safe load average
After the sound of my system thrashing being a better alarmclock than the conventional variety, I started checking my system's load average (among other things) periodically. From what I saw in uptime, my average load average tends to range from 1.0 to 3.0, with occasional spikes up to 6.75 (which I doubt is healthy for this P100), and a minimum of about 0.5. I know that uptime(1) says that Load averages that are consistently high (that is, all three number reported by uptime are higher than 1) indicate an overtaxed system. (I just noticed a typo in the manpage). And I know that 1 is the maximum processor performance on a single processor system.. The thing is, is that this system is a single processor system, and I think it has a slight temperature problem. As it is, this system is a laptop with the heatsink directly below the keyboard (so using ice or or adding fans would be somewhat inconvenient), and by how hot my PCMCIA cards are when I remove them, I half expect CPU0 on fire to come up. The maximum operating temperature listed in my manual is 35C (internal or external, I'm not sure which, but I'm hoping external), while the cards feel like they're at least 45. So, something I was wondering about would generally be considered a maximum safe load average. I have had some problems with some hardware that look almost as if this machine wants to become a toaster (hopefully the hardware wasn't damaged, only confused, by the heat). -- Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Installation problem
On Tue, Apr 30, 2002 at 12:33:47PM +0200, Miroslav Mazurek wrote: I'm trying to install Potato (2.2r3) on PC with 386/387, 16MB, 2GB SCSI IBM, Advansys SCSI, ATAPI CDROM and NE2000 compatibile. Does this make any sense? Would that configuration be usable for anything? Any suggestions? I run SCO on such configuration years ago. It should be usable, just watch what you install. I'm writing this on a system that is a little newer (a P100 laptop) but also with 16M RAM, a 2GB harddrive, and a SCSI CDROM. This system is used (but a little slow) for everything that I do. I merely have an average load average of about 1.5 - 3.0, with occasional spikes to ~7 (don't let this scare you... I'm on a network that _requires_ java to use the internet, and I'm online while compiling kernels, and in X, and doing homework with programs that recommend a system with 2-3 times the processing power of mine, and running a terminal off my serial port, and half a dozen other things at the same time). Problem is: I have Debian on CDROM but I can't boot from CDROM. I made rescue and root disc and 4 or 5 driver discs. (I also tried with compact which have one driver disc - it should also support advansys controller) After beginning of instalation I am ofered to load modules from disc, I supply driver disc but inst. program says - cannot mount floppy. I remember having some floppy problems when I installed debian a few months ago. I also remember that each time there was a problem with a floppy, it was a bad block or two that caused it. I know from experience that it is somewhat difficult to mount a floppy when its first couple of sectors are bad. After that I cannot do disk partitioning because advansys is not recognized. Any idea? You can always try a more recent set of disks. Checking the kernel-config for the 2.2r4 installation floppies, advansys is compiled in. The current potato is 2.2r6 -- Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
ethernet problem, driver or hardware
I've only just gotten this system onto the network (after 4 months (that long?!) of trying). The probem of the past was my borked ethernet card. It has just been replaced (warranties are good for something), but now I have a possibly not hardware problem. Simply stated, after this system is online for a few hours I have problems. A few different error messages come out, but they all seem to be related. Some lines from the logs: Apr 27 11:26:10 icosagon kernel: NETDEV WATCHDOG: eth0: transit timed out Apr 27 11:26:10 icosagon kernel: eth0: Tx timed out, lost interrupt? TSR=0x1, ISR=0x80, t=101. Apr 27 11:26:10 icosagon kernel: Hw address read/write mismap 1 Apr 27 11:26:10 icosagon kernel: Hw address read/write mismap 2 Apr 27 11:26:10 icosagon kernel: Hw address read/write mismap 3 Apr 27 11:26:10 icosagon kernel: Hw address read/write mismap 4 Apr 27 11:26:10 icosagon kernel: Hw address read/write mismap 5 Apr 27 11:26:14 icosagon kernel: eth0: Tx timed out, cable problem? TSR=0x1 ISR=0x0 t=322 Apr 27 11:48:13 icosagon kernel: device eth0 entered promiscuous mode Apr 27 11:48:14 icosagon kernel: device eth0 left promiscuous mode Apr 27 19:53:45 icosagon kernel: eth0: MII is missing! It also dropped its interrupts a couple of times, followed immediately by it reconnecting. So far, three times today I've been hit by a flood of the first seven lines (and the corresponding hundreds of log entries). The thing that makes me wonder about the cause, is that these messages stop after I ping another system. So right now I am wondering about what I need to fix to stop this. I am currently using my 2.4.18 kernel with the driver (axnet_cs) compiled as a module. The hardware side is a Linksys Etherfast PCMPC100v3. The rest of the network side is a collective of windoze boxen. Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
a closed bug reappears
About a month ago a bug (not yet archived) that I had filed was closed. I couldn't reproduce the bug after using a newer upstream version. The problem is, is that the bug is back with the packaged version of this newer source, and I'm not entirely sure as to the best way of sending in the information. Do I send a message to the old bug report, directly tell the maintainer, or do I just file a new bug report. Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: sh: /usr/X11R6/lib/X11/xkb/xkbcomp: No such file or directory
Charlie Grosvenor wrote: I am using Debian woody and am using the xserver-s3 I can go into X ok but in the information that gets logged to the console I get the following: sh: /usr/X11R6/lib/X11/xkb/xkbcomp: No such file or directory Couldn't load XKB keymap, falling back to pre-XKB keymap Is xbase-clients installed and configured? _FontTransSocketUNIXConnect: Can't connect: errno = 2 Could not init font path element unix/:7100, removing from list! If you don't have a font server, you can go into /etc/X11/XF86Config and comment out the line: FontPath unix/:7100 Please wrap your mail. Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: fakeroot
Sean Perry wrote: On 09-Apr-2002 Theo Bierman wrote: Hi All I am trying to build a .deb package using fakeroot from any *.tgz file. And I kepe on getting : fakeroot debian/rules binary dh_testdir make: dh_testdir: Command not found make: *** [thread-stamp] Error 127 Any ideas quick hint, look in the debian/control file of the package you want to build. There is likely a line Build-Depends. make sure you install every package in that list. Is debhelper installed? dh_testdir is in the package debhelper. Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: confused: primary and extened partition
lilifan2 wrote: Almost all books and docs say that primary partitions can contain extended partitions. And they can. My system has 4 primary and 4 extended partitions, with the extended partitions contained in hd3. 1. I don't think primary partitions can physically contain extended parttitions, b/c we can set up only one primary partition as 20mb and only one extended partition 30 mb which is larger than the primary one. So let us do not say primary partitions contain extended ones, let us say extended parttions are associated w/ primary ones. The extended partitions _are_ contained in a primary partition, a larger one that you don't usually see. 2. Still, my question is: how do know which extended partitions are associated to which primary ones? e.g. hd1 20mbprimary hd2 34mbprimary hd5 234mb extended hd6 100mb extended Is hd5 associated to hd1 or hd2? Or we are not supposed to know this? The system will keep track of it? hd5 looks like it is associated to hd3, which would contain hd5 hd6. Type dmesg|grep hda on the command line and see the partitions listed. The extended partitions are the ones contained between the '' '', and are contained in the preceeding primary partition. Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Modeline Calculator
Phil [EMAIL PROTECTED] wrote: I need to calculate valid modelines for a configuration. Does anyone know where in the world (or on the web) that modeline calculator is??? /usr/share/doc/xserver-common-v3/VideoModes.doc.gz It shows how to calculate a modeline. If needed, I can send it directly to you... it's a bit large to be posted (~23k). Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LILO doesn't work!
black.fish wrote: hi! i've got a problem with my debian woody. when i installed it there was an error message like: writing lilo... hmm, i think you're writing a fs ... this might be an important bug it came when i the installer was just writing the base of woody. since i can't boot In /usr/sbin/liloconfig: Hmm. I think you're configuring the base filesystem, and I'm therefore simply going to exit sussessfully without trying to actually configure LILO properly. If you're not doing that, this is an important bug against Debian's lilo package, and should be reported as such Since you were just writing the base, liloconfig didn't bother to properly configure LILO. Check /etc/lilo.conf for any problems. linux normally. when i start my pc without a boot disk, the screen looks like this: L 40 40 40 40 40 40 40... and it doesn't stop printing 04 on the screen until i reset with a boot disk. i looked at /boot and noticed that the boot.b is uncomplete. there's a error message like error while writing... at the end of the file. i tried to rewrite lilo. then i downloaded the latest version of lilo and tried again but it didn't work. what shall i do? please help me! A couple months ago, I got the L 40 40 40 40 ... after rearranging some partitions. It was because LILO didn't like the the line boot=/dev/hda5 (how that was set, I can't remember, but I think that I used (for the last time) liloconfig), which refered it to a logical partition. It works with boot=/dev/hda. All I can say is, check /etc/lilo.conf for errors. And for a statement about the error, I refer you to /usr/share/doc/lilo/Manual.txt.gz: Disk error codes [...] 0x40 Seek failure. This might be a media problem. Try booting again. (Yes, that's all there is for that error code) Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: ps is not working
Rainer Sigl wrote: On a debian machine I have the problem that after a reboot ps and similar commands as 'top' aren't working. The command simply hangs and does no response. Every else is working well beside openafs didn't work too after reboot because openafs-client (updated by dselect) was a newer version as the kernel module. What reason is possible for this? Obviously the /proc directory has something to do with the malfunction of the program. I have Reiserfs, Kernel 2.4.13 Sounds similar to a problem I once had. If you umount /proc (using -l) then ls /proc, are any files listed? Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: kcalc math bug?
Mike Fedyk wrote: On Sat, Mar 30, 2002 at 12:43:52AM +0100, Vincent Lefevre wrote: On Fri, Mar 29, 2002 at 15:33:36 -0800, Mike Fedyk wrote: Package: kcalc Priority: optional Section: utils Installed-Size: 392 Maintainer: Debian QA Team [EMAIL PROTECTED] Architecture: i386 Source: kdeutils Version: 4:2.2.2-5 run these calculations: 1x2+5x50 (in this order, not in normal precedence order that computers and etc use...) And it outputs 252 not 350! It should be 252 anyway. Any calculation tool should use the conventional precedence. NO. Not if I enter it in a specific order. 1x2 = 2 2+5 = 7 7x50 = 350 Not 252! That is how a calculator should work when you are typing in the calculation by *hand*! I shouldn't have to worry about precedence then. This app is supposed to replace the calculator I can buy for $10. And that is how it should work. And 252 is what I get on my $20CDN (~$10US) calculator. If you want 252, remember BEDMAS. Remember, we are working with hand entered calculations that go into the app in a certain order, and *that* is the order that it should be calculated... And hand entered calculations on my calculator are calculated in the proper order, contrary to what you would want. It looks like you wanted left-to-right operator precedence rather than using order of operations. Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: L 80 80 80 80, etc. on boot
Martin Edward John Waller wrote: I've done a potato install but can't boot from the hard disk (have the floppy boot working OK tho'). I get the above message (L 80 80 80 continuosly). Is there a way to fix this? I installed lilo diecrtly onto the mbr. Does the information in /etc/lilo.conf match your system? What do you get when you run lilo? From /usr/share/doc/lilo/Manual.txt.gz: L error ... The first stage boot loader has been loaded and started, but it can't load the second stage boot loader. The two-digit error codes indicate the type of problem. (See also section Disk error codes.) This condition usually indicates a media failure or a geometry mismatch (e.g. bad disk parameters, see section Disk geometry). [...] Disk error codes [...] 0x80 Disk timeout. The disk or the drive isn't ready. Either the media is bad or the disk isn't spinning. If you're booting from a floppy, you might not have closed the drive door. Otherwise, trying to boot again might help. Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Vim help -- not finding help.txt
Rohan Deshpande [EMAIL PROTECTED] wrote: In Debian unstable, I tried to do :help, but got this error: Sorry, help file /usr/share/vim/vim60/doc/help.txt not found So, I tried gunzipping the help.txt.gz in that directory, and then Vim complained it could not find tags. I figured it was because the files in /doc were not gunzipped, but Vim then complained it couldn't find help.txt.gz! Any ideas what to do? Did you just upgrade from vim 5.6 (a couple people had problems with the helpfile after the upgrade (its in the archives))? What is in your /etc/vim/vimrc? Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How to reconfigure fetchmail?
Stan Brown [EMAIL PROTECTED] wrote: I need to reconfigure fetchmail on one of my woody machines. I tried using dselct to delte reinstall it. That did not prompt me for configuration data, so I tried dpkg-rconfigure fetchmail. This just returned a prompt. How can I do this? Install and run fetchmailconf. Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Problems with CD-ROM and mail server
When I try to use a CD Player program, it works fine under root, but under a normal account it tells me that it can't access /dev/cdrom. I believe I have the /etc/fstab file configured correctly so I'm pretty clueless on how to fix this. I have pasted my fstab file on the end of this email. Are you part of group cdrom? Seneca [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: vim help file Error detected while processing BufRead Auto Commands for*.gz
John F Davis [EMAIL PROTECTED] wrote: I have the latest vim. When I try to issue the :help command I get the following error: Error detected while processing BufRead Auto commands for *.gz: E21: Cannot make changes, 'modifiable' is off E434: Can't find tag pattern Hit ENTER or type command to continue What gives? This issue came up on monday night (EST). vim has changed. http://lists.debian.org/debian-user/2002/debian-user-200203/msg03259.htm l Seneca [EMAIL PROTECTED]
Re: No puedo bootear Debian.
Bill Moseley [EMAIL PROTECTED] wrote: From: Daniel [EMAIL PROTECTED] To: debian-user@lists.debian.org Subject: No puedo bootear Debian. Date: Fri, 01 Jan 1904 01:00:41 -0600 Ok, a few hours delay in delivering messages is understandable, but this is a bit much. No wonder the subject seems to mean that debian cannot boot. My systems can't even _pretend_ to be that old, my oldest's BIOS only goes back to 1980, (a 286), and it can't boot debian. If a circa 1980s system cannot boot debian with 640k RAM, then how can the Analytical Engine (the only computer-relative I can think of that is that old) be expected to. Seneca [EMAIL PROTECTED]
Re: debianized vim
news [EMAIL PROTECTED]; on behalf of; Harry Putnam [EMAIL PROTECTED] Setup: Woody/testing After installing vim with apt-get, I find that the debianizations that So, vim 6. have been done cause sections of my .vimrc, built up over time on several different platforms to be unusable. [...] Some thing in this code cause a mess with debians setup. The helpfiles when accessed in the normal way inside vim with :h subject give this warning and then appear in binary. Error detected while processing function GZIP_read: line3: E21: Cannot make changes, 'modifiable' is off E434: Can't find tag pattern When I hit `enter' the file appears in binary format. This looks familiar... Maybe my code can just be edited somehow so there is no conflict. Otherwise I'd like to gunzip the helpfiles and redo the tags stuff. Now I remeber where I saw it... I saw it after I had upgraded from vim 5 to vim 6. To get the helptext back, I looked at the default vimrc provided for vim 6. The ending of my vimrc now is: if has(autocmd) filetype plugin on Seneca [EMAIL PROTECTED]
Re: can't start x after woody upgrade
Chris Jenks [EMAIL PROTECTED] wrote: I know that this has been covered recently, or at least I think it has, but I was not able to find it in the archives after 2 hours of reading (probably just couldn't find the right thread). I went from 2.2r3 to 2.2r5 to Woody, and installed Xfree86 4.1.0-14 after going to Woody (no x before that). startx doesn't seem to be on the box. I'm thinking that I'm missing a package, but if I am, I don't know what one. A pointer to either would be appreciated if someone knows what I'm missing. Here is what I have: [a long listing missing the desired package] I did take things like x-chat and man page entries out of this list. At the bottom of http://www.debian.org/distrib/packages, under the heading Search the contents of packages, you can put in a file name and be told what package it belongs to. [EMAIL PROTECTED]:~$ dpkg -S startx xbase-clients: /usr/X11R6/man/man1/startx.1x.gz xbase-clients: /usr/X11R6/bin/startx [EMAIL PROTECTED]:~$ Seneca [EMAIL PROTECTED]
update problem
I just did my semi-frequent update of my system. Because of a hardware problem, I currently rely upon floppies to do all of my file transfers (the affected system has not connected to another system since 5 years or so), so I can only do a large transfer every couple of weeks or so. Before I could install the large packages, I encountered some difficulties: - every time I start up the system, fsck tries to run on root file system, which is mounted - /dev/hda2 (/usr) and /dev/hda7 (/var) never unmount cleanly - xdm is not running while top reports it as running, and it looks and functions like it is running - ps catches Signal 11 every time I try to use it, and while shutting down - chkrootkit reports: [...] Checking `lkm'... Signal 11 caught by ps (procps version 2.0.7). Please send bug reports to [EMAIL PROTECTED] You have21 process hidden for ps command Warning: Possible LKM Trojan installed [...] - pcmcia never starts - nothing goes to xconsole anymore - root's path statement is ignored umount: /dev/hda2: not mounted umount: /usr: Illegal seek umount2: Device or resource busy umount: /var: device is busy done. The list of packages that have been updated and newly installed is below. Does anyone have _any_ idea (other than downgrading these packages or reinstalling my entire system) on how to deal with this? At the moment, I am most concerned about the problems with my root account, mount, ps, and the output from chkrootkit. For some reason, I have a sneaking suspicion that there is some connection between them. parted_1.4.24-1 tiger_2.2.4-19 john_1.6-17 acct_6.3.5-30 bastille_1.2.0-6 debsums_2.0.1 aide_0.7-11 integrit_2.03.02-1 harden-tools_0.0.12 harden-servers_0.0.12 harden-localflaws_0.0.12 harden-environment_0.0.12 harden-doc_0.0.12 update_2.11-4_i386.deb tasksel_1.15_i386.deb tar_1.13.25-2_i386.deb sysvinit_2.84-2_i386.deb syslinux_1.66-1_i386.deb sysklogd_1.4.1-10_i386.deb slang1_1.4.4-7_i386.deb pcmcia-cs_3.1.31-7_i386.deb mount_2.11n-4_i386.deb modutils_2.4.13-3_i386.deb makedev_2.3.1-57_all.deb libpam-runtime_0.72-35_i386.deb libpam0g_0.72-35_i386.deb libpam-modules_0.72-35_i386.deb klogd_1.4.1-10_i386.deb gettext-base_0.10.40-1_i386.deb fdutils_5.3-5_i386.deb elvis-tiny_1.4-18_i386.deb dpkg_1.9.19_i386.deb bsdutils_2.11n-4_i386.deb base-passwd_3.4.1_i386.deb base-files_3.0.2_i386.deb base-config_1.33.9_all.deb binutils_2.11.92.0.12.3-6_i386.deb binutils-h8300-hms_2.9.5.0.37.4_i386.deb gettext_0.10.40-1_i386.deb make_3.79.1-12_i386.deb nqc_2.3.r1-3_i386.deb tclreadline_1.2.0-4_i386.deb binutils-doc_2.11.92.0.12.3-6_all.deb cfi-en_3.0-4_all.deb developers-reference_2.9.0_all.deb doc-base_0.7.11_all.deb doc-debian_2.2.5_all.deb funny-manpages_1.3-4_all.deb gcl-doc_2.5.0.cvs-1_all.deb gnupg-doc_2000.10.01-1_all.deb info_4.0f-1_i386.deb libpam-doc_0.72-35_all.deb lskb_08-2000-1_all.deb man-db_2.3.20-13_i386.deb parted-doc_1.4.24-1_all.deb fnlib-data_0.5-6_all.deb imlib-base_1.9.11-3_all.deb imlib-progs_1.9.11-3_i386.deb cpp-2.95_2.95.4-1_i386.deb expect5.24_5.30.1-11_all.deb expect_5.32.2-4_i386.deb expectk5.24_5.30.1-11_all.deb expectk_5.32.2-4_i386.deb gcl_2.5.0.cvs-1_i386.deb imlib1_1.9.11-3_i386.deb libcupsys2_1.1.14-2_i386.deb libfreetype6_2.0.8-1_i386.deb libgd-gif1_1.3-2.deb..deb libgmp3_3.1.1-13_i386.deb libjpeg62_6b-5_i386.deb..deb libldap2_2.0.23-1_i386.deb libmysqlclient10_3.23.47-6_i386.deb libpaperg_1.1.5_i386.deb libparted1.4_1.4.24-1_i386.deb libxaw6_4.1.0-13_i386.deb t1lib1_1.3.1-1_i386.deb tcl8.3_8.3.3-6_i386.deb tk8.0_8.0.5-9_i386.deb tk8.3_8.3.3-8_i386.deb zlib1g_1.1.3-19_i386.deb eximon_3.34-1_i386.deb exim_3.34-1_i386.deb kernel-package_7.85_all.deb mysql-common_3.23.47-6_all.deb iptables_1.2.5-1_i386.deb lpr_2000.05.07-4_i386.deb mime-support_3.14-1_all.deb snort_1.7-9_i386.deb telnet_0.17-17_i386.deb gnupg_1.0.6-2_i386.deb libnasl1_1.0.10-1_i386.deb libnessus1_1.0.10-1_i386.deb mutt_1.3.27-2_i386.deb nessus_1.0.10-1_i386.deb nessusd_1.0.10-1_i386.deb nessus-plugins_1.0.10-1_i386.deb mtools_3.9.8-7_i386.deb sash_3.4-8.1_i386.deb catdoc_0.91.4-7_i386.deb groff-base_1.17.2-15_i386.deb groff_1.17.2-15_i386.deb gv_3.5.8-25_i386.deb ibritish_3.1.20-21_i386.deb ispell_3.1.20-21_i386.deb less_373-2_i386.deb magicfilter_1.2-49_i386.deb dpkg-dev_1.9.19_all.deb git_4.3.20-2_i386.deb patch_2.5.4-9_i386.deb sysutils_1.3.8.5.1_i386.deb lynx_2.8.4.1b-3_i386.deb Thanks for any help, Seneca [EMAIL PROTECTED]
Re: Laptop keyboard and switching X resolutions
Jeremy Whetzel [EMAIL PROTECTED] wrote:
I searched the archives for this one, but maybe I didn't look for the
right thing. I'm trying to figure out how to do the on-the-fly X
screen resolution changing (via Ctrl-Alt-+ or -) with a laptop
keyboard. I know that on a regular keyboard that you're supposed to
use the grey keys, but how is one supposed to do it on a laptop? Any
info and pointers are appreciated...
It seems to depend upon the laptop. On my laptop (with no convenient
NumLock) I would use Ctrl-Alt-Fn-; or p (';' being '-' and 'p' being
'+'). Fn (or as I usually call it, blue (most of the labels for the
alternate keystrokes it causes are in blue)) is used on my laptop's
model to access NumLock, the numeric keypad, right Ctrl, right Alt,
volume, monitor settings, and a few more keyboard keys (including some
that require Shift).
It isn't too bad, I just can't do some key combinations. And some that I
can do require me to push almost every key on the left side of the
keyboard while I type in numbers one at a time with the numeric
keypad.
The only annoyance that I have found at this point is that xterm (and
rxvt, and xvt) don't accept some of the shortcuts that I can normally
use outside of X (Alt+Fn+u, sometimes also requiring Shift in that mix
also, (Alt+4 on the numeric keypad) to logout).
Seneca
[EMAIL PROTECTED]
Re: pcmcia
Andreas von Heydwolff [EMAIL PROTECTED] wrote: Just today I installed potato from floppies on a 486 SL 25 laptop built in 1993. I have never used a pcmcia card in my life, don't own one and don't know anyone who has one. It looks as if the 486 can hold two cards in one slot (or would it be called a double slot). So here's the question: Sounds like you have the option of using it as either one or two PCMCIA slots (the number of cards depends upon physical size). There are some type numbers for these... I think that it is type II that that double slot of yours can hold two of (what my laptop from 1995 can hold two of). Could I use the a card that I would buy today or have standards changed and I would have to look for a vintage pcmcia network and/or modem card? I think that you would be limited to the 16-bit PCMCIA cards, and not be able to use the 32-bit CardBus cards. Seneca [EMAIL PROTECTED]
Re: 486 SX
Michel Loos [EMAIL PROTECTED] wrote: Em Qua, 2002-02-13 às 15:40, Mark Janssen escreveu: On Wed, 2002-02-13 at 19:16, Gerard Robin wrote: hello, I have the oppotunity to get hold of few omputers 486 SX (33Mz, 25Mz, hard disk = 89 Mb or 127 Mb) before they go to the rubbish. It will be a really painful and slow task... but it'll work... provided they have some decent amount of memory (8 being the bare minimum... 16 works good) I must disagree here, worked for me on 386 with 4MB RAM, just forget about any graphical interface and it really works fine. Do think it is possible for me to get it onto my 386 with 1M RAM? I can do my own compiling on a pentium, but I would need to get it onto 1.2M floppies (I can't get into my 386's BIOS, so it can't boot off of its 1.4M drive (only the 1.2M or harddrive)). I know that the results could be somewhat slow, but anything is faster (and more stable) than watching it labour through windoze. I don't even have most of the DOS commands (fixing a corrupt floppy, and the harddrive was overwritten with floppy info instead of the floppy). Seneca [EMAIL PROTECTED]
Re: What is a good, small, web browser?
Karsten M. Self kmself@ix.netcom.com wrote: on Fri, Feb 01, 2002 at 11:29 AM +0100, Karsten Heymann ([EMAIL PROTECTED]) wrote: * Seneca Cunningham [EMAIL PROTECTED] [020201 11:05]: David Moore [EMAIL PROTECTED] wrote: I have a very good reason for not wanting to install GNOME or another desktop manager. I tried GNOME about a month ago, and it's performance was similar to that of my 286 when windows 3.0 was installed on it. You really could doze off while it was loading. Well, galeon depends on the gnome libs installed, but it does not depend on gnome running. If your disk space is not a problem, you could try it out. Apart from that, w3m and links are very cool too (and don't forget netcat :-)) I'll also strongly plug Galeon. If you're not violently allergic to GNOME libs (and yes, it does suck in a whole mess of them, along with all of Mozilla), it's an ass-kicking browser. It's not for older/slower boxen though. I'd recommend *NO LESS* than a PII-233, and think you'll be happier with a PIII-600+ CPU. For memory, 128 MiB minimum, 256 strongly recommended. Particularly under intensive use (I can easily get over 100 tabs/windows open) it sucks both CPU and memory. But it does good things with both. I think I'll stay away from galeon with my current system, it is a P-100 with 16M ram. Otherwise, Browse-X is the best full-featured, lightweight, SSL-enabled browser I've run across. I decided to try Browse-X, and I have been taking a look at how it handles my local files. It doesn't take too long to load itself, or some local documents, but it duplicates the address in the links. If the address is file:///usr/share/doc/foo.html in the file, it is interpreted as file:///usr/share/doc/foo.html/usr/share/doc/foo.html. Is there some configuration that I haven't done, and couldn't find out about? Thanks for any help, Seneca [EMAIL PROTECTED]
Re: X configuration problem
Kapil Khosla [EMAIL PROTECTED] wrote: I could not install my base system due to network problems and stopped installation and did reboot. I then installed xserver-vga16, xf86setup. When I do startx i get the following error. dbe : Unknown error loading module Config error :/etc/X11/xf86config:48 Subsection extmod Module section keyword expected giving up Xinit : Connection refused (errno 111) unable to connect xserver xinit : no such process (errno 3) Server error Is your Subsection extmod inside Section Module? What does your XF86Config say? I have a old Toshiba laptop 4000CDT, and the video card is of Chips technologies 6. I was wondering what would be the right steps to install a working X from scratch. When I first installed X, I just downloaded the packages, installed them, and spent a while trying to find some working monitor settings for an LCD. Or maybe someone could point me to where the error is in this case, In this case it could just be where many errors with X are, in XF86Config. Hope this helps, Seneca [EMAIL PROTECTED]
a problem with pcmcia-modules-2.4.17 in sid
I grew a little tired of compiling kernels with pcmcia support, and then not being able to use my pcmcia hardware, so I decided to take a break from it for a while and try a pre-compiled kernel (kernel-image-2.4.17-386_2.4.17-1_i386.deb). I noticed in the description for pcmcia-cs that: The actual kernel modules required for this package are contained in the pcmcia-modules-package, where is the version of the kernel for which the modules have been compiled. So I downloaded the appropriate version of pcmcia-modules, and attempted to install it. But it didn't work: icosagon:~# dpkg -i /floppy/*.deb Selecting previously deselected package psmcia-modules-2.4.17-386. (Reading database ... 20503 files and directories currently installed.) Unpacking pcmcia-modules-2.4.17-386 (from .../pcmcia-modules-2.4.17-386_ 3.1.31-3k1_i386.deb) ... dpkg: error processing /floppy/pcmcia-modules-2.4.17-386_3.1.31-3k1 _i386.deb (--install): trying to overwrite `/lib/modules/2.4.17-386/pcmcia/pcmcia_core.o', which is also in package kernel-image-2.4.17-386 Errors were encountered while processing: /floppy/pcmcia-modules-2.4.17-386_3.1.31-3k1_i386.deb icosagon:~# dpkg-deb: subprocess paste killed by signal (Broken pipe) After that happened, I took a closer look at the description for pcmcia-modules: This package contains the set of loadable kernel modules for the PCMCIA Card Services applications program interface. They have been compiled to be compatible with the kernel in the kernel-image-2.4.17-386 package version 2.4.17-1. My system is mostly a combination of testing and sid, with a few non-essential potato packages not yet upgraded. What should I do about this? Should I just force it? All of these packages are the version that is in sid. Thanks for any help, Seneca [EMAIL PROTECTED]
x11 installation problems
Brian Wiese [EMAIL PROTECTED] wrote: I can't seem to get X to install correctly no matter what I do. What version? Everything else seems to work fine and installing without X works fine. I've been using the simple install and have tried various combinations of the X window system and X windows (core) selections but I get errors or the the screen ends up about 4 times the size of my monitor. xdm doesn't ever seem to come up either. I am new to debian also. Any advice would be greatly appreciated. Not much advice can be given without knowing a bit more about your system. What are the error messages, what's in your XF86Config or XF86Config-4 file, what's in the log files, and what's your hardware? Seneca [EMAIL PROTECTED]
Re: I boot and get: L40 40 40 40 40...
Joey Hess [EMAIL PROTECTED]; on behalf of; Joey Hess [EMAIL PROTECTED] Seneca Cunningham wrote: I have just been rearranging all my partitions, and now it looks like the most recent change is messing my system. A shortened version of the first couple post-post lines is: L 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 Lilo manual: If the BIOS signals an error when LILO is trying to load a boot image, the respective error code is displayed. The following BIOS error codes are known: [...] 0x40 Seek failure. This might be a media problem. Try booting again. Hmm. That's not very helpful. It is if you boot the next time with a boot floppy (then identify the problem). With this problem, as it turned out, in lilo.conf the boot device was set to /dev/hda5, and that is a logical partition. It seems like my computer didn't like that. It booted properly after changing boot back to /dev/hda. Seneca [EMAIL PROTECTED]
Re: I boot and get: L40 40 40 40 40...
Iván Filpo [EMAIL PROTECTED] wrote [with attributions corrected]: I'm getting insted of 40 a lot of 01's u think is probably the same ?. I'm booting hda5, that's the root. the other partition has win2k installed so i think that if i set in the lilo.conf to boot /dev/hda it will return an error . Any comments on that ? I have MS-DOS 7.0 installed on /dev/hda1, and lilo is set to boot /dev/hda, and it works fine. But the 01's you're getting has a different meaning than my 40's. To quote the lilo manual: Seneca Cunningham wrote: Joey Hess [EMAIL PROTECTED]; on behalf of; Joey Hess [EMAIL PROTECTED] Lilo manual: If the BIOS signals an error when LILO is trying to load a boot image, the respective error code is displayed. The following BIOS error codes are known: [...] 0x40 Seek failure. This might be a media problem. Try booting again. 0x01Illegal command. This shouldn't happen, but if it does, it may indicate an attempt to access a disk which is not supported by the BIOS. See also Warning: BIOS drive 0xnumber may not be accessible in section Warnings. Warning: BIOS drive 0xnumber may not be accessible Because most BIOS versions only support two floppies and two hard disks, files located on additional disks may be inaccessible. This warning indicates that some kernels or event the whole system may be unbootable. From what you wrote, I don't know if you have more drives than hda1 and probably a floppy and cd of some variety. I have spent some time looking at the Lilo manual. Since it is a lot of 01's, a severe problem (according to the manual) is occurring. One can also tell approximately when and a possible cause by what letters (if any) come up before the error code. If I knew the letters, then I could come up with more information. Hope this helps, Seneca [EMAIL PROTECTED]
What is a good, small, web browser?
I have a small system (100MHz pentium, 900M /usr, 1024K video ram) that I want to access the internet on. A problem that I have is that when the network that I use was set up, the gateway software that was decided upon requires the browsers used to have java support. I can't get away from it, I need a browser with java support to access the internet (the main reason why I still use windoze). So I was wondering about if there are any good, _small_ browsers like that. I tried mozilla on my system, and I had enough time to eat lunch while it started up (I have since removed mozilla). I have X4 installed and working, but I don't want to install any desktop managers. Thanks for any help, Seneca [EMAIL PROTECTED]
Re: What is a good, small, web browser?
David Moore [EMAIL PROTECTED] wrote: On Thu, 2002-01-31 at 20:44, Seneca Cunningham wrote: I have a small system (100MHz pentium, 900M /usr, 1024K video ram) that I want to access the internet on. A problem that I have is that when the network that I use was set up, the gateway software that was decided upon requires the browsers used to have java support. I can't get away from it, I need a browser with java support to access the internet (the main reason why I still use windoze). So I was wondering about if there are any good, _small_ browsers like that. I tried mozilla on my system, and I had enough time to eat lunch while it started up (I have since removed mozilla). I have X4 installed and working, but I don't want to install any desktop managers. Thanks for any help, Seneca [EMAIL PROTECTED] Try Opera. It's non-free, but it seems like what you want. I personally would prefer Galeon, but that won't work if you don't want to install GNOME. I have a very good reason for not wanting to install GNOME or another desktop manager. I tried GNOME about a month ago, and it's performance was similar to that of my 286 when windows 3.0 was installed on it. You really could doze off while it was loading. It seems like a possible spare-time (like I'll get any with 4A literature OA chem calculus) project of mine will be learning more about the internet java and (possibly) making my own browser that fits my requirements. I would prefer not to load anything that's non-free. Thanks for any help, Seneca [EMAIL PROTECTED]
Re: Compiling pcmcia-source
Nathan E Norman [EMAIL PROTECTED] wrote: On Tue, Jan 29, 2002 at 10:00:30PM -0700, Rick Macdonald wrote: [ moved prefix reply to end (hi Karsten)] On Tue, 29 Jan 2002, Seneca Cunningham wrote: I need the pcmcia module to get access to my ethernet card and cdrom, but I just can't get pcmcia-source to compile. I've tried to do it with woody and sid, but I just can't seem to get my pcmcia module to compile. I get warnings about how malloc.h is deprecated throughout the entire process, but at the end... [ snip ] This particular version is of my most recent attempt, with the pcmcia-source from sid, and downloaded today. The previous versions were all similar, with an identical result. Please _don't_ suggest that I change kernels to 2.4.17, I had enough trouble floppying 2.4.16 over, earlier this month, and I need the pcmcia module to even attempt to connect to the network (pcmcia ethernet card). I can't burn a cd and use it, my cdrom is an external one, connected through pcmcia. I can't do SLIP or PLIP because of a combination of a hardware and software problem (windoze combined with semi-functional (works when it isn't needed, stops when its needed) hardware). In spite of having to floppy everything (packages 1.6M are made into spanning archives, then reassembled), I generally download new versions of the more important packages if I find them in testing or sid. Probably the only exception to that would be the kernel-source... I don't have enough good floppies to do that. Well, I'm going to say 2.4.17 anyway... If you have the 2.4.16 source, there's no need to grab the whole 2.4.17 source! Grab the patch from kernel.org. I have, but I haven't been able to get it to work. patch claims that it can't patch the 2.4.16 source, applypatch does nothing except output messages (only when set to verbose) about how the patch kit and source directory are apparently okay, and patch-kernel just outputs the current version of the kernel source. 2.4.17 is mostly solid. I like it. My number one question is, why are you trying to compile pcmcia-source at all? 2.4 kernels have _built in_ pcmcia module support; I ditched pcmcia-source and started using the kernel pcmcia modules back at 2.4.7; I haven't looked back. My laptop is a Tecra 8100, with a 3Com ethernet card (can't remember the model cos I almost never use it); I also have a Lucent ORiNOCO 802.11b card. Works great. At this point, without compiling pcmcia-source, my pcmcia hardware is dead. As of now I haven't been able to compile a kernel that works with pcmcia. I set pcmcia support to be compiled into my kernels. Last time pcmcia worked was with the 2.2.19 kernel in potato, and now only when I boot 2.2.19 from floppy. Thanks for any help Seneca [EMAIL PROTECTED]
Re: libz1 package
Doug Jolley [EMAIL PROTECTED] wrote: Hi -- I'm new to Debian. Can anyone please tell me what the deal is with the libz1 package? It seems to be needed by quite a few other packages. Accordingly, I was surprised when I couldn't find it on my distribution CDs. I was even more surprised when I couldn't find it in the packages section of the Debian web site. I'm sure I'm missing something pretty basic. I wondered about that myself, when I first started with Debian. zlibg1 provides libz1, so you need to install it. Hope this helps, Seneca [EMAIL PROTECTED]
Re: I boot and get: L40 40 40 40 40...
I managed to get the system to boot with my old potato rescue disk. It seems that lilo was looking for the MBR or something similar in the wrong location. Without putting out any error messages, when I had last run lilo before the problem, it put the boot device to /dev/hda5, an extended partition. I got an error message with dpkg-reconfigure lilo, but not when I ran lilo. A later attempt with dpkg-reconfigure set the boot device to /dev/hda3, which left me with a prompt requesting the location of the MBR every time it booted up. My system works now (without the MBR prompt), with boot=/dev/hda and root=/dev/hda5. Thanks for any help, Seneca [EMAIL PROTECTED]
Compiling pcmcia-source
I need the pcmcia module to get access to my ethernet card and cdrom, but I just can't get pcmcia-source to compile. I've tried to do it with woody and sid, but I just can't seem to get my pcmcia module to compile. I get warnings about how malloc.h is deprecated throughout the entire process, but at the end... In file included from 3c575_cb.c:98: /usr/src/kernel-source-2.4.16/include/linux/malloc.h:4: warning: #warning linux/ malloc.h is deprecated, use linux/slab.h instead. 3c575_cb.c:879: conflicting types for `wait_for_completion_R02032293' /usr/src/kernel-source-2.4.16/include/linux/completion.h:30: previous declaratio n of `wait_for_completion_R02032293' make[4]: *** [3c575_cb.o] Error 1 make[4]: Leaving directory `/usr/src/modules/pcmcia-cs/clients' make[3]: *** [all] Error 2 make[3]: Leaving directory `/usr/src/modules/pcmcia-cs' make[2]: *** [build-modules] Error 2 make[2]: Leaving directory `/usr/src/modules/pcmcia-cs' make[1]: *** [kdist_image] Error 2 make[1]: Leaving directory `/usr/src/modules/pcmcia-cs' Module /usr/src/modules/pcmcia-cs failed. Hit return to Continue This particular version is of my most recent attempt, with the pcmcia-source from sid, and downloaded today. The previous versions were all similar, with an identical result. Please _don't_ suggest that I change kernels to 2.4.17, I had enough trouble floppying 2.4.16 over, earlier this month, and I need the pcmcia module to even attempt to connect to the network (pcmcia ethernet card). I can't burn a cd and use it, my cdrom is an external one, connected through pcmcia. I can't do SLIP or PLIP because of a combination of a hardware and software problem (windoze combined with semi-functional (works when it isn't needed, stops when its needed) hardware). In spite of having to floppy everything (packages 1.6M are made into spanning archives, then reassembled), I generally download new versions of the more important packages if I find them in testing or sid. Probably the only exception to that would be the kernel-source... I don't have enough good floppies to do that. Thanks for any help, Seneca [EMAIL PROTECTED]
I boot and get: L40 40 40 40 40...
I have just been rearranging all my partitions, and now it looks like the most recent change is messing my system. A shortened version of the first couple post-post lines is: L 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40 These lines continue not to the point of ad nauseum, but to the point of infinity. I can boot into my DOS partition with a boot disk, but I can't use the boot disk that I make earlier for my kernel (a custom compilation of 2.4.16), the partitions have been changed. I can't alter the settings on my boot floppy as the system with the problem is the only linux system that I have any access to. My system is a combination of potato, woody, and sid, that is mostly woody. The original partitions were somewhere along the lines of: /dev/hda1 FAT16 1000M /dev/hda3 Linux ext2 800M /dev/hda2 Linux swap 200M The altered partitions are something like: /dev/hda1 FAT16 100M /dev/hda2 Linux ext2 900M(this is /usr) /dev/hda5 Linux ext2 200M(this is /) /dev/hda6 Linux ext2 200M /dev/hda7 Linux ext2 200M /dev/hda8 Linux ext2 200M /dev/hda4 Linux swap 100M The problem did not happen after partitioning, or shifting the files from the old hda3 to the newer hda2, 5-8. It happened after I relocated files from what would now be /usr/usr to /usr. I had had some problems with this process earlier on, all being corrected within minutes. To my very untrained eye, it looks like some crucial file was corrupted in that final move (why is it _always_ the last change that kills the system). Do you know of any way to fix this without reinstalling the entire system? The only way I can do any installation is by floppy. Thanks for any help, Seneca [EMAIL PROTECTED]
Re: What part of X causes major display problems?
Moments before I was going to downgrade back to X3, I got X4 working... sort-of. I finally got rid of the pixel lines and black screens with random white pixels by setting changing colour depth. The colour depth that gave those problems is 16. At 24, X could not find any working screen modes. At 15, I got a pixel lines with some colour variation. At 8, I got a black screen with random purple pixels. And at 4, it works. Thanks for any help, Seneca [EMAIL PROTECTED]
Re: Questions about parted
As I previously wrote: Matthias Wieser [EMAIL PROTECTED] wrote: Am Sonntag, 13. Januar 2002 03:29 schrieb Seneca Cunningham: I am want to resize all of my partitions, and from what I have read, parted is a way to do it. I originally tried the version in potato, but it didn't yet do what it needs to for my first step: reducing the FAT16 partition by about 900M. So I upgraded to the woody version and now I get this: icosagon:/sbin# ./parted invalidate: busy buffer [many lines of invalidate: busy buffer] (parted) quit [14 more lines of invalidate: busy buffer] icosagon:/sbin# parted worked for me on unstable. I might try that later. I have just tried that, and I get the same message as before. So now I ask, is it safe to use in it's current state (beyond the normal risks of partition editting)? And what other tools could I use instead of parted? Thanks for any help, Seneca [EMAIL PROTECTED]
Re: What part of X causes major display problems?
ben [EMAIL PROTECTED] wrote: On Tuesday 15 January 2002 03:38 pm, Seneca Cunningham wrote: (EE) Error from xf86HandleConfigFile() Fatal server error: no screens found this is basically telling you that none of the values that might run can. Yes, I had this Fatal server error when I installed X3. My solution was adding a modeline, and it worked. The file that caused this problem is my X3 config file that worked properly with X3. what kind of output were you getting from x3? greater that 640x480? at what kind of depth? My X3 output was 640x480, with virtual 800x600. I think my computer can do higher resolution, I don't have any unused monitors that do anything higher. The colour depth was 16 bits, but I could probably do higher at the cost of speed. I fixed the syntax, and I started getting the pixel-lines with my old XF86Config file. I have attached the output of diff after fixing the syntax. There's not much difference between the the versions of XF86Config. given that it did run x3, why not settle for that, since it sounds like pretty old and inefficient hardware? i guess the question is why are you so determined to get the latest incarnation of x to work on the least capable system? I am determined to get X4 to work on my machine because... a) it is there. If X4 wasn't there, I wouldn't try to get it to work. b) I wanted to see if it would use .xsession without changing runlevel. In X3, the only way on my system to get X to use .xsession was to change to runlevel 5. I prefer runlevel 2. I knew X read the file, because if I put something bad into it, X wouldn't run. I have decided that I _really_ hate the people who send their systems out with insufficient documentation. so you've google-searched for all the info that might be there pertaining to the hardware? which raises the next question, what are you working with, After all my years with the machine, I learned what it could and could not do. I mightn't have been explicitly told everything, but I knew its basics and I have checked for some information about my system (some from internet, some from physical checks (and measurement) of the machine, and some from the manual). trs80's, att xt's or what? how old is the machinery and what manufacturers are involved? it kinda sounds like some part of your hardware can't take the My hardware isn't too old... it's only 7 years old (I think (and that's the newest stuff I'm allowed to customise)). Anyway, all that I have been able to find out about the video setup is that it has a Chips and Technologies 65545 video chip. pounding it's getting from x4. are you still getting log messages about 'left-alt' and 'meta' keys? I stopped getting the log messages about left-alt and meta after commenting out the problem line. Thanks for any help, Seneca [EMAIL PROTECTED]
Re: What part of X causes major display problems?
ben [EMAIL PROTECTED] wrote: On Wednesday 16 January 2002 01:49 pm, Seneca Cunningham wrote: [snip] My hardware isn't too old... it's only 7 years old (I think (and that's the newest stuff I'm allowed to customise)). Anyway, all that I have been able to find out about the video setup is that it has a Chips and Technologies 65545 video chip. i used to have a toshiba laptop with a 65545 v-chip that balked at the virtual screen settings. the only way it would run was with no virtual screen settings specified. that was in x3, but maybe you could check that. Mine worked with the virtual screen settings in X3. The only times I had problems with that was when I set the settings too high, and that was because I only have 1024K video RAM. Thanks for any help, Seneca [EMAIL PROTECTED]
Re: What part of X causes major display problems?
ben [EMAIL PROTECTED] wrote: On Wednesday 16 January 2002 03:28 pm, Seneca Cunningham wrote: ben [EMAIL PROTECTED] wrote: On Wednesday 16 January 2002 01:49 pm, Seneca Cunningham wrote: [snip] My hardware isn't too old... it's only 7 years old (I think (and that's the newest stuff I'm allowed to customise)). Anyway, all that I have been able to find out about the video setup is that it has a Chips and Technologies 65545 video chip. i used to have a toshiba laptop with a 65545 v-chip that balked at the virtual screen settings. the only way it would run was with no virtual screen settings specified. that was in x3, but maybe you could check that. Mine worked with the virtual screen settings in X3. The only times I had problems with that was when I set the settings too high, and that was because I only have 1024K video RAM. don't know any more than that. did you configure the kernel you're running? maybe a google search for x4 + 65545 might reveal something pertinent. After checking that, I didn't find much of anything about the problem... but I did find some sites that I could order a 386SX on a board with the 65545 for its video. Thanks for any help, Seneca [EMAIL PROTECTED]
Re: What part of X causes major display problems?
ben [EMAIL PROTECTED] wrote: On Monday 14 January 2002 06:30 pm, Seneca Cunningham wrote: [snip] I tried my old X3 config file, and X4 couldn't parse it... the logfile says: Parse warning on line 43 of section Keyboard in file XF86Config.bak Ignoring obsolete keyword LeftAlt. Parse error on line 43 of section Keyboard in file XF86Config.bak Meta is not a valid keyword in this section. (EE) Problem parsing the config file (EE) Error from xf86HandleConfigFile() Fatal server error: no screens found The file that caused this problem is my X3 config file that worked properly with X3. here's my config file that was generated with xf86config and gets parsed by x4. you'll need to edit probed hardware settings to whatever your system needs but at least you'll be able to compare for syntax errors that x4 doesn't like. i hope that helps. I fixed the syntax, and I started getting the pixel-lines with my old XF86Config file. I have attached the output of diff after fixing the syntax. There's not much difference between the the versions of XF86Config. Earlier I wrote that I couldn't decide who I really hated, people like you who didn't have problems, or the people who consider it to be alright to send out a computer with only one manual for the entire system, specifically a tiny manual that concentrated on installing a mouse driver, and leaving the entire section on the display a mention of the chip used in the introduction. I have decided that I _really_ hate the people who send their systems out with insufficient documentation. Thanks for any help, Seneca [EMAIL PROTECTED] diff.out Description: Binary data
Login problem
I have a problem that seems like login is working too slowly for my computer, or my computer is too slow for login (a little more likely). Occasionally I get results similar to the results for this fictional user. Debian GNU/Linux 3.0 icosagon tty2 icosagon login: foo baPassword: Login incorrect As is quite guessable, the password for user foo is bar. foo typed in bar, but only the 'r' makes it into the password entry. The ba is merely output to the screen. If foo were to see that ba was visible, and knew that it hadn't been entered into the password entry, foo could have logged in if the full bar had been typed after the ba, leaving the total password typing of foo at babar. I downloaded the source so that I could see if I could do anything about it, but gzip is complaining that it isn't in gzip format. Is there any special package I need to get to be able to decompress the source? I have gzip version 1.3.2-3, tar 1.13.25-1, and login 2902-8. Thanks for any help, Seneca [EMAIL PROTECTED]
Re: Login problem
Cameron Matheson [EMAIL PROTECTED] wrote: I'm probably missing something here... but can't you just wait for the password prompt? I could, but I don't want to always check my timing. Normally I just type at full speed like foo did, and the password prompt comes up before the password gets in. But sometimes the password prompt takes a moment before it comes up, and some characters of my password get exposed. about the source... are you sure that the archive is good? Does it have a .gz or .tgz extension? If it's a .bz2 then you will need to use bunzip2 The filename is shadow_2902.orig.tar.gz instead of gunzip. On Tuesday 15 January 2002 08:16 pm, Seneca Cunningham wrote: I downloaded the source so that I could see if I could do anything about it, but gzip is complaining that it isn't in gzip format. Is there any special package I need to get to be able to decompress the source? I have gzip version 1.3.2-3, tar 1.13.25-1, and login 2902-8. Thanks for any help, Seneca [EMAIL PROTECTED]
Re: What part of X causes major display problems?
ben [EMAIL PROTECTED] wrote: On Sunday 13 January 2002 02:11 pm, Seneca Cunningham wrote: [snip] X -configure is... (==) CHIPS(0): Min pixel clock is 11.000MHz (--) CHIPS(0): Max pixel clock is 56.000MHz ... and from the xf86cfg -textmode method... (==) CHIPS(0): Min pixel clock is 5.500MHz (--) CHIPS(0): Max pixel clock is 28.000MHz i don't have a solution but have you noticed that the x-configure values are exactly double the textmode values. i have no idea why this is but i thought it might be useful to point that out. And each of range includes, or almost includes, the working clocks value from my old X3 config of 28.32. And both config files had to have a modeline manually entered before I could get anything to appear, other than an error message. The X3 xf86config program says _not_ to probe clocks, but X4 looks like it is probing clocks. btw, are you running a 2.2 kernel? i couldn't get x4 to run on 2.2, and went I am running a 2.4 kernel. with the 2.4 kernels for that reason. don't know exactly why--i was more interested in the functionality than the detective work. Detective work is how I survive computer science while forgetting to hand anything in... I get some credit for testfiles that crash the same number of (or sometimes more) programs than the teacher's (most of the caught problems are null pointers), while my programs run unscathed. Thanks for any help, Seneca [EMAIL PROTECTED]
Re: What part of X causes major display problems?
ben [EMAIL PROTECTED] wrote: On Monday 14 January 2002 04:56 pm, Seneca Cunningham wrote: The X3 xf86config program says _not_ to probe clocks, but X4 looks like it is probing clocks. my understanding of clock probing is that the card is either physically capable of standing the probe or it isn't, and that probing a card that isn't capable can actually cause damage. i missed the start of this thread so i don't know what kind of hardware you've got. did x ever run in this Well, I have an old pentium 100, a 640x480 monitor, 16M RAM, 2G harddrive, a Chips Technologies 65545 video chip and 1M video RAM. Not much, but it works (except when it doesn't). environment in any version? x4 should make use of any config file generated to run under x3, and you should keep in mind that a blank screen doesn't I tried my old X3 config file, and X4 couldn't parse it... the logfile says: Parse warning on line 43 of section Keyboard in file XF86Config.bak Ignoring obsolete keyword LeftAlt. Parse error on line 43 of section Keyboard in file XF86Config.bak Meta is not a valid keyword in this section. (EE) Problem parsing the config file (EE) Error from xf86HandleConfigFile() Fatal server error: no screens found The file that caused this problem is my X3 config file that worked properly with X3. necessarily mean that x isn't running--just that it's not running the way that you want it to. I had noticed, I could do stuff in X, I just couldn't see what I was doing, all I could see was narrow vertical lines, or a black screen with a random smattering of white pixels. do you have all the necessary specs for your hardware? i'm one of those that No, I don't have all the specs for my hardware. The manual for the entire system doesn't have the specs beyond the type and size of 4 different monitor possibilities. And yes, the manual is the _only_ one for the system, and it is for everything that can be manualised in it, from the mouse to the monitor. (I hate pre-packaged-commercially-available-from-an-office-store-near-you computers, but in the case of some of my systems, they're all that land on my doorstep) you are probably close to hating who's never had any problems with running x. I'm still deciding who to hate, the people like you, or the people who decide that a system should only have one _tiny_ manual, most of which is the instructions for installing the sound and mouse drivers on various windoze versions. consequently, i know jack about it. all i've ever had to do is run xf86config, enter the relevant specs, and run it. which xserver are you running? By way of xservers, I have xserver-xfree86 version 4.1.0-11. Just as I was checking various logs, I realised why there were two different sets of pixel clock info... (--) CHIPS(0): Clocks scaled by 2 This was a line that I had missed in the 5.500MHz-28.000MHz log. Thanks for any help, Seneca [EMAIL PROTECTED]
Re: Questions about parted
Matthias Wieser [EMAIL PROTECTED] wrote: Am Sonntag, 13. Januar 2002 03:29 schrieb Seneca Cunningham: I am want to resize all of my partitions, and from what I have read, parted is a way to do it. I originally tried the version in potato, but it didn't yet do what it needs to for my first step: reducing the FAT16 partition by about 900M. So I upgraded to the woody version and now I get this: icosagon:/sbin# ./parted invalidate: busy buffer [many lines of invalidate: busy buffer] (parted) quit [14 more lines of invalidate: busy buffer] icosagon:/sbin# Can you give a better discription of your system and what you have installed. The system's processor is an old pentium 100, it has 16M memory. I have a 2gig harddrive, that is split down the middle. Since noone has told me to re-install windows on my debian system, I want to free up 900M for my use, leaving 100M for what little DOS stuff I use. My debian installation is part potato, part woody. The kernel is a custom compilation of 2.4.16. Although X is installed, I don't use it. I haven't been able get it work since I upgraded to X4 (display problems). Do you have used parted with root-user or not? I use parted logged in as root. did you updatedb run some time (that is why it probably was trying to find parted at an other place Oops... no. I have now. parted worked for me on unstable. I might try that later. Thanks for any help, Seneca [EMAIL PROTECTED] installed.packages Description: Binary data
Re: What part of X causes major display problems?
It seems that old bits of memory have been put onto the screen (as I am probably the last person to realise), when the screen is in pixel-lines. The two examples of this I have are the xf86cfg graphics going into the pixel-lines after using it without adding the -textmode. The other, more recent example is a banner of repititions of the win95 startup logo in pixel-lines (yuck). This one happens if I just reboot after being in DOS7.0, even though the win95 logo doesn't come up in DOS (I don't have the actual logo file, but a backup of it in io.sys). But even after all this time playing with XF86Config-4, I _still_ haven't been able to get X working. I still only get pixel-lines, black with random white pixels, or an error message about no screens found. I have tried doing about all I could think of. If this continues much longer, I think I'll try to downgrade. But one thing that is _really_ getting to me is the difference between the logfiles. As I had previously wrote: It's somewhat interesting to notice a difference between the logfiles when testing the configurations from each method. In the logfile from testing the X -configure is... (==) CHIPS(0): Min pixel clock is 11.000MHz (--) CHIPS(0): Max pixel clock is 56.000MHz ... and from the xf86cfg -textmode method... (==) CHIPS(0): Min pixel clock is 5.500MHz (--) CHIPS(0): Max pixel clock is 28.000MHz The lines preceded by (==) are default values and the (--) lines are probed values. This is interesting, I would have thought that the values would be the same. What sort of effect could this have? I have been spending my time with XF86Config-4, is there some other file that I should try changing? Thanks for any help, Seneca [EMAIL PROTECTED]
Questions about parted
I am want to resize all of my partitions, and from what I have read, parted is a way to do it. I originally tried the version in potato, but it didn't yet do what it needs to for my first step: reducing the FAT16 partition by about 900M. So I upgraded to the woody version and now I get this: icosagon:~# parted bash: /usr/sbin/parted: No such file or directory icosagon:~# whereis parted parted: /sbin/parted /usr/share/man/man8/parted.8.gz icosagon:~# cd /sbin icosagon:/sbin# parted bash: /usr/sbin/parted: No such file or directory icosagon:/sbin# ./parted invalidate: busy buffer [many lines of invalidate: busy buffer] (parted) quit [14 more lines of invalidate: busy buffer] icosagon:/sbin# With all this invalidate: busy buffer stuff, is it safe to use (beyond the normal risks involved in playing with partitions)? And if it isn't, then what should I use instead? Thanks for any help, Seneca [EMAIL PROTECTED]
Re: XFree86 upgrade to 4.1.0
stefao melchior [EMAIL PROTECTED] wrote: I was wondering if any of you can help me suggesting how to upgrade perl from 5.004 to 5.6. I don't know of any easy way... the way I did it was with brute force and ignorance (of conflicts) with dpkg. There are a few awkward dependencies. I did it with: # dpkg -i --force-conflicts packagename.deb This method is _not_ recommended for general installations, using this can break your system if you're not careful. I only used it when I had to. Try using the safer: # dpkg -i packagename.deb before forcing the package. Also, write down the dependencies and conflicts of all the packages being installed. This way, you know more about what order to install and replace files. When forcing a package because it conflicts with an older package, remember to replace the older package soon thereafter. Until all of the requisite packages are installed, some packages may be brokenly installed (namely not configured due to dependencies). The easy way to configure all unconfigured packages is with: # dpkg --configure -a but if you want to be selective about the order that packages are configured, substitute packagename for -a. Use it at your own risk. Seneca [EMAIL PROTECTED]
Re: What part of X causes major display problems?
Kent West [EMAIL PROTECTED] wrote: Seneca Cunningham wrote: I have just upgraded XFree86 from 3.3.6 to 4.1.0. After completing the upgrade, I changed from tty1 to X's display and saw the image on the screen consisted of vertical lines running down my display, the top of the lines were 1 pixel in width and the bottom width was 3 pixels, with 2 wide black horizontal lines running across it. Only the highlighted edges of the login box appeared. No mouse cursor or text is visible on the screen. This continued after rebooting, but the colours of the lines sometimes changes after rebooting. Almost certainly a problem with the video settings in XF86Config[-4]. And that's where I have been concentrating my efforts. You can also try running xf86cfg or xf86cfg -textmode (or is it --textmode?) for a non-GUI setup application. It's xf86cfg -textmode, and I still ended up with the pixel-lines screen. D-Man [EMAIL PROTECTED]; on behalf of; dman [EMAIL PROTECTED] wrote: Try X -configure instead. X4 has auto-configuration built-in to it. It didn't work for my card, but it gave me a skeleton config to fill in correctly (with my values from X3). I got a plain black screen with some random white pixels. It's somewhat interesting to notice a difference between the logfiles when testing the configurations from each method. In the logfile from testing the X -configure is... (==) CHIPS(0): Min pixel clock is 11.000MHz (--) CHIPS(0): Max pixel clock is 56.000MHz ... and from the xf86cfg -textmode method... (==) CHIPS(0): Min pixel clock is 5.500MHz (--) CHIPS(0): Max pixel clock is 28.000MHz The lines preceded by (==) are default values and the (--) lines are probed values. This is interesting, I would have thought that the values would be the same. In my XF86Config there is a line Clocks 28.32 28.32 28.32 28.32 but it didn't alter anything in any of my XF86Config-4 attempts (presumably it isn't needed anymore with the driver used, and is ignored). For each of the two configuration methods, the results when tested remained the same, even after changing the contents. Thanks for any help, Seneca [EMAIL PROTECTED]
What part of XFree86 causes major display problems?
I have just upgraded XFree86 from 3.3.6 to 4.1.0. After completing the upgrade, I changed from tty1 to X's display and saw the image on the screen consisted of vertical lines running down my display, the top of the lines were 1 pixel in width and the bottom width was 3 pixels, with 2 wide black horizontal lines running across it. Only the highlighted edges of the login box appeared. No mouse cursor or text is visible on the screen. This continued after rebooting, but the colours of the lines sometimes changes after rebooting. When Ctrl+Alt+Bksp is pushed, I see tty1 for a moment, the screen goes black, then, a few seconds later, the screen goes back the way it was. It is possible to login and out when the screen is like this, the cursor is visible as 2 short horizontal lines with a width of 1 pixel. How can I find out what part of X is causing this problem. All of the bug reports of similar problems that I have seen and checked are at least half a year old, and all say that it was fixed, and these are in older package versions than mine. Thanks for any help, Seneca [EMAIL PROTECTED] ps. If this message was received earlier, please tell me. From my end it looks like it was sent directly to the windoze equivalent of /dev/null.
Re: What part of X causes major display problems?
Jason Wood [EMAIL PROTECTED] wrote: On Friday 04 January 2002 10:13 pm, Seneca Cunningham wrote: I have just upgraded X from 3.3.6 to 4.1.0. After completing the upgrade, I changed from tty1 to X's display and saw the image on the screen consisted of vertical lines running down my display, the top of the lines were 1 pixel in width and the bottom width was 3 pixels, with 2 wide black horizontal lines running across it. Only the highlighted edges of the login box appeared. No mouse cursor or text is visible on the screen. This continued after rebooting, but the colours of the lines sometimes changes after rebooting. If I am imagining your screen correct;ly, this sounds to me like you have configured X to use the wrong driver for your graphics card. I had a similar problem when I first set up X for the very first time. I am using the same settings that worked in X3 (including copying my only working modeline), when possible. When it is not, the only difference is that I use the driver considered to be equivalent (chips instead of ct65545). Try using xf86config, of xf86cfg to reconfigure X, and see what happens. xf86config changes my old X3 config file, and xf86cfg messes up my screen. It spews out some messages that I cannot see because the screen goes into graphics mode, then it draws an image of a monitor, a videocard, a computer, a keyboard, and a mouse, in addition to the outlines of a few boxes and some 1 pixel wide lines at the top of the screen, with a number of random pixels to round out the mix. It's a little amusing in a way... if I start X, at the top of the screen I see two miniature images of that screen at the top. Thanks for any help, Seneca [EMAIL PROTECTED]
What part of X causes major display problems?
I have just upgraded X from 3.3.6 to 4.1.0. After completing the upgrade, I changed from tty1 to X's display and saw the image on the screen consisted of vertical lines running down my display, the top of the lines were 1 pixel in width and the bottom width was 3 pixels, with 2 wide black horizontal lines running across it. Only the highlighted edges of the login box appeared. No mouse cursor or text is visible on the screen. This continued after rebooting, but the colours of the lines sometimes changes after rebooting. When Ctrl+Alt+Bksp is pushed, I see tty1 for a moment, the screen goes black, then, a few seconds later, the screen goes back the way it was. It is possible to login and out when the screen is like this, the cursor is visible as 2 short horizontal lines with a width of 1 pixel. How can I find out what part of X is causing this problem, all of the bug reports of similar problems that I have seen and checked are at least half a year old, and all say that it was fixed, and these are in older package versions than mine. Thanks for any help, Seneca [EMAIL PROTECTED]
Re: File Transfers
frank [EMAIL PROTECTED] wrote: Assumming you can get some sort of ethernet connection, install apache/some web server on the computers you want files off. This how I transfered files off my brother's winblose box, just needed him to enable tcp/ip, and install the windows apache off a computer mag CD, and copy the files to the right place on his computer. It worked for me. You might meed a web browser too... Setting up samba is a black art, and I dont have it (why?). The problem with this setup is 'you are not allowed to alter the win98 setup in anyway'. Have they removed the on switch? Just install apache in your space, and as it only runs when you start it in a terminal (bos box), how will they know? Tcp/ip is probably already on, so just run ipconfig.exe to find out the ip address of the winblows box, and point your linux browser at it. They have already threatened to take the only system that I can use on our ethernet away... the windows one, it has a built-in ethernet port. It is the only one that I can access the internet with. I have no space on it. The entire harddrive is shared. If my brother, on another machine, wants to delete my homework, he can (but wouldn't get away with it). Because of broken hardware, no cash, and windows acting like windows after a brick hits them, the standard routes of file transfer seem to be closed to me. I have just recently noticed a possible method to link these computers, but I have yet to work it all out. I hope it works, yesterday I couldn't access the howto's that I wanted to refer to for this connection, so I started collecting the packages I would need to compile a new kernel. I had to transfer 22 M by _floppy_. My possible method is to connect the systems with a serial cable, and setup the windows computer for a dial-in connection using an exterior modem that uses the same serial port that I have the cable on, and and select SLIP for the dial-up server type. On the other end, I respond like a modem would for the AT sequences (each time a sequence is received, transmit OK). After the phone number is dialed, initiate SLIP. I hope this will work. Its current problems during the initial testing process are: it kills the windows computer's connection to the internet, and I haven't figured out yet the best way of automating the modem responses. I already know to send OK after 3 characters arrive, send OK 8(I think, it's in a logfile) characters later, and so on Time for a byte to eat, Seneca [EMAIL PROTECTED] ps. how long do you think it took to transfer the kernel source by floppy
Re: File Transfers
Dan [EMAIL PROTECTED] wrote: I would suggest that you look into making a ethernet-serial connector. I won't swear that this can be done...but I would expect that you can find the pin-outs for a standard 10bT connection and match them to the pin-outs for the DB9 connector and make a cross over cable as such. As long as you are running SLIP on the debian end...the windows box shouldn't know anything was different. (see http://www.pccables.com/01910.htm) for a picture of a commercial solution) Again YMMV. I haven't actually tried this but I have had to do transfers over a regular serial cable. You might also want to look at: http://www.linuxdoc.org/HOWTO/Serial-Laplink-HOWTO/ Interesting idea, I'd be able to try that if I had a second connection to the network or the windows system. The serial-laplink howto is a document that I tried to get to yesterday and the day before that, but I kept getting socket errors. Today I can get through. Right now, because of a lack of resources (such as cables), I'll try my dial-in method, and hope it works. Part of this is because on the last day of class before the winter break, I told the networking teacher (I never took his networking courses) that I would try to get a semi-functional network going with my systems. I now have some possibilities to connect two of the four systems (now, how will I go about the 286 that doesn't recognise its floppy drive and the 386sx with 1M memory and a half-installation of MSDOS 3.3 and doesn't boot if its hardware is changed). I _really_ want to at the very least get the two newest computers connected. Thanks for any help, Seneca [EMAIL PROTECTED]
make-kpkg exists, right?
Right now my system is half-working and is a combination of potato and woody. I am trying to compile a new kernel after having a number of problems with my system. The instructions for compiling a new kernel say to enter in make-kpkg clean, but that doesn't work on my system. icosagon:/usr/src/kernel-source-2.4.16# make-kpkg clean bash: /usr/bin/make-kpkg: No such file or directory icosagon:/usr/src/kernel-source-2.4.16# make-kpkg --help bash: /usr/bin/make-kpkg: No such file or directory icosagon:/usr/src/kernel-source-2.4.16# make-kpkg bash: /usr/bin/make-kpkg: No such file or directory icosagon:/usr/src/kernel-source-2.4.16# whereis make-kpkg make-kpkg: /usr/bin/make-kpkg icosagon:/usr/src/kernel-source-2.4.16# cd /usr/bin icosagon:/usr/bin/# ls make-kpkg make-kpkg First it claims that make-kpkg isn't a file or directory, then it tells me where it is. Unlike make-kpkg, the manpage can be accessed. I have reinstalled kernel-package three or four times, with the most recent time being from a second download. I am attempting to use the version of kernel-package that is in woody. What do I need to do to get further in my attempt to compile a kernel? Thanks for any help, Seneca [EMAIL PROTECTED]
Re: make-kpkg exists, right?
Sorry about the delay, but I left the computers for a moment, and everyone wanted me to have a byte to eat. So in the order that I received the questions and responses, I have my response and/or output. J.A.Serralheiro [EMAIL PROTECTED] wrote: just compile the kernel the generalw way. make-kpkg will create a package that u install as if some some software package. cd to /usr/src/linux and read README file. There u will find all the instructions u need to compile the kernel without relying on the debian weay Right now, I want to try to get the debian way to work for me. D-Man [EMAIL PROTECTED]; on behalf of; dman [EMAIL PROTECTED] wrote: Is /usr/bin in your PATH? Is /usr/bin/make-kpkg executable? /usr/bin is in my PATH icosagon:/usr/src/kernel-source-2.4.16# ls -l /usr/bin/make-kpkg -rwxr-xr-x1 root root30651 Dec 21 16:27 /usr/bin/make-kpkg ben [EMAIL PROTECTED] wrote: use the full path, ie /usr/bin/make-kpkg clean icosagon:/usr/src/kernel-source-2.4.16# /usr/bin/make-kpkg clean bash: /usr/bin/make-kpkg: No such file or directory Steve Mayer [EMAIL PROTECTED] wrote: make-kpkg is a perl script. Make sure that you have perl installed in the location specified on the first line of the make-kpkg script. icosagon:/usr/src/kernel-source-2.4.16# whereis perl perl: /usr/lib/perl /usr/share/perl first line of make-kpkg is #! /usr/bin/perl icosagon:/usr/src/kernel-source-2.4.16# ls /usr/bin/perl ls: /usr/bin/perl: No such file or directory icosagon:/usr/src/kernel-source-2.4.16# ls /usr/bin/perl* /usr/bin/perl-5.6 /usr/bin/perl5.6.1 /usr/bin/perldoc.stub Rick Macdonald [EMAIL PROTECTED] wrote: No, by the error message it's finding make-kpkg fine or you'd get bash: qwer-qwerq: command not found. It's a perl script, and it's also finding perl fine, or you'd get bad interpreter: No such file or directory. I think the perl script is executing a command or loading or accessing something that isn't being found. I don't know perl yet, so I'm not sure about it. If you want, I can attatch my copy of make-kpkg, but it is 978 lines long. I want to get my system fully working eventually, so I might try the non-debian way later on, but for now I want to continue trying with the debian way. Should I see about downloading kernel-package from potato? Thanks for any help, Seneca [EMAIL PROTECTED]
Re: make-kpkg exists, right?
D-Man [EMAIL PROTECTED]; on behalf of; dman [EMAIL PROTECTED] wrote: I think you can just hardlink /usr/bin/perl5.6.1 (or /usr/bin/perl-5.6) to /usr/bin/perl. # ln /usr/bin/perl5.6.1 /usr/bin/perl I did it, and it works now. Thanks. Seneca [EMAIL PROTECTED]
Is http://www.linuxdoc.org up?
Is http://www.linuxdoc.org up? I've tried to get through, but each time I've tried in the past little while my connection timed out. Seneca [EMAIL PROTECTED]
Re: The only non-beep sound I can get is a SQUAWK
ben [EMAIL PROTECTED] wrote: have you considered upgrading the kernel? you might also need to reinstall gcc. it's sounding more and more like you've got a busted system. when configure fails, what's it telling you in the output? Even before this message I had thought of reinstalling gcc, and did. I still couldn't compile Hello, world, which I consider a little embarassing, but not without precedent (on one of my systems, a Windows box, I have compilers and interpreters for 9 (never realised how many) different languages, and only the qbasic interpreter, that came with MSDOS 5.0, and compiler for turing work properly (I think)). I have thought of upgrading the kernel, but I want to first figure out how to connect this system to my Windows box (the only machine with access to the network, and the internet, and I'm not allowed to install Linux on it) to make file transfers easier (all my file transfers are by floppy). Of course, being a Windows machine with typical Windows problems, that system is difficult to reach (SLIP PLIP cannot be done for various reasons, and neither can ethernet). Anyway, I have the output here. The output was the same both before and after the reinstallation of gcc, so there is only this one file. All of the text except for expr: syntax error was on stdout. Those lines were on stderr. Thanks for any help, Seneca [EMAIL PROTECTED] alsa_config.output Description: Binary data
Re: The only non-beep sound I can get is a SQUAWK
As I had previously wrote: This is getting to the point that I am actually thinking of trying ALSA Well, I had tried that, and found that the package alsadriver is not available for the version of the kernel that I am using (2.2.19). I downloaded the drivers, followed the installation instructions, and .\configure failed. I wasn't about to attempt to go through the compilation manually. The reason: I can't even get Hello, world to compile (gcc can't find the header file). Now I have the kernel modules set up, and squawking when I give them the chance, the base alsa packages, and the source for the appropriate (I hope) alsadriver. I am wondering more about the specific causes of my problem, and how to deal with them. I'm tired of always getting nothing. Thanks for any help, Seneca [EMAIL PROTECTED] alsa_config.output Description: Binary data
Re: The only non-beep sound I can get is a SQUAWK
ben [EMAIL PROTECTED] wrote: run dmesg, optionally piped through more, to see how the boot treats your soundcard. if the card isn't recognized on boot, not much of anything else can happen with it. post the output of dmesg and any other output generated in the course of trying to get the card to do what you want it to. I've run dmesg, and the full output from it is here. The part that seems immediately relevant is: Soundblaster audio driver Copyright (C) by Hannu Savolainen 1994-1996 SB 3.01 detected OK (220) ESS chip ES1688 detected I had read that the ESS chip is sometimes incorrectly detected by sb, but my manual specifically lists the detected chip. In response to the question asked by Simon Eilting [EMAIL PROTECTED] Why don't you use the kernel driver for the es1688 AudioDrive? I am saying this: What is its name? I haven't been able to find an obvious one, and this is with kernel 2.2.19, so I used sb.o as the HOWTO suggested. Thanks for any help, Seneca [EMAIL PROTECTED] dmesg.out Description: Binary data
Re: The only non-beep sound I can get is a SQUAWK
ben [EMAIL PROTECTED] wrote: if every method of generating sound produces the same output and given that the system does acknowledge the card, it would seem that the driver is at fault. it may have come up before. check the archives. I have checked the archives, and I couldn't see a similar problem. This is getting to the point that I am actually thinking of trying ALSA, and that would mean floppying packages over, and manually checking their dependancies (not much fun). Thanks for any help, Seneca [EMAIL PROTECTED]
The only non-beep sound I can get is a SQUAWK
I've RTFM (with the 'R' pronounced red not reed), checked the manpages info, the HOWTO's, Google, and some other places over the past two days. With all those resources, all my soundcard will put out, other than the standard beep, is a squawk. The squawk length has nothing to do with format, just the file size. My machine has an ESS ES1688 AudioDrive. According to the HOWTO, I can use the SoundBlaster driver sb.o. My manual gives slightly conflicting signals as to if it's PnP, but from what I've read over the past two days, it probably is, although pnpdump reports No boards found. When any output is attempted through /dev/sequencer, nothing is heard and I get a message stating that the device is not configured. Other audio devices squawk, or claim that it does not exist, even though I see it under /dev. Thanks for any help, Seneca [EMAIL PROTECTED] sound.notes Description: Binary data
What causes segmentation faults?
On my machine, two or three different programs have given me segmentation faults. The two that I remember are Enlightenment and bwBASIC. The segmentation fault in Enlightenment only happened once, and Enlightenment was removed soon thereafter. In bwBASIC, however, I was able to reproduce the fault, and figure out what I was doing that caused the segmentaion fault to happen, but I don't know why it happened. I couldn't find anything when I tried to look it up using man because /usr/man/man1/bwbasic.1.gz is a dangling symlink on my system, and I don't know how to fix it at this point. What I did that caused the segmentation fault was change a line that was causing an error, without deleting it first, then running the program. If anyone is interested, the attachment contains the contents of my search for how I caused the segmentation fault. Thanks for any help. Seneca [EMAIL PROTECTED] find_fault Description: Binary data
Re: File Transfers, taken from Re: an XFree86 problem, an xscreensaver problem, followed by anEnlightenment problem
Karsten M. Self kmself@ix.netcom.com wrote: on Fri, Dec 14, 2001 at 09:02:07PM -0500, Seneca Cunningha([EMAIL PROTECTED]) wrote: Before anyone tells me to convert the Windows computer, I am not allowed to change _anything_ about that computer. Define change. If you want to set yourself up so there are no successful solutions, that's your own affair. So I will define change. For that system change means altering anything it in any way other than adding programs, or by external connection. Its on a network, and the admin has setup a like Big Brother's, not even allowing me a password on my files, while their entire drives have passwords. They claim that it is merely so they can run virus scans from a central location, but I suspect that in addition to being able to monitor every website that I go to, they use it to check for any games (another condition is no games). If I don't abide by their limitations, all my connections to the outside world of computers are removed. Seneca [EMAIL PROTECTED]
File Transfers, taken from Re: an XFree86 problem, an xscreensaver problem, followed by anEnlightenment problem
I am looking for a fast and reliable method for transferring files between systems. David Maze [EMAIL PROTECTED] wrote: Seneca Cunningham [EMAIL PROTECTED] writes: SC and other such difficulties, could it just be a few bad transfers SC by floppy, as large packages (1.4M) have to be made into a SC spanning archive, transfered, reassembled, then installed. You really have no other way to move files between the machines? There are lots of options besides floppies; you could make Debian CD images if you had a CD burner (http://cdimage.debian.org/), or assemble a small Ethernet-based network, or even if all else failed attach the two computers with a parallel cable (see the PLIP-HOWTO off of http://www.linuxdoc.org/). Or if you're just trying to install, you could order a CD set; http://www.cheapbytes.com/ has Debian 2.2r4 CD sets for $10, plus shipping. In response to the listed options, I have no access to a CD burner, I have no working Ethernet cards, and currently it is assorted packages as I setup the computer that this concerns, but later it will be datafiles 10M, so CD sets don't fit my needs. To the option of using a parallel or serial cable as described in the HOWTOs, it cannot be used for a few reasons. 1. the computer with internet access is a Windows 9x machine a) the computer's Ethernet card is built-in b) the PLIP-HOWTO says that it has not worked with Windows 9x c) the Serial-Laplink-HOWTO requires the Direct Cable Connection program d) Windows fried its files a while ago, killing Direct Cable Connection e) I cannot re-install that piece of fried software, the CD drive does not work 2. my Debian machine has only one parallel port a) this computer is connected to my printer because the Windows computer fried its print drivers Before anyone tells me to convert the Windows computer, I am not allowed to change _anything_ about that computer. I can't buy any replacement hardware as I am a student on a tight budget. Normally I'd spend more time than I have attempting to write my own solution, but I have exams to study for, and final assignments to complete. If anyone cares to know, my answer to this problem works in some DOS environments perfectly using a serial cable, except that the files don't tranfer, well... they don't transfer into a file, they just comes on screen leaving the destination files empty. I know fried is the technical term for electronics, not computers, but it fits... Seneca [EMAIL PROTECTED]
Re: File Transfers
I wish that the solution were so simple as this.. Osamu Aoki [EMAIL PROTECTED] wrote: Hi, $0.00 solutions below: On Fri, Dec 14, 2001 at 09:02:07PM -0500, Seneca Cunningham wrote: I am looking for a fast and reliable method for transferring files between systems. ... In response to the listed options, I have no access to a CD burner, I have no working Ethernet cards, and currently it is assorted packages as I setup the computer that this concerns, but later it will be datafiles 10M, so CD sets don't fit my needs. To the option of using a parallel or serial cable as described in the HOWTOs, it cannot be used for a few reasons. Option 1. split files into small files and move by floppy. dos program should be available somewhere or use cygwin Merging shall be something like: $ cat file1 file2 merged-file copy /b file1 file2 merged-file Option 2. Open box and unplug harddisk. For deb - win, make small vfat directory and drop files. move drive to win machine and boot and read. For win - deb move hdd to deb side and mount it and read Anyway, you seem to have a lot of hardware problems. Fix it first by moving/borrowing cdrom drive. Slow cdrom and ethernet card cost less than $30 only each. 4 hours of work for each at humberger shop. ... but Option 1 is similar to what I do now, and some of my datafiles would need 10 floppies at their current size, and as I build algorithms to get past the limitations of data types, the number of floppies will increase, and Option 2 cannot be done with my systems because of different drivebays. The machine with the bad CD drive uses a standard 3.5 drivebay, but the destination computer uses a 2.5 drivebay (and I thought I finally found a use for the 100M harddisk that is laying on my desk). The amusing part about getting an ethernet card is that I have 2 that are physically compatible with my Debian computer, but both of them are broken. Normally I wouldn't be going on too much about this, but these problems are delaying progress on all of my final assignments for this semester. And it gets a little tedious recording and matching all of the dependancies. Any more ideas for two awkward computers in an awkward situation? Seneca [EMAIL PROTECTED]
Re: File Transfers
Hardware and budgetary issues add difficulty to finding a fast and reliable method for transferring files between systems... Carl Fink [EMAIL PROTECTED] wrote: On Fri, Dec 14, 2001 at 10:21:15PM -0500, Seneca Cunningham wrote: [snip] The amusing part about getting an ethernet card is that I have 2 that are physically compatible with my Debian computer, but both of them are broken. [another] Any more ideas for two awkward computers in an awkward situation? You do realize that two 100BaseT Ethernet cards, compatible with Linux, can be obtained for $25? Why fool around with makeshift stuff, unless you're so poor you can't get the cards by skipping lunch three times? Lunch? What's lunch? Oh... I remember now. That's when I'm in the lab working on my computer science assignments because I don't have money for the cafeteria. And $25? If I were to spend that much, I would not be able to even go to school except by walking for 8km on poorly lit streets at night (transit costs me between $40-$70/month). My budget is extremely tight. But enough about the tight budget of a student. The Debian computer happens to use PCMCIA cards, unlike the connected computer, a Windows98 machine, with a built in ethernet port, that I am not allowed to alter in any way, shape, or form (although I was sorely tempted when Windows zapped its own ability to startup). The question asked before still holds. Any ideas? Seneca [EMAIL PROTECTED]
Re: File Transfers
Carl Fink [EMAIL PROTECTED] wrote: Does the Windows box have a modem? Does your Debian laptop have one? I've done modem-to-modem file transfers before. Okay, this was so long ago they were both running MS DOS 3.3, but it should still work. Can you *borrow* two external modems for a few weeks? Lots of people (e.g. me) have modems we rarely or never use, because we have broadband now. The amazing thing is I located two modems in my area! But the arrangement is incompatible. The modem that can be used with the laptop is a cellphone modem, and the other modem is a cable modem, although I'm not sure about permission to use it. The older phone modems that used to be here are gone. I think they were donated to a daycare centre. If I could get a schematic for a modem, I could build it if it only uses standard, commercially available parts, maybe even getting extra credit in electronics in the process. If this goes on much longer, after exams I'll attempt to fix my possible answer to this problem that partially works in some DOS environments with a serial cable. All I need to do is get the files to transfer properly. Currently, the information doesn't transfer into a file like its supposed to, it just comes on screen leaving the destination files empty. I think its a problem with the timing. I hope Santa brings me some new hardware for Christmas. Seneca [EMAIL PROTECTED]
an XFree86 problem, an xscreensaver problem, followed by an Enlightenment problem
These are three problems that occured in relatively quick succession in the listed order: 1.XFree stopped using its global xinitrc file. It used it a few times, then I changed it from using xserver-vga16 to xserver-svga. Any way to get my computer to start using it again? 2. As root, I do not have the authority to start or chang any screensaver settings, however when I am an ordinary user, I have the authority to start and change the screensaver settings. The first time I used the screensavers I was an ordinary user. Could that have anything to do with the problem? 3.As an ordinary user, I changed window manager to Enlightenment from Blackbox for the first time not as root. Seconds after the menus finished generating, the message: Enlightenment caused Segment Violation (Segfault) This means that Enlightenment or support library routines it calls have accessed areas of your system's memory that they are not allowed access to. This is most likely a bug. It is recommended to restart now. If you wish to help fix this please compile Enlightenment with debugging symbols in and run Enlightenment under gdb so you can backtrace for where it died and send in a useful bug report with backtrace information and variable dumps etc. came onto my screen. What should I do? A slight comment about my Debian system is that all file transfers must be done by floppy, from another computer. After cat messed up the ASCII tables for a terminal until rebooting, impossible to find commands (I can only view man pages using a text editor), and other such difficulties, could it just be a few bad transfers by floppy, as large packages (1.4M) have to be made into a spanning archive, transfered, reassembled, then installed. Thanks for any help, Seneca [EMAIL PROTECTED]
invisible information in dselect
Something that I have noticed in my time using dselect is that in the main package listing, there are some lines that only have partial information on them. This is when it is set to verbose. The information that seems to be invisible is found on the dividing line, but does anyone know why I get lines like: stalled holdhold talled install install when the dividing line has: not installed ; hold (was: hold) not installed ; install (was: install) This only happens when the package is not installed, but these are mild examples. The segment of invisibility continues into the package name at times. Any ideas? Seneca [EMAIL PROTECTED]
XFree86 problem
I have just installed XFree86 onto my computer, tried it with a few configurations that look like what should work, and Fatal server error: No valid modes found and X connection to :0.0 broken (explicit kill or server shutdown keeps coming up. I have the rest of the output on file. About 45 lines come up saying VGA16: ... Deleted. are in that file. Anyone know how to stop this? Seneca [EMAIL PROTECTED]
RE: XFree86 problem, with output
-Original Message- From: Jeff [mailto:[EMAIL PROTECTED] Sent: December 8, 2001 19:50 To: debian-user@lists.debian.org Subject: Re: XFree86 problem Seneca Cunningham, 2001-Dec-08 09:24 -0500: I have just installed XFree86 onto my computer, tried it with a few configurations that look like what should work, and Fatal server error: No valid modes found and X connection to :0.0 broken (explicit kill or server shutdown keeps coming up. I have the rest of the output on file. About 45 lines come up saying VGA16: ... Deleted. are in that file. Anyone know how to stop this? Seneca [EMAIL PROTECTED] You haven't configured a mode that the video can use, so you are getting those messages. You need to reconfigure with the proper information. Make sure you have the following info on your video: - chip set: you can go to www.xfree86.org and find the driver to use for your card/chip set. - monitor hsync and vsync - keyboard and mouse types * I think that's all you need Now, run... # dpkg-reconfigure xserver-xfree86 This will run you through the configuration asking for the above information. After that, go to your normal user mode and run startx to see if it worked. Give it a go, and good luck...jc -- Jeff Coppock Systems Engineer Diggin' DebianAdmin and User I tried # dpkg-reconfigure xserver-xfree86, but it just gave me the message debconf: package xserver-xfree86 is not installed or does not use debconf I have been playing with the modes all day trying to find one that works, and there are no longer 45 lines of settings that are deleted, now there are 2, and still I haven't found anything that works, although when I tried XF86Setup it claimed that you've got a running server. I like having to figure some stuff out for myself, but what hsync and vsync settings to use for an LCD, when everywhere that I have looked claimed that they were ignored with an LCD. Seneca [EMAIL PROTECTED] startx_output Description: Binary data
