[SLUG] VPN'ing and a 2.2.18pre21 kernel...

[Anthony O'Hara]

Hi all, 

I'm trying to get Freeswan running on my debian box, and I've
run into a problem.. :) :) 

It would seem my efforts at patching the kernel so ip_masq_pptp
would work has left me in a position of not being able to patch it
for IPSec support... (Needed for Freeswan)

Can anyone point in the direction of a 2.2.18pre21 kernel source 
tarball? I've looked in all the usual places, but I may be going
blind.. :)

My plan is to patch the original Config.in (which is the file causing
me grief) from the tarball, see whats changed, and incorporate that
into a Config.in that already has the ip_masq patched and happy.. 

Any additional clues are greatly appreciated...

Anthony...

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] setting permissions for web pages, chmod what ?

[Voytek Eymont]

** Reply to note from Rick Welykochy <[EMAIL PROTECTED]> Mon, 07 Oct 2002 14:00:15 
+1000


> The user that httpd runs as (usually httpd, amazingly enough). 
> Do:  ps uxwa | grep http to find out, 
>
> or look in httpd.conf for the following: 
>
> User XXX 
> Group YYY 
>
>
>   
> > what permissions do I need to set for public web pages ? 
>  
> Readable by the httpd process. 
>  
> Suggestion: login to your server as the same user that the httpd runs 
> as, and try to access the web files. Any problems will show up immediately ;)

thanks, Rick

httpd.conf is set for apache/apache

apache was created by the install/setup stuff, is there a way to determine
it's password, or, do I just reset it ? 

if I creata a user, and, do NOT set a password for that users: is there
some default password ? or, UNTIL I set a password, that user can not log
in ?

sorry, I'm very new to multi user file system, takes a while to get
used...

ok, chown and chmod is fine, but, what tells me who owns what, what are the
current permissions ?

tia,


Voytek Eymont
SBT Information Systems Pty Ltd
http://www.sbt.net.au/links/
phone +61-2 9310-1144 fax +61-2 9310-1118 
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] setting permissions for web pages, chmod what ?

[Voytek Eymont]

** Reply to note from Voytek Eymont <[EMAIL PROTECTED]> Tue, 8 Oct 2002 18:19:05


> httpd.conf is set for apache/apache 
>  
> apache was created by the install/setup stuff, is there a way to determine 
> it's password, or, do I just reset it ?

I've changed apache's password, now, when I try to login as 'apache' it
says something very breifly, then, I ;m out at the login prompt, can't
login as apache.

where does that log to failed local login attempts ?

Voytek Eymont
SBT Information Systems Pty Ltd
http://www.sbt.net.au/links/
phone +61-2 9310-1144 fax +61-2 9310-1118 
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] setting permissions for web pages, chmod what ?

[Rick Welykochy]

Voytek Eymont wrote:

> apache was created by the install/setup stuff, is there a way to determine
> it's password, or, do I just reset it ?

You can get access as user apache sans password as follows:

su -

su - apache


 
> if I creata a user, and, do NOT set a password for that users: is there
> some default password ? or, UNTIL I set a password, that user can not log
> in ?

Depends on your system. As a general rule you should add the user,
then explicitly set the password for the user.



> ok, chown and chmod is fine, but, what tells me who owns what, what are the
> current permissions ?

man ls
the 'ls' command does it all :)


cheers
rickw


-- 
_
Rick Welykochy || Praxis Services

Politicians and diapers have one thing in common. They should be changed
regularly and for the same reason.
-- Gerry Brooks (in the Toronto Globe & Mail)
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] hd error, how to test ?

[Voytek Eymont]

just spotted this on the console:

[xxx@xxx root]# hda: dma_intr: status=0x51 { DriveReady
SeekComplete Error }
hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
ide0: reset: success


how to test my hd, is that a big worry, the above message ?

is there a SMART client app in RedHat/Linus to assess drive's SMART status
?

Voytek Eymont
SBT Information Systems Pty Ltd
http://www.sbt.net.au/links/
phone +61-2 9310-1144 fax +61-2 9310-1118 
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] setting permissions for web pages, chmod what ?

[Terry Collins]

Voytek Eymont wrote:
> 
> I've changed apache's password, now, when I try to login as 'apache' it
> says something very breifly, then, I ;m out at the login prompt, can't
> login as apache.

My 2c on this/AIUII.

You don't login as "apache"
On my system, user "apache" doens't have a shell that you can log into.
from /etc/passwd:  apache:x:48:48:Apache:/var/www:/bin/false
It is a security feature.

To start and stop your apache server /etc/rc.d/init.d/httpd
start|stop|status|???

Then in your browser http:localhost or ip or whatever.


-- 
   Terry Collins {:-)}}} Ph(02) 4627 2186 Fax(02) 4628 7861  
   email: [EMAIL PROTECTED]  www: http://www.woa.com.au  
   Wombat Outdoor Adventures 

 "People without trees are like fish without clean water"
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] How to test RAID speed ?

[cmackenzie]

On Mon, Oct 07, 2002 at 06:07:55PM +1000, Jeff Waugh wrote:

> > I'm in possession of a Compaq Proliant ML370 with a DAC960 raid controller
> > installed and I was wondering what I can use to test the through put speed
> > of this card & drives?
> >  
> > I've tried hdparm, but it doesn't like the raid devices (/dev/rd/c0d0 etc)
> > Is there some other tool similar to hdparm that I'm not aware of that
> > supports raid drives ?
> 
> bonnie++: Hard drive bottleneck testing benchmark suite.
> 
> Kicks the crap out of your I/O subsystem. Lots of fun for all the family.

Thanks Jeff, that works a treat. and yes I did stuff up my email 
configuration by changing usersnames on my local box. I've told mutt to 
munge the replyto header until I configure sendmail to put a correct from 
field in :-)


-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] How to test RAID speed ?

[cmackenzie]

On Tue, Oct 08, 2002 at 12:55:00PM +1000, Matthew Hannigan wrote:

> > btw, your From: address is a black hole.
> 
> I think Optus recently changed all their subscribers
> email addresses and blackholed existing ones.
> 
> Delightful behaviour.

They did no such thing. I just forgot to correct sendmails configuration 
afetr changing my username on my local linux box. I've bodged up a replyto 
header within mutt until I modify sendmails configuration.

Rgds,
Chris MacKenzie

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Mail

[Paul Davies]

Dear Sluggers

I am new to linux and I want to be able to send and receive email from my 
linux box.

When I log in as root my email address is [EMAIL PROTECTED] and I 
can use sendmail to send an email to my hotmail account.  However I want to 
be able to send email as say [EMAIL PROTECTED]  How do I do this if I log on as 
root?

I don't understand how you set up your mail user agent in unix. In windows I 
tell the email user agent what my pop server and smpt server is.  The client 
communicates with these servers.  Easy.

I am trying to use pine.  I can't see where pine tells me what my pop and 
smtp servers are.  Am I blind?  Is pine and elm merely mail readers as 
opposed to an email user agent.  I can read roots mail with the mail 
readers.

All I want to do is be able to send an email and receive and email as 
[EMAIL PROTECTED]  I have looked at the sendmail site and it immediately trails 
off into how to set up pop servers etc.  It is not immediately obvious where 
it explains how to do the trivial stuff!

Paul





_
Join the world’s largest e-mail service with MSN Hotmail. 
http://www.hotmail.com

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Mail

[Jeff Waugh]

> I don't understand how you set up your mail user agent in unix. In windows
> I tell the email user agent what my pop server and smpt server is.  The
> client communicates with these servers.  Easy.
> 
> I am trying to use pine.

I'd strongly recommend Evolution, KMail, Sylpheed, or similar full-featured
GUI clients whilst you're starting. Most of the text-based clients require
you to set up your mail transfer agent (sendmail, postfix, exim, etc), which
requires more knowledge and work than is really required.

- Jeff

-- 
 "Funny, I have no trouble distinguishing my mobile phone from the  
   others because it's in my _own fucking pocket_!" - Mobile Rage   
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Mail

[Kevin Saenz]

you will need to let us know.
if the machine is registered in DNS as an MX record.
it's all fine and well you installing either sendmail, postfix, or some
other mail server on your box. but if DNS doesn't recognise it then you
have a few problem.



> 
> Dear Sluggers
> 
> I am new to linux and I want to be able to send and receive email from my 
> linux box.
> 
> When I log in as root my email address is [EMAIL PROTECTED] and I 
> can use sendmail to send an email to my hotmail account.  However I want to 
> be able to send email as say [EMAIL PROTECTED]  How do I do this if I log on as 
> root?
> 
> I don't understand how you set up your mail user agent in unix. In windows I 
> tell the email user agent what my pop server and smpt server is.  The client 
> communicates with these servers.  Easy.
> 
> I am trying to use pine.  I can't see where pine tells me what my pop and 
> smtp servers are.  Am I blind?  Is pine and elm merely mail readers as 
> opposed to an email user agent.  I can read roots mail with the mail 
> readers.
> 
> All I want to do is be able to send an email and receive and email as 
> [EMAIL PROTECTED]  I have looked at the sendmail site and it immediately trails 
> off into how to set up pop servers etc.  It is not immediately obvious where 
> it explains how to do the trivial stuff!
> 
> Paul
> 
> 
> 
> 
> 
> _
> Join the world’s largest e-mail service with MSN Hotmail. 
> http://www.hotmail.com
> 
> -- 
> SLUG - Sydney Linux User's Group - http://slug.org.au/
> More Info: http://lists.slug.org.au/listinfo/slug


--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Mail

[Mary]

On Tue, Oct 08, 2002, Paul Davies wrote:
> I don't understand how you set up your mail user agent in unix. In
> windows I tell the email user agent what my pop server and smpt server
> is.  The client communicates with these servers.  Easy.

Some mail clients in UNIX/Linux, notably the graphical ones, do behave
like this. Jeff's mail mentions several. I don't know which Windows mail
reader you were using, but if it was Outlook, Evolution is probably the
most similar. Netscape Mail you can probably find as either Netscape or
Mozilla Mail on Linux. I don't know which one is closest to Eudora, but
most of the graphical UNIX mail clients are very similar in features to
Windows ones.

> All I want to do is be able to send an email and receive and email as
> [EMAIL PROTECTED]  I have looked at the sendmail site and it immediately
> trails off into how to set up pop servers etc.  It is not immediately
> obvious where it explains how to do the trivial stuff!

I'd strongly suggest switching to a graphical mail reader rather than
trying to configure a mail transport agent, such as sendmail, at this
stage. They're designed to be configured by people who are intending to
run large, publically accessible mail servers. It can be quite difficult
to find documentation on how to set them up as a simple local program
that ferries messages to your ISP - it isn't really what they are
designed for, although they can do it. It's also easy to spend a long
time setting them up with no visible progress for a long time. Very
annoying and off-putting.

-Mary
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] hd error, how to test ?

[Jon Biddell]

At 08/10/02  18:37, you wrote:
>just spotted this on the console:
>
>[xxx@xxx root]# hda: dma_intr: status=0x51 { DriveReady
>SeekComplete Error }
>hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
>hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
>hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
>hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
>hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
>hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
>hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
>ide0: reset: success

Ewww... What type of h/d is it ?  There are several freely-downloadable 
testing tools from some of the drive manufacturers, but unfortunately the 
only ones I've used run under that "other" O/S...

You could try doing an fsck, I guess, but be prepared for almost anything.

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] hd error, how to test ?

[Grant Byers]

* Voytek Eymont ([EMAIL PROTECTED]) wrote:
> hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
> hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
> hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
> hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
> ide0: reset: success
> 
I believe this is caused by an old ATA33 drive operating over an 
ATA66 cable. 

I've had the same messages since about 2.4.8 and other than the messages, 
never had data corruption or any instabilities whatsoever.


Cheers,
Grant
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] hd error, how to test ?

[Andrew Bennetts]

On Tue, Oct 08, 2002 at 06:37:36PM +, Voytek Eymont wrote:
> 
> is there a SMART client app in RedHat/Linus to assess drive's SMART status
> ?

In Debian, there's a package called "ide-smart", and also "smart-suite",
which have this functionality.  I expect they are also available for Red
Hat.

-Andrew.

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

RE: [SLUG] Redhat 7.3, 8.0 & P4 2.5 gig, any ideas ? - Solution

[Richard Neal]

DDR333 has just become stable (well relatively) under the other OS, I
would say it should be becoming stable around now for Linux.

WARNING don't use DDR400 its totally unstable even under that other OS

On Tue, 2002-10-08 at 09:03, Brett Olsen wrote:
> Hi Guys,
> I thought I would post back the solution to the problem I was having with
> Linux on new hardware.
> 
> Problem   : Kernel would hang at the point where it says "Uncompressing
> Linux..."
> 
> Solution  : Add the following parameters to the bootloader :
> mem=exactmap mem=640K@0 mem=510M@1M (where 510=amount of mem-1M)
> 
> Hardware  : P4 2.5gig, 845G chipset, 512M DDR Ram, Geforce 4 Ti4200
> 128MB
> 
> Tested On : Redhat 8.0, Mandrake 9.0 both now working fine.
> 
> Cause : Probably the chipset, the 845G is quite new, could also be
> affected by 333Mhz DDR ??
> 
> Thanks for you suggestions and help
> 
> Brett
> 
> 
> 
> 
> Hi Guys,
> Thanks for all the suggestions, Erik van Pienbroek got back to me with this
> suggestion, I will post the results of my testing tomorrow, his symptoms
> match exactly those that I am having now.
> 
> ---
> Hi,
> 
> Sounds just like my problem.
> I've just got an new P4 2,4Ghz box and when I tried to boot Linux (various
> Distro's tried) the kernel would hang at the point where it says
> "Uncompressing linux...". After monitoring the Linux Kernel mailing lists,
> I've found out the problem could be solved by passing the following
> parameters to the bootloader : mem=exactmap mem=640K@0 mem=510M@1M Since
> then I haven't had any troubles running linux on my new box.
> 
> Hope this works for you too.
> 
> Erik van Pienbroek
> ---
> 
> I will post back again if this works, if so, perhaps someone else won't have
> to spend days trying to get linux going again
> 
> Thanks Guys
> Brett
> 
> -Original Message-
> From: Richard Neal [mailto:[EMAIL PROTECTED]] 
> Sent: Thursday, 3 October 2002 11:08 PM
> To: [EMAIL PROTECTED]
> Subject: Re: [SLUG] Redhat 7.3, 8.0 & P4 2.5 gig, any ideas ?
> 
> 
> quick note here the Epox 4g4a+ mobo is a great update board for an older ATX
> box and PSU.
> 
> I talked recently with a friend who has one of these and he updated to the
> 2.4.19 after some stability problems with the 2.4.18 RH default kernel then
> noted "this machine rips" , so much the thing is running a bit hot and he
> had to upgrade the HSF; Ive noticed this too when doing allot of compiling
> on a P4 the stock HSF doesn't cut it and the cpu runs a bit on the toasty
> side.
>   
> > On Thu, 2002-10-03 at 14:43, Matthew Hannigan wrote:
> > > On Thu, Oct 03, 2002 at 12:37:46PM +1000, Brett Olsen wrote:
> > > > Hi Guys,
> > > > I have just upgraded to a P4 2.5 and I do not seem to be able to
> > > > install/run Linux anymore :-(
> > > [ .. ]
> > > > Epox 4g4a
> > > > Intel P4 2.5
> > > > 2 x 256MB DDR
> > > 
> > > 
> > > > Leadtek 128MB Ti4200
> > > 
> > > I
> -- 
> 
> * Hey if you're going to get mad at me every time I do something   *
> * stupid, then I guess I'll just have to stop doing stupid things! *
>   
> 
> -- 
> SLUG - Sydney Linux User's Group - http://slug.org.au/
> More Info: http://lists.slug.org.au/listinfo/slug
> -- 
> SLUG - Sydney Linux User's Group - http://slug.org.au/
> More Info: http://lists.slug.org.au/listinfo/slug
> -- 
> SLUG - Sydney Linux User's Group - http://slug.org.au/
> More Info: http://lists.slug.org.au/listinfo/slug
-- 

* Hey if you're going to get mad at me every time I do something   *
* stupid, then I guess I'll just have to stop doing stupid things! *
  

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] hd error, how to test ?

[Heracles]

On Wednesday 09 October 2002 4:37 am, Voytek Eymont wrote:
> just spotted this on the console:
>
> [xxx@xxx root]# hda: dma_intr: status=0x51 { DriveReady
> SeekComplete Error }
> hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
> hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
> hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
> hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
> ide0: reset: success
>
>
> how to test my hd, is that a big worry, the above message ?

Not sure about the Linux way, but if you can get a seatools floppy, you can 
boot off it and either - if it is a seagate drive - use the seagate specific 
tools or otherwise use the generic setting. it will do a test of the 
mechanism and give you a report.

Stay well and happy
Heracles
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] enabling bash auto completion

[Ken Foskey]

Gday,  I just enabled autocompletion under bash.  It upgraded on my
debian system but my .bashrc did not contain the lines to load it.

Check for a file /etc/bash_completion.

check your ~/.bashrc for lines similar to the following:

if [ -e etc/bash_completion ]; then
source /etc/bash_completion
fi


Thanks
KenF



-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Decided to go it "alone"

[Mick Boda]

I have decided to configure my own Linux server using Redhat 7.2 rather than 
the "click next" esmith server solution I have been using.

I decided on this route because the only way to learn is to DO... not let 
something do the configs for you.  I have a temp server with esmih to provide 
internet sharing while I get in "Knee deep in depths of hell".  

If I need help (when I need help) I promise to call wish me luck.

Mick

Anyone interested in forming a Wollongong Linux community drop me line.
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] rsync'ing in a script

[David]

I want a script to automate file back-ups, and I've tried the following
with no success:

rsync -a -e ssh --password-file=file /etc me@elsewhere:/backup 2>
/dev/null

This works ok, except that it asks for a password on the command line.

 contains the password on a single line, and nothing else. I've
tried it with and without \n and neither work without manually entering
the password.

Any suggestions?

David.

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] rsync'ing in a script

[Jeff Waugh]

> rsync -a -e ssh --password-file=file /etc me@elsewhere:/backup 2>
> /dev/null
> 
> This works ok, except that it asks for a password on the command line.

> Any suggestions?

I'd suggest using a passphrase-less ssh key.

- Jeff

-- 
"I can't imagine anyone telling Emma Bunton to shut up. It would be 
rather like slapping Bambi."
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Linux in the lab

[Richard Neal]

Hey Tux people

Ive just read this article posted on slashdot but the following
paragraph caught my eye-

"Future supercomputers must be cost-effective, efficient and easy to
enhance and scale. Scalable supercomputing systems that run proprietary
operating systems clearly are a thing of the past," said Bill
Feiereisen, leader of Los Alamos' Computer and Computational Sciences
Division. "Instead of buying a complete proprietary computing system, we
are looking toward a future in which a robust set of integrated, open
source software tools enables us to assemble a truly scalable
supercomputer from components that best meet our needs. We are excited
by the new partnership with Linux NetworX to help advance these goals."

Now when you have people like Los Alamos National Laboratory saying
Linux rules it makes the whining from Richmond seem pretty pathetic

ooh and the URL

http://www.lanl.gov/worldview/news/releases/archive/02-106.shtml

Regards

Richard (were is my DVD copy of mdk 9.0) Neal


-- 

* Hey if you're going to get mad at me every time I do something   *
* stupid, then I guess I'll just have to stop doing stupid things! *
  

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] CDRW problem

[James Gregory]

On Tue, 2002-10-08 at 17:10, Tom wrote:
> Hi,
> 
> Could SLUG do a better job of advertising The Linux Documentation
> Project?  How about adding a 'Linux Help' page to www.slug.org.au,
> which would simply say "Please go to www.tldp.org for help.  If they
> don't have a HOWTO, then you don't need to know how to."?

And something for newbies wanting to do a first linux install on
dual-boot configs. parted is a really neat tool for chopping up windows
partitions. I spent about two days messing around with defrag on a
windows computer recently, and eventually found parted, downloaded the
boot images and it worked, first time. Couldn't be easier.

http://www.gnu.org/software/parted/

James.

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] perl tutorials (thursday nights)

[Angus Lees]

The weekly perl tutorials are continuing this thursday night at the
PCAN shop in Parramatta (http://www.cat.org.au/pcan/), 6-8pm.

In the previous two weeks, we've covered basic perl data types and
control structures.  This week we'll go into regular expressions,
writing functions and variable scoping.  The next few weeks will cover
references, nested data structures, file I/O, CPAN and using modules.

Unfortunately, space is quite limited (only really desks/computers for
10 people), so please email Matthew Davidson <[EMAIL PROTECTED]>
beforehand to confirm your spot.  Preference will be given to those
who have attended before (you should probably mention this explicitly
in your email).

In order to help PCAN raise much-needed funds, we'll be charging a $5
fee at the door (or probably slightly inside the door for practical
reasons).

-- 
 - Gus



msg26845/pgp0.pgp
Description: PGP signature

Re: [SLUG] Mail

[Amanda Wynne]

- Original Message - 
From: "Mary" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, October 08, 2002 7:50 PM
Subject: Re: [SLUG] Mail


>I don't know which one is closest to Eudora, but

That would be Kmail. I use both at home. 
At work I use Outlook Express, and I hate it.

My 2 cents worth.

Amanda

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] rsync'ing in a script

[Gareth Walters]

- Original Message -
From: "David" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, October 08, 2002 10:59 PM
Subject: [SLUG] rsync'ing in a script


>
> I want a script to automate file back-ups, and I've tried the following
> with no success:
>
> rsync -a -e ssh --password-file=file /etc me@elsewhere:/backup 2>
> /dev/null
> This works ok, except that it asks for a password on the command line.

I did something like this a while ago and as I recall it was a bit of a pain in
the neck but the final solution worked out somehting like this...
#!/bin/sh
eval `/usr/local/bin/ssh-agent2`
echo "passphrase"| ssh-add2 -p pathtokey
rsync -vrzC --progress /path1
username@ipaddress:/path2 --rsh="/usr/local/bin/ssh2"

This mysteriously stopped working one day and I just created a new key without a
passphrase
as I didn't have the time to work out why it died.


---Gareth Walters

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] killing a runaway console; mouseless mc [^] command line history

[Voytek Eymont]

1. 
if I get a 'problem' on a console, such as:

I can't get out of an editor...
and, after a while, I've tried every possible sequence I can think off,
and, I'm still inside an editor

how do I kill whatever on say console 2 ?

2. 
if I'm running mc without a mouse, mc has a command line history as
'[^]' , how can I get that without mouse, what key combo ?

TIA, 


Voytek Eymont
SBT Information Systems Pty Ltd
http://www.sbt.net.au/links/
phone +61-2 9310-1144 fax +61-2 9310-1118 
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Help wanted - graphic artists :-)

[David Guest]

Hi Sluggers

I am a general practitioner involved with a little project to write an 
open source patient management program for doctors in community practice 
and hospitals. Gnumed is GPLed and written in python/wxpython and postgres.

The CVS moved from sourceforge to savannah a few months ago and Richard 
Stallman has politely suggested that we replace the words "open source" 
in our logo (see website logo, http://gnumed.org/). I think he prefers 
"free". :-)

The original designer of the logo is no longer contactable. A few of us 
could cut it up in the gimp but none of us have much experience or skill 
with graphics.

Any takers will be paid at the usual free software rates. :-(

David

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

defrag ? was Re: [SLUG] CDRW problem

[Voytek Eymont]

** Reply to note from James Gregory <[EMAIL PROTECTED]> 09 Oct 2002 01:56:45 +1000


> partitions. I spent about two days messing around with defrag on a 

what does one use to assess ? and, defrag ? Linux ext3 hd ?



Voytek Eymont
SBT Information Systems Pty Ltd
http://www.sbt.net.au/links/
phone +61-2 9310-1144 fax +61-2 9310-1118 
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: defrag ? was Re: [SLUG] CDRW problem

[Jeff Waugh]

> > partitions. I spent about two days messing around with defrag on a 
> 
> what does one use to assess ? and, defrag ? Linux ext3 hd ?

There is a utility 'defrag' that works on ext2 (which ext3 is based on), but
it's a very, very rare occasion that you'd need to use it. ext2/3 are pretty
tidy on their own.

- Jeff

-- 
  "Anyway - I need something more James Bond than Banana Man, if you know   
   what I mean..." - Tom Gilbert
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

RE: defrag ? was Re: [SLUG] CDRW problem

[Rowling, Jill]

You don't bother to defrag a Linux filesystem. Like most Unixes, Linux
partitions don't get defragmentation problems.
There is a permanent fragmentation which is quite small (varies between OS's
but might be about 0.2% or so) and an algorithm that keeps it neat and tidy.

Some Unixes (and I think Linux is in this category) vary the allocation
algorithm according to how full the disk is: on a fairly empty disk it's
optimised for speed and on a fairly full (~95%) disk it optimises for space.
So full filesystems run slower.
The rules will vary for each filesystem, so EXT2 will be different to EXT3,
XFS, ReiserFS, etc.

Cheers,

Jill.

-- 
Jill Rowling, System Administrator
Eng. Systems Dept, Aristocrat Technologies Australia
Level 2, 55 Mentmore Ave Rosebery NSW 2018
Phone: (02) 9697-4484 Fax: (02) 9663-1412 Email: [EMAIL PROTECTED]
 


-Original Message-
From: Voytek Eymont [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, 9 October 2002 20:02
To: [EMAIL PROTECTED]
Subject: defrag ? was Re: [SLUG] CDRW problem


** Reply to note from James Gregory <[EMAIL PROTECTED]> 09 Oct 2002
01:56:45 +1000


> partitions. I spent about two days messing around with defrag on a 

what does one use to assess ? and, defrag ? Linux ext3 hd ?



--
CONFIDENTIALITY NOTICE
--
This email is intended only to be read or used by the addressee.
The information contained in this e-mail message may be confidential
information. If you are not the intended recipient, any use, interference
with, distribution, disclosure or copying of this material is unauthorised
and prohibited. Confidentiality attached to this communication is not waived
or lost by reason of the mistaken delivery to you.

If you have received this message in error, please delete it and notify us
by return e-mail or telephone Aristocrat Technologies Australia Pty Limited
on +61 2 9413 6300.
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] hd error, how to test ?

[Voytek Eymont]

** Reply to note from Jon Biddell <[EMAIL PROTECTED]> Tue, 08 Oct 2002 20:17:45 +1000


> Ewww... What type of h/d is it ?  There are several freely-downloadable  
> testing tools from some of the drive manufacturers, but unfortunately the  
> only ones I've used run under that "other" O/S...

it's a 6GB IDE Seagate originally from a Compaq Deskpro P2-400, now in a
P2-266 clone.

fwiw, all these hd test tools boot off the floppies, with a run-time dr/pc
dos

I'm running SeaTool/Ontrack right now, short test was 'good', long test in
progress.


Voytek Eymont
SBT Information Systems Pty Ltd
http://www.sbt.net.au/links/
phone +61-2 9310-1144 fax +61-2 9310-1118 
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Linux, New Hardware & Me

[Brett Olsen]

Hi Guys,
I have Linux running great again at home, I thought I would post the steps
that I went through to get it working, I know you all probably know this,
but anyway  

Hardware : Intel P4 2.5gig, Intel 845G Chipset, 512MB DDR 333Mhz, Leadtek
Ti4200 128MB DDR

Problems : Linux would not boot, X would not recognise my video card, My
Chipset was not fully supported and thus DMA was not enabled.

Solutions :
Installing Linux - Firstly as you all know, I could not boot Linux at all, I
found through other groups that adding mem=exactmap mem=640K@0 mem=510M@1M
(where 510M=total memory -1) fixed the issue.
To get the installer of your favourite distro working I appended this line
to the Linux start-up command, so it would be : Linux mem=exactmap
mem=640K@0 mem=510M@1M.
If you X configuration fails, don't worry, just set it to the correct
settings and set the boot mode to text, this will make it easier when
loading the driver latter.
Once you have installed, you need to modify your boot settings to include
the above parameters, I used grub because I could not get LILO to accept
mem=exactmap (complained about requiring a numerical value).
When booting press e to edit your boot options, append the above line, once
booted remember to modify your appropriate grub config file, its menu.lst in
Mandrake.
* Note to mandrake(& Redhat?) users, I could not get lilo working, so be
careful using the rescue mode off the install CD as it will re-install lilo
and not grub if you replace the boot loader, any ideas on getting lilo to
work anyone ?

X Configuration - X was unable to use the nv driver for my card, this was
easy as I was going to replace the driver with NVIDIA's anyway.
Go to www.nvidia.com, download 1x Kernel and 1x GLX driver, they must be
matching, I recommend getting the source from the page.
Make install both the kernel and the GLX driver, the module should load and
complain about tainting your GPL license, this is good as it means that the
kernel module is ready to go.
Modify you XF86config-4 file and replace the driver line nv with nvidia,
save and you should be able to now start X.

Cool, Now I had Linux installed and video, but it really performed sluggish
(no pun intended).

DMA, Chipset Support - The version of the kernel I was using is
2.4.19-16mdk, this version was unable to detect my IDE chipset correctly,
this means no DMA, not a good thing with a modern machine.
To test for DMA support I downloaded hdparm from
http://www.ibiblio.org/pub/Linux/system/hardware/ you can check to see the
status of your drive by running hdparm /dev/hda (substitute this for your
HDD), to test the drive run hdparm -tT /dev/hda, it only was getting
3.2mb/sec before changing my kernel :-(
After some trawling of the newsgroups I found a post by Alan Cox, suggesting
to use 2.4.19-ac4, I don't know if this is the best version to use (any
feedback?)
Now the 2.4.19-ac4 patch was against the official release of 2.4.19, so I
download the release version of the Kernel from www.kernel.org and then Alan
Cox's patch from the same site.
Then, since it has been awhile since I last recompiled my Kernel, I read the
instructions at http://www.tldp.org/HOWTO/Kernel-HOWTO.html :-), this
document covers everything you need to know, read it carefully.
The kernel patch was applied using patch -p1 (something like this, zcat 
patchfile.gz | patch -p1 or patch -p1 < patch if it is uncompressed), Alan
uses kernel.vanilla as his base directory so remember the -p1 option.
Since I was recompiling I took the opportunity to choose the correct
processor type and make some other minor changes.

Now with the new kernel installed, as per the instructions at www.tldp.org,
hdparm was returning a result around 10x higher (I can't remember the exact
figure), and the system finally started to feel like a 2.5 gig P4 Linux
Beast :-)

Things left to do : find some nice LM monitoring software, tweak my video
card (I think its registry.c ?), load my windows games using wineX, work out
how to make grub look pretty (graphical) and install my LCD I ordered from
CrystalFontz

Thanks
Brett
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] setting permissions for web pages, chmod what ?

[Voytek Eymont]

** Reply to note from Terry Collins <[EMAIL PROTECTED]> Tue, 08 Oct 2002 18:18:16 +1000


hi Terry, long time no talk

> My 2c on this/AIUII. 
>
> You don't login as "apache" 
> On my system, user "apache" doens't have a shell that you can log into. 
> from /etc/passwd:  apache:x:48:48:Apache:/var/www:/bin/false 
> It is a security feature. 

which leads me to another Q:

if I have users that simply have web server and mail server hosted, and,
all they get is ftp access, is that the way to edit their 'entitlements',
as per apache sample above ?

> To start and stop your apache server /etc/rc.d/init.d/httpd 
> start|stop|status|??? 
>  
> Then in your browser http:localhost or ip or whatever.

yes, that part is OK, it's the permissions I need to understand, the pages
are not served based on insufficient permission

so, I should give 
files: rw r r
directories: xrw xr xr
?

I see every user I've made gets a group made as well, should I make a
group, say, web_users, and, put all users there ? then , manage it by that
group ?



Voytek Eymont
SBT Information Systems Pty Ltd
http://www.sbt.net.au/links/
phone +61-2 9310-1144 fax +61-2 9310-1118 
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] upgrading/installing wirh rpm, webmin

[Voytek Eymont]

RH7.3, webmin 1.

If I'm upgrading say Apache or MySQL through Webmin:

as far as I can tell, Webmin takes care of shutting stuff down, before
copying files, then, restarting, (I think..)

if I upgrade stuff from a command line with rpm:

does rpm takes care of shutting say Apache or MySQL;
OR, am I supposed to do that before I run rpm ?



Voytek Eymont
SBT Information Systems Pty Ltd
http://www.sbt.net.au/links/
phone +61-2 9310-1144 fax +61-2 9310-1118 
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Granville TAFE notes

[Paul Copeland]

Hi All,

At the last SLUG meeting a lecturer from Granville TAFE approached me 
telling me his Linux lecture notes were freely available on the net.  If 
said lecturer is on the list could he please give me the location again, or 
could someone tell me what his e-mail address is if he is not on the SLUG 
list.  Any information would be greatly appreciated.

Best to all
Paul

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Mount CD image as drive

[Doug Foskey]

Newbie Q: How do I mount an image file so I can access the files? (ie I want 
to mount '/~.../crap.img' to '/mnt/img1' )
This is so I can access my Mdk CD images to load some more s/w, without 
burning CDs etc.
Also if I wish to do this frequently can I add it to /etc/fstab?
Thanks, Doug
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Mount CD image as drive

[Tom Massey]

* Doug Foskey <[EMAIL PROTECTED]> [2002-10-09 12:04]:
> Newbie Q: How do I mount an image file so I can access the files? (ie I want 
> to mount '/~.../crap.img' to '/mnt/img1' )

mount -o loop /~.../crap.img /mnt/img1

> Also if I wish to do this frequently can I add it to /etc/fstab?

I'm not sure, but it might work if you put 'loop' in the options
field - this is the fourth field.
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] rsync hanging unless verbose

[David]

I've been playing with rsync as a back up tool
(yes, that's right.. I lost some important config files, didn't I..
doesn't that focus the mind!)

When i did this:

rsync -a -e ssh /etc me@elsewhere:/backup

.rsync hung during the process. I decided to use the -vvv option to
figure out what was going on, and lo! the damn thing worked perfectly.
Several tests suggest that -vvv slows the transfer down, thus fixing what
ever makes it hang. Is this perhaps a bug? Any other suggestions?

David.

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Mount CD image as drive

[Simon Wong]

On Wed, 2002-10-09 at 12:36, Doug Foskey wrote:
>   Newbie Q: How do I mount an image file so I can access the files? (ie I want 
> to mount '/~.../crap.img' to '/mnt/img1' )

This a really cool thing to be able to do.

You may need to be root to mount it (I do):

root@host# mount ~/crap.img /mnt/img1 -t iso9660 -o loop

The key is the loopback device.

>   Also if I wish to do this frequently can I add it to /etc/fstab?

Should be no problem.



-- 
**
* Simon Wong *
**

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Telnet hangs after message and before login

[Jessica Mayo]

I've got a problem at work that's hurting my head... I have found a single
mention of it through google, but no solutions, perhaps I'm not searching
for the right thing. SSH is not an option, sorry. :)

Box is Redhat6.1, recently had an FTP security patch installed.
Telneting from some IP addresses works, but not others.
hosts.allow is set up correctly and works fine with FTP.
This used to work fine, but I'm not sure exactly when it stopped working.


Symptom: No login prompt (telnet hangs)

/home/user> telnet 203.47.39.xx
Trying 203.47.39.xx...
Connected to 203.47.39.xx.
Escape character is '^]'.

Red Hat Linux release 6.1 (Cartman)
Kernel 2.2.12-20smp on an i686  
_


Cause:
'login' process spawned by telnetd dies and sits there as a zombie.


Solution:?
Help! :)


-- Jess.
(Everything with a Grin :)

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] killing a runaway console; mouseless mc [^] command linehistory

[Michael Lake]

Voytek Eymont wrote:
> 1.
> if I get a 'problem' on a console, such as:
> 
> I can't get out of an editor...
> and, after a while, I've tried every possible sequence I can think off,
> and, I'm still inside an editor
> 
> how do I kill whatever on say console 2 ?

1. Open a new console or a terminal session.
2. In that terminal type "ps ax" and look for a line that contains the
name of the editor program
   lets says its vim [well if it was nedit or any other gui one would
know how to get out of it :-) ]
3. note down the 'process number' of that program
4. type "kill -9 process_number"

One might also be able to type instead:
killall -9 vim
(if it was vim)

Once you do this the offending ediitor should promptly vanish from that
terminal.

You can always test this out beforehand, just open a vi session with a
file called crap ie "vi crap" in a terminal and practice killing it.

Mike
-- 
Mike Lake
Uni of Technol., Sydney


UTS CRICOS Provider Code:  00099F

DISCLAIMER

This email message and any accompanying attachments may contain
confidential information.  If you are not the intended recipient, do not
read, use, disseminate, distribute or copy this message or attachments.
If you have received this message in error, please notify the sender
immediately and delete this message. Any views expressed in this message
are those of the individual sender, except where the sender expressly,
and with authority, states them to be the views the University of
Technology Sydney. Before opening any attachments, please check them for
viruses and defects.



-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Fwd: Linux gaining acceptance in Canberra

[Rick Welykochy]

This bodes well for Linux boffins. It will eventually mean be realised as
more jobs for those familiar with Linux.


 Original Message 
Subject: [LINK] Linux gaining acceptance in Canberra
Date: Wed, 09 Oct 2002 11:54:03 +1000


Linux gaining acceptance in Canberra
BY STEVEN DEARE
9 October, 2002 9:18
SYDNEY, AUSTRALIA
Australian ComputerWorld


The Linux operating system is finding a welcome home in Canberra as an
increasing number of Federal departments and agencies look towards
cost-effective open source solutions.

Agencies are keen to trial Linux, eager to explore the costs and benefits,
said Steve Alford, business strategies branch general manager for the
National Office of the Information Economy (NOIE). He cited existing
procurement policy that states agencies are free to use open source
software where it provides a value-for-money solution to their needs.

While it was hard to gauge which agencies' IT systems would gain most from
Linux, savings alone meant government interest, said Alford. "As Linux is
implemented by more users both within government and industry, we will have
a better idea of where [in government] Linux can be implemented with most
benefit."

Peter Gigliotti is the assistant director of computing at the Bureau of
Meteorology. He has had no problems using Linux for about two months on a
development cluster for one of the government's largest Web servers.
Gigliotti is typical of government IT managers turning to Linux.
"Everybody's looking at the bottom line these days. I'd estimate we've made
a cost saving of about 30 per cent, that's hardware and software," he said.
"I suppose what else made us use Linux was confidence of stability, it's
supported by the commercial channels now. There's been a lot of money spent
on the development of it by companies like IBM. It can certainly provide a
fair amount of computer power, too."

The scalability of Linux is also attractive for the government because it
runs smoothly on mainframes, such as IBM's z Series mainframe used by the
Department of Veterans Affairs.

Following an infrastructure upgrade, the Department of Veterans Affairs
(DVA) moved to a thin client operation with one central data centre. It
moved file and print servers to the mainframe to run under Linux. The
network supports about 3000 PCs.

The reason behind the Linux move was "basically cost", said Alan Parker,
information manager at the DVA. "We will reduce floor space and less
operator support will be required, especially in managing growth." Parker
said the Department had not been actively seeking to replace Microsoft
software, but had SuSE Linux offered to them by IBM.

As open source supporters like IBM and Sun are also Commonwealth
Government-endorsed suppliers, Linux has become more available to
government agencies.

A lack of credible information on the total cost of ownership for Linux
systems has so far prevented agencies from trialing Linux, said Alford.
However, this will be addressed when NOIE briefs government officials on
Linux before the year's end.

"Also, where agencies have outsourced their IT operations, using Linux may
require contractual variations, and this may not be a simple exercise," he
said.

This has been a huge wall for open source acceptance that is finally
crumbling, said Cybersource managing director Con Zymaris. The Victorian
professional services company specialises in open source solutions and
training. "What's traditionally worked against Linux in the government is
that if something can't be used everywhere, they won't use it.

"The problem is that 80 per cent of people in government departments could
use Linux desktops, but Microsoft contracts in effect lock the government
down further than the contract period because it's difficult to get other
software working on it."

He agrees mainframes are a big growth area for Linux with its scalability.
"Now that mainframes can run Linux, you can use its full power."

Zymaris referred to Telia Net in Sweden, one of Europe's first major
commercial Linux installation which last year doubled the capacity of its
initial mainframe Linux installation. This ran up to 1500 virtual Linux
Internet servers simultaneously. A major commercial player, IBM, was
involved here, too.

Now Linux is being used by government and with the NOIE seminar to come, it
will gain a permanent place in government IT, said Zymaris. The NOIE
seminars will be held for Federal government officials by the end of the
year. While Zymaris acknowledged the problems of integrating Linux in
government IT systems,  he said the NOIE seminars were evidence government
agencies could no longer ignore the benefits. "Them being so conservative
means that Linux has finally made it."

-- 
A message from the system administrator: 'I've upped my priority. Now up
yours.'
-- unknown
-- 
SLUG - Sydney Linux

Re: [SLUG] upgrading/installing wirh rpm, webmin

[Michael Lake]

Voytek Eymont wrote:
> if I upgrade stuff from a command line with rpm:
> does rpm takes care of shutting say Apache or MySQL;
> OR, am I supposed to do that before I run rpm ?

rpm won't shutdown the services. 
Best to stop the services first, then upgrade then restart em.
Apache have a utility called 'apachectl' that takes stop and start as
parameters and is quite useful.
It's prob on your system already.

Mike
-- 
Mike Lake
Uni of Technol., Sydney


UTS CRICOS Provider Code:  00099F

DISCLAIMER

This email message and any accompanying attachments may contain
confidential information.  If you are not the intended recipient, do not
read, use, disseminate, distribute or copy this message or attachments.
If you have received this message in error, please notify the sender
immediately and delete this message. Any views expressed in this message
are those of the individual sender, except where the sender expressly,
and with authority, states them to be the views the University of
Technology Sydney. Before opening any attachments, please check them for
viruses and defects.



-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] [OT] looking for place to rent

[evilbunny]

Hello slug,

  I'm currently moving from my existing location and am looking
  for a 1 bdrm place in the St George area (pref close to Arncliffe)
  for about ~$120-130/week 

  Gave notice on existing place Saturday just past, and have 2 and a
  bit weeks to find somewhere...

  Know of anything let me know... thanks...

-- 
Best regards,
 evilbunny  mailto:[EMAIL PROTECTED]

http://www.cacert.org - Free Security Certificates
http://www.sydneywireless.com - Telecommunications Freedom



smime.p7s
Description: S/MIME Cryptographic Signature

Re: [SLUG] setting permissions for web pages, chmod what ?

[Terry Collins]

Voytek Eymont wrote:

...snip.

> if I have users that simply have web server and mail server hosted, and,
> all they get is ftp access, is that the way to edit their 'entitlements',
> as per apache sample above ?

...snip.

> yes, that part is OK, it's the permissions I need to understand, the pages
> are not served based on insufficient permission
> 
> so, I should give
> files: rw r r
> directories: xrw xr xr

AIUII

There are two aspects here, user & group ownership and user/group/world
permissions.

The way I approach this is to set the user ownership (of the directory
and all files and sub-directories) to the login user that can edit the
files. So each company would have a user that logs on to upload/change
their WWW pages and has "ftp" access to their companies file areas.

I set the group ownership to your server process (apache, nobody, ???).

I set the user permissions to 7/rwx
I set the group/world permissions to 55/r_xr_x for all directories and
most files.
I set images to 750 (and anything else I don't want spidered)

I make sure that all subdirectories that I don't want spidered at 750
(especially don't forget the ones listed in robots.txt)


> I see every user I've made gets a group made as well, should I make a
> group, say, web_users, and, put all users there ? then , manage it by that
> group ?

It really depends on what your client wants and what you are providing.
If you are providing a common access area, then having all logins from
one company in the one group makes sense, so that group has write access
to that area.

However, a login can have multiple groups (check /etc/group & "man
group"), so it might be easier (best security), that user "fred" is in
group fred, and group "freds-companyX" and group company-x-webmeisters"
(if need be). 

-- 
   Terry Collins {:-)}}} Ph(02) 4627 2186 Fax(02) 4628 7861  
   email: [EMAIL PROTECTED]  www: http://www.woa.com.au  
   Wombat Outdoor Adventures 

 "People without trees are like fish without clean water"
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Help wanted - graphic artists :-)

[Terry Collins]

David Guest wrote:
> 
> Hi Sluggers
> 
> I am a general practitioner involved with a little project to write an
> open source patient management program for doctors in community practice
> and hospitals. Gnumed is GPLed and written in python/wxpython and postgres.

Do you know about previous Slug assisted efforts?
Archives, Graham Kelly and someone from North Sydney Hospital whose name
I can not remember.

-- 
   Terry Collins {:-)}}} Ph(02) 4627 2186 Fax(02) 4628 7861  
   email: [EMAIL PROTECTED]  www: http://www.woa.com.au  
   Wombat Outdoor Adventures 

 "People without trees are like fish without clean water"
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] setting permissions for web pages, chmod what ?

[Tony Green]

On Wed, 2002-10-09 at 13:29, Terry Collins wrote:

> I set the user permissions to 7/rwx
> I set the group/world permissions to 55/r_xr_x for all directories and
> most files.
> I set images to 750 (and anything else I don't want spidered)
> 
Just in case you want to execute the images??  0640 should be find for
'non-cgi' type stuff
-- 
Tony Green <[EMAIL PROTECTED]>
Tel   :   +61-(0)2-9500-9996



signature.asc
Description: This is a digitally signed message part

[SLUG] am i leaking memory ?

[Voytek Eymont]

I've installed RH 7.3, Apache, MySQL on P2-266 96MB, 6GB IDE
after it starts, top shows

2:45pm  up  3:23,  2 users,  load average: 0.00, 0.00, 0.00
59 processes: 57 sleeping, 2 running, 0 zombie, 0 stopped
CPU states:  1.1% user,  0.7% system,  0.0% nice, 98.0% idle
Mem:94304K av,   70844K used,   23460K free,   0K shrd,4700K buff
Swap:  224868K av,   0K used,  224868K free   27832K cached

at this point, all the stuff runs, but, not in production

next day, top shows all ram used, and, we're into swap a little
it seems, I'm leaking ram...? how to identify what's leaking ?



Voytek Eymont
SBT Information Systems Pty Ltd
http://www.sbt.net.au/links/
phone +61-2 9310-1144 fax +61-2 9310-1118 
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] am i leaking memory ?

[Jeff Waugh]

> I've installed RH 7.3, Apache, MySQL on P2-266 96MB, 6GB IDE
> after it starts, top shows
> 
> 2:45pm  up  3:23,  2 users,  load average: 0.00, 0.00, 0.00
> 59 processes: 57 sleeping, 2 running, 0 zombie, 0 stopped
> CPU states:  1.1% user,  0.7% system,  0.0% nice, 98.0% idle
> Mem:94304K av,   70844K used,   23460K free,   0K shrd,4700K buff
> Swap:  224868K av,   0K used,  224868K free   27832K cached
> 
> at this point, all the stuff runs, but, not in production

That's okay, Linux reports memory in an odd fashion if you're not used to
it. There are lots of emails in the SLUG archives about this (I've sent a
couple, Andrew Bennetts has, etc).

Basically, a lot of the RAM being used above is for cached files. Linux will
free this up if it needs it for software.

> next day, top shows all ram used, and, we're into swap a little
> it seems, I'm leaking ram...? how to identify what's leaking ?

It's probably just swapped out stuff that isn't running. I doubt to you have
a leak anywhere, these kinds of results from free(1) and top(1) are normal.

- Jeff

-- 
  My VHS is better than your VHS.   
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Mount CD image as drive

[Matthew Hannigan]

> mount -o loop /~.../crap.img /mnt/img1

Does the loopback driver allow writing?

If so, and you don't want to write, use the
readonly (ro) option too:

mount -o loop,ro /~.../crap.img /mnt/img1

Probably a lot faster, too, since it won't
update access times.

Regards,
Matt
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Mount CD image as drive

[Colin Humphreys]

On Wed, Oct 09, 2002 at 03:00:52PM +1000, Matthew Hannigan wrote:
> 
> > mount -o loop /~.../crap.img /mnt/img1
> 
> Does the loopback driver allow writing?

It does, but only if the fileystem type supports writing. (e.g iso9660
will be read only, but a ext2 image will be writable)


-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] via vt8235 ide support in 2.4.18

[Mark Munro]

I’m trying to setup a linux kernel with via vt8235
support

I can find reference to a patch and a text document claiming
to BE a patch but I don’t know how to install it

The patch is from:

 http://lwn.net/Articles/8035/

But I am having no success using the “patch”
command with it

Anyone got any ideas??

Thanks in advance

Mark Munro

 

Re: [SLUG] via vt8235 ide support in 2.4.18

[Matthew Dalton]

> I’m trying to setup a linux kernel with via vt8235 support
> 
> I can find reference to a patch and a text document claiming to BE a patch but I 
>don’t know how to install it
> 
> The patch is from:
> 
>  http://lwn.net/Articles/8035/
> 
> But I am having no success using the “patch” command with it
> 
> Anyone got any ideas??

Copy the text from the line:

  diff -Nru a/drivers/ide/via82cxxx.c b/drivers/ide/via82cxxx.c

to just before the row of equals signs. Paste in a file, which I'll call
'vt8235.diff'.

Change directory to your linux source tree:

cd /usr/src/linux

Run patch, using -p1 to remove the a/ from the paths in the vt8235.diff
file.

cat vt8235.diff | patch -p1

Compile away!

Matthew
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

RE: [SLUG] via vt8235 ide support in 2.4.18

[Mark Munro]

Thanks Matthew
But unfortunately ended up with this:

patching file drivers/ide/via82cxxx.c
Hunk #1 FAILED at 1.
Hunk #2 FAILED at 68.
Hunk #3 FAILED at 126.
Hunk #4 FAILED at 377.
4 out of 4 hunks FAILED -- saving rejects to file
drivers/ide/via82cxxx.c.rej
patching file include/linux/pci_ids.h
Hunk #1 FAILED at 981.
1 out of 1 hunk FAILED -- saving rejects to file
include/linux/pci_ids.h.rej


fyi
My current kernel is 2.4.19
Any more information needed?

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] am i leaking memory ?

[Voytek Eymont]

** Reply to note from Jeff Waugh <[EMAIL PROTECTED]> Wed, 9 Oct 2002 14:57:39 +1000


> That's okay, Linux reports memory in an odd fashion if you're not used to 
. 
> It's probably just swapped out stuff that isn't running. I doubt to you have 
> a leak anywhere, these kinds of results from free(1) and top(1) are normal. 
  
thanks, Jeff

my current prod. system says this:

  4:40pm  up 15 days,  2:11,  1 user,  load average: 0.00, 0.00, 0.00
48 processes: 47 sleeping, 1 running, 0 zombie, 0 stopped
CPU states:  0.9% user,  0.5% system,  0.0% nice, 98.4% idle
Mem:62284K av,   60300K used,1984K free,   0K shrd,9016K buff
Swap:  200772K av, 424K used,  200348K free
8604K cached

 total   used   free sharedbuffers cached
Mem: 62284  60168   2116  0   9016   8604
-/+ buffers/cache:  42548  19736
Swap:   200772424 200348

so, how can I tell when I really start needing more RAM ?



Voytek Eymont
SBT Information Systems Pty Ltd
http://www.sbt.net.au/links/
phone +61-2 9310-1144 fax +61-2 9310-1118 
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Redhat 8 - user menus in panels

[Norman Gaywood]

I'm probably looking right at it but I can't see it!

Has anyone figured out how to add your own menus to a panel in RH8. I
would have thought:

Right click panel -> Add to Panel -> Menu

But menus do not seem to be things you can add to the panel. You can add
a "GNOME menu", but you can't add/delete things from it.

You can add a "drawer" which allows you to build up some panels of
icons. But it's not a menu.
-- 
Norman Gaywood -- School of Mathematical and Computer Sciences
University of New England, Armidale, NSW 2351, Australia
[EMAIL PROTECTED] http://turing.une.edu.au/~norm
Phone: +61 2 6773 2412 Fax: +61 2 6773 3312
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] via vt8235 ide support in 2.4.18

[Matthew Dalton]

Mark Munro wrote:
> 
> Thanks Matthew
> But unfortunately ended up with this:
> 
> patching file drivers/ide/via82cxxx.c
> Hunk #1 FAILED at 1.
> Hunk #2 FAILED at 68.
> Hunk #3 FAILED at 126.
> Hunk #4 FAILED at 377.
> 4 out of 4 hunks FAILED -- saving rejects to file
> drivers/ide/via82cxxx.c.rej
> patching file include/linux/pci_ids.h
> Hunk #1 FAILED at 981.
> 1 out of 1 hunk FAILED -- saving rejects to file
> include/linux/pci_ids.h.rej
> 
> fyi
> My current kernel is 2.4.19
> Any more information needed?

These errors mean that your versions of the files via82cxxx.c and
pci_ids.h are different enough to the ones the diff was created against,
that patch can't figure out where to make the changes.

The diff itself is not very complicated... you could probably just make
the changes manually. Make backups of the files before you start though.

Understanding diff files is easy, especially with diff files created in
'unified' mode, as this one is.

Lines starting with '-' are to be deleted. Lines starting with '+' are
to be added. Whenever a line is changed, it will be represented in the
diff by a deletion followed by an addition. This section is an example
of that:

 -   via_print("Driver Version: 3.34");
 +   via_print("Driver Version: 3.35");


Moved lines are similar, but will not appear adjacent to each other:

 -   { "vt8235", PCI_DEVICE_ID_VIA_8235, 0x00, 0x2f...
  #endif
 +   { "vt8235", PCI_DEVICE_ID_VIA_8235, 0x00, 0x2f...

(I've trimmed these lines in the interest of keeping them < 72 chars so
they don't wrap)


Lines with no '-' or '+' are existing lines that shouldn't be modified.
They are in the diff file to give context. It would be these lines that
cause your patch command to fail -- they must appear in the file exactly
as they are in the diff, otherwise patch will not recognise them. You
might be able to spot why patch failed when you make the changes
yourself (this information is also what patch puts into the .rej file)

There's only two other code changes to worry about, so I'll leave them
up to you. Most of the diff is comment changes.

The @@-bracketed parts indicate line numbers. Don't worry too much about
them, although they are useful in finding the correct line to change if
you are manually applying a diff. The diff section after each "@@ ...
@@" line is the 'hunk' your error message refers to.

Matthew
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] via vt8235 ide support in 2.4.18

[Matthew Dalton]

Matthew Dalton wrote:
> 
>  -   via_print("Driver Version: 3.34");
>  +   via_print("Driver Version: 3.35");

[cut] 
 
> Lines with no '-' or '+' are existing lines that shouldn't be modified.
> They are in the diff file to give context. It would be these lines that
> cause your patch command to fail

Er... no.

Lines starting with '-' can also cause patch to fail. For example, if
your version number is different to 3.34, the above section of the diff
won't work.


Matthew, re-reading his own post and disagreeing with it.
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] How to test RAID speed ?

[Jeff Waugh]

> Thanks Jeff, that works a treat. and yes I did stuff up my email
> configuration by changing usersnames on my local box. I've told mutt to
> munge the replyto header until I configure sendmail to put a correct from
> field in :-)

Oh, that's mutt's job though:

  my_hdr  From: Jeff Waugh <[EMAIL PROTECTED]>

Other useful mutt bits:

  http://perkypants.org/dotfiles/

:-)

- Jeff

-- 
   "PHP, when it first came out, didn't really have any merits, and many
claim it still doesn't, but it filled a void where a simple tool to 
perform a simple task was needed." - Rasmus Lerdorf 
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug