrees with both
menu.lst and fstab?  Is that information stored in a 3rd place, too?
 
 I've manually fixed up the hda7 menu.lst file, but before I went any
 further I thought I'd ask about this, in case.
 
 I should and could copy the menu.lst file to both drives again, but
 thought I'd ask which one grub is really using, and how it picks it.
 
 I still wish grub had a mode that let you check the config before you
 rebooted.
 

The easiest way to set where grub looks for the stage* images might be
to just re-install grub from the new partition.
try unmounting /dev/hda6 and commenting it out in fstab, then from the
new partition do:
# sudo grub-install /dev/hda
# sudo update-grub
and it should find the new menu.lst


or checkout the '--root-directory=' option
install grub-doc
info grub


-- 
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html


Re: [SLUG] A Ubuntu/Debian and a grub question

2006-07-19 Thread Tony Lissner
Tony Lissner wrote:
 [EMAIL PROTECTED] wrote:
 
 1) How will grub work out where to find menu.lst?  How will it 'know'
to use the updated /dev/hda7/boot/grub/menu.lst rather than the out
of date /dev/hda6/boot/grub/menu.lst version?

 2) I used apt to update the kernel image, and it thought I was still
using the / mounted on hda6 (it clobbered all the extra hda7 entries
I'd made and used to boot from).  How do I tell apt that I'm running
on hda7, not the hda6 which I was using when I first installed?

Check where the kernel is looking for root with

rdev

If it is hda6 it should change to hda7 after a reboot

What made apt decide that hda6 is / - that disagrees with both
menu.lst and fstab?  Is that information stored in a 3rd place, too?

 I've manually fixed up the hda7 menu.lst file, but before I went any
 further I thought I'd ask about this, in case.

 I should and could copy the menu.lst file to both drives again, but
 thought I'd ask which one grub is really using, and how it picks it.

 I still wish grub had a mode that let you check the config before you
 rebooted.

 
 The easiest way to set where grub looks for the stage* images might be
 to just re-install grub from the new partition.
 try unmounting /dev/hda6 and commenting it out in fstab, then from the
 new partition do:
 # sudo grub-install /dev/hda
 # sudo update-grub
 and it should find the new menu.lst
 
 
 or checkout the '--root-directory=' option
 install grub-doc
 info grub
 
 

-- 
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html


Re: [SLUG] How to set runlevel in Breezy

2006-06-06 Thread Tony Lissner
On Tue, 2006-06-06 at 17:44 +1000, Kirti wrote:
 Fedora has neat way of conserving resources by changing from init 5 to 
 init 3.
 Ubuntu (breezy) comes with default runlevel 2 with GUI. How to change 
 this to CLI and stop most of the inessential services - I only need to 
 have database and apache servers running? What would then be an option 
 to change back to GUI when required?

Stop the gui
sudo /etc/init.d/gdm stop 

To prevent it starting on next boot
sudo update-rc.d -f gdm remove

To put it back
sudo update-rc.d gdm defaults

Or for an easier way to manage what
programs will run at boot
sudo apt-get install rcconf

When you need the gui
startx 

man update-rc.d

-- 
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html


Re: [SLUG] SLUG: lan and ppp[0] clash

2006-05-22 Thread Tony Lissner
On Mon, 2006-05-22 at 15:22 +1000, Nicholas Tomlin wrote:
 Sluggers,
 
 I've used Linux for some years now but I've struck an issue on which I need a 
 quick fix to my ignorance on - I ask of your knowledge on the following:
 
 I have on this machine [195.168.0.1] Mandrake 10.1 on an AMD 64 bit Athlon 3k 
 cpu with .5g ram and 2 80g HDD's and typically run kde 3.2.3 as my desk top.
 
 I have an 8 port 10/100 ethernet network switch which permits me to connect 
 computers with each other, I have this machine, a Tosh A10 Satellite 
 [195.168.0.2] and a dual boot mandrake / windoesn't box [195.168.0.3] which 
 MYOB's my accounts [please tell me there is a linux replacement to MYOB].
 
 network picture:
 
 domain tac.com.au
 [nameserver 223.225.0.3]
 [nameserver 223.225.0.6]
  |
  |
 dialup  -- Modem [ppp0]  195.168.0.1  --blue cable--[switch] 
   
   |
   
   |
 Laptop  ethernet port-- 195.168.0.2  --blue cable--[switch] 
   
   |
   
   |
 accounts -ethernet port-- 195.168.0.3  --blue cable--[switch] 
 
  So far I am able to transfer files between the laptop and this machine, this 
 is done where I share photos and the like between the laptop and this 
 machine, for this I use gftp and log in to the user on the Toshiba and 
 control the transfer from 195.168.0.1. ie, I start gftp from 195.0.168.0.1, 
 then specify the connection to 195.168.0.2:21 user nick pswd x and 
 push enter and I can see the contents of 195.168.0.2 from 195.0.168.0.1, and 
 can freely transfer files both ways using 195.168.0.1. I would like to be 
 able to do this from konqueror instead of gftp [despite its brilliance] 
 because there are other [ugh] windows users who need to do the same thing 
 though they do not grok any of this, gftp would be 'too hard' for them. How 
 can this be done? How can it be done from [ugh] windoesn't?
 
 All machines have an ethernet connection available and I am using the blue 
 cables, can ping each computer from 195.165.0.1, but cannot necessarily ping 
 from, eg, 195.168.0.2 back to 195.168.0.1, it tells me the host is 
 unreachable despite 'ifconfig 195.168.0.1 and 2 up' on both computers - any 
 ideas what I have done wrong here?

Some things to play around with.

Are the nic lights blinking on 0.1 when pinging from 2?
Run tcpdump on 0.1 to see if ping gets there

ping might be getting to 0.1 but not making it back to 2/3?

# iptables -L  check if firewall is not up (policy ACCEPT)

What are the results from 'netstat -rn' from all machines?

192.168.0.2 and 192.168.0.3 should point to gateway 192.168.0.1

To get across interfaces eth0 and ppp0 on 0.1
echo 1  /proc/sys/net/ipv4/ip_forward

 
 When I have the lan - ethernet - network running I seem to be precluded from 
 using the internet on this machine 195.168.0.1, and I can't see this machine 
 from 195.135.0.2, nor the internet connection via ppp0. I can have ppp0 
 running and eth0 running but it will only see the eth0 interface. When I do 
 'ifconfig 195.168.0.1' down' and then try the internet connection again I can 
 see the internet and can send / receive emails any ideas on what I've 
 done wrong with this set up? screen captures of what to do would really help 
 me with this..
 
 Sometimes I run the laptop in windoesn't for the purpose of downloading 
 pictures from my LG880U camera phone [no, it won't work in linux, no drivers, 
 dammit]. Any ideas on how to get around this issue? It does have a bluetooth 
 so I might be able to get a bt card and do it via that.. ideas please?
 
 tia
 
 Nick Tomlin.
 www.alarmist.com.au
 We scare for you...

-- 
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html


Re: [SLUG] FireFox 'open with' csv excel

2005-11-25 Thread Tony Lissner

Roger Barnes wrote:

Hi Stu,
 


I am wondering if anyone else has had this problem?



I have. :)


When we create a CSV (comma seperated values) text file in 
our app and firefox passes it to Excel (yes I know, immovable 
object currently but working on it! OO _is better), all of 
the data is thrown to column one in a completely boneheaded manner.


Normally, one would expect Excel to start the whole import 
text file stuff ie how the text should be imported 
(delimited/fixed) whatever.



If you haven't already, try naming the file from the web browser with a .csv extension 


For example, if your application is a cgi, eg 
http://somehost.com/sendsomecsv.php?param1=value1, try adding a csv filename 
after the script name, eg 
http://somehost.com/sendsomecsv.php/export.csv?param1=value1

Also, try experimenting with the content-type header that you return:
Content-type: application/vnd.ms-excel
Content-type: application/csv
Content-type: text/csv

One or both of these approaches has worked for me in different situations, but 
I'm not sure how well it applies for firefox.



This works in Firefox for our application using php4.
Excel doesn't ask you what to do with it because it already 
knows (and always gets it wrong). Different versions of Excel 
don't handle this the same so it might need tweaking.


If this doesn't do what you want there are programs to write 
proper Excel files and you shouldn't have to do any 
escaping/fugding the data.


header('Content-type: application/octet-stream');
$filename = date(d-M-Y-) . 'file.csv';
header(Content-Disposition: attachment; filename=$filename);
header('Pragma: no-cache');
header('Expires: 0');

If there are any fields that start with 0 (zero) it will drop 
the leading 0 unless you fool it by adding an equal sign and 
double quote the value ,=\$fieldname\,


I have found that it doesn't matter what spreadsheet program 
you use, they are all braindead when used this way, including 
00 and Gnumeric.

--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html


Re: [SLUG] debian

2005-10-23 Thread Tony Lissner

Paul Maloney wrote:

Hi all.
Well I finally got denian up and running and found it to be a great OS. 
My problem is that Kppp will not run, when I click on Kppp nothing 
happens. How can I fix this.

Thanking you in advance


Kppp might not be installed?
If you install it, it may depend on a heap of other kde stuff 
OR use 'pppconfig' which should already be installed.

--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html


[SLUG] Sarge ssh dropping connection after upgrade

2005-06-18 Thread Tony Lissner


This setup has been running without problems
for the last 3 - 4 years until the upgrade.

I have been having trouble with ssh dropping the
connection ever since the server was upgraded to
Sarge last week. The upgrade went ok with no 
errors or warnings.


Two computers on an internal 192.168.xxx.xxx 
network not connected to the internet.


Normally the client machine(still woody) ssh's to
the server (upgraded to sarge) with key based
authentication and stays connected until the next
power failure.

Now since the upgrade every night the client
loses the connection and has to be re-connected
manually every morning.

The ssh config files seem to be identical (diff).
The only thing that has changed is the upgrade.
The client is still running woody and hasn't been
touched for a year or two.

A few days ago I was connected to the server with
two xterms at the same time and one dropped out
but the other remained connected?

So just wondering if any one else is having this
problem with Sarge?

I'll go through the logs later, and if anyone
has any ideas what to look out for?

In desperation I could always try telnet(yea I
know, boo hiss spit) to isolate
ssh from the likely candidates.

in sshd_config:
KeepAlive yes
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html


Re: [SLUG] Linux/Windows shared partition question

2005-06-01 Thread Tony Lissner

Leslie Katz wrote:
I have a computer which, until recently, had one hard drive. My 
operating system was Windows XP Professional. The drive had two 
partitions, one the usual C:\, of file type NTFS, and the other, of file 
type FAT32, a hidden one from which, as I understand it, I can restore 
Windows, if necessary.


I recently installed a second hard drive in the computer and, on it, 
installed Fedora Core 3. On the second drive, I created a /boot 
partition and a / partition, both of file type ext3. I also created a 
swap partition.


I knew I could create on the second drive a partition of the vfat file 
type and so I did that as well, understanding that that partition would 
be capable of being used by both OSs.


However, that partition does not appear when I run Windows.


Have you formatted it in Windows?



Was my understanding wrong? Have I simply created an unnecessary partion 
accessible only by FC3?




--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html


Re: [SLUG] St George banking (again...)

2005-03-18 Thread Tony Lissner
Sonia Hamilton wrote:
YASGQ (Yet Another St George Question):
I can't get St George internet banking going, has anyone got this
working? Pointers on how to do it?
I've searched the slug archives, lot's of (justified) kvetching but I
couldn't find a solution. I've looked at [1], installed prefbar [2],
tried every agent string. I'm running Ubuntu Warty, I've got SunJava 1.5
installed.
I've written to STG, I get the usual response... blah, blah supported
browsers, blah IE blah :-)
I have been using St George Internet Banking
for about four weeks now with Debian Sarge.
Firefox:
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.5) Gecko/20050110 
Firefox/1.0 (Debian package 1.0+dfsg.1-2)
about:plugins
Java(TM) Plug-in 1.5.0_01-b08
(I had trouble with the Java version before
this one, it wouldn't work(TM)).

Even though I had '*.stgeorge.com.au ALLOW' in
the 'Web Features - Block Popup Windows - Allowed Sites'
I still had to turn off Popup Blocking to get it to work
and just used Firefox default user agent and it works OK.
Mozilla:
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.5)
Gecko/20050105 Debian/1.7.5-1
Java(TM) Plug-in 1.5.0_01-b08
Using teh default User Agent here too and
'*.stgeorge.com.au' in the Allowed Sites
it works with blocking on:)
Other than the above nothing else was done.
Tony


--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html


Re: [SLUG] dpkg-reconfigure does not write a new XF86Config-4 file

2004-12-06 Thread Tony Lissner
Michael Lake wrote:
Hi all
I have a strange problem. Im trying to create a new XF86Config-4 file as 
after an upgrade X does not work anymore. I get no screens found.
But hen I do a dpkg-reconfigure xserver-xfree86 a new XF86Config-4 
config file is not written.

There are no errors during the reconfiguration, I answer all the dialogs 
 using the default values from last time but no new config is written at 
the end. The config would normally be in /etc/X11 or maybe /root

I have searched with find for XF86Config-4 and XF86Config-4.new
I have even tried removing any old XF86Config-4 but to mo luck.
I had the same thing happen when my XF86Config-4
file was altered by another program.
From Sarge /etc/X11/XF86Config-4
# If you have edited this file but would
# like it to be automatically updated
# again, run the following commands as root:
#
# cp /etc/X11/XF86Config-4 /etc/X11/XF86Config-4.custom
# md5sum /etc/X11/XF86Config-4 /var/lib/xfree86/XF86Config-4.md5sum
# dpkg-reconfigure xserver-xfree86
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html