Re: server socket directory has suspicious ownership, aborting.
On Sat, Jun 24, 2000 at 11:40:15AM +0800, Destrius wrote: Or rather: # X(or startx, initx, etc.) X: server socket directory has suspicious ownership, aborting. I suddenly got the above error message last night. I don't recall doing anything to my X installation at all. But you upgraded your box, right? Anybody know what's wrong? I have no idea where the server socket directory is. Set the group ownership of /tmp/.X11-unix to root. Worked for me... -- Graham Ashton coms.com: 07080 81
Scanning emails for viruses (exim)
Can anybody help me to use exim to filter incoming emails for viruses, using a virus scanning package such as McAffee's VirusScan? Any success stories, or alternative techniques, would be helpful - I'm still at the how should I go about this, and what should I get hold of? stage. I've seen the question/answer in the exim FAQ[1], but haven't the first idea about what ought to go in the script it mentions. Thanks. [1] http://www.exim.org/FAQ.html#SEC182 -- Graham Ashton
Re: New modules.conf + sound...
I'm not on the debian list at the moment, but have just been searching the archives. I noticed your messages which looked like they would solve my (exact same) problem [potato + 2.2.13 + all updates]; -- In article [EMAIL PROTECTED] you wrote: /lib/modules/2.2.10/misc/sound.o: invalid parameter parm_io After a quick search, I have not yet found where sound.o is handed a parm_io parameter (I could just be totally misinterpreting the error). the parm_io refers to the io=0x220 in your modules options alias sound sb options sb io=0x220 irq=5 dma=0 dma16=0 mpu_io=0x330 ahhh. it seems the new version of modutils now tries to pass io=0x220 directly to sound.o (which it doesn't directly support) Try commenting out the alias #alias sound sb and running update-modules, to see if the module loads -- I removed the alias as suggested, and did modprobe sound. No errors. So I did modprobe sb too and, lo and behold, it worked (the sb options were still listed in my conf.modules). I mention it because Chris reported failure. And I'm now left wondering how I should get the sb module loaded automatically (the proper way) at boot time. Thanks for the info... Copied to debian-user in the hope that it'll accept my post... -- Graham
Re: wterm
On Monday 07 June, Pedro Bastos wrote: hello everybody :) where can i find wterm in .deb format ? cause i tried to install it in .tar.gz format, and the compile process just failed. not sure, but it compiled fine for me on slink. what were the errors (don't reply to me, I might not understand it, post it to the list). you're probably just missing some header files, or something. -- Graham
Re: AIC7xxx not recognised
On Sunday 30 May, Roy Coates wrote: Hi, I'm having a real problem getting a custom kernel to recognise an Adaptec 2940UW-Pro (AIC7xxx) controller on bootup. The kernel (2.0.36) has been configured with only the AIC7xxx driver as an internal module - no other scsi drivers have been selected. On bootup it simply refuses to recognise the scsi card, causes a kernel panic (unable to mount root FS) and I'm back to the drawing board. is your root file system on the SCSI disk? if so, you've probably got a chicken and egg problem. if the module that tells the kernel how to access the SCSI module is only accessible via the SCSI disk, how is it supposed to load it? try compiling support for the SCSI card into the kernel proper, and see what happens. I did the same thing myself once - rather embarrassing! -- Graham
Re: Re: about Linux Books questions ?
On Tuesday 25 May, David B.Teague wrote: Q1. O'reilly Corp have a book Running Linux, 2/e Can teach and help us know Debia/Linux structure and detail information ? Running Linux just has commands. Useful to remind you about command syntax. No structure of Debian, little detail. I think you're confused. Linux in a nutshell is basically commands and a terse explanation of a few other things - handy as a quick reference. Running Linux is a quite different kettle of fish. I don't have it, but I would have it if I didn't think that I already know much of it's content, or where to find out more online. If it had been around when I'd been a newbie I'd definitely have bought it, and I'd have been glad I did. -- Graham
Re: strange msg in bootup
On Friday 21 May, Dan Willard wrote: The route command now requires a netmask. Just add one in to both of the route commands and it should go away. if you read the 2.2 docs in /usr/src/linux/Changes (or whatever it's called) then you should find a bit that tells you that routes for local interfaces are added automatically by the kernel. comment out the route... lines in /etc/init.d/network and the problem should go away. Following is a listing of my /etc/rcS.d/S40network file, hope it will help: unusually, I was right - they're uncommented - stick a # infront of the two route... lines (S40network is just a link to the init.d/network file by the way). #! /bin/sh ifconfig lo 127.0.0.1 route add -net 127.0.0.0 IPADDR=144.122.246.42 NETMASK=255.255.252.0 NETWORK=144.122.244.0 BROADCAST=144.122.247.255 GATEWAY=144.122.246.1 ifconfig eth0 ${IPADDR} netmask ${NETMASK} broadcast ${BROADCAST} route add -net ${NETWORK} [ ${GATEWAY} ] route add default gw ${GATEWAY} metric 1 -- Graham
Re: gtk-themes problems
On Thursday 20 May, Martin Bialasinski wrote: GA but GIMP still refuses to respond to the GTK themes - do I need to GA recompile GIMP to make it work? Check dpkg -s gimp. Themes don't work for programms compiled for libgtk1.0 or libgtk1.1 earlier than 1.1.16 or such. okay, I did, and it appears to be compiled against gtk 1.0.6; shandy% dpkg -s gimp [snip] Depends: aalib1 (= 1.2-7), libc6 (= 2.0.7u), libgimp1 (= 1.0.2-2), libgpmg1 (= 1.14-3), libgtk1 (= 1:1.0.6-2), libjpeg62, libmpeg1, libncurses4, libpng2, slang1 ( 1.3), slang1 ( 1.2.2-0), xlib6g (= 3.3-5), xpm4g (= 3.4j-0), zlib1g (= 1:1.1.3) I've installed the version of gimp that I found in potato, and now gtk themes work perfectly in gimp. Thanks for the tip. shandy% dpkg -s gimp [snip] Depends: aalib1 (= 1.2-7), libc6, libgimp1, libglib1.2 (= 1.2.0), libgpmg1 (= 1.14-3), libgtk1.2 (= 1.2.0), libjpeg62, libmpeg1, libncurses4, libpng2, slang1 ( 1.3), slang1 ( 1.2.2-0), xlib6g (= 3.3-5), xpm4g (= 3.4j-0), zlib1g (= 1:1.1.3) -- Graham
Re: gtk-themes problems
On Wednesday 19 May, Martin Bialasinski wrote: GA I've installed gtk-themes-0.5 from source, but can't get it to GA work properly. Try the gtk-engines-* packages available at deb http://www.debian.org/~jim/debian-gtk-gnome/gnome-stage-slink unstable main brilliant - it works a treat, thanks. I've installed GNOME from there too - rather swanky. but GIMP still refuses to respond to the GTK themes - do I need to recompile GIMP to make it work? -- Graham
gtk-themes problems
I've installed gtk-themes-0.5 from source, but can't get it to work properly. plain themes work fine but pixmap themes don't. I tried using the BeOS theme by copying /usr/local/share/themes/BeOS/gtk/gtkrc to ~/.gtkrc, but this is what happens when I fire up a GTK app; Gtk-WARNING **: /usr/lib/gtk/themes/engines/libpixmap.so: undefined symbol: gdk_imlib_set_cache_info Segmentation fault Any ideas? Any themes that do work don't appear to work with the Gimp either - any ideas? I'm running slink, and Gimp was installed from the 1.0.2-3 package. Thanks. -- Graham
install floppies: digging around
I'd like to check the versions of some of the modules (e.g. the Tulip NIC driver) used on the slink installation floppies - can anybody tell me which files to dig around in, and on which disk? -- Graham
ePerl + Apache problem
I just installed mod_perl and ePerl from slink today, and am having problems configuring ePerl (Apache is also from slink). I have added the following to access.conf; PerlModule Apache::ePerl Directory /var/www/nelson Files *.epl Options +ExecCGI SetHandler perl-script PerlHandler Apache::ePerl /Files /Directory Perl $Apache::ePerl::Config-{'BeginDelimiter'} = '?epl'; $Apache::ePerl::Config-{'EndDelimiter'} = '?'; $Apache::ePerl::Config-{'CaseDelimiters'} = 0; $Apache::ePerl::Config-{'ConvertEntities'} = 1; /Perl as suggested by the Apache::ePerl docs, and it works some of the time. If I comment out the PerlModule... line and the Perl.../Perl section then it works all of the time, otherwise apache dies almost immediately after I've started it up with /etc/init.d/apache start, or /etc/init.d/apache reload. Nothing of any relevance appears in /var/log/apache/error.log. Any ideas? This is driving me nuts! Thanks. -- Graham
Re: debian-user-digest Digest V99 #802
On Tuesday 04 May, Tommy Malloy wrote: Wouldn't it be nice if I could tell the following to everyone to type command keyword or command string, and a list of applications related to that keyword or string would be provided. try this; % apropos foo where foo is your command or string. the apropos command is part of the man-db package. -- Graham
Re: proc line in fstab
On Thursday 22 April, Angus Claydon wrote: Could anyone who's online just now let me know what the proc line should look like in /etc/fstab ?. Unfortunately I messed it up and none of modules are loading correctly. like this; proc/proc proc defaults 0 0 -- Graham
2940U2W drivers in slink?
Just a quickie - can anybody verify that the slink boot floppies contain drivers for the adaptec 2940U2W scsi card? I don't want to go out and buy 6 boxes that I can't install debian on! I've been reading on dejanews that you need the 2.2 kernel to get a recent driver for the U2W... Thanks. -- Graham
Re: 2940U2W drivers in slink?
On Thursday 22 April, Gary L. Hennigan wrote: Graham Ashton [EMAIL PROTECTED] writes: | Just a quickie - can anybody verify that the slink boot floppies contain | drivers for the adaptec 2940U2W scsi card? It might, but you'd be better off getting a copy of the boot disks with a more current version of the aic7xxx drivers. You can pick them up at: http://www.debian.org/~adric/aic7xxx superb. thanks a lot (super speedy response too). -- Graham
file size limit on sparc
very quick question - am I right in thinking that if I install slink on a sparc I won't be limited to 2GB files as I am on a 32bit intel box? -- Graham
Re: Kernel Upgrade: Why?
On Tuesday 20 April, Richard Harran wrote: I am currently running kernel 2.0.36 in my slink system. Would there be any advantages to upgrading my kernel? If so what version would be a good (stable) version to which to upgrade? Are there any good reasons not to upgrade? I'm no expert, but I think it depends quite a bit on what your hardware is, and what you do with it. do you need a really fast TCP/IP stack for example? do you have multiple processors? certain things in 2.2.x are vastly improved, but I imagine that plenty of others are pretty much the same. I've been running 2.2.6 here without any problems on all my boxes. 2.2.4 was also free from problems for me (didn't try 2.2.5). issues with slink are explained here; http://www.uk.debian.org/releases/stable/running-kernel-2.2 I'd do it if I was you. I thought the speedup was noticeable, but I'm running SMP. Would I see a performance increase if I made sure my linux partition was on the outside of my disk? Would I be even better putting my swap partition as far out as possible? the partitioning HOWTO talks about this; http://webdocs.essex.ac.uk/LDP/HOWTO/mini/Partition-3.html#ss3.3 go on, do it - you know you want to... -- Graham
Re: SMBMOUNT won't
On Thursday 15 April, Alec Smith wrote: I'm running kernel 2.2.5 with smbfsx installed from the deb... I get the attached error... Any tips would be appreciated. you need to recompile smbmount with 2.2.x kernels. I've never done it, but have read that it needs doing. -- Graham
Re: compiling kernel
On Wednesday 14 April, Pollywog wrote: If I need to upgrade a kernel on a Debian system, can I do it in the same way as it is done with other distributions or is there some obscure Debian way to do this? you can do it the standard way (and I did it this way for months), or you can follow the advice of many on this list and try the make-kpkg utility. I now use make-kpkg every time. I don't really understand that much of what make-kpkg is capable of, but I use it to make .deb kernel packages (which I can then install on several boxes without needing to faff around copying all sorts of files around). I do something like this; % cd /usr/src/linux % make-kpkg clean % make xconfig % make-kpkg --revision mykernel.1 kernel_image It then makes a package in /usr/src. Job's a good'un... -- Graham
Re: compiling kernel
On Wednesday 14 April, Pollywog wrote: % make-kpkg --revision mykernel.1 kernel_image I am guessing I still need to do: make modules make modules_install cp System.map /boot cp arch/i386/boot/bzImage /boot/vmlinuz lilo no, the modules are compiled by make-kpkg and bundled up into the .deb. when you do dpkg --install kernel-image.blah-blah.deb dpkg handles installing the new kernel for you. it also asks if you'd like to install it with lilo, according to your current /etc/lilo.conf. very nice, and very quick. -- Graham Ashton Futures Testbed, Futures Lab, BT
Re: flying: not enough colours
On Wednesday 31 March, [EMAIL PROTECTED] wrote: Graham Ashton wrote: not enough colors: I need 10 colors + 4 planes. xdpyinfo tells me; class:TrueColor flying runs only in 8bpp . startx -- -bpp 8 oh damn. thanks for the info. -- Graham
Re: libraries documentations
On Wednesday 31 March, Armin Wegner wrote: there are a lot of nice libraries in /usr, but where are the documentations for the c/c++ libraries? humbug% dpkg --list * | grep libc | grep doc pn glibc-doc none (no description available) un glibcdocnone (no description available) un libc-docnone (no description available) pn libc6-doc none (no description available) I think they're in 'info' format. you can get them from http://www.fsf.org/manual/ in other formats. -- Graham
exim - forwarding mail
I'm running exim on my mail server, which is fully DNS registered, and has proper connectivity to the outside world. I'll call it server.domain1.com One of my mates wants me to set it up so that mail sent to [EMAIL PROTECTED] gets directed to his workstation (which doesn't have an MX record), which I'll call workstation.domain2.com I'm sure it's possible to get this stuff to work, but I'm getting a little confused by all the router, director and transport terminology in the manual. I've entered an alias in /etc/aliases as follows; user: [EMAIL PROTECTED] but when I send a message to [EMAIL PROTECTED] it bounces, with the following error: --cut-- A message that you sent could not be delivered to all of its recipients. The following address(es) failed: [EMAIL PROTECTED]: (generated from [EMAIL PROTECTED]): unrouteable mail domain workstation.domain2.com --cut-- I've checked that his machine is capable of accepting mail by telnetting to port 25 and using the VRFY [EMAIL PROTECTED] command. What would be the easiest safe way of getting this working (without getting him an MX entry, which is frowned upon round here for people's desktop machines). Any advice would be much appreciated. I noticed the ROUTERS CONFIGURATION section in /etc/exim.conf. Is this the right place to be looking? Should I put the; literal: driver = ipliteral transport = remote_smtp section before the lookuphost: driver = lookuphost transport = remote_smtp section? Is that the best solution? I'd rather do something like that on a host by host basis if possible, which is a bit beyond me. Thanks. P.S. I'm running the latest version of slink. -- Graham
flying: not enough colours
when I try and run any of the pool type games in the flying package, I get an error message; not enough colors: I need 10 colors + 4 planes. xdpyinfo tells me; class:TrueColor depth:16 planes available colormap entries:64 per subfield red, green, blue masks:0xf800, 0x7e0, 0x1f significant bits in color specification:6 bits any ideas? I really like the pool game, and I haven't played it in ages... -- Graham
Re: Can't get sound module installed
On Friday 26 March, Richard Harran wrote: Could someone please tell me either what I might be doing wrong, or how I could do the 'configure the sound driver with CONFIG_AUDIO option' thing. have you done either make config, make menuconfig or make xconfig before the make dep; make clean ? if not, that's your problem. you need to configure the kernel first so that make knows which bits you want to build in. otherwise you won't be getting a kernel that's customised to your hardware - you'll be getting the default, which doesn't include audio support. make menuconfig and make xconfig are nicer than make config, but they all do the same job. read the documentation in the Documentation subdirectory of the kernel source. I think there's a HOWTO on kernel compilation too - you'll find it at http://metalab.unc.edu/LDP/ -- Graham
Re: Can't get sound module installed
On Friday 26 March, Richard Harran wrote: I've run the install script from the awedrv package, configured the kernel (make menuconfig), sorry - didn't notice that bit when I first read it. still, it sounds like you've missed the audio section out. try it again, and look for the audio section. -- Graham
Re: Can't get sound module installed
On Friday 26 March, Richard Harran wrote: OK, I replying to myself. I didn't select /dev/dsp as one of the sound options in the kernel. Now I've sorted this, I still have a problem: I can't use /dev/dsp, because it doesn't exist. What should have created this, and how do I fix it? cool. try this; cd /dev ./MAKEDEV dsp mine looks like this; humbug% ll /dev/dsp crw-rw-rw- 1 root audio 14, 3 Jul 21 1998 /dev/dsp I think the default permissions are crw-rw, but I couldn't access it as a user that way (and putting me in the audio group didn't work). anybody got any ideas why that is? -- Graham
Re: Can't get sound module installed
On Friday 26 March, Maarten Boekhold wrote: (and putting me in the audio group didn't work). anybody got any ideas why that is? Did you logout/login again after you added yourself to the audio group? I don't think so - I tried using the newgrp command instead. It works fine now with the default permissions, so I think that must have been it. Thanks. -- Graham
Re: SMBMOUNT Problems...
On Sunday 21 March, Michael Beattie wrote: ShadowGate:/root# smbmount //dahouse/c_on_dahouse /dahouse -U BW07442 -P shadowgate.com -I 192.168.0.10 mount error: Operation not supported by device Please look at smbmount's manual page for possible reasons I looked at the man page, and it was no help... :( are you running kernel 2.2.x ? if you are you probably need to recompile the smbmount binary yourself. I've no idea why, I just read it somewhere on dejanews. I've not tried it yet either (too busy). Dont you have to use \\ instead of //??? no. -- Graham
Re: /usr/include/linux - eh?
On Friday 19 March, [EMAIL PROTECTED] wrote: I'm trying to compile vmware (virtual machine under linux), having updated my kernel to 2.2.3. When I did it I just moved the /usr/include/{linux,asm} directories out of the way and then made symlinks to the appropriate kernel headers and then moved the old file back in place when I was done. Worked for my YMMV. thanks for the reply. I've updatd the symlinks but it still fails in the same place. I think I'll wait until vmware come out with a precompiled module for 2.2.3 and try it again then. this is very frustrating though...! -- Graham
rexecd ?
is there a debian package for the rexec daemon anywhere? I've found the client, but the server doesn't come with it. thanks. -- Graham
Re: rexecd ?
On Friday 19 March, J.H.M. Dassen wrote: is there a debian package for the rexec daemon anywhere? /usr/sbin/in.rexecd is in netstd. d'oh! thanks a lot. [sheepishly exits stage left] -- Graham
/usr/include/linux - eh?
I'm trying to compile vmware (virtual machine under linux), having updated my kernel to 2.2.3. vmware needs to compile a module during the installation process, and it takes one look at the version.h file that I have in /usr/include/linux and barfs (wrong kernel version - version.h still reckons I'm running 2.0.36). It appears that the info in version.h that comes with the libc6-dev package suggests that I'm still running 2.2.3. I just removed and re-installed libc6-dev, and version.h is no different. I can't believe that such an oversight is normal, which leads me to conclude that I'm not doing things the debian way (again). Could somebody help me out here? How do I keep version.h up to date? Thanks. P.S. vmware actually says the following; Whoa, something is wrong with the system include files on your machine! The file linux/version.h is for a 2.0.36 Linux system but you are running a 2.2.3 kernel. And fixing the version number in version.h noticably doesn't fix my compilation problems. -- Graham
no local mail delivery
I've just installed a fresh slink box, and selected the Basic profile during installation. I can't deliver mail locally (exim is installed), and there are no files in /var/spool/mail. Sending an email to a valid user has no effect - it just sits in the mailq, frozen. What can be wrong with it? -- Graham
CPAN upgrade - oops
I just upgraded to CPAN 1.48 on my slink box (fully up to date), and now the CPAN shell won't work. It says this; cpan i /Zlib/ Going to read /root/.cpan/sources/authors/01mailrc.txt.gz perl: error in loading shared libraries /usr/lib/perl5/i386-linux/5.004/auto/Compress/Zlib/Zlib.so: undefined symbol: zlibVersion Any ideas? -- Graham
unable to open /dev/mixer
I've just got sound going under 2.2.3, SoundBlaster 64 [OEM] (how happy am I?!!) My little mixer controller complained that it couldn't access /dev/mixer, the permissions on which were 660. It's owned root,audio. I put myself in the audio group, but it didn't fix it. If I do newgrp audio I'm asked for a password. What am I missing? I get the impression it's a real easy one, but I just can't work it out. For a short term fix I just did chmod o+rw /dev/mixer, but I'd like to do it properly. Thanks. -- Graham
UCD snmp package?
I've been unable to find the UCD snmp stuff in .deb format. Does it exist? I've seen the CMU package, but need to get the UCD code instead to make it work with php. I know UCD snmp is really easy to compile from source (that's what I've done), but I like to stick to debian packages where I can... I'm using slink on i386. Thanks. -- Graham
Re: fig2dev (fig - gif)
On Thursday 04 March, Richard Lyon wrote: I'm trying to use fig2dev to convert fig format images into gifs, but I keep getting empty gif files. Your right! It's actually the transfig package as a whole which appears to have no support for the generation of gif. that's bizarre. it specifically says in the fig2dev man page that it can output to gif, or atleast, that it's a valid option. I quote; -L Set the output graphics language. Valid languages are acad (AutoCad slide format), box, epic, eepic, eepicemu, gif, ibmgl, jpeg, latex, mf (METAFONT), pcx, pic, pictex, png, ppm, ps, pstex, pstex_t, textyl, tiff, tk (tcl/tk), tpic, xbm and xpm. If you look at the source code you will find there is support for reading gifs, but none for generating them. fair enough. jpeg it is then (I can't rely on the audience's browsers being capable of reading png files just yet). I suppose I could find something else that can do jpeg - gif conversion, but there doesn't seem to be much point converting from a lossy format to a non lossy format just for the sake of it! thanks for the reply. -- Graham
Re: Hamm--Slink, now fetchmail/exim behaves strangely
On Tuesday 02 March, Mark Phillips wrote: I've just upgraded from Hamm to Slink. Fetchmail seems to work the same as before, except that now, only the first 10 messages get to my mail box straight away. I've got the same problem. Strange, isn't it. -- Graham
fig2dev (fig - gif)
I'm trying to use fig2dev to convert fig format images into gifs, but I keep getting empty gif files. Here's a sample attempt; shandy% cat data 0.3 1.2 0.5 1.34 0.7 1.6 shandy% graph -T fig data file.fig shandy% file file.fig file.fig: FIG image text, version 3.2 shandy% So far so good - I've made myself a fig file. It's the next bit that fails. shandy% fig2dev -L gif file.fig file.gif shandy% file file.gif file.gif: empty shandy% Any thoughts? Other fig2dev output formats work fine (e.g. jpeg, ps, etc) P.S. Is there a problem with the list? I don't appear to have received any messages from it all weekend. Thanks. -- Graham
Re: A couple more strange sound probs...
On Saturday 20 February, Paulo Silva wrote: I have promissed to write yesterday, but I completly forgot. Here goes my awe64 setup information. [snip] that's a very clear description - thanks a lot for the info. I'll try it later on today (maybe tomorrow if I'm too busy), and let you know how it goes. thanks again. -- Graham
Re: A couple more strange sound probs...
On Friday 19 February, rich wrote: From: Paulo J. da Silva e Silva [EMAIL PROTECTED] Could you please send a copy of your /proc/sound in order to make us sure everything is working. I think I'll join in with this thread rather than starting a new one, as my hardware and problems are very close to rich's. I have compiled 2.2.1 too, on an i386 with a soundblaster 64 AWE. I'm running slink. I'm having problems playing wave based audio. it will play about half a second, and then go quiet for half a second, then come back, then go away again. CDs play okay, but the output doesn't sound very clear, and the right channel is much louder than the left one. I'm assuming that I've missed something out whilst compiling my kernel, or have selected something I shouldn't have. Any thoughts on how I ought to proceed would be much appreciated. Here's the output from /proc/sound... ---cut--- humbug:root cat /proc/sound OSS/Free:3.8s2++-971130 Load type: Driver compiled into kernel Kernel: Linux humbug 2.2.1 #2 SMP Thu Feb 18 16:30:49 GMT 1999 i686 Config options: 0 Installed drivers: Type 26: MPU-401 (UART) Type 2: Sound Blaster Type 29: Sound Blaster PnP Type 7: SB MPU-401 Card config: Sound Blaster at 0x220 irq 7 drq 1,5 SB MPU-401 at 0x330 irq 7 drq 0 Audio devices: 0: Sound Blaster 16 (4.16) (DUPLEX) Synth devices: Midi devices: 0: Sound Blaster 16 Timers: 0: System clock Mixers: 0: Sound Blaster ---cut--- -- Graham
proftpd + virtual hosts
is it possible to do name based virtual hosting with proftpd? the documentation on the web site suggests not, and that you need to do IP address based virtual hosting instead. :( -- Graham
Re: not a plain file - apt install with dselect
On Wednesday 17 February, John Stevenson wrote: However apt is having problems getting a few of the files. See the error message below. Is there a problem with apt, the ftp site or my installation? Any clues ?? I'd say that it's a problem with the mirror, as if you login to the site and try and file the file that it's failing on, it doesn't appear to be there. Get ftp://ftp.mcc.ac.uk frozen/main dpkg-http [13.0k] Error ftp://ftp.mcc.ac.uk frozen/main dpkg-http 550 /pub/linux/distributions/Debian/dists/frozen/main/binary-i386/admin/dpkg-http_0. 19.deb: not a plain file. You could try pointing apt at another mirror and seeing if that works. -- Graham
[vim] :se num - underlined numbers
I've got vim installed on both hamm and slink boxes, versions 5.0 and 5.3 of vim respectively, and use both from the same remote rxvt session. When I switch line numbering on in 5.0, everything looks fine. When I switch it on in 5.3, the line numbers themselves are underlined. It's really annoying. I've got syntax highlighting turned off. Can anybody suggest why line numbers are underlined in 5.3, but not in 5.0? I think it must be an entry in /etc/vimrc, but can't figure out which one. -- Graham
mysql password problems
I've just started fiddling with mysql (from the standard .deb files) on my hamm box, and am having problems with permissions/privileges. I can't create databases with mysqladmin. I can't even connect to the mysql privilege tables to alter the permissions, as I don't seem to have the permissions! This is what I've done; host% mysql mysql ERROR 1044: Access denied for user: '@localhost' to database 'mysql' host% mysql -u root mysql ERROR 1045: Access denied for user: '[EMAIL PROTECTED]' (Using password: NO) host% mysql -u mysql mysql ERROR 1044: Access denied for user: '@localhost' to database 'mysql' I seem to be lacking the one important piece of info, and that's the password for the 'root' user. I'm assuming that it's a debian thing, as when I last compiled mysql from source there was no default password for the 'root' user, which meant that I could connect without a password and change it to whatever I liked. I'm sure this must be documented somewhere on my box, and I'm sure it'll tell me the password. It's just that I can't find it, not even with the help of dwww! Any clues on where to look, or what the password ought to be? Thanks in advance. -- Graham
Re: [PHP3] Max TEXT size..
On Wed, Feb 10, 1999 at 12:29:00PM +0100, Chris Chabot wrote: While developing several sites based on a PHP (3.0.6/Apache-1.3.4) engine, ive stumbled across a max. size limit for TEXT fields while inserting them in the MySql (3.22.15) database. The max size was 65535, which sounds awfully much like a 2byte unsigned numeral, now my question is, is this a limit thats set by PhP by the internal variable parser, or is this a external factor (eg a limit set in mysql, etc ..) Ive been looking thru both the php and mysql documentation, and came up blank so far... look in section 7.2 of the mysql manual, and scroll down until you find 'TEXT' at the left hand side of the page. http://www.mysql.com/Manual_chapter/manual_Syntax.html#Column_types -- Graham
Re: [PHP3] Max TEXT size..
On Wed, Feb 10, 1999 at 11:46:56AM +, Graham Ashton wrote: look in section 7.2 of the mysql manual, and scroll down until you find 'TEXT' at the left hand side of the page. sorry about that folks - I replied to the wrong list - wasn't thinking straight. now that's 2 cock-ups in as many days. -- Graham
exim - allowed domains
I've got a bit of a problem with my exim config at the moment, and the exim web site's DNS is refusing to resolve so I can't get to the right part of the manual. Apologies for posting it here, but I'm in a bit of a rush to fix the problem... exim is receiving mail perfectly happily for all mails addressed to [EMAIL PROTECTED] and [EMAIL PROTECTED] what it isn't doing at the moment, is accepting mail to [EMAIL PROTECTED], which is being forwarded from my other account so that I can read it on my own mailhost. I don't have control over the box that's doing the forwarding, so I can't take just delete the route in the mean time! what I'd like to do is to tell exim to accept mail addressed to [EMAIL PROTECTED] as well. so my question is this: should I list domain2.com in the 'local_domains' field in exim.conf? my gutt reaction is that this wouldn't be exactly what I want, as domain2.com isn't really a local domain. any thoughts? TIA. -- Graham
Re: exim - allowed domains
On Tue, Feb 09, 1999 at 02:03:56PM +, Graham Ashton wrote: exim is receiving mail perfectly happily for all mails addressed to [EMAIL PROTECTED] and [EMAIL PROTECTED] what it isn't doing at the moment, is accepting mail to [EMAIL PROTECTED], sorry for replying to my own post, but this now seems to have resolved itself. it appears that exim was quite happily accepting my forwarded messages, but that the system that was sending them out was having problems. sorry for the wasted bandwidth. I'll dig harder next time... -- Graham
Re: mutt with pgp support?
On Fri, Feb 05, 1999 at 12:08:01AM -0800, Stephen A. Witt wrote: On Thu, 4 Feb 1999, Chris Frost wrote: Does the mutt included with debian (hamm) include support for pgp v5 (or gpgp)? I'm not sure. Certainly does pgp2 and pgp3 (whatever that means). I'm thinking of switching over to mutt from pine I've done that, having used postilion inbetween. I like mutt now I'm used to it, and got a bit pissed off with pine - it was very slow with large mailboxes. If anyone here has used both mutt and pine, what are the main things mutt offers over pine (besides the license); is there anything which pine does better? as already mentioned, pine has it's own internal editor. I've you've got an editor that you'd like to write your emails in though, I think this is a 'feature' of mutt. Does mutt have a nice built in address book like pine? It has a list of aliases, from which you can select the one you want. I've not seen anything more swanky than that though. Does mutt do the IMAP thing like pine? Yes. I've not tried it, but I didn't rate pine's IMAP capabilities so mutt might be better (i.e. quicker). Is mutt able to lookup email addresses from an X.500 directory server via LDAP? You can specify the name of an external program to use to lookup addresses for you. i.e. if you have an external program that can look up addresses from an LDAP server, you're all set. I *really* appreciate the message threading you get in mutt for mailing lists too. That's one of it's greatest advantages for me. When I used mutt several years ago, I seem to remember that it was rather confusing to configure, whereas pine has a nice built in configuration utility with on-line help. It took me an afternoon of reading the docs to find out what all the possible options were, and then sitting down and hacking together my .muttrc file. There are some good example .muttrc files available (linked from the FAQ, I think). -- Graham
Re: gnome .deb install probs
On 2 Feb, Syrus Nemat-Nasser wrote: The stuff you downloaded from ftp.gnome.org was old (however those debs are frozen in slink). However, libglib and libgtk 1.1.13 is new. GNOME is rapidly evolving alpha software. [snip] thanks to those who replied. I think I'll see what happens if I download a lot of source, and compile things into /usr/local. for now atleast. as far as a mail client goes, perhaps I should dig up my copy of mutt in the meantime... -- Graham
Re: Counting number of lines in a text file
On 3 Feb, Jay Barbee wrote: for a in file1 file2 file3; do \ echo -n $a seperate_file; \ echo -n ::: seperate_file; \ wc -l $a seperate_file; \ done How would I gather a list of all TXT files in the directory before this and pass each file to this script. replace the first line with; for a in `ls *.TXT`; do -- Graham
gnome .deb install probs
I've downloaded libgnome0_0.30-1_i386.deb from ftp.gnome.org, and have just tried installing it on my hamm box. I got a load of dependency problems with dpkg, even for things that are installed. I installed libglib1.1.13_1.1.13-1_i386.deb libgtk1.1.13_1.1.13-1_i386.deb without any problem. When trying to install libgnome0, dpkg complains that I don't have (amongst other things) the right versions of libglib and libgtk! I think I do, and am a little confused. humbug:root dpkg -i libgnome0_0.30-1_i386.deb Selecting previously deselected package libgnome0. (Reading database ... 39220 files and directories currently installed.) Unpacking libgnome0 (from libgnome0_0.30-1_i386.deb) ... dpkg: dependency problems prevent configuration of libgnome0: libgnome0 depends on gdk-imlib1 (= 1.7); however: Package gdk-imlib1 is not installed. libgnome0 depends on libglib1.1 (= 1.1.3-1); however: Package libglib1.1 is not installed. libgnome0 depends on libgtk1.1 (= 1:1.1.2-1); however: Package libgtk1.1 is not installed. libgnome0 depends on liborbit0 (= 0.3.0-1); however: Package liborbit0 is not installed. libgnome0 depends on libungif3g (= 3.0-2) | giflib3g (= 3.0-5.2); however: Package libungif3g is not installed. Version of giflib3g on system is 3.0-5. dpkg: error processing libgnome0 (--install): dependency problems - leaving unconfigured Errors were encountered while processing: libgnome0 All I really want to do is to get enough of gnome installed to let me use the balsa mail client. Do I need to install lots of packages from slink or potato to get everything I need? Would I be better off just upgrading to slink right now anyway? I'm really not liking this very much, and am starting to think it's too hard... hello tarball, my old friend... Can anybody point me in the right direction (i.e. should I upgrade the whole box, or should I just go and get specific packages from another distribution?) -- Graham
gnome + gtk + dpkg
I've finally decided to bite the bullet, and install GNOME. I'm running hamm, without any packages from slink or potato installed. I've downloaded the .deb files for the basic gnome system from ftp.gnome.org, and thought I'd try and work out how to install them with dpkg. The problem is, it's conflicting with the versions of gimp and gtk+ and glib that I have installed at the moment (snipped output from dpkg -l); gimp1.0.0-1The layers-based, non-Motif GNU Image Manipu libgtk1 1.0.4-1The GIMP Toolkit set of widgets for X libgimp11.0.0-1Libraries necessary to run the GIMP So I'm obviously going to have to update quite a bit if I want everything to run smoothly. Can any of you GNOME + gimp users advise me here? Ought I to move up to gtk+ 1.1.x, or should I stick as closely as possible to the stuff that comes with hamm? Thanks. -- Graham
Re: apache modules in hamm
On 27 Jan, Nathan E Norman wrote: Take a look at /etc/apache/httpd.conf - you'll notice a whole section on modules. Uncomment the lines you want. smashing, thanks. It's still not working, but I suspect I need to revisit the manual to work that one out... -- Graham
apache modules in hamm
I just tried setting up SSI on apache on my hamm box, but found to my dismay that very few modules are compiled in by default; humbug:root apache -l Compiled-in modules: http_core.c mod_so.c I notice that the shared object one is there though, so I thought that maybe all the others would be available as part of the distribution, and that I just hadn't installed them. I can't find anything with dselect. Are they there? I searched for apache and include and came up with nothing. I don't really fancy re-compiling from source in this instance, as I think it's about time I learnt how to use the debian package system properly! -- Graham
scsi cd drive locking up
I'm having trouble mounting CDs in my Toshiba SCSI CD drive. As soon as I put a CD in (data or audio), the little green light on the front flashes about 10 times, and then lights up permanently. The LED only goes out when I eject the disk. When I attempt to mount a (perfectly good) CD, I get; humbug:root mount -t iso9660 /dev/scd0 /cdrom mount: block device /dev/scd0 is write-protected, mounting read-only mount: wrong fs type, bad option, bad superblock on /dev/scd0, or too many mounted file systems I was away on holiday for a week, and when I got back I powered on my machine and everything seemed to start up normally (but I'd left an audio CD in the drive and the Adaptec 2940 BIOS spent a long time trying to determine if the CD was bootable before it went on to detect my hard disk). Before I went away I was using cdparanoia without any problems, and normal CDs were mounting without incident. I tried fiddling with the permissions, changing them from; br--r- 1 root cdrom 11, 0 Dec 1 20:25 /dev/scd0 to; brwxrwxrwx 1 root cdrom 11, 0 Dec 1 20:25 /dev/scd0 but (unsurprisingly) it's not made any difference. Does it sound like a hardware problem, or could it be that something needs resetting? Could it be Linux? A cold reboot doesn't seem to help. TIA - I desperately want to do some more ripping! -- Graham
RE: windowmakers .0.20 and dockit.
On 11 Jan, Person, Roderick wrote: I assume that this is part of wmakerconf, which I have got to yet. no, it's on the attributes menu (the one you get when you right click an application's title bar). I don't know when it was introduced, but it was certainly in there by 0.20.3. it has become obsolete with the advent of the emulate appicon feature in -- Graham
Re: X on DELL laptop hangs
On 15 Dec, David S. Zelinsky wrote: A friend of mine installed hamm on a Dell laptop with Neomagic video. He got X configured with the VGA16 server. I have a Dell Lattitude CPi 266XT (or whatever it's called), with a neomagic chipset. I downloaded the XBF server from Red Hat's site which comes with a sample XF86Config file. It works fine under Red Hat, and I have no reason to believe it won't work fine under debian (I've installed it under debian [it's dual boot], but have no window manager yet so have only seen it working for a second or two - I'll let you know what happens when I get it up if you remind in a week or so). He also tried dropping in the SVGA server from XFree86 3.3.3, which handles the Neomagic chip. The same thing happened. doesn't sound like an X problem to me, but that's just a gut reaction. Ultimately, my friend installed a copy of RedHat 5.2, which worked fine, with none of this problem. I don't know if 5.2 has APM support in the kernel by default. 5.1 certainly didn't. I feel defeated and somewhat guilty, since it was partly at my urging that this friend had tried Debian in the first place, so I'd really like to find out what went wrong. If I was you I'd go into the bios and disable all APM related things, and see if it still does it. * The problem seemed to occurs within a few seconds ( 30? ) after switching to text mode (quitting X, or switching to virtual console). We were able to get a shell prompt, even issue a few commands, and switch back an forth between the v.c. and the X screen a couple of times, before the system would just freeze. not so keen on the sound of that. * The Debian was a 2.0 CD from Cheap Bytes so was mine, but I didn't notice anything different about it from the normal version. * I didn't think of checking the BIOS settings, for things like APM. Could a broken APM cause symptoms like this? Then again, isn't APM disabled in kernel-images that ship with Debian? If it is disabled in the kernel the BIOS will still override it if it feels like it, sometimes leading to a hung box. I recompiled mine with APM support and it works rather nicely (so long as you don't try and suspend it). -- Graham
kernel + module mismatch? (was Re: [FAQ] starting isapnp)
On 15 Dec, Frank Barknecht wrote: [EMAIL PROTECTED] hat gesagt: // [EMAIL PROTECTED] wrote: I've currently reached the stage where I've recompiled my kernel having applied the latest awe-drv patch, and am trying to get the module to load. I've been following the sbawe32 HOWTO to the letter. The link is there. /etc/rcS.d/S15isapnp should be the link to /etc/init.d/isapnp like this: It is. It seems to be working fine. I didn't know what rcS.d was for, but now I've found some documentation on it everything makes perfect sense. humbug:root modprobe -a sound /lib/modules/2.0.36/misc/sound.o: unresolved symbol unload_awe /lib/modules/2.0.36/misc/sound.o: unresolved symbol attach_awe This does not look good. I hope you did use make-kpkg (in the kernel-package) for building your new kernel. It helps getting rid of some mistakes during installation of kernel and modules. I didn't. I downloaded the source for 2.0.36 and the latest awe patch; http://bahamut.mm.t.u-tokyo.ac.jp/~iwai/awedrv/ (version 0.4.3-pre4) Make sure you have: CONFIG_MODULES=y CONFIG_MODVERSIONS=y CONFIG_KERNELD=y I do. Which suggests to me that my modules don't match my kernel. What does the unresolved symbol message actually mean? Could it be that the unload_awe and attach_awe symbols aren't being picked up from my attempts to patch the kernel source? This for now. Please keep asking. Thank you. And thanks for your helpful reply. I suppose my next questions should be am I likely to have more joy using make-kpkg?, and should I get a .deb for my kernel source? -- Graham
Re: Netscape, samba, power, awe64 ?s
On 15 Dec, Brian Morgan wrote: 1.I posted a samba question earlier. Has anyone had a chance to look at that yet? didn't see it, sorry. 2.Downloaded netscape from netscape.com, ran ns-install, and now I get an error: can't load library 'libXpm.so.4' I've had this problem in the past, on another debian machine, but I don't remember how I solved it. Any suggestions? I ran into the same thing a week or so ago; humbug% ldd /usr/local/netscape/netscape libXt.so.6 = /usr/lib/libc5-compat/libXt.so.6 (0x4000b000) libSM.so.6 = /usr/lib/libc5-compat/libSM.so.6 (0x4004d000) libICE.so.6 = /usr/lib/libc5-compat/libICE.so.6 (0x40056000) libXmu.so.6 = /usr/lib/libc5-compat/libXmu.so.6 (0x4006b000) libXpm.so.4 = /usr/lib/libc5-compat/libXpm.so.4 (0x4007d000) libXext.so.6 = /usr/lib/libc5-compat/libXext.so.6 (0x4008b000) libX11.so.6 = /usr/lib/libc5-compat/libX11.so.6 (0x40096000) libdl.so.1 = /lib/libdl.so.1 (0x40134000) libc.so.5 = /lib/libc.so.5 (0x40137000) libg++.so.27 = /usr/lib/libc5-compat/libg++.so.27 (0x401f5000) libstdc++.so.27 = /usr/lib/libc5-compat/libstdc++.so.27 (0x4022d000) libm.so.5 = /lib/libm.so.5 (0x4025e000) you want to know which package contains libXpm.so.4. I was missing some of the libg++ stuff to start with too. I think it's one of the xpm4.7 packages (I seem to have the libc5 ones, even though I thought I'd got the glibc version of netscape)... 4.I think someone offered advice on this before, but I'd like some help getting an awe64 soundcard to work. Isn't there something I have to do to recompile a the kernel? Sounds kind of scarey, for a newbie. Can anyone help? there was a post earlier today from Frank Barknecht to me that was discussing it. the best approach would appear to be to recompile your kernel with the make-kpkg utility, and to apply all the relevant awe .deb packages too. I haven't tried that yet. Check his message for more info. there's also an awe mini howto that you should read, available from http://metalab.unc.edu/LDP/ you should get the isapnp stuff too (explained in howto). come back with more specific questions if you get stuck. if I'm successful (in the near future) I'll post a summary of what I did. -- Graham
cfengine cron job
I've just installed a new hamm system, and am getting mail messages about cfengine's cron job; -- Subject: Cron [EMAIL PROTECTED] run-parts --report /etc/cron.daily X-Cron-Env: SHELL=/bin/sh X-Cron-Env: PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin X-Cron-Env: HOME=/root X-Cron-Env: LOGNAME=root Date: Mon, 14 Dec 1998 06:44:51 + /etc/cron.daily/cfengine: cf:alice:/etc/cfengine/cfengine.conf:26: parse error cfengine:alice::26: Warning: actionsequence is empty cfengine:alice::26: Warning: perhaps cfengine.conf has not yet been set up? cfengine:alice::Execution terminated after parsing due to errors in program /etc/cron.daily/cracklib: 45375 45375 -- I understand that this is because I've not configured cfengine, and don't intend to in the near future (it looks like the sort of thing I'd like to play with before too long though). What would be the most debian friendly way of disabling it? I thought of removing the cfengine package, but then thought that there might be a less heavy handed approach, other than deleting/moving the files from /etc/cron.daily and /etc/cron.weekly. Is there a nice way to do it? -- Graham
Re: cfengine cron job
On 14 Dec, Ben Collins wrote: Execute this command: /usr/sbin/dpkg-divert --local --add --rename --divert /etc/cron.daily/cfengine /etc/cron.daily/cfengine.norun wow. how cool is debian? thanks a lot. [scuttles off to see what else dpkg-divert can do...] -- Graham
Re: Remote printing
On 14 Dec, Kent West wrote: At 02:07 PM 12/14/1998 +, [EMAIL PROTECTED] wrote: I'd like to print directly to a networked laser printer that has its own IP address. Yep, it's possible. Desirable? Depends. If you're the only one printing to it, sure. If others are printing to it also, it's probably better to point everyone to one queue instead of having multiple queues serving the same printer. agreed. some printers don't handle multiple simultaneous requests too well, apparently. 2) how to go about it? I was afraid you were going to ask that. tricky one, and often depends on the model of the printer in question. I'm no expert either, but here we go... I think some network printers run a proper lpd demon, whereas some just accept connections on TCP port 9100. If you have one of the latter the best thing to do is to get a copy of LPRng, which is capable of doing such things (and much more besides). It is a drop in replacement for the standard lpd code, and there's a .deb for it too. I'd get LPRng anyway, actually, as it's the business. There's a HOWTO document as well (but it's huge). To give you a feel for it, here's my /etc/printcap # hplj4-136; HP LaserJet 4 on JetDirect EX card hplj4-136|lp :lp=hplj4-136%9100 :sh :cm=HP LaserJet4 :sd=/var/spool/lpd/hplj4-136 :lf=log :af=acct :of=/usr/local/lib/filters/ofhp: :if=/usr/local/lib/filters/ifhp: :vf=/usr/local/lib/filters/ifhp -c: :mx#0 The filters I'm using are specifically designed to work with HP printers, and work like a charm. They're also available from the LPRng web site. Note that there aren't any \ characters at the end of each line (LPRng doesn't need them), and that the :lp line is pointing to port 9100 on host hplj4-136 (I don't think that you can specify a TCP port with most implementations of lpd). LPRng web site; http://www.astart.com/lprng/LPRng.html -- Graham
Re: fetchmail + procmail setup
On 14 Dec, Richard L. Alhama wrote: But when I choosed smail as my MTA, my used to be working ~/.procmailrc could not sort mail anymore. oh. well I have a very similar setup to dave. my .forward and .procmailrc files seem to be virtually identical (technically). my fetchmail is done slightly differently though (I'm running it from a cron job instead - I know, daft, but I can't be bothered to fix it). what should ~/.forward contain? here's my .forward; humbug% cat .forward |IFS=' ' exec /usr/bin/procmail -f || exit 75 #ashtong humbug% ll .forward -rw-rw-r-- 1 ashtong ashtong60 Dec 8 10:39 .forward the permissions on /var/spool/mail? humbug% ll /var/spool/mail -d drwxrwsr-x 2 root mail 1024 Dec 13 13:37 /var/spool/mail/ humbug% ll /var/spool/mail total 24 -rw-rw 1 ashtong mail23017 Dec 13 17:28 ashtong they may not be right, but they work for me, and are debian's defaults. I'm also using smail, and I just copied the .forward from my old red hat system that was using sendmail. everything works as expected. I've got vanilla hamm. hope that helps you a bit. -- Graham
Re: X is giving me a headache.
On 11 Dec, [EMAIL PROTECTED] wrote: X is totally confusing me at the moment. Or rather, how to get it all working, where are the good window managers, etc. I had problems getting xbase configured, but my previous experience with Red Hat meant I knew how to frig it. I have a security problem now (X server is setuid root), but it doesn't matter where I am (I am the only user). Still, I sympathise. I've about given up on KDE. I have read a lot about KDE and how good it was; how many great product awards it is receiving, and wanted to install it as my WM. I've been there. I used it for a few weeks, then realised that it didn't suit my tastes. I found it almost as ugly as Windows. I've been an avid window maker follower ever since (I came back to it at version 0.14.x, having played with it at version 0.5.x). I don't want prerelease (v1.1) software at this point (because of instability), window maker is very stable on linux. I've never had any problems with it. some earlier versions have crashed about once a month for me, but 0.20.3 has been perfect so far. it's got a very nice gui config app too. the wmaker deb maintainer has just announced new ones on the wmaker mailing list (they're for slink - I don't know if that's a problem with earlier debian systems or not); http://master.debian.org/~mmagallo/packages/wmaker/ So, I thought I'd try Window Maker go on... Maybe I'm expecting too much. All I'm after is a good windowing system for my new Linux. One that I'm not going to have to reboot to fix all the time I've had wmaker uptimes of ~25 days, after which I've just logged it out. Can I just get the include files for X, will that placate the compiles for KDE or Window Maker? yes, it should get you part of the way there. I always compile wmaker from source (always have done, even when I used Red Hat, as I like the newest versions). I had problems getting all the required stuff on board when I first installed debian, but it just came down to trial and error with libraries and header files in dselect. if I was in your position I'd try the wmaker debs at the above URL, and if they don't work well on hamm/frozen/whatever, let me know and I'll see if I can work out which .debs I installed to get wmaker to compile smoothly (you need xpm, tiff, jpg libraries to get the most out of it). it's very beautiful and extremely nice to use though. -- Graham
Re: apache: httpd: cannot determine local host name.
On 12 Dec, Zack Brown wrote: I installed apache, but at the end of the configuration stage it failed to start, giving an error. I ran apacheconfig, and got the following output (essentially the same error): I don't know why it should have trouble finding my hostname. It's right there in /etc/hostname. it's not looking at /etc/hostname. it's trying to work out what hostname it should send to a browser in reply to a request to a given ip address (namely the ip address of your network interface). put a line in /etc/hosts like this; 123.456.789.012 myhost.mydomain.commyhost where 123.456.789.012 is your ip address. if you've got a dynamic ip address, or that still doesn't work, look through your httpd.conf file until you find the line that looks like; #ServerName myhost.mydomain.com replace myhost.mydomain.com with what it should be, and remove the '#' from the start of the line. then try restarting apache. you shouldn't need to use the ServerName directive in simple cases though. apache should be able to work it out for itself if your system is set up correctly. -- Graham
Re: WindowMaker 0.20.3 on hamm
On 12 Dec, Tun Yang wrote: *** Opt graphics libtiff3g-de 3.4beta037- 3.4beta037- tiff development files [libc6] Hi... I was trying to install WindowMaker 0.20.3 that I got from www.windowmaker.org Alright... so I don't think I can build any X program at this point... but I can't find any X header/development stuff I didn't installcan someone tell me the packages? Also, I can't find libtiff-dev and libjpeg-dev packages, so maybe my problem is I need frozen/slink? I've got hamm, and the following packages (might not be a list of everything you need, but hey, I'm just searching through dselect...); libtiff3g-dev libjpegg-dev xpm4g-dev xlib6g-dev I've also got a few libc5 variants of some of the above (similar names but without the g on the end). I'm not sure which I used when I compiled wmaker 0.20.3. I've not installed anything that's not in hamm. -- Graham
Re: fetchmail daemon mode
On 9 Dec, Shao Zhang wrote: If I comment out the line set daemon 300 in the file ~/.fetchmailrc. Then my fetchmail works fine... If I run fetchmail with the above line, then id doesn't work. The process just sits there and doing nothing... are you sure that it's not doing anything in the background, every 5 minutes? you could try /usr/sbin/tcpdump (normally must be run as root) to see if there are any packets going between your box and the mailserver on the right port (e.g. /usr/sbin/tcpdump host client and host server, or /usr/sbin/tcpdump port 143 [the IMAP port - look in /etc/services for other port numbers]). Do I have to start fetchmail in a special place?? no. mine is in my crontab file, because I don't use daemon mode. If I put it in ip-up, do I have to use the fetchmailrc in the root directory?? If I use that, how does fetchmail know to put in my user account rather than root?? I would have thought you could specify which .rc file to use. looking at the man page, I find that you can do; fetchmail -f /path/to/.fetchmailrc As for where it will decide to put it, I'm not sure. A full reading of the man page might be in order. You may be able to fix it by setting the $LOGNAME or $USER environment variables in the script that calls fetchmail, but that would be a bit of a fudge, and I wouldn't recommend it. hope that helps. I've not tried any of that though. read the man page before you do! -- Graham Ashton
2.0.36 compile prob
I've just tried compiling 2.0.36 with SMP support on my brand new and shiny debian box (hamm), but it broke halfway through. I'm asking the question here as I think I'm missing an important system file, and was hoping somebody would be able to tell me which .deb it's in. make menuconfig and make dep; make clean work fine, but after a fair degree of success with make zImage I get; -- gcc -D__KERNEL__ -I/usr/src/linux-2.0.36/include -E -D__ELF__ -D__SMP__ -traditional trampoline.S -o trampoline.s as86 -0 -a -o trampoline.o trampoline.s make[1]: as86: Command not found make[1]: *** [trampoline.o] Error 127 make[1]: Leaving directory `/usr/src/linux-2.0.36/arch/i386/kernel' make: *** [linuxsubdirs] Error 2 -- It compiles cleanly without SMP support (i.e. with SMP=1 commented out in /usr/src/linux/Makefile). Thanks in advance. P.S. not really sure what I'm doing with dpkg yet, as I've just switched from using Red Hat (which may explain a few things). -- Graham Ashton
Re: Network adapters (speed 100MB)
On Tue, 20 Oct 1998, jrivas wrote: Anybody knows some network adapter (speed 100MB) which supports linux drivers? I never expected my first post to the debian list after having just switched from Red Hat would be an answer(!), but... Drivers exist for quite a few of them. I use 3com 3c905 cards and have had no problems with them, but many recommend one of the many alternatives instead (several people have reported great difficulty with them under high loads). Another popular choice is the intel etherexpress. I'd start by looking at Donald Becker's excellent drivers page; http://cesdis.gsfc.nasa.gov/linux/drivers/ -- Graham Ashton