[SLUG] Re: "How To Adopt Gnus" (An Emacs Mailer)
\begin{Jeff Waugh}
> When Gus gets around to his Emacs talk... :-)
> http://people.mandrakesoft.com/~gc/html/howtos/howto-adopt-gnus.html
bah gnus. yay wanderlust.
all i need to do now is learn to read japanese..
--
- Gus
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: ask gcc (compling,linking)
\begin{henry}
> gcc -c image.c -o image.o
> ar rcs libimage.a image.o
>
> gcc -c gif.c -o gif.o
> ar rcs libgif.a gif.o
>
> gcc -c generic.c -o generic.o -L. -limage -lgif <- it seems that I do wrong
> on this line
> ar rcs libgeneric.a generic.o
libraries are pulled in during linking, not during compiling.
"gcc -c" only compiles.
you want:
# compiling: .c -> .o
gcc -c generic.c -o generic.o
# linking: lots of .o and libraries (really just other .o) -> program
gcc -o myprogram myprogram.o other_objects.o -limage -lgif
--
- Gus
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Debian or Redhat
\begin{Jeff Waugh}
> > Can people comment on the differences between the two. Both the lay out
> > of the file system and the its package manager.
[...]
> Hopefully a fairly balanced view.
yep. what he said.
just some free advice:
the usual behaviour for someone new to redhat seems to be to choose the
"install everything" option.
don't do this for debian, you really don't need ~11 web servers ;)
just install the minimum - what you need *today*. since apt-get(*) can
trivially install new software when you realise you needed them.
basically, do a "lazy" install and amortise the cost of answering all
those questions. it'll then do what you want, when you finally want it
and you'll end up liking it even more ;)
(*) or better yet aptitude or deity-{curses,gtk}. apt-get is a
power-users tool and debianites should get out of the habit of
advertising it to people who don't even know what they want to
install yet. (just say "aptitude install spamassassin" instead ;)
(my opinion: redhat is a company, debian is a community. all the
usual pros/cons of each are manifested in their products)
--
- Gus
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Highly Technical Talk Offers / Requests?
\begin{Jeff Waugh}
> Working on ideas for upcoming SLUG meetings, I'd like to guage how many
> people have highly technical talk ideas [1] that they wouldn't normally see
> as appropriate for a SLUG audience. Also interested in *requests* for highly
> technical talks, too.
hmm.. after writing this list, i *would* consider them appropriate for
a SLUG audience, so i guess they don't count as "highly
technical". naturally they could all be ramped up until they became
truly inappropriate.. here they are anyway:
apache (1.3) internals and mod_perl
Embperl (think of it as php, except with perl and more magic
HTML-understanding stuff)
(X)Emacs (and elisp?)
Aegis (configuration management system)
i'm guessing i'll get the LaTeX/METAPOST/pic talks out of my system at
the docfest.. ;)
--
- Gus
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Re: RFC: SLUG Mailing List FAQ Update
\begin{Jamie Wilkinson}
> This one time, at band camp, John Clarke wrote:
> >I agree, but only if we also encourage uploading keys to a public
> >keyserver. There's no point signing a message if the recipients don't
> >have the key and can't get it.
>
> Ever since the keysigning in July last year, I've kept a reasonably up to
> date SLUG keyring at this url:
>
> http://spacepants.org/slug/slug-keyring.gpg
madcow:~> gpg --keyserver wwwkeys.au.pgp.net --verbose --send-keys --no-options
--no-default-keyring --keyring ~/slug-keyring.gpg
[..]
Key block added to key server database.
New public keys added: 3
New userid's added: 1
New signatures added: 15
in future, people should upload the keys they sign / their own key
semi-regularly.
--
- Gus
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Debian Security
\begin{Lester Cheung}
> Just want to know how secure/insecure is a minimal debian install. coz
> the more I read, the more paranoid I am. I have read the debian security
> howto serveral times. Are the suggestions in there enough for a normal
> home machine/regular office gateway?
just be minimal and don't do anything stupid ;)
go through the *entire* list of installed packages (especially
anything with network connotations) and ask yourself if you actually
need that at this particular moment. if not, remove it.
dselect likes to install everything marked "standard", which is good
for a unix desktop / login server but not good for a firewall.
then make sure you keep up to date with debian security updates (by
adding the appropriate apt/sources.list lines).
i have been running my home network off a debian (stable) installation
for many years. i have /etc/hosts.{allow,deny} setup to ban any
outsiders. i don't add silly wildcard entries to /etc/exports and the
like. i use the default ipmasq package firewalling rules (with one
exception for 0.0.0.0 dhcp packets on the local network).
i don't have a separate firewall - my "main machine" also runs
pppd. so far its been a successful experiment in application-level
security. (i've doomed it now though, haven't i ;)
--
- Gus
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: X terminal won't connect
On Fri, Apr 26, 2002 at 12:23:29PM +1000, Jeff Waugh wrote: > XDMCP is comparable to NFS in terms of security. ;-) if you use settle for using xhost security it is, yes. however, the Xauthority mechanism is a whole lot more general than that. (from my XFree86 3.3.6 Xsecurity(3x) manpage) NAME Xsecurity - X display access control SYNOPSIS X provides mechanism for implementing many access control systems. The sample implementation includes five mecha- nisms: Host Access Simple host-based access control. MIT-MAGIC-COOKIE-1Shared plain-text "cookies". XDM-AUTHORIZATION-1 Secure DES based private-keys. SUN-DES-1 Based on Sun's secure rpc system. MIT-KERBEROS-5Kerberos Version 5 user-to-user. i presume its not too hard to add another authentication method, if you have a particular desire to use the new authentication-abstraction- library-of-the-day. the rest of Xsecurity(3x) gives lengthy descriptions on each mechanism and how to use them. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Turning off auto checking PGP signatures in mutt.
At Thu, 25 Apr 2002 16:10:47 +1000, Mary Gardiner wrote: > Since I do occasionally receive signed messages to mailing lists I'm on, > and am not really interested in authenticating the identity of the > sender, or in waiting some seconds for the message to display while the > key is downloaded from the keyservers, is there an option in mutt to > turn off the auto checking, but still allow the key to be checked with a > keystroke or two? an alternative solution would be to add "no-auto-key-retrieve" to ~/.gnupg/options -- - Gus.so.long.little.mutt -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Louis Selvon's problems
louis, your mailer (USANET web-mailer (CM.1201.3.01A)) has a very confusing quoting style. At Mon, 22 Apr 2002 11:40:10 +1000, Louis Selvon wrote: > Angus Lees <[EMAIL PROTECTED]> wrote: > At Thu, 18 Apr 2002 20:17:46 +1000, Louis Selvon wrote: > > When I create a domain called "domain.com" with username "username" > > from the control panel from root that directory is accesible at > > > > /home/virtual/domain.com/ (This is the base directory for the domain) > > > > At the user account level for that domain, it's home is > > > > /home/username (It cannot see /home/virtual/domain.com, only root > > can). > > > > Now root Perl repository is located at > > /usr/bin/perl > > > > The domain.com repository is located at > > /home/virtual/domain.com/usr/bin/perl > > >you have several "chrooted environments". read chroot(8) and chroot(2) > for the details, but basically its possible to change where you (and > your child processes) think / is. once you chroot into a subdirectory, > its very difficult (but not impossible) to get out again. > > *** I don't understand this chroot stuff, but you are most likely right. > > What I did discover is that the control panel that I was provided, has a > template directory. In that template directory there is an outdated version of > perl5 installed which was the one my newly created virtual sites were picking > up. > > I do not know how chroot works. So what I did was as per what I said below. I > removed the perl5 for the template (rm -Rf), and then copied the one from the > main /usr/bin/perl5 over. From what you said if sounds like chroot is a better > way to do this, instead of copy. chroot just moves where the kernel says "/" is for a particular process. you will need to copy any files into the chroot location if you want those files to appear there.. > >using hard links will also remove some of the security benefits of > chroot - it will allow the chrooted user to (potentially) modify the > (real) system files. > > *** Don't want that. With what I just did above, virtual sites user still > cannot modify the main perl stuff from root. Right ?? correct. they can only modify their copy of the files (those in /home/virtual/domain.com/). > > cp -r /usr/bin/perl /home/virtual/domain.com/usr/lib/perl5 > > > > Will this install Perl properly for the user account, or > > it just won't work ? > > >yes, should do. if you are going to have multiple copies anyway, you > might want to just do a fresh install into each chroot to be sure. > > *** What I did is said above. I copied over to the template directory for that > control panel, and now all newly virtual sites perl5 are in line with root. my concern was with modules that do something more clever in Makefile.PL - other than simply compiling or copying files. there are a few CPAN modules around that will try and autodetect (or prompt for) different configurations. these might fail if they are copied into what looks like a different filesystem (without whatever they had just checked for). i'd say about 95% of CPAN modules will work fine with a simply recursive copy, however. > >you could run a cron job which copied the perl trees around (using > rsync, cp -a, tar or cpio - choice is good). > > *** What is "cp -a" (a man on cp does not show any -a options) ? curious. assuming you have GNU cp, thats a bug in your manpage. "cp --help" should give enough information anyway. -a, --archive same as -dpR (-d, --no-dereference never follow symbolic links -p, --preserve preserve file attributes if possible -R, --recursivecopy directories recursively) its basically the "make this as close as practical to that" option. > I hate to ask this, how do I create such a tasks from what you said above in > the commands ? choose one of the methods above. i'll choose rsync, since that should avoid lots of unnecessary copying. "rsync -a /source/directory /destination/directory" will duplicate /source/directory as /destination/directory. you may also want to add the "--delete" option if you want to delete files in /destination/directory that don't appear in /source/directory. so: assuming you have an /etc/cron.daily/ directory: (if you don't, tell us your distro) create the file /etc/cron.daily/copychroots, containing: #! /bin/sh # # script to update perl in already installed chroots # USERS="domain.com otherdomain.net yadomain.org" # automatically update the template directory using what is installed # in the "real" system. # (you may not want to do this automatically) rsync -a /usr/lib/perl5 /tem
[SLUG] Re: RFC: SLUG Mailing List FAQ Update
At Thu, 25 Apr 2002 09:45:04 +1000, Jeff Waugh wrote: > > > don't punish those who do the right thing. > > (aka "don't optimize for the failure case") > > It's only a recommendation. [...] One can always choose to ignore > the recommendation. but my point is that SLUG should be *encouraging* MIME and PGP. amongst other things, that means NOT recommending against their use.. > It's also a pity that we encourage the thoroughly unmodern approach > of "No HTML Mail", but there's a fair amount of agreement that it's > not wanted. where the HTML version adds value, and its a multipart/alternative with a plain text equivalent, i agree with you. the problem is, only a handful of post i've seen would benefit from HTML (posts with itemised lists (usually event announcements) or those with heavy use of URL links). -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: run levels
At Wed, 24 Apr 2002 11:11:02 +1000, Mick Howe wrote: > I need to change the default run level in redhat 7.1 for a while. if you don't want to make it permanent, then: from lilo (or other bootloader), add the runlevel as a kernel command line argument: eg: lilo: linux 5 will boot into run level 5. use "s" for single user. on an already running machine: use telinit(8) to switch runlevels. eg: root_prompt# telinit 5 (again, "s" for single user) -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: ask a common question
At Thu, 18 Apr 2002 11:22:15 +1000, Jamie Wilkinson wrote: > This one time, at band camp, henry wrote: > >Sometimes I make Makefile ,I get lots of error messages so that I cant look > >them in time. > >I try to "make > tmp.log" , but it fail (I cant dump those error message into > >tmp.log). > >Could someone show a good solution ? > > I use shift-pageup to scroll back up in the xterm or console. and remember Ctrl+s and Ctrl+q for XON/XOFF flow control. in other words: Ctrl+s will block the process from writing more to the screen, Ctrl+q will allow it again. very useful if you want to pause rapidly scrolling output. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Debian Apt-get without internet [beginner]
At Tue, 23 Apr 2002 17:51:43 -0700 (PDT), Mark A. Bell wrote: > The problem is, my Linux machine is not connected to the net so I can't > just use 'apt-get upgrade' to install the 'testing' version. My (now > much neglected) windows laptop has a net connection. > > So my question is, in general terms what will I have to do to upgrade > my current version of Dia to the newer 'testing' version? see the "apt-zip" package for a method of using apt over removable media (floppies, zip drives, etc). alternatively, you could use "apt-get --print-uris install dia" or something to just get a list of urls you should download (somehow) and the filenames where you should put them. then just run "apt-get install dia" and it will magically find what it was about to think of looking for. personally, i'd just download the source "apt-get source dia" (or get the files directly (from apt-get --print-uris source dia): http://ftp.au.debian.org/debian/pool/main/d/dia/dia_0.88.1-3.dsc http://ftp.au.debian.org/debian/pool/main/d/dia/dia_0.88.1.orig.tar.gz http://ftp.au.debian.org/debian/pool/main/d/dia/dia_0.88.1-3.diff.gz transfer them over to the linux box and then do: cd /some/directory/wot/i/copied/those/files/to dpkg-source -x dia_0.88.1-3.dsc cd dia-0.88.1 (or whatever the directory ends up being called) dpkg-buildpackage -rfakeroot -us -uc (compile, compile) sudo dpkg -i ../dia_0.88.1-3_*.deb it will probably get a little hairier than that, since the dia package might build-depend on some other woody packages (newer versions of debhelper is a common one). -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: RFC: SLUG Mailing List FAQ Update
At Tue, 23 Apr 2002 17:01:16 +1000, Jeff Waugh wrote: > > > - you mention removing PGP/GPG signatures. Why? I see many complaints > > about HTML formatted email but I've never seen any about signed email. > > They barf up quite a few mail programs, make things harder to read, etc. > They're not complained about because it would be uncool to diss signatures, > whereas posting in remarkably unmodern plain text is cool. [ Okay, I'm being > sarcastic. It's late in the afternoon. ] i don't get this. surely signing email is a Good Thing for distributed communication, and thus SLUG should be setting an example and *encouraging* it. limiting email content to US-ASCII is simply not good enough for any worldly environment. for SLUG to promote such american-centric bigotry is inexcusable - hence SLUG /must/ promote MIME and encourage all mailing list members to seek MIME compliant MUAs. I see SLUG's place as providing a "best practice" example of how things Should Be Done. IMO, MIME and PGP are certainly a part of that. don't punish those who do the right thing. (aka "don't optimize for the failure case") -- - Gus msg22887/pgp0.pgp Description: PGP signature
[SLUG] Re: Sailing / Picnic / Ultimate Day!
At 21 Apr 2002 10:17:45 +1000, Craige McWhirter wrote: > When: Saturday, May 4th (from 11:00) so whats with the short notice? especially when involving family, my weekends are booked *much* further ahead than that. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Louis Selvon's problems (was: Cannot See Perl CPAN Modules for Linux)
At Thu, 18 Apr 2002 20:17:46 +1000, Louis Selvon wrote:
> When I create a domain called "domain.com" with username "username"
> from the control panel from root that directory is accesible at
>
> /home/virtual/domain.com/ (This is the base directory for the domain)
>
> At the user account level for that domain, it's home is
>
> /home/username (It cannot see /home/virtual/domain.com, only root
> can).
>
> Now root Perl repository is located at
>
> /usr/bin/perl
>
> The domain.com repository is located at
>
> /home/virtual/domain.com/usr/bin/perl
ah.
you have several "chrooted environments". read chroot(8) and chroot(2)
for the details, but basically its possible to change where you (and
your child processes) think / is. once you chroot into a subdirectory,
its very difficult (but not impossible) to get out again.
its an admirable security feature on a web hosting server.
to see what the user sees, do "chroot /home/virtual/domain.com" as
root. when you exit that shell, you'll return to "normal".
(chroots have other uses too. eg: the debian package "pbuilder" helps
you setup chroot environments so you can compile packages for "stable"
on an otherwise "unstable" system (or vice-versa))
> The contents of "/home/virtual/domain.com/usr/bin/perl"
> are not symbolic links to the main /usr/bin/perl .
> This will not work anyway as user accounts cannot
> and should not see anything below it. Right ??
correct. symlinks are also affected, since they go off the user-space
view of the filesystem.
hard links will work, however, but they only work between files on the
same device - which is unlikely to be the case in many chroot-setup
cases.
using hard links will also remove some of the security benefits of
chroot - it will allow the chrooted user to (potentially) modify the
(real) system files.
> Now what I would like to know is this, if I manually
> kill all the contents of Perl for the domain.com, like
> this (from root)
>
> rm -Rf /home/virtual/domain.com/usr/lib/perl5
>
> and then do this (still as root)
>
> cp -r /usr/bin/perl /home/virtual/domain.com/usr/lib/perl5
>
> Will this install Perl properly for the user account, or
> it just won't work ?
yes, should do. if you are going to have multiple copies anyway, you
might want to just do a fresh install into each chroot to be sure.
you'll need to worry about any dependencies (perl and non-perl) too,
of course.
> The only issue with this is as I install new CPAN modules
> as root they will not be seen at the user accounts.
>
> Then I probably need an additional step when CPAN install is
> complete. Have a script that somewhow will copy all these new
> modules to the Perl repositories of all the user accounts
> on this server.
>
> Any ideas on how I can do this ?
>
> Or is there a way to have user accounts perl repositories
> be somehow automatically linked to the main /usr/bin/perl ?
you could run a cron job which copied the perl trees around (using
rsync, cp -a, tar or cpio - choice is good).
personally, i'd just install new copies of the perl modules into each
chroot. scripting something using CPAN.pm would probably be reasonably
straight-forward and very useful.
("perl -MCPAN -e shell" is a very easy way to install CPAN modules)
--
- Gus
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: i need a sofware to penetrate in to systems
At 19 Apr 2002 13:39:54 +1000, Peter Hardy wrote: > Now. Just so I know, in case I do it again, is there a relatively easy > way to recover a partition table? gpart. its way cool. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Creating PDFs
At Thu, 18 Apr 2002 15:38:53 +1000, Triet Hoai Lai wrote:
> Terry Collins <[EMAIL PROTECTED]> writes:
> > [terryc@owl 2002]$ pdftex 20020417-perm.tex
> > This is pdfTeX, Version 3.14159-13d (Web2C 7.3.1)
> > (20020417-perm.tex[/usr/share/texmf/pdftex/config/pdftex.cfg]
> > Babel and hyphenation patterns for american, french, german,
> > ngerman, i
> > talian, portuges, russian, spanish, nohyphenation, loaded.
> > ! Undefined control sequence.
> > l.4 \documentclass
> > [a4paper,12pt]{report}
> > ?
> >
> > And it repeats for every latex command
>
> pdftex is "equivalent" to tex and can only run with plain TeX. If your
> LaTeX document doesn't include any EPS/PS, pdflatex should work. Here is
> template so that you can run both latex and pdflatex:
well spotted. terry should be using the "pdflatex" command, not simply
"pdftex". no wonder it barfs on LaTeX macros.
> %% Preamble
> \ifx\pdfoutput\undefined% We're not running pdftex
> \documentclass[a4paper,12pt]{report}
> \else
> \documentclass[pdftex,a4paper,12pt]{report}
> %% More options described in pdfTeX manual
> \pdfcompresslevel=9
> \pdfinfo
> { /Title (Title of Document)
> /Author (Your Name)
> /Subject (???)
> /Keywords (???)
> }
> \fi
> ...
> \begin{document}
i usually end up using the hyperref package with my pdfs, in which
case it does a lot of the above for you (and a few other things like
setting the pdfpagesize, etc). its a useful package to look into if
you plan to do anything exotic with pdftex.
> ...
> %% Example of including figures: pdflatex supports PDF, PNG and JPEG graphics
> \begin{figure}
> \centering
> \ifx\pdfoutput\undefined
> \includegraphics{fig.eps}
> \else
> \includegraphics{fig.pdf}
> \fi
>
> \end{figure}
> ...
> \end{documnent}
and with images, i find it much easier to leave the extension off
completely (ie: \includegraphs{fig}) and let graphics/graphicx find
the most appropriate file for the driver currently in affect. for this
to work you have to tell graphics/graphicx that you want "pdftex" (or
"dvips", etc), either by declaring it as a "global" option (as the
above preamble does), or with: \usepackage[pdftex]{graphics}.
then you just have your Makefile (or whatever) have both .eps and .pdf
versions (for example) of your images available and \includegraphics
will choose the right one.
if you want to know what formats, file extensions, etc are supported
have a look through pdftex.def for where \Gin@extensions is defined.
eg: from my version:
\ifnum\driver@release>3
\ifnum\driver@release>5
\def\Gin@extensions{.png,.pdf,.jpg,.mps,.tif}
\@namedef{Gin@rule@.tif}#1{{tif}{.tif}{#1}}
\@namedef{Gin@rule@.jpg}#1{{jpg}{.jpg}{#1}}
\else
\def\Gin@extensions{.png,.pdf,.jpg,.mps}
\@namedef{Gin@rule@.jpg}#1{{jpg}{.jpg}{#1}}
\fi
\else
\def\Gin@extensions{.png,.pdf,.mps}
\fi
which looks like it supports (in preference) png, pdf, jpg, mps and
tif - depending on the backend driver version.
(by reading comments earlier in the file, you'll see that .mps is
"metapost output", which can be input directly and converted from the
simple postscript metapost produces to pdf automatically. a damn fine
way of programmatically producing both .eps and .pdf, imo)
\include{random plug for documentation fest coming up in june (15th?)}
--
- Gus
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Re: Creating PDFs
On Thu, Apr 18, 2002 at 03:14:12PM +1000, Terry Collins wrote: > Angus Lees wrote: > > On Thu, Apr 18, 2002 at 01:12:29PM +1000, Terry Collins wrote: > > > Angus Lees wrote: > > > > At Wed, 3 Apr 2002 11:39:31 +1000, marty wrote: > > > > > what tools are people using to create PDFs? > > > > > > > > i use pdftex (or rather, pdflatex). > > > > > > Can I ask how? > > > both pdftex & pdflatex just reject all the latex stuff with !undefined > > > control sequence and the doco is no help. > > > > hmm.. it just works for me (unless you do something silly like try and > > use pstricks) > > Okay, that means I'm fundamentally doing the correct thing. Just a few > crinkles involved > > > > > what version of pdftex (--version) ? > > [terryc@owl 2002]$ pdftex -version > pdfTeX (Web2C 7.3.1) 3.14159-0.13d 0.13d is *ancient* (by pdftex standards) my (standard debian "unstable" tetex packages) have: pdfTeX (Web2C 7.3.7) 3.14159-1.00a-pretest-2004-ojmw iirc, the latest pdftex release is 1.00b-pretest and included in tex live. its possible to compile a newer pdftex and drop it over an existing install (i've done it). you have to remember to rebuild the pdflatex format, and install the relevant .pool files - not just the pdftex binary. > > which tex installation? version? (distro?) > > Whatever came on the RH7.1 CDs from Everything Linux. I haven't upgraded > anything yet. redhat use the latest stable release of tetex. everyone else (suse, debian, tex live) are using tetex-beta. the only places where tex is evolving fast enough these days for that to be a problem is pdftex and context. if you need either of these, always look around for a newer version. it looks like thomas esser will be releasing a new "stable" version of tetex fairly soon though. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Newbie
At 16 Apr 2002 09:42:50 +1000, Ken Foskey wrote: > According to the notes recently on slug wine does not support directx so > it probably will not support games very well. wine does directx fine (can even use DGA). it does not do direct3d (but i think the (commercial/kind-of-proprietry/interestingly-licensed) winex stuff might?). directsound works (at least i get sound out of directx games fine). directplay doesn't. with wine, its really a matter of whether someone with enough knowledge and persistance has tried your program before. there is so many win32 system calls (and other related "system" dlls), that the functions are implemented as needed. eg: total anhiliation and star craft run perfectly fine - i had to do a ~3 line patch to get heroes of might and magic 3 to run. the rule of thumb with wine compatibility is: just try it and see. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: [Re: [Re: [Re: Cron Weird Problem from a newby]]]
At Tue, 16 Apr 2002 20:58:00 +1000, Steve Kowalik wrote:
> Feh. Reboot into single-user mode ('linux single' at the LILO prompt), and
> fsck all the hard drives you want.
no need to *reboot* to get to single user mode.
just "telinit s" - then you probably want to do a ps listing and
gracefully shutdown some of your daemons manually (noone ever gets
enough rc*.d/K* links in the right places)
then remount the drives read-only:
mount /dev/hda -o remount,ro
then fsck away:
fsck /dev/hda
then go back to where you were:
mount /dev/hda -o remount,rw
telinit 2 # or whatever run level you were at initially
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Backup Mail servers (OT)
At Tue, 16 Apr 2002 12:16:51 +1000, Ben Donohue wrote: > The secondary MX only stores the email until the primary is back up. > It then fowards the email to the primary. depends entirely on how the "secondary" is setup. there's no reason you can't have an MX over your fast cheap link, then a lower-priority MX over another, less-desirable link - and have both addresses resolve the same host. automatic failover in case the first link is down. or a beefy mail server, and a smaller mail server willing to help out during peak times - but both can do local delivery (because you have NFS mail spool, etc). > the clients check with the primary only for their email. Is this correct? > Is there a way to make some sort redundant mail server where you have two > setup and the client will respond to either one? or is this clustering? no. you try them in MX priority order. there is no "primary" and "secondary". (oh how i wish www clients could do a similar thing) -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: wine (Was: Re: Opinions, please.)
At Fri, 12 Apr 2002 14:22:53 +1000, Andre Pang wrote: > Not to take anything away from the work that CodeWeavers have > done, but I'm still surprised at the number of people who haven't > heard of Win4Lin. > Its downsides are (a) no support for Windows 2000, and (b) > basically no DirectX support. and of course (`) its non-free. note also that codeweavers is only one of the recent commercial contributors to wine. wine has been quite a lengthy project (win3.1 days, iirc) - much longer than any single commercial contributor. -- - Gus (wine.lurker) -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: More on WINE
At Fri, 12 Apr 2002 10:36:01 +1000, Alan L Tyree wrote: > I use Codeweaver WINE successfully to run a proprietary legal research > CD and (seldom now since installing OpenOffice.org) M$ Office. > > However, I have never been able to install these in a Windoze-free > system since the setup programs fall over. Is there some trick to > getting the setup programs to work? the windows installer requires some funky COM stuff which wine doesn't currently implement. there has been several mentions of people working on it over the last 6 months, and what with the new coverage that crossover is giving wine - i'm sure it won't be too far away. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: What is an easy way to see the HTTP headers in a HTTP request?
At Tue, 9 Apr 2002 15:58:29 +0900, Antony Stace wrote:
> What is an easy way to see the HTTP headers in a HTTP request.
the "lwp-request" (aka "GET", "HEAD" or "POST") script that comes with
the perl LWP library is very good for this sort of stuff.
works with all the stuff that LWP does too (https, http auth, etc).
you would be wanting the "-U" option ("-e" to see headers in the HTTP response).
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Linux in the enterprise?
At Mon, 8 Apr 2002 22:31:41 +1000 (EST), lukekendall wrote: > On 8 Apr, Andrew Bennetts wrote: > > Usually they get dumped to /var/log/messages if the system isn't too far > > gone, so that you often don't need to write anything down. > > That should be mentioned in the man page for ksymoops. try klogd(8). klogd gets the messages from the kernel, ksymoopsifies them and then feeds them on to syslogd. the fact they (may) then end up in /var/log/messages is up to /etc/syslog.conf. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Committee meeting minutes - 26th March
At 08 Apr 2002 19:49:36 +1000, Peter Hardy wrote: > Minutes of the SLUG committee meeting 26th March 2002 >Suggestions for future talks: [...] >As always, volunteers for these or any other talk would be most >welcome. still willing to do "intro to emacs", "intro to perl" and "apache internals / mod_perl" talks - if there's interest. yes, the amount of "i like vi for coding and think php is a viable programming language even though i've never known the alternatives" talk thats going around is finally bugging me ;) > SLUG website >Angus Lee's beta website (http://beta.slug.org.au) was considered for (Lees') (i think. i never did get an authoritive answer on that) >the final site, but concerns were raised about future maintainability. >The other alternative is Jeff Waugh's new website, which he will >demonstrate shortly after his return from Spain. It was agreed to >withhold a decision until then. the code was meant to be fairly easy to read, so if there's any particular part you think will be hard to maintain then let me know. so far i've had no comments on what needs to be done to http://beta.slug.org.au/ (other than improving the admin interface - which no-one can actually see anyway). comments actively sought. (somehow i think i can see how this is going to pan out though..) > Other activities docfest in june. saturday 15th unless anyone has any objections. (markup languages (troff, (La)TeX, Docbook), yada yada. proper announce/CFP to follow room booking) can whomever deals with room bookings please book me the (recently) usual slug room at UTS. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: EMACS: auto indent customisations not loading
At 08 Apr 2002 12:52:37 +1000, Simon Wong wrote: > On Sun, 2002-04-07 at 15:34, Angus Lees wrote: > > i'm guessing you are also setting it somewhere else, and that setting > > is overriding the customize option. > > Not intentionally! > > I couldn't find anything in the site wide /etc files that set this i.e. > in /etc/emacs/* or /etc/emacs20/*. > > Would there be something else? you can probably brute force it with something like: (add-hook 'c-mode-hook (lambda () (setq 'c-basic-offset 4))) i'd suggest running emacs with "-no-init-file" first, just to prove that its something in your .emacs (for our peace of mind ;) -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: EMACS: auto indent customisations not loading
At 07 Apr 2002 14:19:50 +1000, Simon Wong wrote: > I am having trouble with the auto indent settings for the C major mode > (c-basic-offset). > I want this to be 4 which is the standard setting. At one point I had > changed it to 2 but I can't change it back. > C Basic Offset: [Hide] 2 >[State]: this option has been changed outside the customize buffer. > Amount of basic offset used by + and - symbols in `c-offsets-alist'. > Parent groups: [C] note "this option has been changed outside the customize buffer" > My .emacs file has a line: > (custom-set-variables > ... > '(c-basic-offset 4) > ) i'm guessing you are also setting it somewhere else, and that setting is overriding the customize option. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Creating PDFs
At Fri, 5 Apr 2002 19:27:50 +1000, marty wrote: > $author = "Angus Lees" ; > > At Wed, 3 Apr 2002 11:39:31 +1000, marty wrote: > > > what tools are people using to create PDFs? > > > > i use pdftex (or rather, pdflatex). it usually produces pdf's of > > higher quality (conformance to pdf standard and general layout) than > > the adobe tools themselves. > > i never used latex before, any good tools or primers on it? LyX is probably the best gui editor for producing TeX, but writing the source directly isn't that hard either. after you've installed the relevant TeX packages (probably something involving the string "tetex"), have a look for a document called "lshort.dvi" (or maybe .ps). if it wasn't installed do a web search for it (its "The Not So Short Guide to LaTeX2e"). that will give you the basics. for further stuff, start looking at individual installed packages (find a texmf/doc directory or use texdoctk (if available)), packages you don't have - but could if you wanted (on ctan.org), and maybe at the mailing lists on www.tug.org. and feel free to ask questions on the slug list too. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Want a 2nd opinion on this plan to upgrade hdd
At Thu, 4 Apr 2002 00:46:23 +1000, Paul Robinson wrote: > What I was planning to do was install the new drives and mount them as /mnt/temp and >/mnt/temp2 > for example and then run cp -a / /mnt/temp so as to copy everything form the root >structure > across.. but I got to thinking.. wouldn't it eventually try to copy the /mnt/temp >content into > itself and start a really bad loop? What is the best way to upgrade hard drives in a >Linux box? cp -ax / /mnt/temp remember to modify lilo.conf and rerun afterwards. -- - Gus (now booting off hdd, due to multiple instances of the above) -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Creating PDFs
At Wed, 3 Apr 2002 11:39:31 +1000, marty wrote: > what tools are people using to create PDFs? i use pdftex (or rather, pdflatex). it usually produces pdf's of higher quality (conformance to pdf standard and general layout) than the adobe tools themselves. > the tool doesn't need to work with .doc's, just any format i can save from > abiword in... i presume abiword can save a plain text file ? -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: PERL Page Breaks
At Thu, 28 Mar 2002 15:55:53 +1100, Daniel Harper wrote: > I have some Page Breaks in a document that I want to get rid of in Perl. perl -pi.bak -e 's/\f//g' a_document -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Hard links and relinking
At Wed, 27 Mar 2002 14:01:00 +1100, Jeff Waugh wrote: > So using cp -lr is a very effective way to speed up and save space when > patching very large codetrees, but I'm interested in what's going on > beneath. > > When patch changes a hard linked file, it relinks it so that it becomes a > new file (reference to an inode). I'm assuming it does this explicitly, > because hard links would be pretty useless if it didn't. :-) nothing magic. its removing the old file before writing the new file. anything that doesn't do this needs a good beating. (my (un)favourite offender is tar. many a time did i hose my ancient non-debian system by untaring a binary libc distribution over a running system and have it merrily overwrite currently-being-demand-paged libraries. by unlinking first, everything would have been good). and no, you can't really do this transparently - at least, not without playing dodgy LD_PRELOAD games. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Ideas for future SLUG meetings....
At Mon, 25 Mar 2002 18:06:41 +1100, Jeff Waugh wrote: > ... If you teamed up with a couple of other people, we could run a stream of > developer talks for new hackers, introducing all of the free software tools, > from project management stuff, compilers to debuggers and packaging systems. > > Anyone interested in this (from either a speaker's or listener's end)? aye. and anthony's got a good point. we haven't had any *really* newbie stuff for a long time (in fact, i'm not sure if i can remember any..) i guess we pushed a lot of that out onto fests, but then never got close to the "an installfest every two months" point that was often mentioned. so now we do two (?) installfest/workshops a year, and there's little talking that happens at them even then. if we do have really introductory talks, i think we'll need to run them in parallel with "normal" talks though, or make them short enough that those who aren't completely new to linux will be able to sit through them without becoming too disruptive (quiet up the back). in fact, 5 minute "cooking show" intros after the Q&A (and again after the break?) might be an interesting way of doing it. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: terrible docs, was: strange ports and strange daemons
At Sat, 23 Mar 2002 13:49:57 +1100 (EST), david wrote: > [...] > It has occurred to me that IT people need to do a basic HUMAN > communication course before being allowed to touch a computer. > is it just me, or is the more "obvious" alternative to the above: "It has occurred to me that (human) people need to do a basic IT course before being allowed to touch a computer." (it takes a tafe course (or equivalent) to use an oxy-acetylene torch competently. thats a single flame, a thin rod of metal and a pair of goggles. a computer on the other hand..) free software authors aren't bad people. if you can summarize the use and function of SNMP in a sentence (or so), i'm sure the original authors (or at least packagers) would welcome the patch. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: I dream in #!/bin/bash (splitting a diff patch)
At Sun, 24 Mar 2002 18:11:30 +1100 (EST), Jessica Mayo wrote: > I was looking at UML this morning. I wanted a Win32 port. > I downloaded the patch to the latest stable 2.4 kernel. > > [ done while looking over the LINE 0.5 code (think Lin4Win :) ] > > But I'll never get started with distractions like these... > > ... I wanted to look at said patch file in a non-linear fashion. :) > > It contains diffs for a large number of files in the kernel tree, and I > wanted an overall picture, not the line-by-line view I was getting. > I wanted to extract those diffs into their proper tree, without patching. > ... enter a monster... :) [...] > ... apologies tools that i've found useful for perusing large source trees/patches: diffstat - limited use. basically your grep solution, except it shows the number of lines added/removed by the patch. useful for getting a real quick overview of the scope of the change. not useful for much else. ediff - emacs diff/patch mode. the excessive colour highlighting makes it immediately clear whats happening, and its easy to skip forward/backward between differences. you can also selectively apply/backout/merge, etc individual pieces, but thats not something you're interested in yet. run (x)emacs. type "M-x epatch". press '?' in the little ediff "control buffer" that pops up for help - you probably only want "n" and "p" (next and previous) (type all ediff commands in the ediff control buffer). note that this will actually apply the patch before showing you the differences, so you should work on a copy if you wanted to keep the original. (note: M-x epatch-buffer won't modify the files, but only works on single-file patches) etags/ctags (etags for emacs, ctags for vi) - *very* useful for browsing unfamiliar non-trivial source trees in general. you run (eg) "etags `find . -name '*.[ch]' -print`", which creates a TAGS file (ctags creates "tags" iirc) and then open up a source file in emacs and hit M-. (find-tag), type in a symbol name (you might have to give the location of the TAGS file) and the editor jumps to the point in the source code where that symbol was defined. its like having lxr or something built in to your editor. read editor docs for details (and vi usage). (note that automake-generated Makefiles have TAGS/tags targets that build the appropriate files - in a better way than the above etags/find too) there's probably some other tools that can work with cvs (or maybe rcs), rather than with a patch file directly. so it might be worth your while to check in the unpatched and patched versions to some revision tool and then look around again (certainly ediff can work equally as well straight from cvs/rcs/sccs). -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: stopping a module from autoloading
\begin{Andy Eager}
> Is there any way to stop a module from autoloading ?
>
> I am trying to get xcdroast to use my ide cd drive as the reader. When
> I put a cd into the IDE drive, it wants to mount the drive. (and it
> does!). When I try to rmmod ide-cd, it just goes and puts it back
> again. Got so sick of it that I physically renamed the ide-cd.o to
> something else just so I could modprobe ide-scsi for xcdroast.
echo "alias ide-cd off" >> /etc/modules.conf
(debian note: don't use modules.conf, chuck it in a file in /etc/modutils
and then run update-modules)
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Opinions, please.
\begin{Bill Bennett}
> I'd like to install elvis, which is a vi/ex clone, but it
> occurred to me to ask if anyone has any experience with others.
> Has anyone any strong feelings they'd like to air?
emacs
M-x viper-mode
;)
it has 5 emulation levels, to ease the rehabilitation process.
(why is there always vi discussions on this list, but never emacs ones?
is there truly that few emacs users in slug?)
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Re: lynx's homepage?
\begin{Andre Pang}
> On Thu, Mar 21, 2002 at 11:32:03PM +1100, Angus Lees wrote:
> > At 21 Mar 2002 10:17:51 +1100,
> > Peter Hardy wrote:
> > > On Thu, 2002-03-21 at 10:11, Jamie Honan wrote:
> > > > Pictures, frames, ssl, and text mode browsing.
> > >
> > > Does w3m do http auth? This is the one sticking point that's stopped me
> > > from replacing lynx with links on all of my machines.
> >
> > w3m does. links does not.
>
> What do you mean by http auth? The login/password stuff set on a
> htpasswd file? (Just making sure that you're not talking about
> general SSL).
yes. HTTP (Basic, but i think w3m can also do digest) auth.
both w3m and links can do ssl, judging by the presence of links-ssl and
w3m-ssl debian packages.
> > i got freaked out by w3m displaying an image in my xterm just a few
> > days ago. since when/how does an xterm display an image?
>
> Escape codes; I think it's G. I don't know if there's
> support for it in the official xterm, but I think rxvt has had an
> experimental graphics mode thing for yonks, and therefore aterm
> would too. Not sure about Eterm and xterm.
i only use Real XTerms(TM). (the others always have some lingering
compatibility issues, ime)
i had to move the cursor in and around it to convince myself w3m
wasn't doing some wierd X11/ImageMagick/Shape hackery or something.
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: wine (Was: Re: Opinions, please.)
\begin{Peter Hardy}
> On Fri, 2002-03-22 at 10:39, Jeff Waugh wrote:
> >
> > > Commercial software, but well worth the price, and the money gets poured
> > > back in to wine development.
> >
> > Proprietary! We use Free commercial software *every* day.
>
> Pedant! :-)
> Free software, with some proprietary commercial glue, then?
note that as of a week or two ago, wine is now LGPL (instead of X11ish)
- mostly because codeweavers(*) (who make the crossover plugin) had become
aware of some nasty proprietariness afoot.
yay for the commercial wine companies and they're willingness to contribute
back, even though they have not been required to (until now).
(*) at least, i think it was them. it was definately one of the 3 (or so)
companies actively/commercially involved in wine.
if anyone thinks license discussions tend to never get anywhere, flicking
back through the wine archives and seeing how this license change was
quite quickly decided is probably worth the effort. alexandre is to be
congratulated. (there was even reference made to a summary of the MPL
license made by anand back in the early mozilla days.)
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: lynx's homepage?
At 21 Mar 2002 10:17:51 +1100, Peter Hardy wrote: > On Thu, 2002-03-21 at 10:11, Jamie Honan wrote: > > Pictures, frames, ssl, and text mode browsing. > > Does w3m do http auth? This is the one sticking point that's stopped me > from replacing lynx with links on all of my machines. w3m does. links does not. i tend to use w3m for viewing local html files, and links for actual web browsing, since links does render-while-downloading, etc which makes it feel a little nicer over a slow link. i got freaked out by w3m displaying an image in my xterm just a few days ago. since when/how does an xterm display an image? there's also w3, which can be textual or graphical, depending on what options you have turned on. w3 is the emacs web browser, of course ;) having said that, i've started using w3m-el, since i've found w3 to be a little slow on my machine. (i wish some of the "popular" web browser groups would have a look at w3 and steal some ideas out of it tho) -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Making Debian easier??
At Wed, 20 Mar 2002 16:27:54 +1100, Richard Hayes wrote: > As a Debian dummy who is trying to change from Red Hat to Debian but > is failling ;) > > I read about the "Storm Administration System" which I believe was a > part of the Storm distribution. > > "It contains the Storm Administration System (SAS). A modular based > remote, interface independant (Graphical and Text based UI), > administration system." > > The question is was it GPL software? > If so, where can I get a copy? > Is anybody doing anything with it? "aptitude search storm" turns up "stormpkg" the description for stormpkg is: Storm Package Manager is a GNOME-based user-friendly replacement for the console-based package manager "dselect". Many useful features are available, such as full dependency management, APT source list editing, and package list filtering. for a "standard debian" gui tool, i would recommend using deity-gtk, and the gnome frontend to debconf. if you can live with curses interfaces, then try deity-curses, or aptitude (which i use). -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: dual IP addresses
At Wed, 20 Mar 2002 08:52:35 +1100,
Alan Vink wrote:
> > Setting up IP Aliasing on A Linux Machine Mini-HOWTO
> > http://www.linuxdoc.org/HOWTO/mini/IP-Alias/
> "IP Alias has been deprecated in 2.4.x and replaced by a more powerful
> firewalling mechanism"
with iproute there is no need for interface alia. each additional
address is "as important" as the first one. a device can have no
address or a device can have several addresses in several address
families (eg: ethernet, IPX, IPv6, Appletalk, etc)
for help on the ip command:
ip help
for help on manipulating addresses with ip:
ip address help
to see current addresses:
ip address show
to add an additional address:
ip address add 192.168.12.99/24 broadcast + dev eth0
will add an address 192.168.12.99, with netmask 255.255.255.0 and
broadcast 192.168.12.255 ('+' means replace host part with all
ones, '-' means replace with zeros, or you can use an actual address)
to device eth0. note that this address is invisible to normal
ifconfig(8).
if you want something that shows up in ifconfig (similar to a 2.[02].x
ip alias) or can be accessed by name in other programs, add a label:
ip addr add 192.168.12.99/24 brd + dev eth0 label eth0:1
if the broadcast is the same as another address on that interface, you
probably won't want to add it again (just don't give the "broadcast
xx" argument).
iproute(8) removes a whole lot of unnecessary restrictions in the
"old" way of setting up networking (with ifconfig and route).
its good, use it.
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: debian.potato-X
At Tue, 19 Mar 2002 02:22:26 +1100, Bill Taylor wrote: > cat didn't work last night, unless I did somthing wrong. the particular > readme's are .gz .does that make a difference? > they seem to be binaries or produce symbols instead of letters with > cat,less or more,ended up locked up again. the 3 cd set is 2.2r5, the > README files are in usr/share/doc/xserver-common after a default install in your .bash_profile, put the lines: export PAGER=less eval $(lesspipe) this sets your default pager to less, and sets up less to automatically view various types of files (including compressed ones). -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: debian.potato-X
At 19 Mar 2002 22:21:16 +1100,
Ken Foskey wrote:
> Power users use vim which will automatically unzip when it opens for
> edit. It will then zip when you save automatically.
>
> You just have to be deranged to understand the command set.
ObEmacs:
M-x auto-compression-mode
(or echo "(auto-compression-mode)" >> ~/.emacs)
and then simply C-x C-f to visit a file, directory, ftp site, etc.
you can even view a .tar.{gz,bz2}, wander around the contents, choose
and modify a file and then save it back inside..
You just have to be able to use menus (esp. with xemacs) to understand
the command set.
--
- Gus.vi.is.for.editing
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Forward: e-tax
From: Sam Varghese <[EMAIL PROTECTED]> Date: Fri, 15 Mar 2002 20:20:48 +1100 Subject: e-tax As you are aware, e-tax was introduced as a mainstream option last year. The lack of software for any other platform other than Windows has led me to write to you. I have begun some correspondence with the tax office in an effort to try and create a groundswell of opinion that would lead to rectifying this situation. One can but try. The efforts I have made are at http://www.gnubies.com/tax/index.html I would be grateful if you could post news of this to your mailing list(s). Those who wish to, could write to the tax commissioner - [EMAIL PROTECTED] - expressing their sentiments. Thank you, Sam -- Sam Varghese http://www.gnubies.com Linux and all free software is not about the smell of money but the improvement of community -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: tar large files
At Fri, 15 Mar 2002 12:54:03 +1100, Alister Waller <[EMAIL PROTECTED]> wrote: > I have a folder with some rather large files eg 300MB + > > I want to be able to tar up the whole folder so that I can burn them onto CD > and put them onto another system. (in another city). > > I have read up a bit on the tar command and think that this should do the > trick. > > tar cvfk files.tar 60 RRBKGS* > > I get the following error > > [root@mail murr]# tar cvfk files.tar 60 * > tar: You may not specify more than one `-Acdtrux' option > tar cvfk files.tar 60 RRBKGS* is the same as: tar -c -v -f -k files.tar 60 RRBKGS* "files.tar" is an argument to the "-f" option, thus they need to be next to each other. this is confusing the argument parsing and giving you that error. it probably thinks you're doing: tar -c -v -f k -f i -l etc.. i'm not sure what you think the -k option does, but (tar version 1.13.25): -k, --keep-old-files don't replace existing files when extracting probably isn't going to make a difference when creating an archive. i don't know what you think the 60 is for - probably something to do with -k. i presume RRBKGS* is a shell glob that matches the files you want to put in the archive.. so. what you wanted is: tar cvf files.tar RRBKGS* the filename has to go next to the "f". and if you have other options, then you can give them as such: tar cvf files.tar -k 60 RRBKGS* but don't do that, because GNU tar doesn't treat -k the way you think. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Combined Desktop talk for the Linux Workshop
At Thu, 14 Mar 2002 20:30:07 +1100, Jeff Waugh wrote: > I'm looking for a KDE person who would like to do a combined GNOME/KDE > desktop talk with me at the upcoming Linux Workshop in May. > [ Yes, this does exclude the amazing choice outside the scope of desktop > environments, but I'm pretty sure that new or migrating users will find > GNOME and KDE to be simpler introductions to Free Software. It would be a > crime if there weren't any Enlightenment or "using Linux on older computers" > talks to go with them. ] given that i've now switched from mutt to an emacs mail reader (wanderlust - see the extremely long User-Agent header) and thus use emacs exclusively for mail/address book/calendaring, i feel qualified to give a talk demonstrating "the emacs desktop". naturally i wouldn't expect gui newbies(*) to attend/understand/cope, so this would depend entirely on demand. I think more advanced users might be interested (and surprised) by the power of some of these relatively unmentioned alternatives to evolution/gnome/kde. Whether such users will be at a linux installfest *cough* workshop is another matter though.. (*) "gui newbies" as opposed to "non-gui newbies" who would probably cope with emacs/emacspeak fairly well. (hi bart, yz) -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: size mismatch when apt-getting
At Fri, 15 Mar 2002 10:39:14 +1100, Anand Kumria wrote: > mirror.aarnet.edu.au and planetmirror.com are maintained by the same group > of people. So if one is broken, both are. from what i've heard from jason, i thought mirror.aarnet end planetmirror.com.au where different machines, with different copies of the debian archive (and i believe slightly different copying strategies/times) i use ftp.au.debian.org (aka planetmirror.com.au) and have never had a problem - although i also have http.us.debian.org as a fallback, which probably helps a lot. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: On the use of databases
On Tue, Mar 12, 2002 at 10:29:54PM +1100, Jeff Waugh wrote: > It feels a bit wrong using a massive hulk of a > thing like MySQL or Postgres for what's essentially logging. apparently there's this thing called O_APPEND, which supports multiple writers and is persistant ... -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: LaTeX, PsTricks and Colour Problem.
On Mon, Mar 11, 2002 at 03:10:32PM +1100, Bill Bennett wrote:
> Using PsTricks, the following command will draw a thick lined
> polygon:---
>
> \rput(46.0,247.0){\PstPolygon[linecolor=MyYellow,unit=55,PolyNbSides=9,PolyOffset=2]}
>
> Unfortunately, it will also fill the polygon with white.
it should be clear..
you aren't setting a fillcolor or fillstyle above?
give me a complete example that fails
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: ask summary of installed programs
On Mon, Mar 11, 2002 at 12:44:36PM +1100, Andy Eager wrote:
> Just so happens I am working on a script to do a similar thing for the
> LPIC course currently being undertaken at Granville TAFE.
>
> rpm -qai | egrep "^Name|^Size" | sed "s/ */ /g" | awk 'BEGIN {FS=" "} \
>$1 == "Name" && $2 == ":" { printf ("%s: %-40s", $1, $3) } \
>$1 == "Size" && $2 == ":" { printf ("%s: %s\n", $1, $3) }' \
> | sort -k 4 -n > packlist
rpm -qai | awk 'BEGIN {FS=" "} \
$1 == "Name" && $2 == ":" { printf ("%s: %-40s", $1, $3) } \
$1 == "Size" && $2 == ":" { printf ("%s: %s\n", $1, $3) }' \
| sort -k 4 -n > packlist
no need to cleanup before passing to awk. awk is cleverer than you think.
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] doc fest in june (was Re: A good troff book (was: RE: troff))
On Mon, Mar 11, 2002 at 01:40:36PM +, Rev Simon Rumble wrote: > My point precisely. Use something that _IS_ interactive, or at least > something that is a bit easier to learn. nothing produces plain text documents as well as troff (see manpages for example). and wysiwyg is a flawed concept that is inappropriate for textual documents. this leads me to mention the documentation fest i plan to hold in june. exact date to be announced, but probably one of the two middle saturdays. would someone willing to speak about troff (intro, pic, more?) please step forward. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Re: 'CUPS' deb pkg name ?
On Sun, Mar 10, 2002 at 02:50:58PM +1100, Luke McKee wrote: > Gnome-apt ? deity-gtk -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Distributed passwords
On Fri, Mar 08, 2002 at 02:02:53PM +1100, Matt Hyne wrote: > Guys, I have number of linux boxes here and I want to only have one place that >username/passwords are stored (for admin reasons). Rather than go the full overkill >and set up NIS, is LDAP (or something else) a better alternative. why is nis or ldap overkill? compared to setting up cron jobs and ssh keys to rsync files around, and then worrying about locking and atomic copying and testing the whole lot to where you're actually comfortable it recovers from errors safely, you may as well just install slapd, libnss-ldap, libpam-ldap and run one of the padl.com scripts to import an existing /etc/passwd. nis is potentially quicker to setup, but by the time you learn all the quirks of nis, you'll take longer. and once you've got it setup, you'll wish you'd chosen ldap instead. go for ldap. its good. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Re: 'CUPS' deb pkg name ?
On Thu, Mar 07, 2002 at 12:04:00PM +1100, Jamie Wilkinson wrote: > This one time, at band camp, Angus Lees wrote: > >if you don't like dselect (and thats understandable), you could try > >deity-curses, deity-gtk or aptitude. > > Aptitude comes highly recommended by a while bunch of top Debian > maintainers, who were really wrapped with the other day. since discovering this morning that aptitude now *does* detect "new" packages, i hereby publicly declare my switch from dselect to aptitude. -- - Gus.dead.diehard.dselect.fan -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: ask perl
On Tue, Mar 05, 2002 at 05:51:16PM +0800, henry wrote:
> I want to print the first 5 characters in front of a definite word
> as follows:
> It's supposed that the definite word is AU.
> INPUT -->aassewrab cdAUwst
> OUTPUT(I hope to get ) - ->ab cd
($output) = $input =~ /(.{5})AU/;
(semi-intentionally obfuscated to prompt you to read some documentation ;)
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: An editor for SGML?
On Wed, Mar 06, 2002 at 12:46:58PM +1100, Jeff Waugh wrote: > > > Does anyone know of a good SGML editor which can hide the tags (but > > continue to insert them for paragraphs etc), and has a spell checker? > Lyx, or, if you're willing to do some hacking, Conglomerate. emacs excellent psgml-mode won't hide the tags, but it certainly helps type them (and syntax check, and fill out defaults, etc) even if you don't otherwise use emacs, i'd suggest its sufficiently good to be worth a try. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: ask perl
On Mon, Mar 04, 2002 at 08:24:08PM +1100, Adam Kennedy wrote: > And having given you two fish, let me suggest at this point that it would be well >worth your while learning to fish. > > A great place to start is perldoc.com, which should contain just about everything >you could need. > Looking through you see "How do I strip blank space from the beginning/end of a >string?", clicking on the link takes you to. > > >http://perldoc.com/perl5.6.1/pod/perlfaq4.html#How-do-I-strip-blank-space-from-the-beginning-end-of-a-string- assuming you have the perl documentation stored locally, another way to find the same information would be: perldoc -q space > A great alternative if you are really stuck would be to go to what is arguably the >"main" perl channel. > > irc://irc.rhizomatic.net/#perl ( assumes you have mozilla ). i presume thats #perl on openprojects? if not, thats where i commonly am, and such questions are certainly the norm. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: 'CUPS' deb pkg name ?
On Sat, Mar 02, 2002 at 11:59:47AM +1100, Craige McWhirter wrote: > A little trick I find useful is using the cache search functionality: > apt-cache search cups please note that apt-get/apt-cache are no substitute for a real user-friendly apt frontend. if you don't like dselect (and thats understandable), you could try deity-curses, deity-gtk or aptitude. these allow searching, browsing, prompt you for package recommends and suggests (something apt-get completely ignores), etc. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: bash shell script variables
On Mon, Mar 04, 2002 at 12:07:30PM +0900, Antony Stace wrote: > > if [ $iexportflag = "yes" ] ; then > > you might want to try using "==" (equality check) instead of "="(assignment) no you most certainly don't. == is a bash extension and will not (necessarily) work under other POSIX shells. i have sh -> ash, and this is one of the most common errors script writers make. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: ask perl & C
On Fri, Mar 01, 2002 at 06:45:53PM +1100, Adam Kennedy wrote: > Or if the return values are more substantial > #!/usr/bin/perl > > # Get the first lot of results > @stdout_from_c_program = `'first-c-program`; > die "Didn't get anything" unless @stdout_from_c_program; die "program returned non-zero exit code ($?)" if $?; > # Some processing ( or not ) > @args = munge( @stdout_from_c_program ); > > # Hand the processed ( or not ) results to the second program ( assuming you > need to pass to the c program ). > system( "second-c-program $arg[0] $arg[1] $arg[2] etc" ); # Assuming you > need to get you will want to read "perldoc -f system" if you want to interpret $?. ** note that system() returns the process exit code. ** ** ie: 0 (false) on success, non-zero (true) on failure. ** -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: bash shell script variables
On Mon, Mar 04, 2002 at 01:47:03PM +1100, Alister Waller wrote: > I have a companies file that contains RR and XX > > What I want is for the script to process the export for each company in the > companies file and check to see if that companies export flag is set to YES, > if so do the export, if not then miss it out. > > I have an example below: > > #!/bin/bash > > RRexportflag=yes > XXexportflag=no > > for i in `cat companies` > do > > if [ $iexportflag = "yes" ] ; then > > ...do some exporting ... > > fi > done eval is what you are looking for (see other answers). can i suggest changing your code to use something like "read" instead: have a companies file that contains RR and XX and some flags for each (lines like: XX yes) while read company export; do echo "doing $company" if [ $export = yes ]; then .. do some exporting fi done < companies you can expand the companies file (and arguments to "read") to cope with extra per-company options. "read" reads from stdin, hence the while loop being redirected from the companies file. see the section on read in the sh manpage. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: ask something about Xlib
On Mon, Feb 25, 2002 at 11:59:39PM +1100, Crossfire wrote: > If your desire to learn how to program using Xlib, you should refer to > souce code from simple X11 programs, the X11 programming manuals, and > to any XLib tutorial material you can find. (There is a book about > XLib and Motif programming by Jan Newmarch which I learnt the basics > of XLib programming from) the (rather extensive) X11 reference docs that come with xfree86 aren't too bad if you don't feel like buying a book. it'll probably take you a few reads to resolve all the forward references tho. (debian package xspecs or xbooks depending on your version) (you only want to read the Xlib and possibly Xt docs, the rest can get a bit scary) -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: How can I get a list of locks in a table in Postgres?
On Fri, Feb 22, 2002 at 04:53:53PM +0900, Antony Stace wrote: > I have run into a problem with postgres. I cannot update some rows in a table while > i can update other rows in the same table. How can I see what locks are on a >table/row in postgres? don't know. but "pgmonitor" could show you what queries were currently executing, which may be enough to find the problem. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: ask perl(install cpan-module)
On Thu, Feb 21, 2002 at 10:16:26AM +0800, henry wrote: > 2. download Net-telnet-3.02.tar.gz from www.cpan.org > 3. a. untar it > b. perl Makefile.PL > c. make test ->pls see tmp.log as attached(that why my >perl-script fail ) > d. make install try "make TEST_VERBOSE=1 test" -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Perl Help
On Wed, Feb 13, 2002 at 12:22:56PM +1100, Jason Rennie wrote:
> but perl doesn't seem to have a setuid function ?
$> = $user_id;
see "$<", "$>", "$(" and "$)" in perlvar(1)
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] (fwd) Inquiry
tony: i've forwarded your mail to the SLUG mailing list, where i'm sure someone will be able to help you out. feel free to post further linux questions/answers to [EMAIL PROTECTED] if you want to subscribe to this list (or view the list archives) see http://lists.slug.org.au/listinfo/slug slug list: please CC replies to tony, as he isn't subscribed to the list. - Forwarded message from Tony Prince <[EMAIL PROTECTED]> - From: Tony Prince <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] Subject: Inquiry Angus Lees, Secretary, SLUG Dear Angus, I'm not sure if you're the right person to bother with my problem (if not, I apologize!), but I couldn't find any other likely source of advice on your website, so I thought I would try asking you. I am a Mac user, of the ordinary, non-hacker variety, but towards the end of last year I read that SuSE had brought out a version of Linux (7.1, PowerPC version) that could be installed on a Mac, reportedly without too much difficulty. So I bought this version, bought a new (discontinued model) iBook, and did in fact manage to partition the hard disk and install both systems successfully. Now however I have struck a problem that I haven't been able to solve. I want to install WordPerfect and Star Office in the Linux partition, and for the last couple of months (off and on) have been trying unsuccessfully to do so from a disk that comes with Next Handbooks' "Learning Linux" (which I bought just because it had these two programs on it). Of course the Linux OS is complicated, and I don't really understand it very well yet, but even so, I've tried to follow the instructions I found on the disk, attempted to use YaST and rpm according to the advice in the SuSE manual, and so on but all to no avail. So now I'm stuck, and that is why I'm writing to you, in the hope that there may be somebody at SLUG that would be able and willing to advise me. If there is, or if you could give me any other advice yourself, could you please let me know? Thank you. Yours sincerely, Tony Prince ([EMAIL PROTECTED]) - End forwarded message - -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Wine vs Lindows
On Thu, Jan 24, 2002 at 08:16:51PM +1100, Peter Hardy wrote: > Cool trick for wine users: > * Make sure you've got support for the binfmt_misc driver[2] in > your kernel. I think it gets compiled in to distro kernels > these days - check that /proc/sys/fs/binfmt_misc/ exists, or > try "modprobe binfmt_misc" > * If you're using a kernel later than 2.4.13, mount the > binfmt_misc filesystem: "mount -t binfmt_misc none > /proc/sys/fs/binfmt_misc" > * "echo ':DOSWin:M::MZ::/path/to/wine:' > > /proc/sys/fs/binfmt_misc/register" > > And you should be able to run Windows binaries from the commandline. > ie, instead of "wine winfile.exe" you can just run "winfile.exe" and let > the kernel sort out how to make it go. for debian users, installing the wine and binfmt-support packages will automatically do this for you (at least with current "unstable" versions) its pretty cool. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Netatalk under Debian 2.2r2
On Fri, Jan 25, 2002 at 09:36:43AM +1100, CaT wrote: > On Fri, Jan 25, 2002 at 12:28:26AM +1100, Michael Kraus wrote: > > Having problems with netatalk. (Using the Debian package as per ftp site.) > > When I run '/etc/init.d/atalkd start' the following occurs: > ... > > Any ideas? > > Do you have appletalk protocol compiled into the kernel (or loaded in as > a module if that's the way you lean)? alias net-pf-5 appletalk .. but /etc/init.d/netatalk should load it (at least in the 1.5.1.1-1 package i'm using here) -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Network Operating Systems Comparisons & Win2000 Dual Boot
On Sat, Jan 26, 2002 at 10:09:13PM +1100, Russell Andrew Willis wrote: > Also trying to set up dual boot for Win2000 & RH7.1, I'd like to use > NTFS but lilo doesn't seem to load no matter what order I load them. I > have gone through the archives & there appears some issue with the NTFS > system, suggestions welcome. i had the nt bootloader thing booting linux once, by copying the dos entry in boot.ini (or whatever file it is) and pointing it at a copy of the normal linux (lilo) boot sector: install lilo onto a partition (eg: hda2), not the MBR dd if=/dev/hda2 of=/mnt/dosc/linux.bin bs=512 count=1 i think i only needed the first 512 bytes, you might want to copy 1k or so. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: wine capabilities.
On Fri, Jan 25, 2002 at 12:19:51AM +1100, Michael Kraus wrote: > G'day Sluggers and Peter, > > On Thu, 2002-01-24 at 21:37, Michael Kraus wrote: > > > a) run Corel Draw, Adobe Illustrator, Quark Express, etc? > > > > There's a Linux-native version of Corel Draw kicking around I think. > > Info about compatibility of other apps can be found at > > http://wine.codeweavers.com > > Hrmm... wine doesn't seem to be so good (not enough app support)... May > have to go the VM solution (as per below). give it a try. if it works, then good - else it hasn't cost you anything. i can run MSWord through wine and it would probably be stable enough to do some work with. Success with wine depends greatly on: 1. how many undocumented features your program takes advantage of. generally this is mostly related to how new your program is (and whether it was written by microsoft). 2. whether other people have tried and explicitly implemented the functions that your application requires. so just give it a go and see. you will probably also want to fiddle with using windows or builtin common dlls (see DllOverrides in wine conf) > > > b) have Windose fonts available to it? > > > > It's a fairly trivial matter making Windows truetype fonts available to > > Linux in general, and by extension, wine. > > Thought so... what about non-truetype? what other formats are there? if the app rasterises them directly, then it should continue to do so - otherwise, it'll display any font that your xserver can draw. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: PDF, ghostscript and printing.
On Wed, Jan 30, 2002 at 02:37:23PM +1100, Terry Collins wrote: > I need to be able to receive PDFs from that other OS and print them here > > First problem is that pdf2ps errors out. > > So I used xpdf to view the mailed pdf and saved it as a postscript. i've had more luck with pdftops (comes with xpdf) rather than pdf2ps (comes with ghostscript). although i haven't done many tests with ghostscript v7, which apparently has much improved pdf support. -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Silly Debian question
On Fri, Jan 18, 2002 at 11:30:20AM +1100, Rob B wrote: > At 09:45 18/01/2002, Richard Hayes sent this up the stick: > >Last night I tried to install Woody on a system without either CDRom or > >networkcard. > > > >I have a number of floppies: driver-1.bin, driver-2.bin, driver-3.bin, > >driver-4.bin, root.bin and rescue.bin > > > >I can easily install the 6 disks but I can not find the other disk images > >to > >install a base systems on either the debian CD or ftp site. > > > >Do I need the eleven other disks and where are they? > > There aren't any base- disks for Woody, only Potato. Your best bet is to > install Potato and apt-get update && dist-upgrade to Woody after fixing > your /etc/apt/sources file. the woody install doesn't use base disks anymore, it downloads/installs the appropriate debs directly. in fact, if you're using compact (at least), it can even suck the other drivers disks over the network (yay. only 2 floppies needed ;) richard: you only need the other disks if you actually want to install something... to install them from floppy disks, you'll probably have to download them yourself and either copy them across to the harddrive (and point the installation at them) or somehow convince the installation to use them off the floppies directly (which is probably going to be a hassle). i'm not sure how to get the list of packages you'll need. this might be good enough: grep-dctrl -nX -sPackage -FSection base /var/lib/dpkg/available but check that list by hand first. i doubt you need 27 kernel-images, for example ;) assuming the root.bin tar is good enough (which is a big assumption, might want to check first), a nice way to copy them all across might be to use the tar "-M" option: grep-dctrl ... > /tmp/package-list.txt vi /tmp/package-list.txt apt-get autoclean cd /var/cache/apt/archives tar cvMf /dev/fd0 $(ls $(sed 's/$/*.deb/' /tmp/package-list.txt)) and then "tar xvMf /dev/fd0" to extract -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Debian 2.3 Config Documentation
On Tue, Jan 08, 2002 at 12:45:45PM +1100, chesty wrote: > > I have found the lack of documentation > > about how Debian configure things after the installation quite a > > surprise. Does anyone know of any decent documentation on post > > installation of a Debian release? > > So you're looking for the "Secret knowledge of the Debian" pack? > > www.debian.org > www.debian.org/doc > apt-get install debian-guide > apt-get install debian-policy > apt-get install doc-debian *everyone* should have a flick through the Debian FAQ (/usr/share/doc/debian/FAQ/index.html from doc-debian) i'd been using Debian for more than a year before i discovered it, and i still learnt several very useful things. if you want to make it all a little easier to browse, consider installing "dhelp", then looking at /usr/share/doc/HTML/index.html (or http://localhost/doc/HTML/) -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Re: netscape on sparc (debian)
\begin{David Fitch}
> On Sun, Jan 06, 2002 at 10:15:50PM +1100, Angus Lees wrote:
> > netscape doesn't exist for sparc-(glibc-version) ?
> > try mozilla or something instead.
>
> h, so why not?
> got mozilla but not really that impressed with it.
ask netscape.com - its proprietary software, remember?
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: netscape on sparc (debian)
\begin{David Fitch}
> ok it's got me! how do you find/install netscape for sparc (debian
> potato but also woody)?
> I can see all the java and spelling etc pkgs but not the one with
> the actual executable!
netscape doesn't exist for sparc-(glibc-version) ?
try mozilla or something instead.
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Slackware anybody?
\begin{Pete Ryland}
> If you don't run Window Maker, put it in your gnome-session through the
> control centre. Your window manager might be able to make it run on a
> particular desktop, or start minimized or whatever.
>
> Lastly, if you don't run gnome, put it in your .Xsession.
.. or run some other x session manager. "xsm" is probably the "standard"
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: A frees/wan question
\begin{Howard Lowndes}
> I am setting up Frees/wan IPSec tunnels between two sites that both have
> dynamic IPs.
>
> I can get both sites to do a dynamic DNS update (both forward and reverse)
> to a DNS server with a static IP before I need the tunnels to come up.
>
> At the left end, basically the listening end, I have no problems because I
> use:
> keyingretries=1
> left=%defaultroute
> leftrsasigkey=%dns
> [EMAIL PROTECTED]
> right=%any
> rightrsasigkey=%dns
> [EMAIL PROTECTED]
> auto=add
>
> At the right end, the sending end, I use what is essentially a Road
> Warrior setting:
> keyingretries=0
> leftrsasigkey=%dns
> [EMAIL PROTECTED]
> right=%defaultroute
> rightrsasigkey=%dns
> [EMAIL PROTECTED]
> auto=start
>
> What I would like to put here is:
> left=%dns
>
> It makes sense to me that that should work, after all it uses the DNS to
> get the KEY record so why not the A record, but it is not valid.
>
> I was wondering if opportunistic keying might be the answer, but apart
> from having difficulty understanding it, I am not sure if it is what I
> want anyway.
>
> Any ideas?
it would simply be:
left=left.domain.name.com
wouldn't it?
do you restart freeswan after each DNS update?
freeswan converts everything into IP addresses at startup (including
%defaultroute settings..), which is nasty for dynamic environments.
i'd just use (x.509) certificates and:
left=%any
right=%defaultroute
leftcert=Mycert.pem
rightcert=Theircert.pem
chucking *cert.pem into /etc/ipsec.d/
(ie: identify by certificate, and ignore DNS altogether)
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: help for script
\begin{Bernhard Lüder}
> I need to convert 2 lines of data to one.
>
> This is the data I can extract from the data base:
> FIXEDdata1,Flexibledata1,Neededdata1
> FIXEDdata1,Flexibledata2,Flexibledata3,Flexibledata4,Flexibledata5,Flexibled
> ata6,Flexibledata7,Neededdata2,Neededdata3,Flexibledata8
>
> This is the data I want to use in a comma separated from:
> Neededdata1,Neededdata2,Neededdata3
>
> How can I strip off the FIXEDdata and the Flexibledata so I only keep the
> Neededdata?
>
> There is always multiple double lines of data with some single lines, like:
> FIXEDdata1,Flexibledata1,Neededdata1
> FIXEDdata1,Flexibledata1,Neededdata1
> FIXEDdata1,Flexibledata1,Neededdata1
> FIXEDdata1,Flexibledata2,Flexibledata3,Flexibledata4,Flexibledata5,Flexibled
> ata6,Flexibledata7,Neededdata2,Neededdata3,Flexibledata8
> FIXEDdata1,Flexibledata1,Neededdata1
> FIXEDdata1,Flexibledata2,Flexibledata3,Flexibledata4,Flexibledata5,Flexibled
> ata6,Flexibledata7,Neededdata2,Neededdata3,Flexibledata8
> FIXEDdata1,Flexibledata1,Neededdata1
> FIXEDdata1,Flexibledata2,Flexibledata3,Flexibledata4,Flexibledata5,Flexibled
> ata6,Flexibledata7,Neededdata2,Neededdata3,Flexibledata8
> .
>
> where do I start?
right, first problem is easy: uniq(1)
you could do that yourself in the scripting language, but its not worth it
unless you're worried about the overhead of execing another process.
(uniq only removes adjacent duplicate lines, so if you need to remove
duplicate lines wherever they appear use "sort | uniq")
joining two lines:
sed is no good for multiple line work. you can do it, but its just silly:
uniq | sed -ne
's/^\([^,]*,\)\{2\}\([^,]*\)$/\2/;h;n;s/^\([^,]*,\)\{7\}\([^,]*,[^,]*\),.*$/,\2/;H;x;s/\\
//p'
(note the "embedded newline". you could make this a whole lot more legible
by writing the statements on multiple lines, but that's just rearranging
deck chairs)
as soon as you have delimited fields, you should be thinking awk (or perl,
etc). especially when you need to keep "state" between lines.
in awk, "NR" is the record (line) number and "$1,$2,$3.." are the individual
fields in the current record ($0 is the whole record).
so:
uniq | awk -F, 'BEGIN {OFS=","} NR%2 == 1 { x = $3 } NR%2 == 0 { print x, $8, $9 }'
just for comparison, perl is fairly similar:
uniq | perl -anl -F, -e '$,=","; if ($.%2) { $x=$F[2] } else { print $x,$F[7],$F[8] }'
or even:
uniq | perl -apl -F, -e 'undef$\;$_=$.%2?"$F[2],":"$F[7],$F[8]\n"'
(who said line noise? ;)
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: how to use kernel-package and patch kernel for xfs?
\begin{Rob B}
>
> OK ... this is what I have so far ...
>
> aylee:/usr/src/kernel-source-2.4.14$ export PATCH_THE_KERNEL=YES
> aylee:/usr/src/kernel-source-2.4.14$ sudo
> make-kpkg --revision=loop.1.0 --added_patches xfs configure
> aylee:/usr/src/kernel-source-2.4.14$ sudo make config
>
> aylee:/usr/src/kernel-source-2.4.14$ sudo make-kpkg kernel_image
>
much of the joy of make-kpkg is that you don't have to be root to build
kernels.
% tar jxf /usr/src/kernel-source-2.4.14.tar.bz2 -C /tmp
% cd /tmp/kernel-source-2.4.14
% export PATCH_THE_KERNEL=YES
% fakeroot make-kpkg --config=xconfig --revision=loop.1.0 kernel_image
% sudo dpkg -i ../kernel-image-2.4.14_loop.1.0_*.deb
since i'm usually upgrading an existing kernel, i copy /boot/config-2.4.13
(or whatever) into .config before running make-kpkg (and leave out
the "--config=xconfig", since "--config=oldconfig" is what i want then)
> make[3]: Entering directory `/usr/src/kernel-source-2.4.14/kernel'
> make[3]: *** No rule to make target
> `/usr/src/kernel-source-2.4.14/include/linux/modversions.h', needed by
> `sched.o'. Stop.
somehow you missed a "make dep". you could try "make-kpkg clean" and
then a rebuild. without seeing the relevant parts of the build output
i don't really know what went wrong.
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] checkout jdub's (r)evolutionary head
http://www.cse.unsw.edu.au/~gusl/Jeff.jpg tee hee -- - Gus -- SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: unwanted klogd messages
\begin{Darrell Burkey}
> I have two Linux boxes connected to separate Bigpond Direct accounts via
> modems that dial in to the same Telstra router. Anytime the two boxes talk
> to each other I get the following messages on the consoles of both machines:
>
> Dec 29 08:45:26 adminserver kernel: Advised path = 203.48.63.65 ->
> 139.130.12.155, tos 00
> Dec 29 08:45:27 adminserver kernel: Redirect from 139.130.12.129 on ppp0
> about 139.130.12.155 ignored.
>
> This is also filling up the system logs and basically it's very annoying. My
> understanding is that the Cisco router at Telstra is simply advising the
> Linux boxes that there is a direct path between them and that the routing
> tables should be updated to reflect this. ???
yep. ICMP redirect messages. normal and proper.
> 1. Eliminate the cause of the messages.
> I don't have control over the router generating the message so that option
> is out. I also thought that perhaps I could add a static route on each box
> to the other but that doesn't seem to work. I find it a bit confusing as
> each box has a network address plus a local and remote ip address being fed
> to pppd via options. These addresses were assigned by Telstra. Anyway, I've
> tried every combination of routing table entries and I still get the
> messages.
>
> 2. Replace klogd with something like megalog that allows regex matching so I
> can filter out the messages. I've read up on this and it looks like a strong
> possibility but I'd rather eliminate the cause of the messages if possible.
>
> So, anyone have any tricks up their sleeves?
a quick grep of my nearest kernel source (2.2.19), showed:
net/ipv4/route.c:
reject_redirect:
#ifdef CONFIG_IP_ROUTE_VERBOSE
if (IN_DEV_LOG_MARTIANS(in_dev) && net_ratelimit())
printk(KERN_INFO "Redirect from %X/%s to %X ignored."
"Path = %X -> %X, tos %02x\n",
ntohl(old_gw), dev->name, ntohl(new_gw),
ntohl(saddr), ntohl(daddr), tos);
#endif
there's numerous reasons the code jumps to reject_redirect:
(from my quick understanding of the if blocks)
new gateway == old gateway, some multicast stuff, new address is 0.0.0.0,
if the new address isn't in the range of the network cards addresses,
its a unicast interface
i'll let you work out why its rejecting the redirect.
so, to not see these messages, do one of the following:
work out why its being rejected, and avoid that. (will probably involve
reading the relevant section of route.c)
undefine CONFIG_IP_ROUTE_VERBOSE (its a kernel config option, iirc)
make sure the kernel isn't logging KERN_INFO messages to the console
directly:
check /proc/sys/kernel/printk and the relevant section in
Documentation/sysctl/kernel.txt
make sure syslog isn't logging kern.info messages to the console
(/etc/syslog.conf and syslog.conf(5))
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: how to use kernel-package and patch kernel for xfs?
\begin{John Ferlito}
> On Fri, Dec 28, 2001 at 07:12:59PM +1100, Jeff Waugh wrote:
> $ fakeroot make-kpkg --revision=pants.1.0 --added_patches freeswan,mppe configure
> this sets everything up and patches everything
"--append_to_version +freeswan+mppe" can also make things a little clearer
later too..
good for running a "with freeswan" and "without freeswan" kernels, which
would otherwise have the same package name (and thus not be installable
simultaneously)
(a lot of this assumes kernel-package >> potato version)
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Linux Pocketbooks
\begin{Matt -}
> I am wondering if anyone has seen any Linux Pocketbooks
> around recently. I have tried several newsagents and many
> say they are sold out till some time next year (March?!).
i've got a free copy of the advanced pocketbook (debian on the cd) here.
it was sent to slug, i just keep forgetting to remember to bring it :(
i'll try (again) to bring it at the next meeting..
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Sunday Night Linux Challenge!
> On Sun, 23 Dec 2001, Jeff Waugh wrote:
>
> > Holy cow! Krazy Sunday Linux Challenge!
> >
> > Your challenge, should you choose to accept it is:
> >
> > Reboot a linux-mipsel machine using only... a running root bash process.
> >
> > You may not execute any program [1], you may not use the switch [2].
> >
> > - Jeff
> >
> > [1] FATAL: kernel too old
> > [2] Miles and miles away, even by one of those fancy-schmancy car thingies.
\begin{Grant Parnell}
> Hmm... the trick would be to find the warm start boot rom address, or even
> the cold start address, then somehow find where the live kernel image is
> in ram and poke into that a whole pile of machine code instructions to
> jump to the warm start address. To do this you'd be using the built in
> echo command and hopefully sticking stuff into something like /proc/kcore.
>
> Finding the address is going to be a matter of looking up the specs for
> the CPU and hoping like hell they didn't use a bootstrap rom (ie one
> that's banked out when the system's finished with it). EG at reset a 6502
A
> looks up a vector at 0xFFFE as a 16 bit address then does a JMP to that
> address.
>
> You're really going to need another live machine to do some peeking on to
> find where you need to poke stuff.
.. i don't get it.
kill is built into bash..
"kill -INT 1" should be equivalent to ctrl-alt-del on the console, and
shouldn't actually need to spawn another process.
(and whats this "kernel is too old" huff?)
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Offline Email Reading that pre-fetches URL's
\begin{Grant Parnell}
> Don't know how many of you are in the habbit of reading your SLUG email
> offline (which for me is on a 2 hour train trip each day when I haven't
> got other work to do) but it frustrates me when somebody posts about
> something potentially interesting and refers to a URL. I'm not saying you
have a look at wwwoffle. not quite what you asked for, but i think its
exactly what you want.
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Shell scripting challenge (or not)
\begin{[EMAIL PROTECTED]}
> On Fri, Nov 02, 2001 at 05:10:54PM +1100, Andrew Foster wrote:
> > thing.mp3
> > pants.txt
> > yer.rtf
> > others.html
> > These files do exist on my system, except they're scattered throughout a
> > directory tree hierachy, and there's no path info in my text file to say
> > where.
>
> To perform all the deleting in one traversal of the filesystem(s) and
> fork'and'exec a single rm ...
>
> file=filelist args= ; cat $file | while read line
> do
> [[ -z $args ]] && args="-name $line" || args="$args -o -name $line"
> done
> eval "find . $args | xargs rm"
(couldn't help myself ;)
args=-false
while read line; do args="$args -o -name $line"; done < filelist
find . $args -print | xargs rm
or even:
find . -false `sed 's/^/-o -name /' filelist` -print | xargs rm
(of course, if you're using GNU tools, then the above should both be:
find ... -print0 | xargs -0 rm
to make sure you survive paths with whitespace, etc in them)
or if your locatedb is up to scratch:
while read line; do args="$args */$line"; done < filelist
locate $args | xargs rm
or even:
locate `sed 's,^,*/,' filelist` | xargs rm
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: chattr dudes?
\begin{Steve Kowalik}
> chattr is used to change the extended attributes of an inode on a ext2
> partition.
> append only (Very useful if you keep using > instead of >>)
and for syslog files (obviously)
> compressed - I'm not sure why this is here, as I haven't seen it used.
there were some patches to implement it. haven't heard about it in a
while, i guess hard disks got cheaper.
> no dump - Can't use file with dump(8)
my #1 reason for preferring dump/restore over most other backup
methods.
> immutable - File is untouchable and unchangeable.
set it on /bin/login, etc. confound your friends.
> undeletable
never did work out how to use this one. i seem to remember some stuff
going in for it in real early 2.3 (?). iirc, it just means the kernel
keeps a reference on the inode until the filesystem is unmounted or
something.
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: LaTeX problem revisited.
\begin{Bill Bennett}
> \usepackage{pstricks,pstcol,pst-poly,pst-node,multido}
> \PstPolygon[unit=7.5,PolyNbSides=9,PolyOffset=2]
>
> and can report that you'll get a nine-sided "overlapping"
> polygon.
>
> Can some kind soul tell me how to make the outline thicker?
> \fboxrule doesn't work.
iirc, the "standard pstricks" method is the "linewidth" option:
\PstPolygon[unit=7.5,PolyNbSides=9,PolyOffset=2,linewidth=2pt]
> Also, could someone tell me how to fill the polgon with a colour?
again, the pstricks convention is a starred function:
\PstPolygon*[unit=7.5,PolyNbSides=9,PolyOffset=2]
failing that you could maybe make it a path and then fill the
path. see chapter VIII of pst-usr[1-4].ps
personally, i'd just use metapost:
beginfig(1);
numeric sides, radius;
path c;
sides = 9; % number of sides
radius = 2cm; % radius (in TeX units)
c = fullcircle scaled radius;
% add "rotated 20" to the above line if you want
% to rotate the polygon by 20 degrees
% change "draw" to "fill" to get a filled polygon
draw
point 0 of c
for i = 1 upto sides-1:
-- point (length(c)*i/sides) of c
endfor
-- cycle;
endfig;
end;
chuck that in poly.mp (or similar) and run:
mpost poly.mp && mv poly.1 poly.eps
(you've gotta love a language where you can use a for loop in the
middle of a statement..)
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: strip email addresses
\begin{Jeff Waugh}
>
>
> > I have a bunch of xls files I need to get the email addresses from
> > anyone have a quick one liner?
>
> [ Simpler: export to csv and use cut. ]
xls2csv file.xls | cut -d, -f 7
(substitute "7" with the appropriate column)
xls2csv comes with the "catdoc" tools.
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: automounters
\begin{Andre Pang}
> On Mon, Oct 29, 2001 at 05:26:29PM +1100, Jeff Waugh wrote:
>
> > 'tag alles,
> >
> > Gus (and others), which was the automounter you said you preferred on
> > Friday? amd or autofs?
>
> I'm no Gus, but I use autofs at home. It's the more 'modern' out
> of autofs/amd I think, and it also seems to be a lot easier to
> configure than amd :).
ya. autofs-triggered shell scripts are fun (think ldap maps, etc)
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Tar over SSH
\begin{Michael Lake}
> I have a few things echoed to the screen by .bashrc
> I can turn them off but there must be a way to ignore output of ssh before
> root runs the tar program.
then thats a bug in your .bashrc
several suggestions for your .bashrc:
only echo if bash is in "interactive" mode (don't know how you check
that off the top of my head. in tcsh you see if $prompt is defined)
only echo if stdout is a tty (test -t 1)
echo to stderr (echo foo >&2) instead
don't echo anything
the only way around it that i can tell is to put some marker in then
strip off everything before the marker:
ssh user@host 'echo __START-HERE__; tar cvf - /path' | \
sed '1,/^__START-HERE__$/d' | dd of=/dev/mt0
but thats bodgy. fix the real problem, otherwise rsync, etc won't work
either.
note that (GNU) tar can do "tar -f host:/path" directly, using
whatever your "rsh" command ends up running (hopefully ssh). see
tar.info and rmt(8)
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: slug digest, Vol 1 #1325 - 13 msgs
\begin{Kevin Saenz}
> Yes I modified /etc/nsswitch.conf I have ldap before files for
> passwd, group, and shadow. I am sure I have forgotten one???
can you do a simple "ldapsearch gidNumber=1045" ?
what DN did you give the posixGroup entry?
does the "parent" entry exist?
how is nss-ldap configured?
are you using LDAP v2 or v3? (what version of openldap?)
are you running nscd?
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Problems with passing $1 to perl script....
\begin{Jobst Schmalenbach}
> I have written a perl script which has somewhere in the code
>
>$rc = ($content =~ s/$regfrom/$regto/);
>
> $regfrom and $regto are passed to the script from the command line.
> I want to do:
>
> ../bin//sitesed.pl -r 'CommonFooter.cgi\?([a-z]*) /somepath/([a-z]*) blah'
>'SomeNew/$1 someother/$2'
> then it finds all occurences but does not replace it instead I get
>
> .
> SomeNew/$1 someother/$2
> .
>
>
> If I now hardcode the stuff into it, like:
>
>$rc = ($content =~ s/$regfrom/SomeNew$1 someother$2/);
>
> it does exactly what I want, eg:
>
> .
> SomeNew/one someother/two
> .
>
$rc = ($content =~ s/$regfrom/eval "\"$regto\""/e);
don't ask. its too late at night.
(if $regfrom will not change during the script, adding the 'o' option to s///
will make things run faster)
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Interesting ICAAN Quote
\begin{Jon Biddell}
> "What lawyers call "intellectual property" is -- as every Latin student
> knows -- no more than theft from the public domain."
> -- Andy Mueller-Maguhn, newly elected ICANN board member for Europe.
i don't get the "latin student" bit.
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Multithreading vs. Forks
\begin{Ken Foskey}
> Mikolaj J. Habryn wrote:
> >
> > 1) If you think your program requires multithreading, 99% of the time,
> >you're wrong.
> > 2) If you /still/ think your program requires multithreading, see point 1.
> >
> Rubbish, threading is needed in a lot of applications. For example
> web servers or even web query utilities so that they can download
> concurrent parts of the page.
hmm..
of the browsers i currently have installed, links, w3, netscape and
mozilla do that funky "render while downloading" and support
background downloading of files.
links does not link against libpthread, and is (i believe) based on
the lynx code, so afaics it doesn't use threads.
w3 is written in elisp and runs within emacs. emacs does not support
threaded elisp macros, hence w3 certainly does not use threads.
netscape is not linked to libpthread, nor do i see extra process
table entries when running netscape. its possible that netscape
uses a user-space threading emulation (i wouldn't put it past it),
but from what i've seen of netscape behaviour, i don't believe thats
so.
mozilla links against libpthread. a quick look at the process table
when running it strongly implies its using threads.
so:
1 out of the 4 "sophisticated" web browsers i have installed uses threads.
this means that threads are certainly not necessary, and the decision is
probably based on something else.
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Re: Hotplug and cached Disk writes
\begin{Jill Rowling}
> I think the short timeout automounter sounds like a good idea. Most 'doze
> users look at the flashy light on the disk before ejecting it.
> So what are the mount options for auto mount when write,
> auto-umount-after-write, assuming (say) vfat?
you need to setup an automounter, mount(8) can't do it alone.
autofs is better (imo) than amd, so we'll use that:
(this assumes debian autofs package conventions. shouldn't be
hard to change the paths to suit tho)
mkdir /var/autofs/removable
/etc/auto.master:
# everything below /var/autofs/removable has a timeout of 3 seconds
/var/autofs/removable /etc/auto.removable timeout=3
/etc/auto.removable:
floppy -fstype=auto,sync,nosuid,nodev,gid=floppy :/dev/fd0
cdrom -fstype=iso9660,ro,nodev,nosuid :/dev/cdrom
then (if you wish) symlink:
/floppy -> /var/autofs/removable/floppy
/cdrom -> /var/autofs/removable/cdrom
ensure /etc/filesystems lists whatever filesystems you want to try for
fstype "auto" (this is a mount(8) trick, nothing to do with automount):
vfat
ext2
minix
msdos
autofs only unmounts if the disk is currently unused (ie: if a normal
umount(8) would succeed). so just waiting for the light to go off
isn't necessarily good enough.
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
[SLUG] Re: Document Management Systems
\begin{Glen Turner}
> Rick Moen wrote:
> > Subversion rocks. I've finally started using it (cautiously), now that
> > it's passed the self-hosting milestone. And boy will I be glad to ditch
> > CVS.
>
> Ironically we'll probably need a cvs/DeltaV gateway before
> we can finally put CVS to rest at the crossroads at midnight
> with a stake through its black, black heart. Programmers are
> a conservative bunch, although they don't like to think of
> themselves as such, and will want to continue to use their
> burned-in "cvs" commands.
from what i've seen of subversion, it has pretty much identical command
line commands. just s/^cvs/svn/g
this will also mean that most cvs frontends (emacs ;) should be able to be
easily ported.
subversion looks really nice.
--
- Gus
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
