Re: ati fglrx and xorg 7.0.0
Ivan Glushkov wrote: Hi all, I tried to install ati driver after upgrading my Debian/etch to xorg 7.0.0. There are three ways to do that as fas I know: [snipped] Flavio posted on the fglrx list that a new 8.24 version compatible with X.org 7.0 is available for testing. http://www.stanchina.net/~flavio/debian-official/fglrx-driver.html A number of people have reported positive results. Regards, ST -- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Apache::ASP problem
Hello,
I'm testing the Apache::ASP on a debian/sid box.
System info:
apache2-mpm-worker 2.0.55-4,
libapache2-mod-perl2 2.0.2-2,
libapache-asp-perl 2.59-2.
By itself, apache2+mod_perl seems to be working well.
However, I got error messages like these
quote
[Sun Apr 16 18:12:02 2006] [error] [client 127.0.0.1] Can't locate object method
get via package APR::Table at /usr/share/perl5/Apache/ASP.pm line 2016.\n at
/usr/share/perl5/Apache/ASP.pm line
2016\n\tApache::ASP::get_dir_config('APR::Table=HASH(0x81f3f20)', 'Global')
called at /usr/share/perl5/Apache/ASP.pm line
275\n\tApache::ASP::new('Apache::ASP', 'Apache2::RequestRec=SCALAR(0x81f3ef0)',
'/var/www/asp/eg/index.html') called at /usr/share/perl5/Apache/ASP.pm line
183\n\tApache::ASP::handler('Apache2::RequestRec=SCALAR(0x81f3ef0)') called at
-e line 0\n\teval {...} called at -e line 0\n
/quote
in the error.log of apache when I tried out the examples that came with in
libapache-asp-perl.
The best information I can get seems to be this FAQ from
http://www.apache-asp.org/faq.html#Error%3A%20no%20reb1d13fcf, which points to a
mod_perl setup error and their advice is to compile a static apache+mod_perl.
But I'm not quite sure and I'd be grateful for any ideas.
Thanks in advance,
ST
--
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bash, cat and EOF problem
Hello, I ran into this problem which I cannot figure out. I must have missed something. I'm using sid with bash version 3.1.5(1). script1 cat EOF outfile Some words EOF /script1 works but when it is embedded inside an if-then-fi construct like, script2 AAA= if [ -z $AAA ]; then cat EOF outfile Some words EOF fi /script2 it fails with error message script2: line 7: syntax error: unexpected end of file What have I missed? Regards, ST -- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Bash, cat and EOF problem
Dave Thayer wrote: On Thu, Mar 02, 2006 at 02:41:08PM +0800, LUK ShunTim wrote: script1 cat EOF outfile Some words EOF /script1 works but when it is embedded inside an if-then-fi construct like, script2 AAA= if [ -z $AAA ]; then cat EOF outfile Some words EOF fi /script2 it fails with error message script2: line 7: syntax error: unexpected end of file What have I missed? It's not the if-then-fi causing your problem, but the indentation within. Lose the leading spaces in the here text (unless you want your outfile to containSome words rather than Some words) and the limit string (EOF). See http://www.tldp.org/LDP/abs/html/here-docs.html#INDENTEDLS HTH dt Hello Dave and Andrew, Thanks very much for the useful information. Regards, ST -- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How to install a new kernel when your / partition is neary full?
Crispin Wellington wrote: On Wed, 2002-03-20 at 15:15, LUK ShunTim wrote: [snipped] Thanks for the idea and also to all people who responded. If that doesn't make absolute sense, then don't do it. Maybe du -sh to find whats taking up all the space. This is the result: /bin 3.0 M /boot 1.3 M /dev 67 K /etc 6.7 M /lib 20 M /lost+found 12 K /root 313 K /sbin 3.3 M / 66 K The whole thing adds up to ~35 M. However, df -h gives 93 M for the / partition space and 83 M as used. Probably lots of small files each eating up 1K (the min. file size in ext2, IIRC) A 100 M / partition is not large enough for present day kernels (if you'd like to keep the old one while installing a new one) may be the approriate conclusion? Good day, ST --
How to install a new kernel when your / partition is neary full?
Hi, I am now on a 2.2.18 kernel (in sid already) and I'd like to upgrade to a 2.4.x but unfortunately I have a small (~100 Mb) / partition when I started out and it's already quite full. Upgrading via the usual manner with apt-get and friends didn't do the job (and screwed things up quite a bit as a matter fact after unpacking). I have /usr, /var, /home etc mounted on different partitions. One obvious way would be to shrink other partitions and grow /, but I don't want to do that, at least not yet before taking a look at other (better) options. Therefore, I am just wondering whether there is any suggestion as to doing it *elegantly*. Thanks a lot and good day, ST --
Re: How to install a new kernel when your / partition is neary f
Sean 'Shaleh' Perry wrote: I have /usr, /var, /home etc mounted on different partitions. One obvious way would be to shrink other partitions and grow /, but I don't want to do that, at least not yet before taking a look at other (better) options. Therefore, I am just wondering whether there is any suggestion as to doing it *elegantly*. if /usr, var and home are on separate partitions, why is a 100mb / not sufficient? That's what I really want to find out :-). A couple of years ago, even a 50 Mb / partition seemed OK. IIRC, with debian 1.3 it was just a 3 or 4 floppy install. From what I gather, when you install a new kernel image using apt-get, the old one is kept (which is a sensible decision) and present-day kernels (plus the modules etc) take up quite a lot of space after unpacking and I have to install the pcmcia modules at the same time as well. I'm not my linux box now so I cannot provide more factual information at the current moment. Good day, ST --
Winmodem in Acer Aspire 6060?
Dear all, I'd be grateful if someone who owns/has experience with the Acer Aspire 6060 tell me whether it comes with a winmodem? I'm offered a second hand one at a good price and would like to strip Windows off for debian. If it is, then that would be useless. I presume that winmodem is still not supported by Linux. I'm not very up-to-date, though. Or has the situation changed? Regards, ST
No command line prompt for root
Hi, Something strange happened to my Debian system. I can login as an ordinary user and it works normally but when I login as root, it printed out the motd on the screen and said that root has been logged in. A 'who' from another user confirms it. It even told me no mail for root but I don't get the command line prompt # and so I can do nothing as root. Su'ing from an ordinary user meets with the same failure. A Ctrl-C will give back the login: prompt. I've used the rescue disk to boot up and checked the .bash_profile, .bashrc and /etc/passwd files and they are all OK. (I upgraded to slink, kernel 2.0.35 from hamm a few days ago.) Any ideas as to what happened and what is the remedy will be very much appreciated. ST
Re: No command line prompt for root
Nikolai Andreyevich Luzan wrote: Your version of sysklogd is? if it is 1.3-29 then you have the bugged up version. [snipped] That indeed is the case. Thanks. May be that's where the fun is -- you never know what interesting things will come up. And you always got somebody willing to help. ST
Re: Help need on man pages
H C Pumphrey wrote: On Thu, 8 Oct 1998, Wilson Tuma wrote: Plase could you give the the files I have to download to be able to install linux man pages so I can get help by typing say man rm or man cp You need the manpages package (manpages_1.19-1.deb) , the man-dbpackage (man-db_2.3.10-65.deb), the libc6 package (libc6_2.0.7t-1.deb), [1] the libdb2package (libdb2_2.3.16-7.deb), the groff package (groff_1.11a-2.deb), andthe libg++272 package (libg++272_2.7.2.8-0.1.deb) Now you begin to see why the slightly counter-intuitive decision to not have the man pages in the base distribution was taken: it would have made it bigger by several floppies. To get some help before you install the man pages, you can try, for example, rm --help Regards, ST --
Re: Can't drag and drop in Netscape
Blazej Sawionek wrote: LUK ShunTim wrote: KDE and Netscape 4.5bPR1. Sorry I don't uderstand what you are trying to do, but 4.5bPR2 is already available - maybe that can help you? Blazej, May be I have not made myself very clear. What I would like to do is something like this. After I read a web page in Netscape, I would like to bookmark it. There is a Location icon in the Netscape browser window to the left of the URL dialog box and to insert a bookmark of the page in the bookmarks file, you can (or are supposed to) drag it into the bookmark file open in another window. What I see is that there is some animation of the dragging process but no bookmark entry was added to the bookmark file. So I just wonder may be I have missed to install some library to make this drag and drop work. Regards, ST
Re: lilo and hdc
M.C. Vernon wrote: Dear all, A friend of mine has a (SUSE) install on hdc, but wants lilo (or similar) to go on hda (which is a dos/w95 partition)? is there a howto/etc for this sort of thing? Matthew -- Elen sila lumenn' omentielvo Steward of the Cambridge Tolkien Society Selwyn College Computer Support http://www.geocities.com/Area51/Chamber/8841/ http://www.cam.ac.uk/CambUniv/Societies/tolkien/ http://pick.sel.cam.ac.uk/ -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null Yes, lilo comes with a very detailed user guide. Just check /usr/doc/lilo. And for a start, I would recommend bootmenu. You can get it from Sunsite Linux under ../system/boot/dualboot/bootmenu. It is quite easy to install and it will tell you how to configure lilo. Hope this helps, ST --
Can't drag and drop in Netscape
Hi, When I wanted to bookmark a page by dragging the location icon into the bookmark file, it just wouldn't work. I'm using hamm, kernel 2.0.34, KDE and Netscape 4.5bPR1. Have I missed to install anything? Regards, ST --
Compiling apache
Hi, I have to add some database support so I get the source tar ball from apache and re-compile it. With some minor problems, I got it worked. I found the that the size of the binary file was much larger that the original one in the distribution. I suspect that some modules are dynamically linked during run time. How can I do the same with my own compilation? Regards, ST --
Re: support for AGP
[EMAIL PROTECTED] wrote: Does Linux (or, more specifically, XFree86) support AGP video boards? The specific one in question is an ASUSTEK V3000 or something of the sort. Andrew Tarr I don't know about V3000 but I have an Asustek V264GT AGP video card and it is using the ATI Rage 3D 64 bit chip. The ATI_Mach64 X-server works fine in my Linux box. May be you can check out the V3000 card information in the Asus site (www.asus.com). Hope this helps, ST --
Re: TEXINPUTS ?
Remo Badii wrote:
Dear Debian TeX/LaTeX users,
I have just installed the revtex library under $HOME/tex/sty/revtex, as
in my previous Linux system (2 years old, S.u.S.E. 4.2), under Debian
2.0. Although I have set the variable TEXINPUTS to include
$HOME/tex/sty and $HOME/tex/sty/revtex, the system finds a file
/usr/lib/texmf/tex/latex/misc/revtex.cls and stops within it at the line
\input{revtex.sty}, saying that it does not find it (correct, since it
is not in that directory): why doesn't latex find my revtex directory
first? I did not have this problem with the old Linux system (teTeX, as
in Debian).
Running texhash may solve the problem if the package you install is in
the
TeX tree hierarchy, which I think tetex is following.
Regards,
ST
--
GPL'd/free driver for Ensoniq AudioPCI sound card
Hi, I would like to know where can I get a GPL'd or free driver for the Ensoniq AudioPCI sound card for x86 architecture. I've checked out the OSS/Free page and it seems that their free version does not support this (and other PCI) card. Regards, ST --
Re: New files from ftp?
Remo Badii wrote: After installing Debian 2.0 last week from CDs, I am now trying to install Netscape 4, pgplot, and tgif using dselect and ftp from the ftp.denian.org site. I also got the message getting ... xfntpex_3.3.2.3 ... xserver-svga Is this because there are new versions of these files? Yes. Because the packages are being constantly updated. That's why Debian is good. I only wanted the three I mentioned above and did not select anythign else. Should I stop the download process? Because dselect selects these for you. You have to de-select them if you don't like to install these newer versions. I think there's a dselect tutorial at the Debian site. Regards, ST --
Re: 3 com(USR) modem
Michael B. Taylor wrote: [snipped] You will need ppp, net_std, and net_base packages. I think ppp is now ppp_pam or something. You will need to set up your ppp options and your chat script. I think there is now a package that automates this task, but I cant remember what it is called. I think you are talking about wvdial. It is available as a deb package. Just run dselect using ftp to get it from one of the mirrors Regards, ST --
Re: Disabling virtual desktops in X
David Wright wrote: On Fri, 28 Aug 1998, LUK ShunTim wrote: Thanks. I got it. This is what I observed. I change the Display subsection of my XF86Config to this: Subsection Display Depth 8 Modes 1024x768 800x600 640x480 Virtual 1024 768 ViewPort 0 0 EndSubsection A frequent reason for the type of question you originally asked is when you have the modes in ascending order. X starts with the first (now lowest) mode, and newbies may not know about C-A-+. I got desktop = screen (that's what I want) as I start up X but when I change to a lower resolution mode by Crtl+Alt+Keypad+ I have a desktop larger than my screen. I found out that while Modes can accept several arguments, Virtual can only accept one pair. I've tried several Display subsections but then I can no longer use Crtl+Alt+Keypad+ to change resolution. What would you expect to see? If the desktop shrank, all the applications would have to be repainted on the smaller piece of real estate. As it is, nothing on the desktop changes, not even the pointer position (though the server has to choose the visible subsection in such a way that the pointer remains on-screen). So we can't have the best of both worlds. [...] Yes you can. You just treat the size of your desktop like the colour depth, and run an X server for each kind. (There's a package to allow one to cut and paste between different server instances.) Where can I get it to have a try? Regards, ST
Re: Disabling virtual desktops in X
Ole J. Tetlie wrote: Virtual resolution is a feature of XFree86. It allows you to work on a larger space than the actual resolution of your display. It works almost exactly as if you had a display with better resolution. I don't think you can really disable it, but you can avoid noticing it by setting the virtual resolution to be the same as your actual resolution. Just put Virtual xdim ydim in you Display entry in XF86Config. (If you don't know the actual dimension of your display you can use xdpyinfo to find out). Virtual desktops if a feature offered my many window managers to allow you several different workspaces with the set resolution. twm doesn't offer this, so I guess you must be talking about the virtual resoultion. | Now back to my problem. What I see after I started X is the desktop | which is *larger* than the size of the screen and I have to move the | mouse a long distance to get to other parts of the desktop. I experience | this phenomenon in every (if I remembered correctly) window manager I | tried: twm, fvwm95, kwm(kde) and I have a small real desktop :) | What I'd like to have is a desktop that has the same size as my screen. Yup. Just set the virtual resolution to the same as the real. If you delete the virtual entry (as you did), it will be set to accomodate all the modelines, which may not be what you want. -- .elOle. Thanks. I got it. This is what I observed. I change the Display subsection of my XF86Config to this: Subsection Display Depth 8 Modes 1024x768 800x600 640x480 Virtual 1024 768 ViewPort 0 0 EndSubsection I got desktop = screen (that's what I want) as I start up X but when I change to a lower resolution mode by Crtl+Alt+Keypad+ I have a desktop larger than my screen. I found out that while Modes can accept several arguments, Virtual can only accept one pair. I've tried several Display subsections but then I can no longer use Crtl+Alt+Keypad+ to change resolution. So we can't have the best of both worlds. Is this observation correct? Regards, ST --
Disabling virtual desktops in X
I'm just wondering whether enabling virtual resolution is compiled default in X servers. Even if I commented out all Virtual options in my XF86Config file, I still get the virtual desktops. I'm using the XF86_Mach64 server. Any hint if I'd like to disable it? Regards, ST --
Re: Disabling virtual desktops in X
Ole J. Tetlie wrote: *-LUK ShunTim [EMAIL PROTECTED] | | I'm just wondering whether enabling virtual resolution is compiled | default in X servers. Even if I commented out all Virtual options in | my XF86Config file, I still get the virtual desktops. I'm using the | XF86_Mach64 server. | | Any hint if I'd like to disable it? I'm a little unsure what you mean here. Virtual resolution is set in the XF86Config file. It does however seem that you problem is with virtual desktops. This is something that is under the control of the window manager. You need to tell us which window manager you use, if it's really virtual desktops you refer to. -- Ole, Please accept my apology if I have confused virtual resolution with virtual desktop and didn't make myself clear. Please explain the difference. Now back to my problem. What I see after I started X is the desktop which is *larger* than the size of the screen and I have to move the mouse a long distance to get to other parts of the desktop. I experience this phenomenon in every (if I remembered correctly) window manager I tried: twm, fvwm95, kwm(kde) and I have a small real desktop :) What I'd like to have is a desktop that has the same size as my screen. Regards, ST --
Re: kde missing icons
BG Lim wrote: I just installed kde-1.0 from ftp.kde.org. I believe that it is the same release 1.0-1 version found in slink. My problem is that the toolbar icons in kfm and other progs that use those icons don't show up. Instead, you get fuzzy pics where the icons are supposed to be. However the buttons themselves work find, only thing is you have to sort of mentally remember what they do! Below is a reply to a similar question I got from Noah L. Meyerhans [EMAIL PROTECTED] posted to this list a few days ago. Hmm...this actually looks like a bug in the packaging KDE. The toolbar pixmaps live in /usr/share/toolbar, and the problem seems to be that they're installed as part of the kdelibs0g-dev package. If the pixmaps are loaded at runtime (as they obviously are), then they should not be part of the dev package! If you install that package, you'll get the pixmaps. I'm gonna send a mail to the maintainer and inform him of the problem. Hope this helps, ST --
Re: Missing toolbar icons on KDE for Debian Linux
Noah L. Meyerhans wrote: -BEGIN PGP SIGNED MESSAGE- On Sat, 8 Aug 1998, LUK ShunTim wrote: I installed KDE 1.0 on my Debian Linux 2.0 with the .deb packages from ftp.us.kde.org and it worked find; except one thing: I get blurred images instead of nice looking icons (i.e., the Up, Home, ... buttons) on the toolbar in kfm. This problem occurs in some other applications, too. Hmm...this actually looks like a bug in the packaging KDE. The toolbar pixmaps live in /usr/share/toolbar, and the problem seems to be that they're installed as part of the kdelibs0g-dev package. If the pixmaps are loaded at runtime (as they obviously are), then they should not be part of the dev package! If you install that package, you'll get the pixmaps. I'm gonna send a mail to the maintainer and inform him of the problem. It is as you say. Thanks. ST --
Re: Netscape: Can't open display
Kent West wrote: Whenever I try to run Netscape it returns an error to the effect of Can't open display Any clues? (Please be as specific as you can because I don't know what I'm doing.) Thanks. Kent West [EMAIL PROTECTED] -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null Netscape must be run in an X environment. Therefore, assuming your system is set up correctly and your are at the command prompt, try startx If you succeed in getting your X environment, then you will see the xterm window (looks like the command shell), then you can try netscape to start Netscape. Hopes this helps, ST --
Missing toolbar icons on KDE for Debian Linux
Hi, I installed KDE 1.0 on my Debian Linux 2.0 with the .deb packages from ftp.us.kde.org and it worked find; except one thing: I get blurred images instead of nice looking icons (i.e., the Up, Home, ... buttons) on the toolbar in kfm. This problem occurs in some other applications, too. The .xsession-errors file showed the following message: KToolbarButton: pixmap is empty, perhaps some missing file I have tried copying .xpm files around (I don't know the exact file names for these icons) in the /usr/share/app/.. directories but the problem persists. Has anybody come across this problem? Are the icons file really missing from the packages? Or do I have to set some environment variables? Have a nice day, ST -- -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: Simple requests
C. Hatton Humphrey wrote: Hey there, I've gotten Debian installed and running, but I have a few problems I need to get taken care of. First off, I accidentally uninstalled dpkg-ftp, but managed to use dftp to get dkpg-http, but it asks for a proxy address or URL and I have no clue what it is. I think you can use ordinary anonymous ftp to get dpkg-ftp (or any package you want) from a Debian distribution site and then just install it by dpkg -i WhateverPackage.deb Also, my e-mail and news are messed up something bad. I've installed smail and it's working, I can send e-mail internally but once I get to sending e-mail to the rest of the world it goes kaplooie and I can't get anything. I am also getting permission denied errors on /var/spool/mail when I login. Also, what mystical trick to I use to get my backspace key going in XWindows, for some odd reason I can use CTRLH to do a backspace, but the actual delete and backspace keys don't work. Include this in your .xsession or .xinitrc file # Makes BackSpace do backpace xmodmap -e keycode 22 = BackSpace Thanks in advance, Hatton Humphrey - [EMAIL PROTECTED] -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null Hope this helps, ST -- -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: dselect problem...
David Puryear wrote: Hi all, With deselect, I have the Access with ftp with right directory and all then I pick update which download new packages files from dists/unstable/main, non-free, and contrib like the way it should. But when I do Install I get this messages: Processing status file... Processing Package files... Couldn't find packages file for dists/unstable/main distribution (re-run Update) Couldn't find packages file for dists/unstable/contrib distribution (re-run Update) Couldn't find packages file for dists/unstable/non-free distribution (re-run Update) Constructing list of files to get... Approximate total space required: 0k Available space in /debian: 109334k Nothing to get. I don't know why dselect is not see he packages file, which is in /var/lib/dpkg/methods/ftp/* with names Packages.dists_unstable_contrib, Packages.dists_unstable_main, and Packages.dsts_unstble_non-free. I can't figure out why this is happening since I have used same steps to update in another machine without this problem. I got this problem as well when I upgrade from Debian 1.2 to 1.3. I overcame it by turning off the passive FTP tranfer mode when selecting the access method. Hope it works for you as well. ST -- +---+ | LUK ShunTim | | | | Civil and Structural Engineering Department | | Hong Kong Polytechnic University | | Kowloon | | Hong Kong| | | | http://www.cse.polyu.edu.hk/~stluk/ | | email: [EMAIL PROTECTED] | | phone: (852) 27666064| | fax: (852) 23346389| +---+ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Installation problem: unable to open an initial console
Hi all, Please point me to the location (and forgive me) if this question has already been asnwered. (I have tried to search the mailing list archive but apparently it is down.) My problem is like this: I recently upgraded from Debian 1.2 to 1.3 and it worked fine. Yesterday there was a power failure and I try to reboot using the Boot Disk created during the installation but the system just kept on rebooting itself. I then use the Rescue Disk and boot using the option rescue root=/dev/hda5 (I had the / partition on an extended logical partition named hda5.) and the message unable to open an initial console appeared. I then tried the default boot option in the Rescus Disk and I was able to mount my previous partitions on the directory /target. However, I cannot do anything else. (I would like, at least, to save my data files!) There does not seem to be a hardware problem because I get the virtual consoles (1 to 4) from the Rescue Disk boot. I would be grateful for your help. Regards, --ST +---+ | LUK ShunTim | | | | Civil and Structural Engineering Department | | Hong Kong Polytechnic University | | Kowloon | | Hong Kong| | | | http://www.cse.polyu.edu.hk/~stluk/ | | email: [EMAIL PROTECTED] | | phone: (852) 27666064| | fax: (852) 23346389| +---+ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
