Re: Strange delays - new to this Please Help! Thanks
Hi, Todd! You should first mail to debian-user for questions like this. d-anounce is definitely the wrong place. I left the other list adresses in for this mail, so that people know that there was an answer to your request. When you reply, please consider to remove some more. I can only guess that you hosed your network setup. Although networking is well structured, it may well look confusing to a newbie at first. The reason for most delays are network packets send to unknown or unreachable hosts. Timeouts then are the only indication for the local host that sth. is wrong and they may take a while (you wouldn't want to loose your packets to mars because of timeouts). The most important network inside your linux box is the unix network. Everything else is for most installations TCP/IP networking. Some files in the /etc directory (where all the configuration for your machine is stored) store the basic networking configuration: hosts - your local 'DNS'. Hostnames to IP-Numbers table. Only used when host.conf - contains 'order host,...', see? The file resolv.conf - at last stores adresses of name servers. If any of your programs calls for a host by its name, this has to be translated (looked up) into the corresponding IP number, as the network itself doesn't know anything about names. If this lookup process is misconfigured, the resulting timeouts make up already for quite a delay. Now all your packets have to find their way through the network. This is known as 'routing'. See the 'route' and 'ping' commands. Most networking commands have an option like '-n', that makes them output (N)umbers instead of names. Trying one and the same command that hangs without -n once again with -n and it works, shows you clearly that you have a problem looking up names. KDE is communicating a lot, fix your network and retry. You found out that networking is very important these days. Try to read up on how this is handled with linux and the tools. You won't have to install obscure tools then, to try to make it, hm, work, as it did under this other non-free os ... There are a lot of good readings on the internet. For a fine catalog with well defined topics you might want to look at 'dmoz.org'. This one is FREE! Have fun! martin
Re: compiling a program
On Saturday 24 February 2001 16:05, Philipp Bliedung wrote: When I compile a program from the sources and it beakes for whatever reason, how can I get rid of all the files that were created so far If it's a debian package, cd to the unpacked src-packages top directory (the one containing a dir named debian) and use the command fakeroot debian/rules clean And how can I find out if a new version of a package, for example libc6, depends on any other package? For example I tried to compile a Newer packages have a Build-Depends line in their control file, olders haven't yet. There are some ways to view the control file, it is located in the debian directory of the source package however. The first entry should describe the source package, the following the binary packages built from it. The source description might have that Build-Depends line ;) greetings, martin -- Cross Building nano HOW-TO V 0.2 http://eisdoc.debox.de
Re: *.gz
On Saturday 24 February 2001 06:47, Chad C. Walstrom wrote: I admit it. I'm a UNIX elitist. I know not all people will not Me too ;-) realize the philosophies of the UNIX operating system, the small-tools approach, but if I can influence just one person into It's the greatest thing. What does mc(1) do? It taps into the libc libraries and cuts off some of the most powerful and flexible features of our system. The pipe is the bread and butter of UNIX. Why else do you think system I'm afraid you haven't seen mc for a while. It has some bugs in newer versions, but is still my everyday tool. It helps me keep an overview of all the small tools. And to a newcomer it opens up the view on her filesystem. Imagine, you're sitting in the dark (empty shell prompt) and type 'mcenter'. Suddenly the FHS in all its beauty shines in front of you, browsable with your fingertips (if you set Lynx-like-motion on). When i was new to all this Linux stuff (not unix) i had trouble to find docs, buried to usr/doc/package and usr/doc/package and man and info and ... you see - mc was THE way for me to get in control. I'm not going into a discussion about that topic. I simply repeat: unix is small tools and pipes, and ... that's what makes it good; a newbie gains an overview using mc! I understand your point of view and personally like it, but ... as above. greetings, martin -- Cross Building nano HOW-TO V 0.2 http://eisdoc.debox.de
Cross Building nano HOW-TO
Regarding recent questions, a hopeful seed of a Cross building micro HT has been laid to http://home.t-online.de/home/eislink/document/crosshto.txt Input | comments, pointers cheeringly appreciated. Be warned though that this seed has yet only been fertilized by partial success in my own efforts towards cross building. greetings, martin
Re: XFree86 4.0.2 - Very low resolution - how to increase? - Includes: How To RTFM :^)
On Saturday 17 February 2001 09:33, Anthony Campbell wrote: On 17 Feb 2001, Martin Albert wrote: On Friday 16 February 2001 09:41, Anthony Campbell wrote: I'm currently confused about X-4. I've been keeping up to date You're actually running the old 3.3.6 server, propably you didn't install a new one. If you dare (no, it works, at least) do apt-get install xserver-xfree86 I've done this but I can't get it to work. I ran xf86config and produced an XF86Config-4 file, but this refuses to run, with numerous error messages (various sections should begin with Driver.. etc.). I'll glady RTFM if there is one. YOU're actually RUNning the OLD 3.3.6 server. My driver is xserver-svga 3.3.6X (man dpkg ;-) dpkg -l x\* shows each package installed on your machine whose name starts with x. (the \ before * is an 'escape character' for the shell; says 'don't use this char with a special meaning you have for it, pass it unchanged'.; here dpkg shall have arguments -l and x*. info bash or man bash). If this gives so much output, that you cannot see the first lines, use dpkg -l x\* | grep server This passes all output, that dpkg produces through a pipe (told the shell with | symbol) to grep, which outputs only lines of input that have 'server' in them. You will propably see one: XF86-xxx3.3.6 You need: xserver-xfree86 4.0.x apt-get install xserver-xfree86 (man apt, if you have info you may read all man pages with it too, if you like it better (i use both). Caveat: I don't understand apt's man page either ;) Most interesting is a section each manpage has near its end: SEE ALSO. This gives references to related programs and config files. Once you installed xserver-, you can find lots of info in the directory /usr/share/doc. Every package isntalls docs there. Mostly just copyright blurb, but sometimes a wealth of more or less usefull info. If indoubt, give it a try. For X there is a lot of info, eg. on the XF86Config file and the Device drivers. greetings, martin
Re: Problem building X 4.0.2 from debian source
Hi, Frederik! Good to hear you keep going .. :) On Sunday 18 February 2001 20:38, Frederik Vanrenterghem wrote: I tried this (now with 4.0.2-3), but the error seems to remain the same: What I did was: debian/rules source.make cd /xc/programs/Xserver/hw/xfree86/drivers mv mga mga_old tar xvzf mga-1_01_05beta.tgz (to replace mga with the matrox drivers from Matrox) cd /xc/config/cf cp xf86site.def host.def I've edited host.def to include #define HaveMatroxHal YES, and also, I removed some of the drivers I will not be needing. After all this, I did make World, followed by touch debian/stampdir/build and debian/rules binary This is the error message I got: In file included from xftint.h:30, from xftcfg.c:28: XftFreetype.h:34: parse error before `_XftFTlibrary' XftFreetype.h:34: warning: type defaults to `int' in declaration of `_XftFTlibrary' XftFreetype.h:34: ANSI C forbids data definition with no type or storage class XftFreetype.h:37: parse error before `FT_Face' XftFreetype.h:37: warning: no semicolon at end of struct or union XftFreetype.h:53: parse error before `}' XftFreetype.h:53: warning: ANSI C does not allow extra `;' outside of a function make[4]: *** [xftcfg.o] Error 1 make[4]: Leaving directory `/usr/src/xfree86-4.0.2/build-tree/xc/lib/Xft' make[3]: *** [install] Error 2 make[3]: Leaving directory `/usr/src/xfree86-4.0.2/build-tree/xc/lib' make[2]: *** [install] Error 2 make[2]: Leaving directory `/usr/src/xfree86-4.0.2/build-tree/xc' make[1]: *** [install] Error 2 make[1]: Leaving directory `/usr/src/xfree86-4.0.2/build-tree/xc' make: *** [debian/stampdir/install] Error 2 Any more ideas, I'd appreciate them :-) (I'm pretty sure I could just copy the generated mga drivers over the standard apt-installed drivers to get a working X 4.0.2, but I would rather compile the whole thing myself... Understand. Me too. Just it was a while ago - don't remember the details. May be i ran into the same problem. Do you have freetype2-dev installed. The headers are required at least. I also installed my own host.def. But if you do that and still want to build debian binary packages, you have to disable/modify some checks in the debian dir. At the end of rules target install is a cmp with a MANIFEST file. Comment that line out. As i've also taken out files from the server, i had to grep the debian/*.files.i386 files for the relevant modules and delete them there. Keep going ..., greetings, martin
Re: XFree86 broken.
On Monday 19 February 2001 10:21, Eivind Arnesen wrote: I was just beginning to feel comfortable with Debian, when my X configuration broke down. I tried to remove XFree and install it again, but there is too many dependency problems. It looks like the package database is really screwed up. Just about whatever I try to do, I get some complaints about broken dependecies. I think maybe some of the dependent files, Don't give up ;) It is clear that a lot of things depend on x-packages when they run under X. As we don't know hear, what you mean by 'X configuration broke down', we follow your way to install from scratch. I don't know how you get your package files. apt from the net - sri, can't help. You have them stored locally - here we go: As always, have a backup ;) and rtfm: man dpkg; You can force dpkg to do sth. even when dependency problems would normally not permit to do so. We assume you know what you're doing :) I always try a command the normal way first, if dependancy-checks fail and i know it's the right thing to do, than use --force-depends. To show what packages you have installed starting with x: dpkg -l x\* now remove what you think is broken (try normal first, then, if you have to): dpkg --force-depends -P package Now reinstall what is missing: dpkg --force-depends -i debfilename You can always check the state of your efforts apt-get check Hopefully the list of unmet dependancies is getting shorter over time you do that. You can not use apt-get before all dependancies are ok. hth, martin
Re: Problem building X 4.0.2 from debian source
On Tuesday 13 February 2001 15:12, Frederik Vanrenterghem wrote: On Mon, 12 Feb 2001, Martin Albert wrote: On Son, 11 Feb 2001, Frederik Vanrenterghem wrote: using 'make World'. The error only seems to show up when I use 'debian/rules build', which is rather odd. Anyway, here is the debian/rules build applies patches found in debian/patches As you may have overlooked in my posting, I only get the error when I use 'debian/rules build' to build a debian package of X. If I execute 'make World' (also after applying the debian patches), everything seems to be OK (I saw the same error messages, but they do not cause the build process to die). Any ideas what might be going on? Might it have something to do with different levels of optimization? If yes, can this be altered? Well, not overlooked - but i might have guessed it ;-) So, you didn't clean the sourcetree after the last try with the debian setup, before doing make World? Then make World cd your_toplevel_that_includes_debian touch debian/stampdir/build debian/rules binary martin
Re: xdm takes ages to start after X upgrade SOLVEDish
On Friday 16 February 2001 13:28, David Purton wrote: I'v solved this prob by changing this line in xdm-config from true to false: ! All displays should use authorization. ! X terminals may not be configured that way, so they will require ! individualized resource settings. DisplayManager*authorize: false but it suggests it should be true - why? and if this is important can I set it up so there isn't the delay starting? Well, after that i recently started to find garbled fileentries in / containing MIT-COOKIES i'ld think this might make a point. martin
Re: Serial ports - how to get them to coexist peacefully...
On Friday 16 February 2001 17:27, Sean 'Shaleh' Perry wrote: Can I make the onboard and oncard ttyS's play nice on same IRQ? ... or should I play musical jumpers until they're on separate IRQs? I would change the jumpers. Hoping the software ca multiplex is a recipe for disaster. And won't work with ISA hardware. martin
Re: Special issues with X on a TFT screen?
On Friday 16 February 2001 15:04, Daniel de los Reyes wrote: I am helping a friend set his X on a Laptop. The graphic chip is .. setup file. I get the X server up,(the gray screen) but when it tryes to switch, the screen becomes black and there is no way to get it back. I suppose it is stepping out of the frecuencies allowed by the TFT screen. I am not very sure home a TFT screen works, it seems to me vertical frecuency is fixed? I haven't been able to get any information from the vendor. Any help? Be very carefull. I killed the first display of my laptop exactly that way. No reply from the vendor too, so still no X. SVGAText package has a utility in the source package to read the graphic chip registers under DOS. I don't know whether this works under Win. Try the linux-laptop http: site. They have a long list of laptops and evtly. an X setup for that model. martin
Re: XFree86 4.0.2 - Very low resolution - how to increase?
On Friday 16 February 2001 09:41, Anthony Campbell wrote: I'm currently confused about X-4. I've been keeping up to date with Testing and now I don't know which version of X I'm running. My driver is xserver-svga 3.3.6X. I tried running xf86config and this generated a XF86Config-4 file which didn't work. I therefore continued with my existing XF86Config. Can anyone explain briefly what the present position is with regard to X-4? I can only explain briefly what your problem is ;-) You're actually running the old 3.3.6 server, propably you didn't install a new one. If you dare (no, it works, at least) do apt-get install xserver-xfree86 martin
Re: XFree86Config bpp depth confusion
On Friday 16 February 2001 19:54, Robin Rowe wrote: Hi. I have a little confusion with XFree86Config. What is the difference between bpp and color depth? Section Screen DefaultColorDepth 32 This selects which of the following Display sections will be used. martin
Re: Antwort: Rebooting is foolish ....
Hi, Martin! On Friday 16 February 2001 15:54, Martin_Tanzer@dvs-berlin.de wrote: As far as I know you only might want to reboot if you change the hostname and want it active. The Linuxcommunity is proud of their uptimes, so we never reboot... martin And not even that is necessary, /proc/hostname and /etc/init.d does it. Remains: - kernel changes. - Evil proprietary, undisclosed display driver modules. - Having no UPS. martin
Re: Burning CD-ROMs as user
On Thursday 15 February 2001 13:07, Joris Lambrecht wrote: if it's a good idea to let all users use it, why not just change the execute permission to world permissions ? Also, i took a look at gcombust (yuck) and couldn't find a way to burn 12x cdrom's does anyone have any advice on this, i'm even thinking about commerial software right now, burning cd's seems (like browsers) rather rudimentary. It seems you have a RTFM problem? Do sth. against it or spend the $$ to have somebody else do it for you. Having setup your /etc/defaults accordingly, there could be nothing easier than 'cdrecord isofile'. Knowing how to use your Desktop-env. you might even have that as an icon or sth. But unitl you find the right item to click, i'm halfway through with burning :) man cdrecord is all you need. btw. cdrecord can not gain real-time privileges when started by a user. greetings, martin
Re: libggi
On Mon, 12 Feb 2001, Pollywog wrote: On Mon, 12 Feb 2001 03:56:51 +0100 (MET), Roberto Diaz said: example but I find I havent ggi.h nowhere :? libggi2-dev but I got it from woody It's actually just the same. (but i'm working on this) To have more output targets you may install libggi-target-x libggi-target-svgalib libggi-target-aa and so on. Have a look at the latter with the fire demo. greetings, martin
Re: Problem building X 4.0.2 from debian source
On Son, 11 Feb 2001, Frederik Vanrenterghem wrote: list, but now I'm fairly certain it's a Debian-specific problem, since I've succesfully build the debian source with modified matrox mga driver using 'make World'. The error only seems to show up when I use 'debian/rules build', which is rather odd. Anyway, here is the error message again, maybe someone knows why the build fails using the debian build process, and continuous using the 'normal' build process? (the error is obviously also encountered, but it doesn't cause the build process to fail). debian/rules build applies patches found in debian/patches greetings, martin
Re: XF 4.0 PROBLEMS
On Mon, 05 Feb 2001, [EMAIL PROTECTED] wrote: XF 4.0 won;t run on my machine. XFCom_Rage128 Version 1.3/ X Window System (protocol Version 11, revision 0, vendor release 6300) Release Date: 6 September 1999 As the msg clearly says, you simply don't run an X4 server. You have to install a server, which is simple now, as there is only one left: xserver-xfree86. Config Error: /etc/XF86Config:150 Section InputDevice ^ not a recognized section name xinit: Server error. This is an X4 config statement _that should be in XF86Config-4_. hth, martin
Re: console-tools installation error
On Fri, 02 Feb 2001, Michael Abraham Shulman wrote: Trying to switch from kbd to console-tools (testing distribution), I get this error when configuring console-tools: /var/lib/debconf/config.53881: null: command not found console-tools failed to configure, with exit code 127 Any ideas what's wrong? Deinstall (dpkg -P kbd) kbd first. martin
Re:Russian porn: offtopic
On Fri, 02 Feb 2001, DSC Lithuania wrote: You know, I'm sitting here using Microsoft Windows and Explorer. (I do use linux, but not for this system.) And one of the things that makes me just a little bit paranoid about those things is the question of whether there might be a new virus attached. After all, there is a huge list of viruses at McAfee, and most of them are first posted to one or another list server. And I know that they say -- if you don't open the attachments or link to the sites -- then you're pretty safe. And in general that is correct. But nonetheless, I Which is NOT true. http://www.heise.de/ct don't think that there is anyone sane who doesn't believe that Microsoft has left themselves a back door to every system. And such a back door can be exploited. It doesn't need any backdoor. Just scripting/ActiveX :) greetings, martin
Re: Still No Cursor in X
On Thu, 01 Feb 2001, D-Man wrote: | I try PS/2, Buttons 5, ZAxisMapping 4 5 - but strange things happen on | the screen then. May be you have it for me. My Logitech wheel mouse (wasn't cheap -- cost ~$40 and worth it) works fine (or at least it did with RH 6.1 and 7.0 I don't have a new enough GNOME on Debian yet to use X much) for me. Try setting the Buttons to 3. Buttons 4 and 5 don't really count since they are used as the wheel. Hm, Buttons 3 just turns the wheel off. The XF..config man says to have the ZAxisButtons added. The mouse I have is the Logitech MouseMan Wheel -- it has 2 normal buttons, 1 thumb button and a scroll wheel that doubles as a button. It is a nice large mouse that has the thumb side raised up in a natural way. It's the only mouse I have that hasn't hurt my hand. It also supports USB and PS/2. I only use it for _very_ short periods, mostly to move that cursor out of the way :) (does Konqueror support the TAB key to navigate the links? the _only_ advantage of IE) greetings, martin
Re: Log Files
On Wed, 31 Jan 2001, Michael Abraham Shulman wrote: How do I get rid of huge files in /var/log like `xdm.log' and `nmb', and/or prevent them from accumulating in the first place? Can/must I just delete or clear them every so often? Yes and OR or man logrotate hth, martin
Re: What is bounce-debian-user ?
On Wed, 31 Jan 2001, USM Bish wrote: Hi folk, Just a small clarification. The mail placed below was received by me today from bounce-debian-user. This was NOT initiated by me. Quite surprised to see my name on the LOG (List of Greats!). Just the queue to return the mail to the subscribers. martin
Re: Still No Cursor in X
Hi, nice people on this list. I hope it's ok for you (and for you, John), when i cc my answers to pm here. So many of you did already take part. So here the story continues. We're getting closer, it seems ;-) On Thu, 01 Feb 2001, you wrote: Martin, here are some of the answers that I found. Window-manager: I selected enlightnment because my Learning Debian GNU/Linux book says that I need X and then GNOME before I insatll StarOffice. I have StarOffice on KDE, no problem, integrates nicely. ls -al /dev/gpm*: Today I found that both gpmctl and gpmdata were there. gpm -v = 1.17.8 yeah! Linux ver, distro etal = Debian from LinuxMall.com obtained while I was at COMDEX (Las Vegas), Linux is ver 2.2.17. XF86_SVGA: This is my X server. The ver I got from either man or info was 3.3.5 Videdo card: diamond Stealth, 3D 3000 with 2048K Mem. Below is some other info: GeForce DDR, Quadro, ET4000, ET4000W32, ET4000W32i, ET4000W32i_rev_b, ET4000W32i_rev_c, ET4000W32p, ET4000W32p_rev_a, ET4000W32p_rev_b, ET4000W32p_rev_c, ET4000W32p_rev_d, ET6000, ET6100, et3000, pvga1, wd90c00, wd90c10, wd90c30, wd90c24, wd90c31, wd90c33, gvga, r128, ati, sis86c201, sis86c202, sis86c205, sis86c215, sis86c225, sis5597, sis5598, sis6326, sis530, sis620, sis300, sis630, sis540, tvga8200lx, tvga8800cs, tvga8900b, tvga8900c, tvga8900cl, tvga8900d, tvga9000, tvga9000i, tvga9100b, tvga9200cxr, tgui9400cxi, tgui9420, tgui9420dgi, tgui9430dgi, tgui9440agi, cyber9320, tgui9660, tgui9680, tgui9682, tgui9685, cyber9382, cyber9385, cyber9388, cyber9397, cyber9520, cyber9525, 3dimage975, 3dimage985, cyber9397dvd, blade3d, cyberblade, clgd5420, clgd5422, clgd5424, clgd5426, clgd5428, clgd5429, clgd5430, clgd5434, clgd5436, clgd5446, clgd5480, clgd5462, clgd5464, clgd5465, clgd6205, clgd6215, clgd6225, clgd6235, clgd7541, clgd7542, clgd7543, clgd7548, clgd7555, clgd7556, ncr77c22, ncr77c22e, cpq_avga, mga2064w, mga1064sg, mga2164w, mga2164w AGP, mgag200, mgag100, mgag400, oti067, oti077, oti087, oti037c, al2101, ali2228, ali2301, ali2302, ali2308, ali2401, cl6410, cl6412, cl6420, cl6440, video7, ark1000vl, ark1000pv, ark2000pv, ark2000mt, mx, realtek, s3_savage, s3_virge, AP6422, AT24, AT3D, s3_svga, NM2070, NM2090, NM2093, NM2097, NM2160, NM2200, ct65520, ct65525, ct65530, ct65535, ct65540, ct65545, ct65546, ct65548, ct65550, ct65554, ct6, ct68554, ct69000, ct64200, ct64300, mediagx, V1000, V2100, V2200, p9100, spc8110, i740, i740_pci, i810, i810-dc100, i810e, Voodoo Banshee, Voodoo3, smi, generic From XF86 Config. Section Screen Driver SVGA Device Diamond Stealth Monitor FairView DefaultColorDepth 8 BlankTime 0 SuspendTime 0 OffTime 0 SubSection Display Depth8 Modes640x480 640x400 512x384 320x240 320x200 ViewPort 0 0 EndSubSection SubSection Display Depth15 Modes640x480 640x400 512x384 320x240 320x200 EndSubSection SubSection Display Depth16 Modes640x480 640x400 512x384 320x240 320x200 ViewPort 0 0 EndSubSection Now, as Nate and Manuel suggest, try the 'Option sw_cursor'. Leave all your other lines in that section as they are, they are for Manuels board. He just posted the whole Section, so you can see WHERE to insert: Section Device Identifier Elsa Winner T3D VendorName Unknown BoardName Unknown VideoRam8192 ChipId 0x8a10 Option sw_cursor --- # Insert Clocks lines here if appropriate EndSection You may also try, what happens, when you change DefaultColorDepth 8 in your Screens Section to 15 or 16. I WILL GO SKIING NEXT WEEK And Thaks again John Hopefully everything is fine until than. (than with an 'a' here?) greetings to all kind souls, maritn, martin. (Did i mention, i hate those Fujitsu keyboards since 1985?)
Re: Security Related Questions
On Thu, 01 Feb 2001, Pietro Cagnoni wrote: Second question. I have a mystery service running on port 1024. I fairly certain I haven't been hacked, but, who knows? When I do a nmap localhost port 1024 shows up as unknown. How can I find out which daemon is bound to 1024? Thanks in advance! fuser 1024/tcp # as root! man fuser tells a lot of cool stuff! read it, and enjoy! pietro. right. And 1024 is the first of unassigned ports, used temporarily for backchannel ... greetings, martin
Re: modules quit loading automatically all of the sudden
On Thu, 01 Feb 2001, Daniel Barclay wrote: I have a strange problem. All of the sudden, various modules don't load when I boot (sound modules for a basic SoundBlaster and ftape). I haven't been messing with my configuration. Things worked last time I booted a couple of days ago. No apt-get, new kernel installed? Sure? Any ideas what might have changed? What I might have changed accidentally? modprobe says: modprobe: Nothing to load ??? Specify at least a module or a wildcard like \* Any ideas about what I should look at? modprobe sb what happens? greetings, martin
Re: What is bounce-debian-user ?
On Thu, 01 Feb 2001, Nathan E Norman wrote: 1) Sending this mail once was a great plenty. Three times seems a bit excessive, don't you think? Lot's of mail came in here doubled, tripled during the last days. May be the slowness of the list during the last days has sth. to do with it ... martin
Re: Still No Cursor in X
On Mit, 31 Jan 2001, you wrote: Albert, I tried the configuration you suggested. I got a cursor in the Please let me introduce myself as a man of wealth taste and the name is martin 8-) text mode but not in X. I was able to lite up menus. In X? Did you already try to find out about your X-server options? You did again not tell what server window-manager you are using After the ls -al /dev/gpm* I only got the one line regarding gpmdata as .. What does it mean that the gpmctl entry is not there. This is a good question indeed. It took me nearly half an hour now to find out and i'm still not quite sure. This device entry 'named socket' is created when gpm is started, for instance on system start or when you manually run '/etc/init.d/gpm restart'. When this does not happen on your system, sth. is strange. As i didn't work with named sockets yet, i could only make wild guesses. It might be that your gpm is simply too old. Check the version with 'gpm -v'. I've got 1.17.8 and it works fine. You've got networking in your kernel (this is silly, nearly impossible you haven't). Do you run a debian system at all? Anyway this is strange. What kind of a system are you runnign there? If you can, try reinstalling gpm. Post where you've got your packages from, ideally the versions of gpm and X-server, it's name and the gfx card. Thanks John [EMAIL PROTECTED] you're welcome greetings, martin
Re: Still No Cursor in X
On Wed, 31 Jan 2001, Andrea Vettorello wrote: As someone else suggested, try stopping gpm before launching X or trying to configure X (via some graphic configuration utility that right now i don't remember the name, where is dexter =). I don't think, that is hte problem. The mouse moves, just no or distorted visible cursor. Do yourself a favor, don't try another distro ;) I already tried to convince him, to at least jsut go skiing instead ;-) And we're all still here. P.S. If all fails, why not spend 10 bucks to buy a cheap OEM Logitech wheel mouse (i became addicted to that darn wheel), so i could give you my configuration that can confirm it's working perfectly right now =) Why not send it to me. I gave up on that. Either the old KDE or Fujitsu sells a crippled Logitech. I try PS/2, Buttons 5, ZAxisMapping 4 5 - but strange things happen on the screen then. May be you have it for me. greetings, martin
Re: Still No Cursor in X
On Wed, 31 Jan 2001, Andrei Ivanov wrote: I'm not sure whether this has been tried yet When I upgraded to X4.0.2, I did lose the cursor (it was moving but invisible). Turned out the xserver was not upgraded, so it was still using the xserver-svga 3.3.6, instead of using xserver-xfree86. Once that was installed, cursor came back. Andrei Couldn't this be the real thing? AFAIK, a simple apt-get dist upgrade dos not install the new X-server. You will have to do an explicit 'apt-get install your-required-x-server' To check the version of your server is a little tricky - mostly because you don't tell us your video card or x-server that you use and we don't even know, whether you've got X3 or X4. (Folks with X4: i don't know this yet, please interfere when i tell rubbish) But let's try: In a terminal, your console for instance, enter 'XF86_' Then, don't hit return yet, hit TAB twice. The first time it may just beep, the second time you get a short list of programs in your path that start with XF86_. One of them is your X-Server ;-) IF you know which one you should use, give enough characters to distinguish from the other names. You may at every time try to hit TAB again, it will then complete as much as is unambigous. That command is completed (the whole name of the program in question on the command line, your cursor behind it)? Ok, type a space and '-showconfig'. Now hit return. The first line of output reads 'XFree86 Version 3.3.6' with my server - i don't have X4 installed - and you now know why ;-) If you don't know your server, you may try 'X -showconfig'. Now, tell us about the ouput and which debian version you are using. When you upgrade over the net with apt-get, mail your '/etc/apt/sources.list' file. Otherwise the CD you try to install from or anything, just to give us any hint - please ;-) So much for today martin
Re: building package with debug options?
On Mon, 29 Jan 2001, [EMAIL PROTECTED] wrote: Is there an easy way to compile a Debian source package with debugging options turned on (removing the -O2 and adding -g to all the make I'm afraid dpkg-buildpackage --debug??? isn't quite ready :) For a 'brute force' solution have a look at the 5liner of the 'pentium-builder' package. By default it builds with -mpentium, but what's in the script is lastly just up to you. HTH, greetings, martin
Re: French local under X
On Son, 28 Jan 2001, Yvon VODOUNON wrote: I need help to set French local under X. Yesterday i added 'export LC_ALL=de_DE to my .profile Most programs honour it. Have a look for the french code, could be fr HTH, greetings, martin
Re: Start up?
On Mon, 29 Jan 2001, Andrew Clark wrote: The easiest way I have found to do SMB mounts at boot time is to put entries in /etc/fstab for them. IIRC: mount -t smbfs //winpc/share /mnt/winpc/ -o username=Administrator,password=blahblah translates to: //winpc/share /mnt/winpc smbfs username=Administrator,password=blahblah 0 2 You must have 'smbmount' installed for this to work. greetings, martin
Re: Debian - Testing network (FS105/FA311/10T-2)? CMD/SiS IDE?
On Fri, 26 Jan 2001, hogan wrote: Have SiS IDE controller in my current Debian box - it says something about bus-mastering disabled in bios, running in a more basic mode. Leave it be? Try to enable it? I can't see any option in the bios that would affect this unless bios is hard coded to disable it. enable at least DMA for IDE in BIOS if possible. /var/log/messages, /proc/ide and hdparm /dev/hda tell you about your drives man hdparm about hdparm greetings, martin -- Is there a key combination to make sure IE will reload a page? Oh, yes and it's shorter than Alt-X O Alt-N Alt-Z ret ret Ctrl-F5!
Re: my NIC
On Sat, 27 Jan 2001, Henry Gomersall wrote: I have a D-Link DFE-520TX ethernet card connecting me, via a cable modem to .. This particular card cannot be selected from the list of NICs during installation (of the base system from the rescue disk). This then means I .. (ftp.dlinknet.com/PUB/Drivers/linux.zip). Unfortunately the base system doesn't have such added bonuses as the kernel source or Make. :) I know nearly to nothing 'bout Network cards - i may be totally wrong here and i even forgot the DFI Cards name i got installed once, but Why not try to insmod/modprobe some of the driver modules from /lib/modules/x.y.zz/net ? The module will tell you when it was able to initialise the card. For this card i'ld start with de4x5.o and tulip.o. wishing success, martin -- Is there a key combination to make sure IE will reload a page? Oh, yes and it's shorter than Alt-X O Alt-N Alt-Z ret ret Ctrl-F5!
Re: Compiling source .debs
Anyway to get this to work with 'apt-get --compile source pkgname'? It would be really nice if there were a way to override the default compile flags on a system wide basis. Have a look at the simple doings of the pentium-builder pkg. The idea is quite extendable. greetings, martin -- Is there a key combination to make sure IE will reload a page? Oh, yes and You Know It!
Re: fstab entry for cd-rw device?
a CDR, not RW, on my box. So the 'ro' option makes sense. it saves you a 'file system is read-only, mounting ro' msg. ;-) karsten, you see? Not only: At the end of your life it is not important, how you've done it, but _that_ you've done it. greetings, martin -- Is there a key combination to make sure IE will reload a page? Oh, yes and it's shorter than Alt-X O Alt-N Alt-Z ret ret Ctrl-F5!
Re: problem playing audio cd's
On Fri, 26 Jan 2001, Harry Barnes wrote: Hi, I had the same problem as Phillip. I was advised to add the user to audio and disk ... I have already undone this, after reading the post. My problem is I do not seem to have /dev/cdrom To get gtcd to play I had to change the device from /dev/cdrom to /dev/hdc. Looking through the /etc/group file I noticed that I have a cdrom group. Any ideas? Intermediate newbie :) Harry, you have taken the user out of group disk and left him in audio? ok (shouldn't this be cdrom? unsure for the moment, my cdroms belong to the cdrom group) anyway this is a policy but its your system. The device file in /dev, eg. /dev/hdc, should be owned by root, but belong to group, say, cdrom. This is to be able to give different access rights on the device. root must do everything, the group may read the cd, but you don't want anybody lurking in your system to do anything with it. $ ls -l /dev/hdc /dev/cdrom -rw-r- root root hdc then do (as root) # chgrp cdrom /dev/hdc $ ls -l /dev/hdc /dev/cdrom -rw-r- root cdrom hdc that is good either you get lrwxrwxrwx root cdrom cdrom - /dev/hdc that is good! or nothing. Then make symbolic link yourself (as root) # rm -f /dev/cdrom; ln -s /dev/hdc /dev/cdrom and the user must be in the group cdrom # adduser user cdrom $ id user, (user),(cdrom) Now he may do what the permissions for the group allow him to. I'm not sure (root (me) grabs and i i play mp3s), but to program the cdrom to play you must also be able to write to it (?) Anyway, user starts a program and the program gets/runs with users permissions - user may read something, so may the program. happy hacking, greetings, martin -- Is there a key combination to make sure IE will reload a page? Oh, yes and it's shorter than Alt-X O Alt-N Alt-Z ret ret Ctrl-F5!
Re: Two Mice and no Cursor
On Fri, 26 Jan 2001, hammack wrote: Long time and still no cursor. I've tried two pointers, Kensington Expert Mouse serial, and a PS/2 Generic. I have tried all the combinations of Protocol and Device that I can and still don't have a pointer in X. Below are files and excerpts of interest. Any Help appreciated John [EMAIL PROTECTED] Today. all the installations seem crazy to me. You really mean in your home ? Where you got that from? dir of home files for the below: What is p and c in the 2nd and 3rd lines? gpmdata appears to be an empty file and psaux is very small. lrwxrwxrwx1 root root7 Dec 20 22:49 mouse - gpmdata prw-r--r--1 root root0 Jan 25 15:33 gpmdata crw---1 root root 10, 1 Jul 5 2000 psaux XF86Config: Section Pointer ProtocolMouseMantry PS/2 first, simplest Device /dev/mouse ok for PS/2 ## Protocol PS/2 ## Device /dev/gpmdata BaudRate1200 Emulate3Timeout 50 Resolution 100 # Buttons 3 Emulate3Buttons EndSection BUT here: gpmconf: device=/dev/gpmdata poor gpm must read own output responsiveness= repeat_type=raw type=MouseMan append= Make that you have: /dev/psaux rm -f /dev/mouse; ln -s /dev/psaux /dev/mouse /etc/gpmconf: device=/dev/mouse type=ps2 If you installed gpm properly /etc/init.d/gpm start (or restart if runnign) Then have fun with: man gpm XF86Config greetings, martin
Re: How can I install VMware inside Debian?
On Fri, 26 Jan 2001, Dr. Aldo Medina wrote: but now it says my kernel-headers don't match my kernel. I got kernel-headers-2.2.18-1 and a custom-kernel compiled from kernel-source-2.2.18-1. What do you think? With modules? And no make clean / distclean after that? make config before :^) I'm using 2.2.17 - fine. Won't compile under 2.4, yet, errors with headers. Only thing, you have to get used, after each problem/kernel change you have to run hmware-config.pl anew. It will happily reinstall/recompile with your saved answers. I now like to have fresh modules in my kernel ;-)
Re: Can't create file larger than 1G on console ?
On Wed, 24 Jan 2001, Tommy Wu wrote: I've try kernel 2.4.0, 2.4.1-pre8 - 10. I can't create file larger than 1G on console (tty1 - tty6 and X11 terminal), but I can create such file from telent. For the same OS, kernel... only a telnet session can create large file more than 1G. Or I switch to single user mode, I can create such file ok. (use the command: dd if=/dev/zero of=largefile bs=1k count=150) I've tried the same kernel to another machine, use the same debian 2.2r2 linux, it's ok to create such large file, so I think that is not a kernel problem. I also try kernel 2.2.18, there is no such problem. So any suggest for this problem ? ulimit ? greetings, martin -- Is there a key combination to make sure IE will reload a page? Oh, yes and You Know It!
Re: fstab entry for cd-rw device?
On Wed, 24 Jan 2001, Stefan Srdic wrote: I'm trying to get my cd-burner working under Debian 2.2. I've configured my system so that the ide-scsi module is loaded at boot time and I've I know that the module is loaded and functional because I can see the However, I'm not sure which paramaters to add in the /etc/fstab file. My thoughts are to add something like the following: /dev/scd0 /cdburneriso9660 defaults,user,noauto00 Will this work? Or is there a better way to mount my cd-rw? This should do fine. I prefer to mount the rw as /dev/hdd, IDE, and when I actually want to use the burner, i have a script to unmount/unload ide and start the scsi environment. The direct ide reads seem to be faster with my drive at least. I just a had a look, that script si too well documented to be mine. Maybe you'll find it in cdrecord/contribution? Else ask me. greetings and fun with yo new? rw, martin .
Re: mkisofs
On Tue, 23 Jan 2001, David B. Harris wrote: To quote Benjamin Pharr [EMAIL PROTECTED], I'm trying to make an iso of a directory structure on my hard drive using mkisofs. I have Rock Ridge and Joliet enabled (along with a few other things), but it gives me an error saying the files with the same name, but in different directories have the same Rock Ridge name. Then it says Unable to sort directory. Anybody know what is going on? Since people seems to be putting down Joilet and Rock Ridge, I figured I'd mention that I've successfully created numerous images where both types of extensions were enabled. David Barclay Harris, Clan Barclay I second that . mkisofs 1.12 made my last bu as well as any other: mkisofs -r -J -V ebtmpbu -o iso ~/bu Do you try multisession? Are you messing with deep directories? martin
Re: Soundcard Problems
On Sun, 21 Jan 2001, Tristan wrote: I have an ES1869 PnP Audiodrive, IRQ 7 DMA 1 DMA 0. I am having problems ... run isapnp, but isapnp doesn't detect my soundcard, and i tried using sndconfig but it couldn't detect any soundcard either, the soundcard is part of the motherboard, and i can get it to work by using the sb 100% How do you come to the conclusion that it is PnP? I had to configure a laptop using that chip with a fixed/bios controlled setup. If it works with the compiled in driver (that needs to know irq/port/dma in advance) modprobing the module with apropriate options should work. There was some weirdness with 16bit dma though. I may have used dma16=-1 or sth. greetings, martin
Re: Linux 2.4.0 root fs problem (*help*, *help*)
Sven - who is desparate to run 2.4.0 # # Multi-device support (RAID and LVM) # CONFIG_MD=y # CONFIG_BLK_DEV_MD is not set # CONFIG_MD_LINEAR is not set # CONFIG_MD_RAID0 is not set # CONFIG_MD_RAID1 is not set # CONFIG_MD_RAID5 is not set CONFIG_BLK_DEV_LVM=y CONFIG_LVM_PROC_FS=y try to leave this out ... ? greetings, martin
Re: emu10k1 module
On Mon, 15 Jan 2001, Benjamin Black wrote: alias sound-slot-1emu10k1 how do you figure this out? i mean where exactly do you look to find out what the standard aliases are for sound card, network card, video card, tv card, etc? where exactly did you read about sound-slot-0? Hi! I know you would like to have an answer on your second question. Me too ;-) Sorry, can only answer 1 3: Leave the aliases out and have a look at your syslog about x-y-z not found at the moment some program tries to load them. That's how i do it. ;^) I log everything to virtual console 12, so it can see it immediately, with the (additional) line in /etc/syslog.conf: *.* /dev/tty12 (actually, to avoid clutter, it reads: *.*;cron.!=info;mark.!debug /dev/tty12) Well, to at least give a little bit to Q 1 3, i just had a look at the kernel source doc (did they CLEAN that 2.4 source tree, man): Documentation/modules.txt, Doc/soudn/introduction at least mention the concept. The definite source to the numbers will be the source - you will have to grep it for them. greetings, martin
Re: Parallel attached CDR - recomendations?
On Sat, 06 Jan 2001, Nate Bargmann wrote: I am looking for a method to archive a bunch of stuff off of a couple different computers and since one is a laptop and one a desktop, it seems a parallel attached CDR drive might be the answer (I'm not Hi, Nate! May be you're better off with some network cards (even 10Mb is faster than most CDRs). Sure, it depends on what else you're going to do with your CDR. The network solution has many advantages, that you have to fix everything to get your CDR going only in one place, is only one. ^-) 73, greetings, martin (DG2AMA)
Re: emu10k1 module
On Sat, 06 Jan 2001, Anthony Fox wrote: I have auto module loading compiled into my kernel. How do I specify that when an application attempts to access the sound hardware that the kernel should install the emu10k1 module into the running kernel? Right now, the kernel installs the soundcore module only. Thanks, Anthony Hi, Anthony! You should edit in /etc/modutils, then update-modules. You should end up with a /etc/modules.conf that includes ... (my lines (for two soundcards)): alias sound-slot-0esssolo1 alias sound-slot-1emu10k1 greetings, martin
Re: buggy vesa framebuffer.
On Thu, 04 Jan 2001, Kent Nyberg wrote: I compiled the kernel with vesa framebuffer and edited lilo to: vga=0x301 append=console=/dev/tty2 CONSOLE=/dev/tty2 ... But when i for exempel, do an 'ls' in a dir with lots of things in it.. then the framebuffer gets buggish.. (dont know how to write buggy in that form.) ... At first.. ls prints out the things as it should.. but when the consol starts to scroll, then its only the bottom of the screen that scrolls. The top of the screen dont scroll at all. I have had this problem before, and not just with the card (TNT2) i have now. The i740 i had before did the same thing in vesa framebuffer. Do any one know how to fix this? Hi, Kent Have a look at the vesafb docs ... Kernel/vesafb options include 'ywrap' which does not work with my TNT2 :( - should be faster. So i have to use 'ypan', slower but works. My cmdline: ... video=vesa:mtrr,ypan,pmipal vga=791 HTH, greetings, martin
Re: Debian checks filesystems at every boot
on Tue, Dec 12, 2000 at 08:35:28PM +0100, Nico De Ranter ([EMAIL PROTECTED]) wrote: 2.2 cd's. However with the new PC every time I boot Linux will complain the filesystems weren't properly unmounted and will do a full check of the filesystems. I do ofcourse always make sure to turn off the machine by typing halt. Do you have 'Delayed write' enabled on your *drives*? May be after playing around with hdparm? Or as a default on the drive? Just another guess on this. greetings, martin
Re: I'm Confused with SBLive!
On Thu, 14 Dec 2000, Javier Sieben wrote: Hi! I tried to install the emu10k1 module from ALSA. Having any troubles with it, I deinstalled it. When I read that the final kernel (version 2.2.17) had these module in it, I download, compile and install it with the modules. But I can't use audio. The kernel starts and don't install the module. When I use cat /proc/modules it's list all the modules loaded at the time. But no the this module. Try loading with insmod. Fine. When I use cat /proc/modules it displays that the module is unused. I tried to many forms for solving this trouble. I read docs at many sites I've found and nothing. All refers at usr/src/linux/Documentation/sound for the specified card docs. But in the latest kernel hadn't included docs for it. Can any help?? Sorry may english, I'm newbie in it... Javier Hi, Javier! In my /etc/modules.conf i have two lines (for i'm using 2 soundcards): alias sound-slot-0 esssolo1 alias sound-slot-1 emu10k1 this works for me, the drivers are drawn automatically when i use the mixers or start kde HTH, greetings, martin
Re: Weird Error
I just installed Debian 2.2 and compiled the 2.2.17 kernel on a 3.2 gig
Quantum Fireball EX. The drive is a slave on the primary IDE cable and
Starting deferred execution scheduler: atdhdb: dma_intr: status=0x51 {
DriveReady SeekComplete Error }
hdb: dma_intr: error=0x84 { DriveStatusError BadCRC }
...
hdb: irq timeout: error=0x84 { DriveStatusError BadCRC }
hdb: DMA disabled
ide0: reset: success
Some older Quantum drives are at least blacklisted for overestimating
their capabilities when reporting their best PIO/DMA-Modes.
Personally, i use an EX 4.5GB ok in DMA Mode, but it also claims to do
UDMA (can't verify, don't have a controller for that).
You should really try turning DMA off with
hdparm -d 0 /dev/hdb
martin
PCI probs with Triton I
Hi, 'users' ;-) i know most of you aren't so i dare to ask: My last Q to this list received no reply (PIIX, 82371 DMA probs). May be i should have added the hexdump of kmem to be more specific? Starring at the intel 82371FB docs and kernel-sources driver/block/ide-* and driver/pci/* for two nights as well as perusing setpci now gave the following, which puzzles me even more: My IDE does not DMA because the Function 1 (IDE) of the 82371FB does not even appear in the list of PCI devices!, only Function 0 (ISA Bridge). Every try with setpci even with direct acces methods, also a small prog using the pcilib, gives no response for PCI 0:7.1 The only other intel computer here is using the PIIX4 with an 82371AB that lists its subdevices correctly. I even installed windows (see what i'm going through?) to see if it recognizes the IDE functions - it doesn't, only ISA Bridge on which it is running a 'generic busmaster driver'. Who is responsible for the PCI Headers - the BIOS (which is surely broken)? Do i have to initialize the chipset to make the subdevices available? There's not any indication of this in the docs. Pls, i really need this. Does anyone know of anyone i may talk to? Among the web sites i searched during the last days were kernel.org, linux-ide suse.de. May be i missed sth.? _smallest font on_ any answer, please :) martin
Re: sound volume
On Tue, 28 Nov 2000, Andrei Ivanov wrote: Well...there's volume knob on back of some soundcards :) Install a mixer app for X...or tehre's a volume controller in xmms. Andrei also lots of console mixer apps out there. try aumix for example. martin
Re: Converting to SCSI. HELP!!!
On Tue, 21 Nov 2000, Brandt Dusthimer wrote: and 2, I'm also usually the file sharing server. so you know you're good ... ;-) However, when everyone's uploading/downloading to/from me, I experience major system lag and if do anything (even simple command prompt stuff) there upload/download time increases a lot. SO, being the all powerful linux user that I am, !!! I have decided to start using SCSI. I am under the assumption that SCSI will help. If I am wrong. . . well I don't exactly have money spurting out my ears ( I bag groceries ), please tell me that I'm wrong. What's the computer LANCard of yours? Are you sure that your system is using IDE BM-DMA or at least DMA? First of all check that the drives are working in their best mode. hdparm -v /dev/hd? will show whether DMA is used. Of interest is also DRIVE read-ahead. You can tell the capabilities of your DRIVEs with hdparm -i /dev/hd?. You will find a entry MaxMultSect. Use that number to set MultSect with hdparm -m number. Second is the LANCard. If it's an ISAtype or a too simple PCI one, expect overhead code executed there that sums up under heavy traffic. Besides other info sources, the boot-messages may already tell you about your drives lancard. A debian 2.2 copies them to /var/log/dmesg A girl in our network has a Fujitsu IDE 15G doing 15MB/sec and an IBM IDE 15G doing 13 MB/sec. Her LANcard is a cheapo ne2kpci and while she's gaming she can only tell by the noise from her hard drives that some guys are using her PC as the compiler station ... greetings, martin
Re: System.map does not match kernel data
On Tue, 21 Nov 2000, Liam Ward wrote: I'm getting the following message when I run ps/top: Warning: /boot/System.map-2.2.17 does not match kernel data. The message comes from a kernel symbol checking routine in (as far as i remember) misc/symbol.c in the ps source tree. I _knew_ for sure my symbols were correct, so i changed the '#if 1 --- #endif' that encloses exactly that statements to '#if 0'. not quite a fix but, guess what, no Warning message anymore 8:) greetings. martin
Re: Star Office (OpenOffice)
on Tue, Nov 21, 2000 at 09:09:22AM +0100, Max Reiss ([EMAIL PROTECTED]) wrote: is there somewhere a Star Office or OpenOffice deb? Should you just be concerned that installing the standard sun download would in any way break your system - that is not the case. Just download, run: xxx0.bin -net and install to /usr/local/soffice52. Everything will be below this dir. Every user can simply install from there (run /usr/local/soffice52/setup) and will have some files copied to its home-dir. KDE users will find menu links in .kde/applnks after that. greetings, martin
PIIX, 82371 DMA probs (again?)
Howdy hackers, i'm stuck ... Searching debian.org, kernel.org, etc. i can not find a solution. search.debian.org gives just some BUG closed msgs and some people giving up on the issue ... Now i already downloaded the intel 82371FB docs, being prepared to initialize the PCISet myself. But as i've never dealt with (modern) PC hardware before, i'm looking forward to some hard times. That doesn't bother me too much, but the machine in question is the server of a little network with, err, little to no hardware for doing backups as they should ... and after all it's the IDE controller we're gonna play with ... So, i would very much appreciate any help or pointers of yours ... Thank You! Computer is IBM 350-P75 with newest BIOS (1996) lspci 00:00.0 Host bridge: Intel Corporation 430FX - 82437FX TSC [Triton I] (rev 02) 00:07.0 ISA bridge: Intel Corporation 82371FB PIIX ISA [Triton I] (rev 02) CONFIG_PCI=y, CONFIG_PCI_GOANY=y CONFIG_PCI_BIOS=y, CONFIG_PCI_DIRECT=y CONFIG_PCI_QUIRKS=y, CONFIG_PCI_OPTIMIZE=y # CONFIG_PCI_OLD_PROC is not set kernel-2.2.17 with ide-patch both from potato-proposed-updates CONFIG_BLK_DEV_IDEPCI=y, CONFIG_BLK_DEV_IDEDMA=y, CONFIG_IDEDMA_AUTO=y CONFIG_BLK_DEV_PIIX=y, CONFIG_PIIX_TUNING=y now dmesg can tell better than i: Linux version 2.2.17 ([EMAIL PROTECTED]) (gcc version 2.95.2 2220 (Debian GNU/Linu x)) #11 Sat Nov 18 05:09:06 CET 2000 Detected 74540 kHz processor. ide_setup: hdd=cdrom ide_setup: ide0=dma ide_setup: ide1=dma Console: colour VGA+ 80x34 Calibrating delay loop... 148.68 BogoMIPS Memory: 63580k/65536k available (668k kernel code, 412k reserved, 836k data, 40k init) ... CPU: Intel Pentium 75 - 200 stepping 05 Checking 386/387 coupling... OK, FPU using exception 16 error reporting. Checking 'hlt' instruction... OK. Intel Pentium with F0 0F bug - workaround enabled. PCI: PCI BIOS revision 2.10 entry at 0xfda51 PCI: Using configuration type 1 PCI: Probing PCI hardware ... apm: BIOS version 1.1 Flags 0x03 (Driver version 1.13) Real Time Clock Driver v1.09 Uniform Multi-Platform E-IDE driver Revision: 6.30 ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx PIIX: IDE controller on PCI bus 00 dev 38 PIIX: chipset revision 2 PIIX: not 100% native mode: will probe irqs later hda: QUANTUM LPS540A, ATA DISK drive hdb: QUANTUM FIREBALL EX5.1A, ATA DISK drive hdc: QUANTUM FIREBALL1280A, ATA DISK drive hdd: 4X4X32, ATAPI CDROM drive ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 ide1 at 0x170-0x177,0x376 on irq 15 hda: QUANTUM LPS540A, 516MB w/96kB Cache, CHS=560/32/59 hdb: QUANTUM FIREBALL EX5.1A, 4892MB w/418kB Cache, CHS=662/240/63 hdc: QUANTUM FIREBALL1280A, 1222MB w/83kB Cache, CHS=2484/16/63 Clearly, the hdx messages are missing the DMA at the end! And yes, the EX5.1A even claims UDMA, all do mwdma2 (for instance on a PIIX4). (What about the PCI Bus speed? Driver docs say for P75 25MHz would be correct? The ibm hw manual says nothing about any particular board.) Looking into the kernel/driver/block/ide-* sources indicate that the test (ide-pci.c::ide_setup_pci_device::565 --- ide-patched version!) if (e-reg (pci_read_config_byte(dev, e-reg, tmp) || (tmp e-mask) != e-val))/*continue*/; /* port not enabled */ is met. So, the driver meets the condition after the loop: if (!at_least_one_hwif_enabled) printk(%s: neither IDE port enabled (BIOS)\n, d-name); This can not be seen in the above dmesg, because this is already from a version, where i tried to initialize the PIIXa set with the PIIX4 routines. The only outcome is another msg: PIIX: IO baseregs (BIOS) are reported as MEM, report to [EMAIL PROTECTED]. ok, it was just a try ... As i said above, after finishing the mail (and compiling qt - what the heck are libgl-dev and libmng-dev it build-depends on?) i'll start reading the manuals, but i'ld rather prefer joining some linux-midi-audio project than to re-invent a solution for a long closed BUG. Though this was a long post, that i hope do not annoy too many of you, i'm still in hope that so. of you might be able to help me out ... greetings, martin
Re: Internet connection
On Mon, 20 Nov 2000, Knud Sørensen wrote: Hi Hi, Knud! I have installed debian 2.2 and have some problems getting the internet to work. I have noticed something strange. /etc/init.d/inetd is a empty file ! So, your on your way - good! if I move it to /etc/init.d/inetd.bak and try to dpkg -i netbase_3.18-4.deb I get a error. Why !! Good question. A snippet of the error-message in question might help. I might also be a problem with my /etc/gateways file Does anybody have a example I can see!! debian now handles all the setup in /etc/network/interfaces. - iface eth0 inet static address 192.168.0.5 netmask 255.255.255.0 gateway 192.168.0.1 - BUT, a temporary default route to the internet will be setup along the workings of ppp (???) so I don't have to do anything. Uhm, you see i'm not quite sure may be should shut up (anybody else?). This is because i have setup ipchains to forward any non-local adress to ppp. 3) Why is there no link to /etc/init.d/networking form /etc/rc5.d ?? Hope you can help ! wish i could've done better ;-) next time. greetings, martin
Re: System.map
Just kinda curious...how important is it to place the new System.map file (created when you compiled a new kernel) in the /boot directory? I have compiled a number of kernels and, after looking at this discussion, I checked my /boot directory to see if I had moved my System.map file after the last compile. I have not. I have noticed no ill effects. What potential problems am I opening myself up to here? AFAIK the System.map just provides symbols for debugging kernel problems. procps uses it to tell you the wait-channels as the names of functions instead just plain numbers ... martin
Re: 386 install
Btw. forget it. I saw a pentium/100 with 8MB RAM, and it was unusable slow. apt-get-installing a 20kb-package took 5 Minutes(!). After upgrading to 24MB RAM (didn't check 16MB), it was a cool server, even able to run small php3-scripts in a fast manner. As always it only depends on what you're going to want your computer to handle. I have had an 8MB-386 Laptop to work as a portable protocoll-analyser, scope, etc. - no X, just SVGALib and console. To compile it's own kernel would take 1,5 hours, lots of swapping, etc. Does it have to do this - any other machine on the network did it faster and so they did it for him - hey, that's linux, where we all work together ;-) But beware, i had no luck with any debian BINARY packages after the bo release (very ancient) to run on a 386, the kernel hung right after unpacking itself. I will try a debian install on _4_ (four) MB 386-25 in a few days. Still looking around for the details of cross-compiling (to 386 and m68k), as i think this is the only possible way to get things going. The 386 is needed as a single client web-server browser for a christmas party in mannheim, roadcase on December 25th, 8pm. Currently it is running an old SuSe-Kernel, decoding radio-wheather faxes all day long. The 32MHz ATARI-TT (greetings respect, Kerstin!) will be my Desktop-PC. So have fun do your thing, but remember, you will have to compile that stuff yourself! greetings, martin
Re: fetchmail won't fetch
On Sat, 18 Nov 2000, ktb wrote: I'm a little farther along with my mail trials. Out going works fine but I can't receive with fetchmail. I'm using exim, fetchmail and pine on Potato. I use sendmail, so i don't understand the implications of your configuration, BUT: I can't see where you give info to the progs that [EMAIL PROTECTED] is [EMAIL PROTECTED] With sendmail i have an Alias '[EMAIL PROTECTED] [EMAIL PROTECTED]' AND fetchmail gets poll smtp.peoplepc.com with proto POP3 user [EMAIL PROTECTED] there with password is [EMAIL PROTECTED] # NOT: [EMAIL PROTECTED] here Your fetchmail log looks good. Try running fetchmail in the foreground with debugging output. The man-page is admittedly somewhat hard to read, but ... martin
Re: connection problem with T-Online
On Wed, 01 Nov 2000, TrEB wrote: I can't get a stable connection with my T-Online provider. ... to Mobilcom's Freenet provider all is working fine. ... t-online disconnects after 4-5 secs Hi, sorry for being late (prob: sendmail 8.9.3-23 compiled with the PICKY_HOST option doesn't go well with kde1.2/kmail1.0.28)! Most providers use Layer2 X75/LAPB protocol. t-online uses HDLC. With the kernel isdn module it's Register 14 S14=0 LAPB S14=3 HDLC You know this is your problem, if the answer from the D-Channel is BUSY (to be found in the syslogs). hope it helps ;-), martin
Re: rawrite broken from DOS-box under W98? (Was: Re: thinkpad install prob)
On Sat, 20 May 2000, Karl M. Hegbloom wrote: Robert Every floppy I've written in a DOS-box so far wrote fine but Robert didn't work, so I strongly recommend booting into real DOS (via Robert F8 when Starting Windows 9x...). Have other people exerienced this? (I've never used Windows = 3.1, so I don't know.) CrossBrand - BIOS upgrade/reference disks corrupted by Windows 95/98 Applicable Countries: Worldwide Service Hints Tips Symptom: Performing a DIR command in the Windows 95/98 environment on a Flash BIOS update diskette or a PS/2 Reference Diskette may cause the diskette to become inoperative. This affects ALL IntelliStations, PS/ValuePoint, IBM PCs and PS/2 systems. Problem Isolation Aid: None ;-) Fix: Microsoft uses an 8 byte block on the diskette for labelling purposes in Windows 95/98. When a DIR command is executed, this 8 byte block is written to by Windows 95. The Diskette Drive A: Icon under the My Computer folder in Windows 95/98 has the same affect as the DIR command. IBM uses this same 8 byte block for Flash BIOS updates and PS/2 Reference Diskettes to ensure that the Flash is occurring in a stable environment. Once the 8 byte block has been written to by the Windows 95/98 DIR command, the Flash diskette or Reference Diskette will display an error indicating that This is not a valid Flash Disketteor This is not the correct Reference Diskette for this system. - roger? greetings, martin
Re: init - autostarts
On Sat, 20 May 2000, w trillich wrote: Dominic all that starting syslogd, starting inetd etc Dominic when you're booting, it's controlled by init, right? Dominic how do i tell it not to start the things I'd rather Dominic not be running? Mostly you will want to have syslogd, initd ... running. syslogd does the logging for you, inetd starts not so often used daemons. The scripts can be found in /etc/init.d, links to them for each runlevel are in /etc/rc[S1-6].d. Read about 'sysvinit' and turn Snn scripts to Knn scripts to adjust your runlevel setup. if you're wondering about things like scsi-initialization and isdn setup, then i say me too. they seem to happen BEFORE the /etc/init.d/* scripts get called from /etc/rc*.d/S* . . . how about it? where can we turn off scsi probing or isdn scans? Build a kernel without those devices / modules. greetings, martin
Re: Stuck... --mc!!
Le sam, 20 mai 2000, Alain Reinhardt a écrit : I always also suggest to try 'mc' (assuming MidNight Commander is installed) because it is always a good feeling for a new user to go as far as a file manager of some sort -in this case a very good one. Nothing means more when you begin than knowing you have something like a copy, edit, delete control over the beast. ! And If you go that far, please use the 'see' command of MC to look at files in /etc. You will immediatly feel you have just successfully made a first step in the new world of freedom. Good luck ! Mais oui, monsieur, d'accord! I second this - i'ld say so: I always suggest to have 'mc' (MidNight Commander) installed, because it is always a good feeling! I wouldn't want to live without it. People always get crazy, when they're having trouble with their X-Windows-98 and the first thing that i do is to pull the nc or mc disk out of my pocket ... Happy computing! martin
?texinfo where?
Hello, to everybody! Call me silly - i want to compile a source package that build-depends (pls. developers, add this line to more pkgs) on 'texinfo'. I can not find a (source-)package with that name? Where will i find it? Thanks to repliers ;-) all the busy developers, greetings, martin
