Re: logging within shell scripts
On Mon, Sep 13, 1999 at 08:18:30PM -0500, Gregory T. Norris wrote: > This isn't really a Debian issue as such, but I thought that someone > here might be able to point me in the right direction... Hope I'm not > too far off base here :-) > > I've got a shell script that I want to have log it's actions on a > selective basis (dependent upon run-time settings). I know it could > re-exec itself with appropriate redirection, but I don't want to rely on > commandline parameters or environment settings which might inadvertently > be set. Is there any way for a shell script to determine where it's > stdout/stderr is going, and to redirect it "from this point on"? > Preferably something which works with a generic /bin/sh, rather than > specific to bash, but I'll work with whatever I can get... > > Thanx! > Not sure if I understood your question. Anyway, if you want to redirect a particular file in a shell you don't have to know where it's been directed to at a certain point. For example, you can do something like this: exec 1>log 2>log.err ... do something (everything going to stdin is been redired to file "log" and stderr to "log.err") ... exec 1>log.a 2>log.a.err ... I hope this helps. Dejan -- Dejan Muhamedagic UNIX and Linux Support Quant-X Service & Ph: (+43) 4212 90555-0 Consulting Ges.m.b.H. Fax: (+43) 4212 90555-20 http://www.quant-x.comEmail: [EMAIL PROTECTED]
Re: Problem mounting custom boot disk
On Tue, Sep 14, 1999 at 09:37:20AM -0500, David Kanter wrote: > Last night I compiled a new kernel and had it sent to a boot floppy instead > of the hard drive (using the standard Debian tools for compiling a new > kernel). Booting from the floppy is fine. However, I'd like to mount the > floppy and get the kernel image off of it, move it to /boot, and then > re-configure LILO to run it. > > I've tried multiple variations of mount: -t vfat, -t msdos, and nothing at > all (i.e., ext2). Each time I get a message that I have either chosen the > wrong file system, or a couple other problems. > > What should I do? > > Thanks, > Dave > Why don't you just go to /usr/src/linux and do: $ make zlilo In case your kernel lives in /boot, then you'll have to move System.map and vmlinuz from / to /boot and run lilo again. Otherwise, it may be pretty difficult to take out the kernel from the diskette since you don't know its size. -- Dejan Muhamedagic UNIX and Linux Support Quant-X Service & Ph: (+43) 4212 90555-0 Consulting Ges.m.b.H. Fax: (+43) 4212 90555-20 http://www.quant-x.comEmail: [EMAIL PROTECTED]
Re: latex: how to output the ¢ (cent) symbol
On Tue, Sep 14, 1999 at 11:54:28AM -0400, Michael Laing wrote:
> I have a potato system with all the usual latex stuff on it.
>
> I am working with a lot of latin1 (ISO-8859-1) text and need to output
> the ¢ (cent) symbol, which is latin1 character number 162 (0xa2). How
> can I do that? Almost all other characters work fine with the latin1
> input encoding.
>
> Here's a little example document.
>
> \documentclass[10pt,letterpaper]{letter}
> \usepackage[latin1]{inputenc}
> \usepackage{times}
> \begin{document}
> Yo Günther, a nickle is \$0.05 or 5¢.
> \end{document}
>
> Running latex gives me this output:
>
> plum:~$ latex cent.tex
> This is TeX, Version 3.14159 (Web2C 7.3.1)
> (cent.tex
> LaTeX2e <1998/12/01> patch level 1
> Babel and hyphenation patterns for american, french, german,
> ngerman, n
> ohyphenation, loaded.
> (/usr/share/texmf/tex/latex/base/letter.cls
> Document Class: letter 1999/02/09 v1.2z Standard LaTeX document class
> (/usr/share/texmf/tex/latex/base/size10.clo))
> (/usr/share/texmf/tex/latex/base/inputenc.sty beta test version
> (/usr/share/texmf/tex/latex/base/latin1.def))
> (/usr/share/texmf/tex/latex/psnfss/times.sty) (cent.aux)
> (/usr/share/texmf/tex/latex/psnfss/ot1ptm.fd)
>
> ! LaTeX Error: Command \textcent unavailable in encoding OT1.
>
> See the LaTeX manual or LaTeX Companion for explanation.
> Type H for immediate help.
> ...
>
> l.5 Yo G^^fcnther, a nickle is \$0.05 or 5^^a2
> .
> ?
>
> Thanks for your help and/or pointers!
>
> ml
Perhaps you should try with output T1 encoding.
Not sure but it should be
\usepackage[T1]{times}
instead of
> \usepackage{times}
Good luck.
--
Dejan Muhamedagic
UNIX and Linux Support
Quant-X Service & Ph: (+43) 4212 90555-0
Consulting Ges.m.b.H. Fax: (+43) 4212 90555-20
http://www.quant-x.comEmail: [EMAIL PROTECTED]
Re: compiling kernel for floppy boot
On Tue, Sep 14, 1999 at 03:31:50PM -0500, David Kanter wrote: > I'm having the opposite problem: I can't get my kernel onto the hard drive! > > When compiled a kernel last night, using the packaging you mentioned, I was > left with my own "custom" .deb file. When I ran dpkg on it, I was given the > option to create a boot floppy. I'm not sure why you didn't get this message. > > I already have a boot manager, so I don't want LILO to overwrite the mbr on > hda, which I think it does by default. I don't know either, but you can check that in /etc/lilo.conf. If the boot variable is set to a disk (e.g. /dev/sda) then it'll overwrite the MBR. If it's set to a partition like /dev/sda1 then it'll go to that partition and leave the MBR intact. > > Dave > > >>> Steve George <[EMAIL PROTECTED]> 9/14/99 2:06:56 PM >>> > Hi, > > Does anyone know if there is a way you can compile a kernel into a debian > package and have that kernel boot from the floppy? > > I normally compile my kernels with make-kpkg but the package it makes > installs onto the hardrive. I want to do the same thing but place the 'new' > kernel on a floppy rather than alter my stable setup on the harddisk. > > Its a pity make-kpkg doesn't ask you where to install the new kernel to! All you should do is find the kernel (it should be in /usr/src/linux/vmlinux) and do: gzip -9 < /usr/src/linux/vmlinux > /dev/fd0 Good luck. -- Dejan Muhamedagic UNIX and Linux Support Quant-X Service & Ph: (+43) 4212 90555-0 Consulting Ges.m.b.H. Fax: (+43) 4212 90555-20 http://www.quant-x.comEmail: [EMAIL PROTECTED]
Re: Mailing problems
Just a thing to check: is everything OK with the DNS? I mean are there proper MX records being published in your zones. From "does not know how to deliver the message" it seems that others are not aware of the fact that your computer is equipped with an SMTP server. On Wed, Sep 15, 1999 at 04:39:49PM +, Nagy Gergely wrote: > Hi all! > > I have installed Debian potato last week (i had a slink box before). Since > then, i have serious mailing problems. I can send mail, but no one can > send mail to me. After a day, an error message comes telling that it does > not know how to deliver the message. Strange is , that I can acces the box > via http, ftp, telnet, ssh, and so on. I can browse the web, I can do > evrything I want, but the emails just can't find my computer. > > If anyone has any idea how to fix this, please email me. (I'm on the list, > but don't have time to read through the digests, so please CC me ...) > > Gergely Nagy <[EMAIL PROTECTED]> > > P.S.:you can reply to the address above, it's NOT on the server I was > talking about :) > > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null -- Dejan Muhamedagic UNIX and Linux Support Quant-X Service & Ph: (+43) 4212 90555-0 Consulting Ges.m.b.H. Fax: (+43) 4212 90555-20 http://www.quant-x.comEmail: [EMAIL PROTECTED]
Re: telnet to my machine
Hi, This may not be a tcpd issue. If it is, then you should experience a delay just before "Connection closed..." message and be able to see one or two connections to your auth port (aka ident) from the target machine in a TIME_WAIT state: % netstat | grep auth Otherwise, in case it's happening *sometimes*, it could be faulty cables or faulty software. From what you wrote it's not possible to discover this. You should check with netstat what was going on on a particular connection. Bye Dejan On Mon, Sep 20, 1999 at 06:50:44PM +0200, Manuel Arenaz Silva wrote: > Hello, > > I have installed Debian on my computer. Surprinsingly, sometimes when I > try to stablish a telnet conection to my machine I obtain the following > message: > > venus% telnet cambados.des.fi.udc.es > Trying 193.144.50.62... > Connected to cambados.des.fi.udc.es. > Escape character is '^]'. > Connection closed by foreign host. > venus% > > I have tried to solve the problem by modifying the files > /etc/hosts.allow and /etc/hosts.deny, but the conection is always > refused. > > Thanks in advance, > > Manuel Arenaz > > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null -- Dejan Muhamedagic UNIX and Linux Support Quant-X Service & Ph: (+43) 4212 90555-0 Consulting Ges.m.b.H. Fax: (+43) 4212 90555-20 http://www.quant-x.comEmail: [EMAIL PROTECTED]
