shellutils on hurd
I tried upgrading my woody system using apt-get, and I noticed that apt-get produces the following strange output: 16 packages upgraded, 0 newly installed, 0 to remove and 3 not upgraded. Inst shellutils [shellutils on hurd] Inst base-files [shellutils on hurd] Inst lilo [shellutils on hurd] Inst man-db [shellutils on hurd] ... What does the "[shellutils on hurd]" mean??? I don't see how Hurd suddenly comes into the picture, as I'm still using the Linux kernel?? T ()() "Do not modify spaghetti code unless you can eat the consequences..." `--'
CD-ROM only plays mono??
I'm using the cdtool package to play audio CD's from my CD-ROM drive. For some reason, playing CD's only outputs MONO sound (ie. only to one of my speakers). It's not a problem with my speakers since .mp3 and .wav files plays as stereo without any problems. Is this a software problem or is it a hardware problem? Any suggestions will be appreciated. Thanx!! My soundcard is an SB16 and I'm using kernel 2.2.12. T - ()() "Don't modify spaghetti code unless you can eat the consequences." `--'
Limiting FTP bandwidth
I'm running an FTP server on a 20K/sec PPPoE connection, and I'm experiencing this problem: when someone with a high bandwidth is downloading from my server, my network response time suddenly becomes very, very slow. I was told that using a traffic shaper would help, but the kernel traffic shaper seems to limit *everything*. Is there a way to only shape traffic on FTP and nothing else? I'm using wu-ftpd (2.6.0-3) and I've tried setting the "throughput" directive, but it seems to help only a little. Any ideas/suggestions? Thanks!! T - ()() "Don't modify spaghetti code unless you can eat the consequences." `--'
Re: writting a Daemon in c
On Wed, 20 Oct 1999, Evan Moore wrote: > I am writting a Daemon in c, I need to know how to disconnect the program > from the terminal so that if that terminal is destoyed it will not cause > my server to shutdown. I have done this with perl scripts, but can't not > figure it out with c. Thanks in advance Use fork() to fork your daemon process. Exit the main process and have the child process perform the daemon functions. The child process will not exit when the controlling tty is gone. T
ncurses driving me nuts
Hi! I'm not sure if this is the right place to ask for help with ncurses programming (pls redirect me if this doesn't belong here), but the behaviour of ncurses is driving me up the wall here... For some silly reason, wmove() and mvwaddstr() doesn't seem to be functioning properly. The first one or two wmove() -- waddstr() sequences seem to work fine, but after that, the x-coordinate seem to get completely ignored and set to zero for all subsequent calls to wmove() or mvwaddstr(). What am I doing wrong??? I'm trying to print a list of strings and want each to appear at a particular x-coordinate within a subwindow. What's the right way to do this??? Currently, it prints the first one or two lines at the right location, but subsequent calls to wmove() or mvwaddstr() seems to totally ignore the x-coordinate and simply place the cursor at the start of the line. T
Re: /dev/audio
On Tue, 19 Oct 1999, Mary Honeycutt wrote: > Hi, > > I'm receiving messages in .xsession-errors, stating: > "sox: unable to open /dev/audio. Device or resource busy." > > All sounds still play, however, everytime a sound file is run, > a new error msg is generated. > > How can I find out what is using /dev/audio? fuser /dev/audio will display the PID (process ID) of the process that is using /dev/audio. To find out what exactly that process is, enter ps lw where is the numerical PID displayed by fuser. T
EsounD trouble (second try)
(repost of my previous message -- accidentally used reply-to to an unrelated thread in the mailing list -- probably missed everyone who has a threaded mail reader.) Can I use EsounD without GNOME/Enlightenment? I installed the esound package but I can't get anything to play via esd. The only thing I can get out of the speaker is the esd startup tone sequence (when I run esd in the background). I tried esdplay, esdcat, esddsp, xmms (with esound plugin) but I can't seem to get esd to produce any sound at all. What am I doing wrong??? T
EsounD trouble
Can I use EsounD without GNOME/Enlightenment? I installed the esound package but I can't get anything to play via esd. The only thing I can get out of the speaker is the esd startup tone sequence (when I run esd in the background). I tried esdplay, esdcat, esddsp, xmms (with esound plugin) but I can't seem to get esd to produce any sound at all. What am I doing wrong??? T
potato: why does new slang1 preclude modconf, whiptail, gimp, ae,...?
I've been getting a strange scenario lately with trying to upgrade to potato... after I did a `apt-get update', I noticed that `apt-get -s upgrade' noted that slang1 was being kept back. Curious as to why slang1 was being kept back, I did a `apt-get -s install slang1' and apt-get responded that the packages ae, gimp, modconf, newt0.30, whiptail would have been *removed*. Apparently there is some dependency problem around newt0.30 and whiptail that caused a cascade of packages to need removal? But what I really *don't* understand is why gimp would be removed. What has it got to do with newt/whiptail anyway??! Also, isn't ae supposed to be a base package? Why is it removing ae? (Or am I missing the fact that ae is now part of some other package somewhere out there?) T
potential problem upgrading netbase
I was upgrading netbase on a remote potato machine, when apt-get just stopped. (This seems to be a problem with the postrm script from the old netbase package, judging by what happened later). I hit ctrl-C and dpkg said something to the effect that the old postrm failed, and tried the postrm from the new package, but then aborted with an error message, something to the effect of "/etc/init.d/inetd: not found". Apparently it was expecting /etc/init.d/inetd to still be there. I had to create a phony (executable) /etc/init.d/inetd to get the upgrade to work properly. This looks like a breakage of the old postrm script... if that is so, is there any mechanism in place to correctly handle this problem during upgrades? If someone less experienced tried to upgrade netbase it could cause a lot of grief. I can't remember the old version number right off (I was too worried about losing my network connection to that machine so I didn't record anything), but it's a recent potato version of netbase. T
Re: potato: procps and bsdutils conflict?
On Thu, 7 Oct 1999, David Coe wrote: > Hwei Sheng TEOH <[EMAIL PROTECTED]> writes: > > [...] > > dpkg: error processing > > /var/cache/apt/archives/bsdutils_1%3a2.9w-3.1_i386.deb > > (--unpack): > > trying to overwrite `/bin/kill', which is also in package procps > [...] > > However, I was just wondering, is it really the case that both bsdutils and > > procps contains the program /bin/kill? Are they the same, or are they > > different versions? Is the alternatives mechanisms in place for /bin/kill? > > Apparently there's some problem with both packages containing the same file? > > There's been discussion about this recently on debian-devel. IIRC, > the bsdutils version of kill is not linux-specific, but is not > well-supported upstream, the util-linux version is well-supported > upstream but doesn't work on non-linux debian systems (e.g. the Hurd). > The respective maintainers will presumably work out a solution that > works "best" for everyone. > > (Note that this problem only affects unstable/potato). > > You did the right thing by forcing one of them; at the moment it > doesn't appear to matter much which one you let override the other. Hmmm, only one combination seems to work though... installing procps first will cause dpkg to complain when bsdutils is being installed. For some reason, installing procps on top of bsdutils seem to simply overwrite /bin/kill with no warning. I'm not sure how good this situation is, but I suppose both versions of /bin/kill does the same thing. (It doesn't really affect me because I use tcsh and it has a built-in kill command.). T
potato: procps and bsdutils conflict?
I was just upgrading my potato system, and apt-get stopped when processing bsdutils. There was an error message from dpkg: dpkg: error processing /var/cache/apt/archives/bsdutils_1%3a2.9w-3.1_i386.deb (--unpack): trying to overwrite `/bin/kill', which is also in package procps Errors were encountered while processing: /var/cache/apt/archives/bsdutils_1%3a2.9w-3.1_i386.deb I managed to fix this by (forcibly) uninstalling procps, using `dpkg --remove --force-depends', and then `apt-get install bsdutils' followed by `apt-get install procps'. Everything seems to be OK after this... However, I was just wondering, is it really the case that both bsdutils and procps contains the program /bin/kill? Are they the same, or are they different versions? Is the alternatives mechanisms in place for /bin/kill? Apparently there's some problem with both packages containing the same file? Also, the weird thing is, `dpkg -S /bin/kill' only turns up procps, not bsdutils. And also, *both* packages have the man1/kill.1.gz manpage file -- one in /usr/man/ another in /usr/share/man/. Hmmm, interesting, which manpage gets displayed for /bin/kill then??? Are these two manpages supposed to coexist?? This does sound like a problem with these two packages... only I'm not quite sure for which package I should file a bug report. T
BuddyPhone for Linux?
Is there a Linux version of BuddyPhone? I can't seem to find anything on the BuddyPhone webpage... T
ipchains: zeroing built-in chain counters
I'm just wondering, why does `ipchains -Z' only clear the counters for the rules in each chain, but not the counters for the chain policy? Is there a way to clear the counters for the builtin chains' policies? Thanks! T
esound woes... (Was: Multiplexing /dev/dsp?)
I installed esound and tried to get amp to work with esd... but I can't figure out how. Plus, now I found out that esdplay doesn't even work. I can play sounds with esdplay when no esd daemon is running in the background... but whenever esd is running, it just exits immediately and no sound is played. Also, esdctl with any command simply exits without printing a single message. Documentation is scarce, so I've no idea what options I'm supposed to use for esd. Also, my original intention is simply to multiplex /dev/dsp... I'd appreciate it if someone could give me more detailed instructions than the scarce docs I can find, for doing this. What exactly must I pass as options to esd, what environment variables are required, etc..? I'm using kernel 2.2.12... does anybody know if this is actually compatible with esd??? (Though I've no reason to believe it's incompatible, since esdplay works fine when esd isn't running in the background.) T
Re: Multiplexing /dev/dsp
Hmm, I installed the ESound daemon, but I can't get my MP3 player (amp) to use esd -- it insists on going to /dev/dsp directly. I tried the esddsp script but it still doesn't work. Any clues? T
Multiplexing /dev/dsp
I'm just wondering, is there a way for multiple (unrelated) processes to share my DSP port? Is there any driver/modules/etc that multiplexes the DSP device? I'd like to have different programs that use DSP be able to share it. Is this even possible?? T
Re: Library glibc
On Tue, 21 Sep 1999, Manuel Arenaz Silva wrote: > Hello, > > I want to install an MP3 player. I want to download it from > "www.freeamp.org", but I have to chose between to configurations: > > * FreeAmp_1.3.1_for_Linux2.0.x_glic2.0_Intel_x86 > * FreeAmp_1.3.1_for_Linux2.0.x_glic2.1_Intel_x86 > > How can I know which version must I download? I have installed Linux > Debian 2.1 with Kernel 2.0.36. > > Is there a site where I can download this MP3 player in ".deb" format? http://www.debian.org/Packages/stable/sound/freeamp.html T
X server 3.3.4 is slow!
Hi! I'm running potato and just upgraded to the SVGA X server (3.3.4). I discovered that it was *extremely* slow -- the screen repaints are very noticeable and every time it repaints, it takes so much CPU that my background MP3 player (or is it the sound driver) jitters horribly. I'm using a SiS 6326 AGP chipset. I know this isn't very well supported by X, but the previous X server I used (3.3.3) didn't exhibit this slowdown problem. As a side note, I had to use `no_bitblt' and `sw_cursor' for either server to work properly (otherwise there would be strange color shifts when an area of the screen is rapidly updated). The 3.3.4 server worked fine without `sw_cursor' but exhibited the terrible slowdown problem. The 3.3.3 server required `sw_cursor' but was otherwise OK. I've temporarily switched the default X server to a copy of the 3.3.3 server. But I'd like to know why the 3.3.4 server is exhibiting this problem? I'd like to use the 3.3.4 server if I could. Also, does anybody know if XFree86 is planning to release a server that can actually support *accelerated* modes (specifically bitblt) on the SiS 6326?? I checked their website but didn't find much info. T
Re: Block stupid/annoying sites
On Thu, 2 Sep 1999 [EMAIL PROTECTED] wrote: > On Tue, Aug 31, 1999 at 10:26:28PM +0200, andreas palsson was heard to state: > > I am using Debian GNU/Linux as a nameserver, and I wonder how do I > > modify it to reject all lookups for stupid sites like > > "ad.doubleclick.net" or any other annoying banner-site? > > I've been told to use something called "junkbuster" but I rather not run > > anything extra on the host, I simply would like to change something in > > the bind-configuration. > > I know it's not a direct answer to your question, but I seem to recall > having seen doing this exact thing using the IP tools. I'm not exactly > sure, but I *think* it was in the IPChains HOWTO. The example was > blocking a site (ad.doubleclick.net) with ipchains. Read the IPCHAINS HOWTO. I think you can do something like: ipchains -A input -s ad.doubleclick.net -j DENY ipchains -A output -d ad.doubleclick.net -j DENY You probably want to tailor the above to meet your needs. This will block *any* kind of connection to/from that site, although DNS lookups will still work. Your browser will probably complain of timeout connecting to that site or something. I'm not quite sure about a cleaner solution. T
Re: Talk
On Fri, 27 Aug 1999, Stephen R . Gore wrote: > On Fri, Aug 27, 1999 at 11:52:10PM +0200, Tony Schonfeld wrote: > > > > Since i use Debian i can't use talk command > > i've always this message : > > > > [Your party is refusing messages] > > > ---end quoted text--- > > Debian's default config is "mesg n". To enable talk, the user must > enter "mesg y" at the command line or in .bash_profile (or whatever > shell they happen to prefer). Another possibility is that you are trying to talk from a Debian machine to a non-Debian machine... keep in mind that there *are* different talk protocols out there; if the other machine isn't using the same protocol, it won't work. T
Re: make-kpkg and 2.2.12
On Thu, 26 Aug 1999, Hans van den Boogert wrote: > At 06:25 PM 8/26/99 -0700, Eric G . Miller wrote: > >make (menu)config > >make-kpkg clean > >make-kpkg --revision=custom.1.0 kernel_image > >dpkg -i ../kernel-image-2.2.12_custom.1.0_.deb > > > >These four steps work fine for me (though I'm still using 2.2.10). > > Shouldn't there a make dep after make clean? Either `make-kpkg clean' or the `make-kpkg ... kernel_image' itself already performs a `make dep' automatically. T
Re: Oracle
On Thu, 26 Aug 1999, Daniel Yang wrote: > I am not sure what exactly you are looking for. Oracle doesn't provide free > Oracle DB and its tools for Linux and any other OS. However, you can get 30 > days trial CD for $4.95 each on some Oracle products across different > platforms, including Oracle8 for Linux. Check Oracle website for details. > Daniel > -Original Message- > From: ICON ICON <[EMAIL PROTECTED]> > To: debian-user@lists.debian.org > Date: Thursday, August 26, 1999 5:53 AM > Subject: Oracle > > > >Does anyone know of any free software for oracle development?? [snip] Try looking at http://technet.oracle.com/. They have free downloads (as long as you use it for personal purposes only) for various Oracle products, including Oracle 8i (8.1.*) which just came out recently. All you need is to sign up as a Technet member and you'll get access to the downloads. T
Strange behaviour of Apache JServ
Hi! I'm currently running a Java servlet that I wrote on Apache JServ. I'm getting this strange behaviour that every time I restart Apache, and then access the servlet, Apache returns a page complaining of an internal configuration/ runtime error. However, if I persistently reload the page, at some point the servlet suddenly starts to work, and from that point on, everything functions as normal, and the servlet can process queries without any errors. The annoying thing is that none of this internal errors show up in the Apache log files, so I've absolutely no idea if this is a bug in the servlet itself, or did I misconfigure something in Apache/JServ, or is it a bug with Apache?? I even tried having my servlet open a file in /tmp/ and writing diagnostic output into it -- but that file simply shows up *empty*. Any ideas what might be causing this problem and how to fix it? I'm using JDK 1.1 (Debian package 1.1.7v2-2), JSDK 1.0 (downloaded straight from Sun's website), the apache package 1.3.6-15.2, and libapache-mod-jserv 1.0-1. Also, possibly related is that fact that I'm using the MM MySQL JDBC driver (manually installed) which is loaded by the servlet's constructor. I know for sure it's NOT the actual database queries that cause this problem, because I tested it once with all database queries disabled. It might be the actual loading of the JDBC driver that causes the problem, though... (The driver *does* eventually load properly, because after the servlet suddenly decides to start working, all DB queries work perfectly.) T
Compiling servlets for Apache JServ
Hi! I probably missed something obvious, but I can't figure out how to compile servlets for Apache JServ. I have a file called `ServletTest.java' and that imports the javax.servlet.* packages. When I type `javac ServletTest.java', it complains that it cannot find the javax.servlet.* packages. I'm using jdk1.1-dev, and I've downloaded jsdk.jar from Sun's website. My guess is that I need to somehow specify the path to this .jar file?? I placed it in /usr/share/java because that's where Apache JServ expects it to be (I'm using the Debian package libapache-mod-jserv). How do you specify this path?? I tried setting CLASSPATH but obviously that doesn't work. Or perhaps I need to download something more than just jsdk.jar?? I tried looking through the docs but can't find anything that helps (got really frustrated after spending about an hour or more just looking for docs, both locally and on Sun's website.) T
Configuring PNP card *before* loading /etc/modules
Hi! I'm running a system with a PNP network card that needs to be up before /etc/init.d/network executes. I need to execute `isapnp /etc/isapnp.conf' *before* /etc/modules is read by the kernel (since the network driver cannot do anything until isapnp configures the network card). So when is /etc/modules read/executed?? Is it possible to arrange stuff in the right order so that things happen in exactly this order: (1) isapnp executes, (2) the network driver module loads (ie. /etc/modules loads); (3) /etc/init.d/network runs and configures the network. T
Re: boot messages too fast to read.
On Tue, 24 Aug 1999, Mark Wagnon wrote: > On Mon 08/23/99 11:31PM, David Teague wrote: > > > After boot, dmesg will give a good bit of the boot messages, but > > sometimes it fails to give parts of the boot message that I want :( > > I have this problem, and it drives me crazy. Sometimes I want to cut > and paste the info into my pleas for help, but end up resorting to > using a pencil and paper and retyping the message. > > The man page for dmesg says something about increasing the ring > buffer size, but does anyone know how to do this? > > inquiring minds want to know. It seems that certain messages are not recorded by dmesg: such as isapnp messages. (I've never been able to see the board ID and "activated OK" message in dmesg -- only on the console.) Anybody can explain why?? I'd like to have dmesg record *every* bootup message, so that I don't have to avoid switching VC's right after bootup in order not to erase important messages that may not appear in dmesg. T
Re: Kernel panic!
Hi all! Thanks for the replies... I managed to figure out what was wrong: my custom-compiled kernel had the IDE drivers as *modules* (and I didn't list them in /etc/modules) -- so after the kernel finished initializing, it discovers that it didn't know *how* to read /dev/hda1 because it hadn't loaded the IDE drivers yet!! That was what the kernel panic was all about. :-) Silly, silly, me... don't know why in the world I would configure the IDE drivers to be modules rather than compiled-in, but after managing to boot into the system with the older kernel on my Rescue Disk (which, of course, had built-in IDE drivers, not modules), I managed to recompile the kernel with the right config. Now the server is up and running. Thanks anyway for the help... T
Kernel panic!
Hi, please CC replies to my email (my mailbox can't handle the traffic on this list so I'm not subscribed, but I desperately need help.) I'm trying to set up a server box with Debian... I needed to use kernel 2.2, so I first plugged my HD into an existing Debian system and installed from there to get the required versions of the packages. Everything seemed to work -- I can even boot my new installation from LILO by adding the appropriate entries in lilo.conf (on my other machine). But once I moved the HD to the machine it's supposed to run on, it doesn't boot properly. I used LILO to install the boot sector on /dev/hda and root set to /dev/hda1, and the kernel seems to boot fine, but after detecting my Ethernet card, it gives this error: VFS: cannot open root device 03:01 Kernel panic: VFS: Unable to mount root fs on 03:01 What does this message mean?? The kernel booted from /dev/hda1 so why can't it mount the root partition now? I've tried giving Linux the different combinations of the "root=..." parameter but it doesn't seem to work. Does it matter which version of LILO I use to install the boot sector?? This system is running kernel 2.2.10 (custom build) and most packages are from potato. Due to some problems I had to use an older (slink) version of LILO -- will this screw it up? It seemed to be fine when I booted this same HD on my other machine (with the LILO setup for the other machine, of course). Also, if LILO is the cause of the problem, how would I run a glibc2.1 version of LILO from a slink rescue disk??? I don't have a potato boot disk handy... Thanks for any help/suggestions T