Re: chroot()ing a user's login
Hi Robert, > > > > And this works: > > > > expanse:~# chroot /usr/remote/ su - ookhoi > > > > ookhoi $ > > > > > > > > Of course bash is there: > > > > ookhoi $ /bin/bash > > > > ookhoi $ > > > > > > And is it in the chrooted /etc/shells? > > > > Thanx you for your response! Yes, it is: > > > > okhoi $ cat /usr/remote/etc/shells > > # /etc/shells: valid login shells > > /bin/ash > > /bin/bash > > /bin/csh > > /bin/sh > > /usr/bin/es > > /usr/bin/ksh > > /usr/bin/rc > > /usr/bin/tcsh > > /usr/bin/zsh > > /bin/sash > > > > But if not, then a normal login wouldn't work either I think? > > I will try to reproduce it at home It seems to be quite useful for ISP > setups... In the meantime, did you try giving > > chroot /usr/remote /bin/bash > > as your login shell? Of course enter it into /etc/shells as well. I did > not try it yet, but it could even work. You mean in /usr/remote/etc/passwd ? That wil not work, because it then can't chroot to /usr/remote In /etc/passwd it is not nescessary because the chroot works already. If I do a login in stead of a su, I see the motd from /etc/motd at the first login, and the /usr/remote/motd at the second login (you have to type in your password twice for a chroot login). Ookhoi
Re: chroot()ing a user's login
Hi Robert, > > > On Sun, Dec 12, 1999 at 12:04:09PM -0500, Nagilum wrote: > > > > I had read some docs which mentioned that on SysV, you can specify a * > > > > in > > > > the 7th field of the passwd file (thisis from memory, I may be off) and > > > > that user's login will then be chroot()ed to his home directory. > > > > > > > > I was hoping to find a similar functionality in Debian, so I tried the * > > > > in the 7th field and that didn't work. So then I grabbed the source for > > > > login (shadow package) and grepped the source for chroot. In > > > > libmisc/sub.c I found it, along with some commentary: > > > > > > > > /* > > > > * subsystem - change to subsystem root > > > > * > > > > * A subsystem login is indicated by the presense of a "*" as > > > > * the first character of the login shell. The given home > > > > * directory will be used as the root of a new filesystem which > > > > * the user is actually logged into. > > > > */ > > > > > > > > So, I tried changing a user's login shell to '*/bin/bash' to no avail. > > > > When I attempt to login, I am asked for the username.. and then I am > > > > asked > > > > for the password twice and booted out. > > > > > > > > I also tried replacing /bin/login with a re-compiled version from the > > > > (slink) source but the same thing happened. > > > > > > The documentation specifies: > > > > > > 1) Once the user has logged in they are chrooted and asked to login via > > > that password file _in the chrooted directory_. > > > > > > 2) The shell must be available in the chrooted env (as well as all needed > > > bianries). > > > > > > So for this to work, you must have a complete working filesystem in each > > > home directory (/home/foo/dev /home/foo/bin /home/foo/usr/bin > > > /home/foo/etc > > > ...). > > > > I have a password entry at my system like this: > > ookhoi $ grep ookhoi /etc/passwd > > ookhoi:x:1000:1000:Ookhoi:/usr/remote:*/bin/bash > > > > Under /usr/remote I have a complete and working filesystem which I use > > for nfs boots. > > Now I try to login: > > > > ookhoi $ su ookhoi > > Password: > > Password: > > No shell > > > > The first password is the password on my system, the second password is > > the password I use when I do a nfs boot, so the change root works. But > > for some reason I get the "No shell". Can you help me with that please? > > > > The password entry at the nfs boot system is: > > ookhoi $ grep ookhoi /usr/remote/etc/passwd > > ookhoi:x:1000:101::/home/ookhoi:/bin/bash > > > > And this works: > > expanse:~# chroot /usr/remote/ su - ookhoi > > ookhoi $ > > > > Of course bash is there: > > ookhoi $ /bin/bash > > ookhoi $ > > And is it in the chrooted /etc/shells? Thanx you for your response! Yes, it is: okhoi $ cat /usr/remote/etc/shells # /etc/shells: valid login shells /bin/ash /bin/bash /bin/csh /bin/sh /usr/bin/es /usr/bin/ksh /usr/bin/rc /usr/bin/tcsh /usr/bin/zsh /bin/sash But if not, then a normal login wouldn't work either I think? Ookhoi
Re: chroot()ing a user's login
Hi Ben, > On Sun, Dec 12, 1999 at 12:04:09PM -0500, Nagilum wrote: > > I had read some docs which mentioned that on SysV, you can specify a * in > > the 7th field of the passwd file (thisis from memory, I may be off) and > > that user's login will then be chroot()ed to his home directory. > > > > I was hoping to find a similar functionality in Debian, so I tried the * > > in the 7th field and that didn't work. So then I grabbed the source for > > login (shadow package) and grepped the source for chroot. In > > libmisc/sub.c I found it, along with some commentary: > > > > /* > > * subsystem - change to subsystem root > > * > > * A subsystem login is indicated by the presense of a "*" as > > * the first character of the login shell. The given home > > * directory will be used as the root of a new filesystem which > > * the user is actually logged into. > > */ > > > > So, I tried changing a user's login shell to '*/bin/bash' to no avail. > > When I attempt to login, I am asked for the username.. and then I am asked > > for the password twice and booted out. > > > > I also tried replacing /bin/login with a re-compiled version from the > > (slink) source but the same thing happened. > > The documentation specifies: > > 1) Once the user has logged in they are chrooted and asked to login via > that password file _in the chrooted directory_. > > 2) The shell must be available in the chrooted env (as well as all needed > bianries). > > So for this to work, you must have a complete working filesystem in each > home directory (/home/foo/dev /home/foo/bin /home/foo/usr/bin /home/foo/etc > ...). I have a password entry at my system like this: ookhoi $ grep ookhoi /etc/passwd ookhoi:x:1000:1000:Ookhoi:/usr/remote:*/bin/bash Under /usr/remote I have a complete and working filesystem which I use for nfs boots. Now I try to login: ookhoi $ su ookhoi Password: Password: No shell The first password is the password on my system, the second password is the password I use when I do a nfs boot, so the change root works. But for some reason I get the "No shell". Can you help me with that please? The password entry at the nfs boot system is: ookhoi $ grep ookhoi /usr/remote/etc/passwd ookhoi:x:1000:101::/home/ookhoi:/bin/bash And this works: expanse:~# chroot /usr/remote/ su - ookhoi ookhoi $ Of course bash is there: ookhoi $ /bin/bash ookhoi $ Thanx for any pointers! Ookhoi
Re: diskless
Hi Marcin Kurc,
> Anyone have any experience with diskless debian?
> What would be the best way to create root directories for the diskless
> machines?
Install Debian, mount server:/usr/remote /mnt, copy / to
/mnt, set up bootp, config your kernel with
IP: kernel level autoconfiguration
CONFIG_IP_PNP
BOOTP support
CONFIG_IP_PNP_BOOTP
RAM disk support
CONFIG_BLK_DEV_RAM
Initial RAM disk (initrd) support
CONFIG_BLK_DEV_INITRD
NFS filesystem support
CONFIG_NFS_FS
Root file system on NFS
CONFIG_ROOT_NFS
then
dd if=/usr/src/linux/arch/i386/boot/bzImage of=/dev/fd0
mknod /dev/nfs c 0 255
rdev /dev/fd0 /dev/nfs
and you can boot every computer with or without hard disk in it. It's
cool, because you can show Linux to windows users, or you can prepare a
hard disk with al the tools a full system has.
Good luck!
Ookhoi
Re: netfilter
Hi Marcin Kurc,
> Does anyone compiled netfilter on debian? (0.1.11 or 0.1.12)
/me, yes.
Ookhoi
Re: PPP not working after reinstall
> > I'm sending this e-mail using Windowze because my Linux box was > > reinstalled (the primary HD was broken) and PPP isnt working. I used > > pppconfig > > to configure. I estabilish the connection with pon but the connection's > > velocity doesnt pass above 0 bytes, even pinging or ftping. > > What can be wrong? > > Thanks for any help. Using Windows is getting me nervous. :) > > May be You need "defaultroute" option for pppd ? > Take a look at route output after connection and see if there a deafult route > through > ppp0 ? Last week I installed Potato, and I didn't get PPP to work. I check and changed a lot in the ppp config, but nothing helped. It indeed doesn't seem to create a defaultroute (but of course defaultroute was 'on'). It could still be silly me, but now I read this, it could also be a Potato problem? Ookhoi
Re: frontpage extensions for apache
Hi Gareth, > > > Has anyone succesfuly install the extensions on a debian box befoe? > > > If someone has I would appreciate any help and or advice on the subject. > > I installed it on Solaris. Can you post your error messages and > > versions? > > Its frontpage 98 server extensions version3.0 I installed fp 2000 (that's version 4 I believe). > I checked the config file exists I even added the Port and > ServerRoot directive > (although I am not sure on the values) to the default config file when it > didn't work. > ok this is the error message I get from the script > > Installing root web into port ... > installing server / on port > Will chown web to root as part of install. > Will chgrp web to daemon as part of install. > Server config file "/usr/local/frontpage/version3.0/we.cnf" has no Port > or ServerRoot directive. Please check that you have the correct config file > for your server type. > ERROR: / installation failed. Ah, so you want to install a _web_. I thought you wanted to install the module in the webserver. Ehm, I don't know what is wrong then.. :-( Sorry. (on Solaris it is a binary btw, not a script). Ookhoi
Re: frontpage extensions for apache
Hi Gareth, > I am trying to install the Frontpage extensions for apache > I haven't been able to get the install script to work so I have been hacking > it to bits. > > however it seems that it gets stuck on the config file and I don' know > and can't find what exactly it is looking for/trying to do. > > Has anyone succesfuly install the extensions on a debian box befoe? > If someone has I would appreciate any help and or advice on the subject. I installed it on Solaris. Can you post your error messages and versions? Ookhoi
Re: potato and 2 processors
> Luckly, I'll have a new machine this week. It's a dual pentium > III. > How good is potato, mainly 2.2.x (x>= 13), in SMP? > I means: how stable is it? (kernel and potato) with this config. > I'm already running potato in my personal workstation. It's fine. > But the above machine will be a server. It is not Potato, but the kernel, I think. We run potato at a dual PII 450 with kernel 2.2.7, and it is 91 days up now (= a webserver). Ookhoi
Re: secure pop3 via ssh
> How can I set a secure pop3 server using ssh?
> I read the man page for fetchmail and there is an example of a
> secure conection via ssh.
> How do I implement the server?
If you use ssh to connect to the pop3 server (as I do), you don't need
special things for the pop3d.
If you want to secure the pop3d, you might want to look for ssl I
_think_ (but not all clients are able to fetch mail via ssl?).
Ookhoi
Re: Compiling ATA/66 support into the kernel?
> >> use 2.3.x kernel for ata/66 > > Does it support booting off a ata/66 drive? The patches for 2.2.13 say > > that booting off of ata/66 drive is not supported. > > The driver author made a mistake. You can boot off drives connected to > the HPT366 controller. Take a look here: > > http://www.csie.ntu.edu.tw/~b6506063/hpt366/ > > However I still suggest you connect your disk to the UDMA/33 channels > for the time being, because /dev/hde to /dev/hdh do not exist on the > boot floppies, thus there will be no easy way installing Debian on to > a disk on HPT366. I installed Debian with the new bootfloppies on the HPT366 controller. You just open the second virtual console, fdisk the /dev/hde, mkfs it, and mount the partitions. Then go back to console 1, choose unmount, say cancel there (you see them mounted) and you can go on with the install. Why don't the boot floppies support /dev/hde and up? Ookhoi
Re: can I list just the directories, executables?
> > I have > > alias dir='ls -al | grep ^d | grep -v grep' > > > > I take it you copied that alias from one involving ps, > where you need the last bit to stop it seeing itself. > You don't want it here. Shit ;-) Thanx! Ookhoi
Re: can I list just the directories, executables?
> >> Hi. I would like to know if there is a command that will list just > >> directories? > > > > Do a > > > > man ls > > > > There, press / (for searching in the manual). > > > > Type -d (the text to search). > > Type n twice (for next occurence of the searchtext). > > > > See? The option -d is it. > > I don't think this is what is wanted. -d just keeps ls from listing the > conents of the directory when you specify the directory name on the > command line. > > To list just the directories in the current directory you can do > > find . -type d -maxdepth 1 > > or you can feed the output to ls to format it better > > find . -type d -maxdepth 1 | xargs ls -d > > and add other args to ls as needed. I have alias dir='ls -al | grep ^d | grep -v grep' in my /etc/profile. I like it. :-) Ookhoi
Re: boot messages (where are they?)
> The original question was about suppressing verbose messages at boot.
My question: I want to know what the computer said at boot time.
Something like dmesg. dmesg isn't useful anymore after a few months
uptime, because it only contains new information which has pused the old
boot messages out of its memory. I _thought_ that there was a file
called dmesg in /proc somewhere, but I can't find it, nor something
similair. Can somebody puss my nose in the right direction? Thanx!
Ookhoi
Re: keymap error when closing X
Hi Brandon, > In message <[EMAIL PROTECTED]>, Ookhoi writes: > +- > | > I get the following when closing Xwindows. > | > > | > What does this mean? > | > How do I fix it? > | > > | > System: `/usr/X11R6/lib/X11/xkb/xkbcomp -w 1 -R/usr/X11R6/lib/X11/xkb > -xkm > | -m us -em1 "The XKEYBOARD keymap > | > compiler (xkbcomp) reports:" -emp "> " -eml "Errors from xkbcomp are not > fa > | tal to the X server" > | > keymap/xfree86 /var/tmp/xfree86.xkm' > | > xinit: connection to X server lost. > | > > | > I wonder if part of the problem is that 'us' is an empty file in > /usr/X11R6 > | /lib/X11/xkb/symbols > | > | I have the same message when I _start_ X. As soon as the XDM loginscreen > | is there, or when X is running, the mouse and keyboard don't work. They > | do during boot. Don't know if the message is related with the malfunctioning > | of the keyboard and mouse. > +--->8 > > It's not related. Also, it's not an error; if you look closely, the "error" > strings are *arguments* to the command. > > It's just the X server telling you that it is executing a command. Said > command has arguments that cause it to document its log output *if it has > any*, which it apparently doesn't because the "Errors from" line only occurs > as an argument and not by itself. > > As for the "xinit: connection to X server lost", that usually indicates that > you used CTRL-ALT-BackSpace to abort the X server (although it *could* > indicate the server died for some other reason). I killed the X server via a remote login. Of course I couldn't kill the server with a ctrl-alt-backspace, because keyboard didn't work. > In any case, X server problems are not appropriate for linux-kernel; X > doesn't have a kernnel component. I know, but as I wrote in my previous message, with kernel 2.3.21 prepatch 2 the keyboard and mouse die, and with kernel 2.3.19 X works fine. That is the reason I cc'ed it to linux-kernel. You did cut that part of my message out of your reply. Ookhoi
Re: keymap error when closing X
Hi Lance, [cc to kernel mailinglist] > I get the following when closing Xwindows. > > What does this mean? > How do I fix it? > > System: `/usr/X11R6/lib/X11/xkb/xkbcomp -w 1 -R/usr/X11R6/lib/X11/xkb -xkm -m > us -em1 "The XKEYBOARD keymap > compiler (xkbcomp) reports:" -emp "> " -eml "Errors from xkbcomp are not > fatal to the X server" > keymap/xfree86 /var/tmp/xfree86.xkm' > xinit: connection to X server lost. > > I wonder if part of the problem is that 'us' is an empty file in > /usr/X11R6/lib/X11/xkb/symbols I have the same message when I _start_ X. As soon as the XDM loginscreen is there, or when X is running, the mouse and keyboard don't work. They do during boot. Don't know if the message is related with the malfunctioning of the keyboard and mouse. This is with kernel 2.3.21 pre-patch-2.3.22-2 Kernel 2.3.19 is fine (didn't test another) Mouse and keyboard are PS/2 No gpm (disabled) Debian Potato: XFree86 Version 3.3.5 / X Window System (protocol Version 11, revision 0, vendor release 6300) Release Date: August 23 1999 If I kill X via a remote login, keyboard works again (at console thus) I would be happy to supply more info if nescessary. Ookhoi
Re: FW: FTP Install
Hi Cory, That's strange. I _always_ install via ftp, and it works very well. I started with Debian as a newbie, and found the installation manual very clear and valuable. Nowadays I can only find this one: http://www.debian.org/releases/stable/i386/install It has several pages; quite difficult to print, but I guess the old long one is still alive? Anyway, it's as easy as an install via cd. The only thing is that you need to get your modem up and running, which can be difficult for a newbie (like me in those days). Good luck! And feel free to mail me (us :-) if you want to know something. Groetjes, Ookhoi > -Original Message- > From: Cory Rudder > Sent: Tuesday, August 31, 1999 12:13 > To: 'Patrick Olson' > Subject: RE: FTP Install > > > Thanks for your response, For the info of everyone it seems that FTP install > capabilities are not yet a reality for debian. In my experience installing > debian is no longer a reality. I have in my possession three different > debian CD's. One I created on the ftp site, one from cheap bytes, and one > from the Debian GNU / Linux; Guide to installation and usage. I have tried > to install all of these CD's on three different machines. I have yet to be > successful. All of my machines get hung up on the "now booting the kernel.." > message. I am at a lose for what to do. Maybe I have just been spoiled by > the ease of which freebsd installs itself. > -Original Message- > From: Patrick Olson [mailto:[EMAIL PROTECTED] > Sent: Tuesday, August 31, 1999 11:31 > To: Cory Rudder > Subject: Re: FTP Install > > > On Mon, 23 Aug 1999, Cory Rudder wrote: > > I'm a debian newbie and I would like to ftp install a stable version of > > debian what file do I need to download to do this? Also recommendations on > > different versions would be helpful. > > Thanks. > > It looks like no one answered your question, so I'll give it a try. I > don't actually know anything about FTP install, but here's a couple of web > pages that look like they might help: > > http://www.debian.org/releases/slink/i386/ch-install-methods.en.html > http://www.debian.org/releases/slink/i386/install > > Hope this helps, > Patrick > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: more RAM = more speed?
Hi Patrick, At home, I run a P133 with 48 meg, but most of the time only text, because of my bad monitor and card. At work, I run a P133 with 128 meg SDRAM. _That_ is just perfect, although I still need swap (and 28 meg of swap is not enough.. :-( ). Netscape (4.61 I believe) needs a restart almost every day, or else swap gets full, and the machine starts to freeze. A P133 is fast enough for X with Netscape (and StarOffice, but I only tried it (quite cool)) and al kinds of servers like web, ftp, dns, irc etc, so I guess you should buy the extra memory if you think your computer is worth it (especially the monitor/vcard imho). HTH. Groets, Ookhoi > > In your case of course, more memory would show *huge* speed increases. It > > looks like another 32 megs would stop the immediate swapping problem. If you > > want to run StarOffice, I imagine another 16 to 32 megs above that is > > required. I don't have star office installed myself, so I can't check for > > you. :( (one more week... :) > > I have StarOffice on a very similar machine. This is the report it > gives: > > total used free sharedbuffers cached > Mem: 30264 29616648 24884128 17600 > -/+ buffers/cache: 11888 18376 > Swap: 128484 5192 123292 > > I got there by this: > 1. boot Linux (was running DOS game, sorry) > 2. log into xdm and start StarOffice > 3. open a new (empty) text document > 4. switch to tty1 and login to run free > > It looks like StarOffice is quite a memory hog too. It sounds like 64MB > is enough to make Netscape happy. > > Based on the free output above, do you think 96MB would be enough to keep > swapping to a minimum with StarOffice thrown in? If you're not sure, I'd > be happy to wait a week or so until you have StarOffice installed. > > > If you can afford it, the new mobo+cpu+ram would probably be the right thing > > to do. > > Actually, my crazy idea was running my existing CPU in a new board for a > while until I want to (and can afford to) upgrade it. But unless the new > board has 3 ISA slots, I don't think I want to do that. Adding a net > card, sound card and 56K modem to about $280 for the board, CPU and RAM is > a bit to much. > > Thank you, > Patrick Olson > > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: change cursor
Hi Cuno, > How can I change my cursor. > I've looked in /X11/cursorfont.h. > > And I've put something like this in bash_profile. > > xsetroot -cursor_name left_ptr > > But that didn't work. > Does anybody know how I can change the cursor from X to an arrow I use fvwm2 and have this: Exec xsetroot -cursor ~/X/custom.cursor1.xbm ~/X/custom.cursor1.xbm in the file: ~/.fvwm2/post.hook That file looks like this btw: EdgeScroll 100 100 EdgeResistance 100 100 Execxsetroot -cursor ~/X/custom.cursor1.xbm ~/X/custom.cursor1.xbm Execxv -root ~/X/background.gif -quit HTH! Groetjes, Ookhoi
Re: partition problem
Hi Hammy, > I'm having a problem partitioning my hard drive with the install program. > Once I run install.bat, I can setup the colour monitor and the keyboard. The > next step is to partition the hard drive and this is where I am having > problems. > I keep getting the following error: > 'Fatal error: Bad primary partition. Press any key to exit.' > > I have a 1.2 GB hard drive that is formatted with fat16. I also tried the > install from floppy but I get the same error. > > I hope someone can help me. I'm really interested in getting this working. Maybe you have more luck with the Linux fdisk? I had problems too with the partitionprogram that is used with the installation. Groetjes, Ookhoi
Re: minicom -> dumb terminal?
> Has anyone used minicom to provide a dumb terminal using a null modem cable? Yes. I wanted to be able to compile and test a kernel on a remote Linux server, so I had to be able to use Lilo in case the fresh kernel couldn't boot the machine. > If so, any tricks to that? If you configure your kernel, say yes to: Support for console on serial port CONFIG_SERIAL_CONSOLE If you say Y here, it will be possible to use a serial port as the system console (the system console is the device which receives all kernel messages and warnings and which allows logins in single user mode). This could be useful if some terminal or printer is connected to that serial port. Even if you say Y here, the currently visible virtual console (/dev/tty0) will still be used as the system console by default, but you can alter that using a kernel command line option such as "console=ttyS1". (Try "man bootparam" or see the documentation of your boot loader (lilo or loadlin) about how to pass options to the kernel at boot time. The lilo procedure is also explained in the SCSI-HOWTO, available via FTP (user: anonymous) in ftp://metalab.unc.edu/pub/Linux/docs/HOWTO.) If you don't have a VGA card installed and you say Y here, the kernel will automatically use the first serial line, /dev/ttyS0, as system console. And I have this in /etc/lilo.conf on the remote machine: [cut] vga=normal delay=20 default=2.2.7 serial=0,9600n8 append="console=ttyS0 [cut] Hth. Good luck! It's a nice feature. Groetjes, Ookhoi
Re: Somebody's scanning my ports or what?
> > im getting this kind of things about 2 or 3 times a week... > > with some netbus and bo try. > > > > i sent lot of emails to their isp but still receive 2 or 3 attack per > > week from other ips... > > > > what should i do?, is there a way to protect me against this? currently i > > have fakebo. > > > > Benoit Joly > > > > You can not prevent your system from being scanned, all you can do is > control what is learned from the scan. It is not a crime, as far as I > know, to simply attempt connections to a machine on the public internet. We (as a ISP) do report portscans to the provider of the scanner. And almost always we get the reply that the scanner has been warned or someting similar. A portscan can only be used for criminal actions, so providers do see is as a crime. Groetjes, Ookhoi > First thing I would do is only have ONE machine exposed directly to the > internet. Use this machine as a firewall/gateway for all the other > machines. Turn off all services on this machine that you are not using. > Carefuly plan and put into place a set of packet forwarding / masquerading > rules for traffic between your internal protected network and the public > internet. For services that you wish to provide to outside hosts, make a > separate network different from the internal network. This is commonly > called a "DMZ" in network documents. > > So your internet firewall / gateway will probably have THREE interfaces if > you wish to provide public access to some services: > > 1. The interface to the external internet. > 2. The interface to the private local network. > 3. The interface to the internal network with public services (www, > ftp, mail, news, etc.) > > The whole world can access certain ports in your public access net ( 80, > 21, 23, 25, 119, etc) Nobody in the outside world has direct access to > your internal net. Nobody on the public access net has access to your > internal net and your internal net has access to everything.
Re: screen corruption
Hi Carl, > Question: when I cat a binary file to the screen, all the characters > get replaced with graphical symbols. > > Sometimes, 'reset' will fix this, and sometimes it won't. > > Is there some utility that can restore such a corrutped screen > in every case? > > What happens is that I slowly lose virtual terminals due to this > process until I need to reboot to have enough. Does this message help? Groetjes, Ookhoi Date: Thu, 15 Apr 1999 14:47:59 +0200 From: Ookhoi <[EMAIL PROTECTED]> To: debian-user@lists.debian.org Subject: Re: how to reset the console? > I read it in some HOWTO, I can't remember witch one. > It is an ANSI escape sequence to reset the terminal. It is handled by the > terminal. > > To type ^[c press control-v esc c. Cool. :-) So now we have: reset stty sane -v c to restore a messed up text interface. Groetjes, Ookhoi > > > did you tried 'echo ^[c' ? > > > > What does this do? Or where can I find docs about this? Tia.
Re: Disk geommetry, was Re: Kernel Upgrade: Why?
> Think of it as a record (ya' know those old odd looking vinyl things). > It spins at 33.3 rpm but the sound/music doesn't change from outer to inner. > Same deal with hard drives although the outside is 'spinning' faster, it > still picks up the same amount of data per rotation as it would near > the center. I don't think this is true. This is a part from http://www.quantum.com/src/storage_basics/c3.5_part2.html#geometry In earlier hard drive designs, the number of sectors per track was fixed and, because the outer tracks on a platter have a larger circumference than the inner tracks, space on the outer tracks was wasted. The number of sectors that would fit on the innermost track constrained the number of sectors per track for the entire platter. However, many of today's advanced drives use a formatting technique called Multiple Zone Recording to pack more data onto the surface of the disk. Multiple Zone Recording allows the number of sectors per track to be adjusted so more sectors are stored on the larger, outer tracks. By dividing the outer tracks into more sectors, data can be packed uniformly throughout the surface of a platter, disk surface is used more efficiently, and higher capacities can be achieved with fewer platters. The number of sectors per track on a typical 3.5-inch disk ranges from 60 to 120 under a Multiple Zone Recording scheme. Not only is effective storage capacity increased by as much as 25 percent with Multiple Zone Recording, but the disk-to-buffer transfer rate also is boosted. With more bytes per track, data in the outer zones is read at a faster rate. Quantum Corporation is a pioneer in Multiple Zone Recording, and was the first manufacturer to implement Multiple Zone Recording on 2.5-inch disk drive products. Groetjes, Ookhoi > > -Original Message- > > From: Richard Harran [SMTP:[EMAIL PROTECTED] > > Sent: Friday, April 23, 1999 8:35 PM > > To: Jonathan Guthrie > > Subject:Re: Disk geommetry, was Re: Kernel Upgrade: Why? > > > > I think that you are missing a very important point here. A hard disk, > > unlike, for example, an audio CD-ROM, spins at a fixed angular > > velocity. Thus the 'linear' speed over the disk surface is faster > > towards the outside of the disk than towards the centre (as in v=wr). > > Thus at the > > very least, it would seem likely that the reading of a large amount of > > data which was continuous would be quicker from the outside of the disk > > than from the innner. (This probably represents a larger performance > > increase with the large web-server than with the average single-user > > system). > > > > That said, I take your point about the seek time depending on the > > distance that the heads have to move, and it is evident that multihead > > drives would be expected to improve performance. > > > > Rich > > > > Jonathan Guthrie wrote: > > > > > > On Thu, 22 Apr 1999, Ookhoi wrote: > > > > > > > Well, there was a discussion here about a benchmark Linux vs NT, and > > > > some people here said that the preformance of Linux could have been > > > > affected by the fact that Linux was near the center, and NT on the > > outer > > > > side. > > > > > > Probably not the reason. > > > > > > > And the data on the outer side passes the heads much faster than the > > > > data on the inner side. But then, there is much more data on the outer > > > > side, and a piece of data on the outer side will go round in the same > > > > amount of time as a piece of data on the inner side.. > > > > > > I am not aware of any disks that use a higher density recording format > > for > > > the outer tracks than they do for the inner tracks. As far as I am > > aware > > > (and I really haven't paid much attention to such things since ST-277's > > > were state of the art) the bit density of the outer tracks is LOWER than > > > the bit density of the inner tracks. That's because the outer tracks > > are > > > physically larger, but they hold the same number of bits. > > > > > > Not that it matters. The whole disk spins as a single unit so even if > > > there were more bits on the outer tracks, you'll still wait the same > > > amount of time (on average) for the sector you want to come around. > > Read > > > on, and I'll explain. > > > > > > > So, is there an advantage if whe put for example swap at the outer > > side > > > > of a disk? > > &g
Re: Disk geommetry, was Re: Kernel Upgrade: Why?
> > And the data on the outer side passes the heads much faster than the > > data on the inner side. But then, there is much more data on the outer > > side, and a piece of data on the outer side will go round in the same > > amount of time as a piece of data on the inner side.. > > I am not aware of any disks that use a higher density recording format for > the outer tracks than they do for the inner tracks. As far as I am aware > (and I really haven't paid much attention to such things since ST-277's > were state of the art) the bit density of the outer tracks is LOWER than > the bit density of the inner tracks. That's because the outer tracks are > physically larger, but they hold the same number of bits. There is more data on the outer tracks nowadays: http://www.quantum.com/src/storage_basics/c3.5_part2.html#geometry > Not that it matters. The whole disk spins as a single unit so even if > there were more bits on the outer tracks, you'll still wait the same > amount of time (on average) for the sector you want to come around. Read > on, and I'll explain. [knip] > An extreme example of this would be where you dedicate an entire drive to > a (fairly small) swap partition. That's how the news servers I use do it. > For something less extreme, I kind of like the recommendation made by OS/2 > gurus: Their advice was to put the swap file in the most used partition > on the least used drive. You might try something like that where you put > the swap partition in the middle of a disk that isn't used for very much. > > In short, my recommendation for boosting the performance of a computer > that uses a significant amount of swap is to add RAM to the computer. Thanx a lot for your explanation! Groetjes, Ookhoi
Re: Beep
> >> > > How to make beep? > >> > > I don't want to do peintf "\a" but i have a process which has no > >> > > controlling > >> > > terminal (from atd/cron) > >> > >> Ookhoi: > >> > I just echo \a to /dev/tty8 It doesn't bother me there, and it makes a > >> > beeb (from out of a .procmailrc :-) > >> > >> You could also use the "wall" command - it notifies all logged-in users. > >> > >> Advantages: > >> * also displays a message > >> * notifies people even over telnet etc > >> > >> Disadvantages: > >> * distracting message presented to users > >> * does nothing if no-one is logged in > > > > Yeah, but the disadvantage _is_ that it actualy writes something to your > > screen. I just want the sound. :-) > > But thanx of course! > > You could play a wav or ulaw file. 'play /path/to/file.wav' would > work. This doesn't care about consoles or anything, as long as the > account that the command is run under has write access to the /dev/dsp > device you should be fine. I don't have a soundcard installed... :-( > This has the advantage of letting you > select the sound for easy descrimination between some other random > beep Well, it is an annoying beeb every second for one minute, so it just can't miss my attention. :-) Groetjes, Ookhoi
Re: Beep
Hi! > > > How to make beep? > > > I don't want to do peintf "\a" but i have a process which has no > > > controlling > > > terminal (from atd/cron) > > Ookhoi: > > I just echo \a to /dev/tty8 It doesn't bother me there, and it makes a > > beeb (from out of a .procmailrc :-) > > You could also use the "wall" command - it notifies all logged-in users. > > Advantages: > * also displays a message > * notifies people even over telnet etc > > Disadvantages: > * distracting message presented to users > * does nothing if no-one is logged in Yeah, but the disadvantage _is_ that it actualy writes something to your screen. I just want the sound. :-) But thanx of course! Groetjes, Ookhoi
Re: SSH 2
> where can I find ssh 2, please? ftp.cs.hut.fi Groetjes, Ookhoi
Re: Disk geommetry, was Re: Kernel Upgrade: Why?
> > > Would I see a performance increase if I made sure my linux partition > > > was on the outside of my disk? Would I be even better putting my swap > > > partition as far out as possible? > > > > the partitioning HOWTO talks about this; > > > > http://webdocs.essex.ac.uk/LDP/HOWTO/mini/Partition-3.html#ss3.3 > > > > go on, do it - you know you want to... > > The HOWTO, and the person responding to the Mindcraft survey, both assert > that the part of the disk furthest from the spindle is the fastest, and that > one can use this to optimise performance. While I don't doubt the first part > of the statement, I was under the impression that once you hit a drive's > onboard logic, geometry is pretty much up for grabs these days, and > attempting to put something on the "outer edge" of the disk is an excerise in > self-delusion. The HOWTO is dated 1997, and the information therein may be > older. Does anyone out there have some knowledge of current hard drive > manufacturing? Or better yet, some numbers from formal or informal > experiments in drive partition performance? Well, there was a discussion here about a benchmark Linux vs NT, and some people here said that the preformance of Linux could have been affected by the fact that Linux was near the center, and NT on the outer side. And the data on the outer side passes the heads much faster than the data on the inner side. But then, there is much more data on the outer side, and a piece of data on the outer side will go round in the same amount of time as a piece of data on the inner side.. So, is there an advantage if whe put for example swap at the outer side of a disk? Groetjes, Ookhoi
Re: The case of the missing 64 meg...
> >> Linux won't recognize memory above 64 megs by itself. You need to add > >> 'mem=128m' in your lilo.conf file. (or rather append = 'mem=128m'). Check > >> out the lilo.conf man page for better details, I can't remember the exact > >> syntax at the moment. > > > > Just want to add that another option would be to switch to a 2.2.x > > kernel - they automatically recongize memory >64. > > So they say. I have a AOpen AX6B MB with the latest bios and I am > running 2.2.5 and I still have to add the mem= line to get my full 128M > recognized. My 2.2.3 does find the 128 meg ram without any help from lilo.conf (AMI bios I believe..).
Re: Beep
> How to make beep?
> I don't want to do peintf "\a" but i have a process which has no controlling
> terminal (from atd/cron) and it has to make some beep /dev/beep ,/dev/pcsp
> don't work.(no "char-modules-1{0,3}) but in kernel configuration I haven't
> found anything like this
I just echo \a to /dev/tty8 It doesn't bother me there, and it makes a
beeb (from out of a .procmailrc :-)
Re: how to reset the console?
> I read it in some HOWTO, I can't remember witch one.
> It is an ANSI escape sequence to reset the terminal. It is handled by the
> terminal.
>
> To type ^[c press control-v esc c.
Cool. :-)
So now we have:
reset
stty sane
-v c
to restore a messed up text interface.
Groetjes, Ookhoi
> > > did you tried 'echo ^[c' ?
> >
> > What does this do? Or where can I find docs about this? Tia.
Re: how to reset the console?
Hi Giuseppe,
> did you tried 'echo ^[c' ?
What does this do? Or where can I find docs about this? Tia.
Groetjes, Ookhoi
Re: ATI RAGE IIC PCI
> -> > i did the same (d/l mach64 Xserver from xfree86.org) > -> > but the Xserver eats lot of my memory, top says it's 50% (i have 32 megs > of > -> > RAM) > -> > > -> > probably something's wrong... > -> > -> My Xserver (XF86_SVGA) eats 17388kb mem. I believe that is normal. > > i wouldn't say that :) > my SVGA ets only 5 MB with communicator executed and on 2MB video card > (1024x768x64k cols) > > this ate ~16 MB on 1152x864 in 32bit mode Mine is stable at the 17 meg, 1280 x 1024. How can I determine the mode it is in? Groetjes, Ookhoi
Re: ATI RAGE IIC PCI
> i did the same (d/l mach64 Xserver from xfree86.org)
> but the Xserver eats lot of my memory, top says it's 50% (i have 32 megs of
> RAM)
>
> probably something's wrong...
My Xserver (XF86_SVGA) eats 17388kb mem. I believe that is normal.
Groetjes, Ookhoi
Re: Any way to convert Word 7 files?
> I've a nasty feeling the answer is going to be "no", but I wonder whether
> there is any way to convert or read a Word 7 file on linux?
>
> Or do I have to borrow a Windows machine and convert it to txt?
strings document.doc > document.txt works fine and fast.
Groetjes, Ookhoi
Re: Apache
> I tried to configure apache. Apparently, everything is OK (at boot time, > it appears to start), but > in fact it doesnt'run. > I get the following message in /var/log/apache/error.log: > > [emerg] httpd: Your kernel was built without CONFIG_SYSVIPC > [emerg] (38) Function not implemented : could not call shmget > > I recompiled the kernel, but I don't know to what option > "CONFIG_SYSVIPC " corresponds. > > I'm running debian 2.2.4 kernel. > > Can somebody help me ? >From the docs: System V IPC CONFIG_SYSVIPC Inter Process Communication is a suite of library functions and system calls which let processes (running programs) synchronize and exchange information. It is generally considered to be a good thing, and some programs won't run unless you say Y here. In particular, if you want to run the DOS emulator dosemu under Linux (read the DOSEMU-HOWTO, available via FTP (user: anonymous) in ftp://metalab.unc.edu/pub/Linux/docs/HOWTO), you'll need to say Y here. You can find documentation about IPC with "info ipc" and also in section 6.4 of the Linux Programmer's Guide, available via FTP (user: anonymous) from ftp://metalab.unc.edu/pub/Linux/docs/LDP/programmers-guide. Saying Y here enlarges your kernel by about 7 KB. Just say Y. HTH. Groetjes, Ookhoi
Re: ATI RAGE IIC PCI
> -> > I tried to set up XFree for given card. > -> > I even don't know what card should I define, the given card was not in > any > -> > list in docs... can some1 help me ? > -> > > -> > I think It's Mach64 chipset, true ? > -> > -> What does /usr/X11R6/bin/SuperProbe says? The output from SuperProbe > -> might help you. > > First video: Super-VGA > Chipset: ATI (chipset unknown) (Port Probed) > Signature data: 3a004756 (please report) > Memory: 0 Kbytes > RAMDAC: Generic 8-bit pseudo-color DAC > (with 6-bit wide lookup tables (or in 6-bit mode)) > Attached graphics coprocessor: > Chipset: ATI Mach64 > Memory: 16384 Kbytes > > ehm. that's (as i said) ATI RAGE 2C with 4 megs of RAM and it seems now it's > not supported in XFree, right ? Take a look at http://www.xfree86.org/cardlist.html It doesn't say 2C, but maybe it helps? Groetjes, Ookhoi
Re: How to determine actual modem speed
Hi Dave, > > Thanx! I have an internal Dynalink and an external E-tech, both 56k. Do > > you have any advice for me how to tune them so that they communicate as > > fast as possible? Most likely I have to replace the Dynalink with an > > identical E-tech, but then? What mru and mtu should I try? And what mnp > > block size? There are three exchanges between the modems. > > I haven't spent much time messing around with these parameters, maybe a > dejanews search would yield some relevent info. IIRC, you can't get a 56K > connection unless one of the modems is directly connected to a digital phone > line (i.e. on an ISP's rack), so if you are getting 33.6 that may be as good > as it gets. I heard that before, so I asume that is true. :-[ Have to look at the parameters. > > Maybe a stupid question, but how can I send the command to the modem? I > > connect with the pppd and a chat script. After that minicom -o doesn't > > let me in: "Device /dev/ttyS3 is locked." > [...] > > That's why installed uucp, because cu doesn't do this. Later, I figured out > that if you go into the "Modem and Dialing Parameter Setup" screen I could > add the info command (I6) ahead of the Reset command (Z) I would get my info > once as minicom sets up. Also while the device is locked? I'll try the uc command. Thanx for your help and information! Groetjes, Ookhoi
Re: ATI RAGE IIC PCI
> I tried to set up XFree for given card.
> I even don't know what card should I define, the given card was not in any
> list in docs... can some1 help me ?
>
> I think It's Mach64 chipset, true ?
What does /usr/X11R6/bin/SuperProbe says? The output from SuperProbe
might help you.
Groetjes, Ookhoi
Re: chat serial port configuration
Hi Tomas, > Hello, I am trying to set up a PPP connection to my ISP. > Using Minicom I can dial, log in and start PPP, so far so good. > But, how do I specify which COM port the chat program should use? > I cant find a word about it in the man page. Maybe you want to try pppd? I believe this works... /usr/sbin/pppd 115200 /dev/ttyS3 connect chat "" ATZ OK ATDT CONNECT "" ...but you want to check. /usr/bin/pon /etc/chatscripts/provider /etc/ppp/peers/provider (others?) HTH! Groetjes, Ookhoi
Re: vi GUI for X ?
> > Also, is the only way to avoid "broken words" at the end of a line? > > If you want to wordwrap automatically, then some of the vi variants can > do that for you. In vim, say ":set tw=72" in command mode. I use :setwrapmargin=8 and have the file ~/.exrc which contains: set wrapmargin=8 set tabstop=3 But this is plain vi (or does it work for vim too?). Groetjes, Ookhoi
Re: How to determine actual modem speed
> > Is there a way to measure the actual speed at which a modem is > > connected? As modems negotiate the speed al the time, the CONNECT string > > at the start of the connection is not up to date some time later. > > Depending on your modem model, there's probably an AT command which will > give this information. For my USR Courier, it's ATI6. My older modem was a > Cardinal internal which IIRC used a Rockwell chipset, and used the command > AT&V1. Thanx! I have an internal Dynalink and an external E-tech, both 56k. Do you have any advice for me how to tune them so that they communicate as fast as possible? Most likely I have to replace the Dynalink with an identical E-tech, but then? What mru and mtu should I try? And what mnp block size? There are three exchanges between the modems. > If I reset the Courier (ATZ), it will forget the speed. To get around this > you can either use cu from the uucp package, or modify the modem init string > in minicom from ATZ to ATI6Z. Maybe a stupid question, but how can I send the command to the modem? I connect with the pppd and a chat script. After that minicom -o doesn't let me in: "Device /dev/ttyS3 is locked.", although the manuals says: -o Do not initialize. Minicom will skip the initializa- tion code. This option is handy if you quitted from minicom without resetting, and then want to restart a session. It is potentially dangerous though: no check for lock files etc. is made, so a normal user could interfere with things like uucp... Maybe this will be taken out later. For now it is assumed, that users who are given access to a modem are responsible enough for their actions. Thank you for your reaction! Groetjes, Ookhoi
How to determine actual modem speed
Hi, Is there a way to measure the actual speed at which a modem is connected? As modems negotiate the speed al the time, the CONNECT string at the start of the connection is not up to date some time later. Thanx in advance! Groetjes, Ookhoi
