Re: AGP video card (fwd)
Hi, > page. One thing they don't tell you though that you have to do under > Debian is delete the 'X' bin that should be in /usr/X11R6/bin and then > just link the Elsa_GLoria bin (XSuSE_Elsa_GLoria I believe) to 'X'. I really don't think that is a good idea. Debian created that "X" as an suid wrapper for X servers so that the server wouldn't have to be suid. I have the same card (Fire GL 1000 Pro) and am using the server the "standard" Debian way. Instead of deleting Debian's "X," I would leave that X there and modify /etc/X11/Xserver to make Debian use the new X server (XFCom_3DLabs). You will also probably need to use SuSE's xf86config program. The last time I checked there was no new XF86Setup program that would work with this server, and the current stable XF86Setup program doesn't seem to work with this server (at least for me). When you run SuSE's xf86config, you should probably answer "no" to the question of whether you should install a symbolic called "X" to XFCom_3DLabs since this functionality is handled by Debian's "X" and the /etc/X11/Xserver file. As some one else on this list sayed, this X server is not very stable yet. I have had some problems with it. Again, I really don't think you should erase Debian's "X." If you do erase it, you may have problems later on when the Permedia2 X server becomes stable and becomes part of the XFree86 distribution meaning that it will eventually become a Debian package. When this happens, the lack of Debian's "X" will probably mean that future installations of XFree86 will not install properly. -Ossama -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: resolv.conf
You might also want to check /etc/nsswitch.conf. You should have something like the following for the "hosts" entry: hosts: files dns -Ossama -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
2 CPUs and ps/top -- question
Hi, I remember that this question was asked before but I can't seem to find it in the list archives. My question is how can I tell what processes are running on a given processor using ps or top, or some other utility. SMP support has been enabled in my kernel (2.0.35). Will I need SMP profiling for anything? TIA, -Ossama -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Password to single user boot.
Why don't you just restrict booting by using LILO's "password=" and "restricted" in /etc/lilo.conf? Here is what the man page for "lilo.conf" says: password=password Protect the image by a password. restricted A password is only required to boot the image if parameters are specified on the command line (e.g. single). To prevent modification via a rescue boot floppy you can probably disable booting via floppy in the BIOS and then password protect the BIOS using the BIOS's own security features. -Ossama On Wed, 5 Aug 1998, Liran Zvibel wrote: > Hi. > > We consider putting Linuxes (or should it be Linuces? )in the classes, but > don't want the students to be able to boot single user and then rm -r *, > or open accounts and try to get into the network. > > Is there an easy was to put a password on single user boot? > > If not, I have and idea how to prevent it: > 1. Repartition that way that there is another primary partition of about 1 > MB in size. > 2. dd the kernel to that partition (after it was 'rdev'ed to the real root > partition. > 3. Make this partition the only one bootable. > 4. Disable the possibility to boot from the floppy drive in the (password > protected) bios. (And the computers are locked so they (the students) > can't clear the bios settings with the jumpers) > > I think that in this was the students won't be able to boot single user. > > What do you say? > Is there an easier way to do it? > > > TIA, > Liran. > --- > http://www.math.tau.ac.il/~liranz/ > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null > > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Password to single user boot.
No problem. Have fun! :) By the way, perhaps lilo.conf should be readable by root only since a password will be listed inside it, in your case. This shouldn't be a bad thing, from what I can tell. I am glad that you asked this question since I wanted to do the same thing but never actually tried to figure out how to do it till now. -Ossama -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Password to single user boot.
> As David Maze points out, it'd also be best to NOT use mbr as the master > boot record since that will allow one to boot off a floppy and/or > another partition. > > I prefer to let LILO be the master boot record on my servers since they > only run Linux ... one less piece of software to worry about. Ditto, but as I stated in my original post :), one can disable booting by floppy in the BIOS and then password protect the BIOS. Since booting by floppy isn't a common thing (at least for me), this shouldn't be too much of an inconvenience. It is easy to re-enable floppy booting in the BIOS. It may be a pain but better to be safe than sorry. By the way, I don't think I understood what you and David were talking about concerning mbr and the master boot record and floppy booting. Could you please elaborate? Thanks, -Ossama -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: 2 CPUs and ps/top -- question
I found an answer to my question from one of my colleagues. This
seems like a good thing to know so I'll post it in case anyone is
interested. Here is he how checks to see how well his SMPing is working:
(incidentally, these are benchmarks for a Debian 2.0 system running kernel
2.0.35 on a dual PII-333)
Basically, we make Perl count to one million in two seperate processes,
simultaneously. For two processes on a two processor machine, the CPU
usage should be close to 100% for each process, assuming that the machine
is lightly loaded, meaning that each process is running on a seperate CPU
since it isn't possible to have 200% usage on one CPU. :)
time perl -e 'for($i=0;$i<1000;$i++){}' &
*** Results
1.200u 11.550s 0:12.78 99.7%0+0k 0+0io 227pf+0w
time perl -e 'for($i=0;$i<1000;$i++){}' & \
time perl -e 'for($i=0;$i<1000;$i++){}' &
*** Results
0.000u 12.750s 0:12.81 99.5%0+0k 0+0io 227pf+0w
12.730u 0.070s 0:12.91 99.1%0+0k 0+0io 227pf+0w
time perl -e 'for($i=0;$i<1000;$i++){}' & \
time perl -e 'for($i=0;$i<1000;$i++){}' & \
time perl -e 'for($i=0;$i<1000;$i++){}' & \
time perl -e 'for($i=0;$i<1000;$i++){}' &
*** Results
0.000u 12.750s 0:25.26 50.4%0+0k 0+0io 227pf+0w
12.130u 0.640s 0:25.28 50.5%0+0k 0+0io 227pf+0w
0.690u 12.070s 0:25.58 49.8%0+0k 0+0io 227pf+0w
12.720u 0.080s 0:25.58 50.0%0+0k 0+0io 227pf+0w
-Ossama
--
Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Password to single user boot.
> The default Debian setup allows you to select which partition to boot > from, before selecting a default. On my system (with only /dev/hda2 > marked bootable), I see a "2FA:" prompt. If I press any keys here, I > can press "2" (boot from hda2), "F" (boot from floppy), or "A", which > changes the prompt to "1234F:" and allows me to boot from any > partition. This is a useful feature for Debian, and completely > undesirable behavior for the system you describe. Aahh! I had no clue what you guys were talking about until now. I completely forgot that the default Debian install does the above. :) Thanks for the elaboration. I don't have the above problem since my lilo isn't set up like the above. Thanks again for the details. -Ossama -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Automatic X startup
Hi, > Could anyone tell me how to get X automatically started. Now I have to > write 'startx' every time. Try configuring xdm to start at boot. You can probably use xbase-configure to configure it. If that doesn't work check /etc/init.d for a script called xdm. If you don't have xdm but have something like "xdm-new.dist," than move that file to /etc/init.d/xdm. The next time you boot, xdm should start X for you and provide you with an X login prompt/window. -Ossama -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
[OFF TOPIC] cvsweb and CVSROOT
Hi, Sorry for the off topic question. How can I get cvsweb to show the existing directories in the repository without listing CVSROOT? I've already restricted access to CVSROOT via /etc/cvsweb.conf but I would like to prevent it from being listed in the web page that cvsweb creates, altogether. Thanks, -Ossama __ Ossama Othman <[EMAIL PROTECTED]> --- PGP Keys --- Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
automount trying to mount /home/.htaccess
Hi, I recently installed apache (debian package) on my hamm system. For some reason automount keeps trying to mount /home/.htaccess. However, there is no .htaccess in my auto.home map. Here is the error I get: automount[193]: attempting to mount entry /home/.htaccess automount[9518]: lookup(file): looking up .htaccess automount[9518]: lookup(file): lookup for .htaccess failed My /home directory has directories that are automounted but I never setup the automounter to try to mount .htaccess. I get the above error whenever my web server is accessed. -Ossama __ Ossama Othman <[EMAIL PROTECTED]> --- PGP Keys --- Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: xterm-debian
Hi, Read /usr/doc/X11/README.Debian on your systems. That file has explanations and solutions for your problems with xterm-debian. -Ossama
Re: Big-endian/little-endian
Hi,
> The authors concern was that on a big endian system network order and
> byte order are the same so hton* and ntoh* do nothingso how do you
> do the swapps...
You could just cast the address of the variable to a "char *" and swap
things around using pointer arithmetic or arrays. Here is a sample piece
of C++ code (it should be easy to modify for C):
// "number" is the size of the array being byte swapped
swap4(void *dest, void *src, int number) {
unsigned char *t = (unsigned char *)dest;
unsigned char *s = (unsigned char *)src;
if( t == s ) {
for (int i = 0; i < number; ++i, t += 4) {
unsigned char tmp;
tmp = *t; *t = t[3]; t[3] = tmp;
tmp = t[1]; t[1] = t[2]; t[2] = tmp;
}
}
else
for (int i = 0; i < number; ++i, t += 4) {
t[3] = *s++; t[2] = *s++; t[1] = *s++; *t = *s++; }
}
This may not be the best way to do the byte swapping but it is fairly
straight forward to do it this way. However, it may not be very
efficient. Of course, if you're not swapping entire arrays then this may
be a decent solution.
-Ossama
I forgot... Re: Big-endian/little-endian
Hi again,
> // "number" is the size of the array being byte swapped
> swap4(void *dest, void *src, int number) {
I forgot to mention that the code excerpt I posted was meant for swapping
variables that are 4 bytes in size, e.g. floats and such.
-Ossama
Re: PDF file generation
> How does one create a PDF (Acrobat) file? In searching through the > packages I found a number of ways to view a PDF file, but didn't seem to > find any way to create one. The only ways I am aware of besides using Adobe Acrobat is to use "ps2pdf," found in the ghostscript package or by using using "pdftex" in the teTeX package. Note that if you use ps2pdf, make sure you are using the one that comes with ghostscript 5.10+ since the ones before it are somewhat broken. -Ossama
Re: threads package for libc6
Hi, > Where can I find the threads package for libc6 - I have looked all through > the dselect list? All I find is the one for libc5 in the old libs section. A specially adapted version of Xavier LeRoy's implementation of kernel-level POSIX threads (pthreads) is already part of libc6. Some libc6 distributions may have a seperate package for libc6 pthreads. However, Debian 2.0 has pthreads already integrated into libc6. -Ossama
Re: threads package for libc6
Hi, > Well that explains why I couldn't find anything. What about the man pages? > Which package has those? When the guys do a man on pthread_create nothing is > found It seems that I didn't have the libc6 man pages either. :) Anyway, I found the docs package. Not surprisingly, it is: libc6-doc "man pthread_create" now works for me. Enjoy. :-) -Ossama
XF86 thread safe?
Hi, Does anyone know if Debian's XF86 package is thread safe (i.e. patches in /usr/doc/libc6-doc/README.linuxthreads.gz have been applied)? Thanks, -Ossama __ Ossama Othman <[EMAIL PROTECTED]> --- PGP Keys --- Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc
Re: Motif
Hi, > What versions of MOTIF will work with Debian 2.0? Can't help you with Motif. Have you considered the LessTif package? I believe that it aims for Motif 1.2 "compliance." It may still be a little buggy, however. -Ossama
Re: telnet break-in
You might want to consider disabling all incoming connections if you are
never going to login to your machine via remote.
As George suggested, you can edit your host.{allow,deny} files or edit
/etc/inetd.conf and comment out things like:
telnet
ftp
login
exec
finger
shell
...you get the idea...
and then restart the inetd daemon with the HUP signal (kill -HUP pid_of
_inetd ... i.e. fill in the process ID of inetd). All of your outgoing
telnets, ftps, etc. would still work just fine but incoming telnets, for
example, would be completely disabled.
-Ossama
Re: telnet break-in
Hi, > > telnet > > ftp > > login > > exec > > finger > > shell > > ...you get the idea... > > And dont forget the r services ... rlogin, rsh, rexec, etc. Yep, the "r" services are covered by login, exec and shell. Here is what inetd.conf has to say: shell stream tcp nowait root/usr/sbin/tcpd /usr/sbin/in.rshd login stream tcp nowait root/usr/sbin/tcpd /usr/sbin/in.rlogind exec stream tcp nowait root/usr/sbin/tcpd /usr/sbin/in.rexecd See :-) -Ossama
Hardware monitoring software?
Hi, Is there such a thing as software that will check environmental conditions on a motherboard, CPU fan, etc? My motherboard has a hardware monitor on it but I don't always want to go to the BIOS to check the conditions. Thanks, -Ossama
Re: not all memory is recognized
> i have a debian system with 128 megs of memory. > i have the following line in lilo.conf: > > append="mem=128" It should be: append="mem=128M" It works fine for me. -Ossama
Re: telnet break-in
Hi, > Interesting ... after reading the above messages, I decided to check out > my /var/log files and found a whole suite of "in.telnetd[XXX]: connect" > statements. Bummer! It appears that it is time for you disable all of those incoming services we mentioned. Have you considered using Secure Shell? If you can't use Secure Shell, you might want to try One-Time Password packages, such as OPIE. I believe that even WU-FTP is starting to support one time passwords, too. -Ossama
Re: telnet break-in
> > can't use Secure Shell, you might want to try One-Time Password packages,
> > such as OPIE. I believe that even WU-FTP is starting to support one time
> > passwords, too.
>
> Uhm, secure shell is not the answer to all problems. It WILL revert to an
> unsecure protocol if the other end can not establish an encrypted session.
The idea is to make sure all machines use Secure Shell. I know this isn't
always possible, but we don't live in an ideal world. :)
> Also, it is still just as easy to log in with a cracked password. About
> the only thing it is really great at is making sure that the other machine
> really has permission for a password-less login. Still, even THIS can be
> hacked if someone has gotten onto the machine in the past and snarfed a
> copy of your public ssh key.
Hence the One-Time Password suggestion. Either way, better to have/use
SSH than use telnet/ftp/r{login,sh,exec}.
-Ossama
CVS 1.10 is out...hamm?
Hi, Will CVS 1.10 be available in hamm? -Ossama
Re: CVS 1.10 is out...hamm?
Hi, > > Will CVS 1.10 be available in hamm? > > No. Hamm is released. It will be available in slink, it's successor. Doh! I didn't realize that "released" distributions (e.g. hamm) don't get periodic updates to packages. Does this mean that only big fixes go into releases? -Ossama
Re: telnet break-in
> > > Hence the One-Time Password suggestion. Either way, better to have/use
> > > SSH than use telnet/ftp/r{login,sh,exec}.
> >
> > I have both SSL-Telnet and SSH installed. I don't type root passwords over
> > clear connections unless it is an emergency.
>
> Hmm - why is it that emergencies always happen when I'm away from
> Cambridge? ;(
Well, one Debian user seems to have had a break-in. However, the break-in
wasn't due to any Debian security hole. The break-in was due to
standard/known UN*X security holes. For example, allowing incoming
telnets is one (typing clear text passwords over the net).
-Ossama
Re: Logitech TrackMan & CD Changer
Hi, > Does anyone have any experience with using the TrackMan+ or TrackManFX > with hamm? I'm thinking about buying one but part of my decision will > be based on it's compatability. I have a TrackMan Marble and it works just fine. Assuming that there isn't anything different with the connectors (PS/2 right?) then there should be no reason why the others wouldn't work. Just make sure you have the psaux kernel module installed and that you select /dev/psaux as your X mouse device. You may also have to choose a generic PS/2 mouse protocol when you run XF86Setup, for example. -Ossama
Re: Beowulf cluster
> from what I remember Beowulf uses PVM, and that is already packaged. > They basically seemed when I read their pages to be makeing "add ons" which > make PVM more powerfull >From what I recall, PVM is being superseded by MPI. There is already a Debian packaged implementation of MPI called "mpich." I'd suggest using MPI instead of PVM. By the way, has anyone adopted MPICH since it was orphaned (poor little mpich :))? If not, I'll adopt it. -Ossama
Re: X11 remains black (fwd)
Hi, > using the PCMCIA drivers), but if I start X (version included in Debian > 2.0), the screen remains black. SuperProbe does not work neither. Screen > black, machine locked ? > When I reboot with 2.0.30, no problems, not at all. Please note that this > 2.0.30 kernel was made with gcc 2.7.2.. while the 2.0.34 with gcc 2.8.1. Don't use gcc 2.8.1 or egcs to compile the 2.0.x kernels. This is a known issue. Compiling the kernel with those compilers causes problems with X11 starting up. Use gcc 2.7.x to compile the kernel. By the way, it is for this reason (at least one reason) that Debian 2.0's default C compiler is gcc 2.7.2.3. I believe that the latest development 2.1.x kernels can be compiled using the new compilers. I am not suggesting that you use the development kernels, however. -Ossama
Re: X11 remains black (fwd)
Hi, > > I believe that the latest development 2.1.x kernels can be compiled using > > the new compilers. I am not suggesting that you use the development > > kernels, however. > > Why not? :) > I got 2.1.119 last night...man it is sweet :) > My first development kernel :) I didn't want to be responsible if something went wrong. :) I plan switching to the development kernels after I get my motherboard back from ASUS. -Ossama
Re: driver for 3com fast etherlink XL ?
> I need to find a driver for my net card in win nt is name is driver for > 3com fast etherlink XL. You want the driver for 3Com 3c90x cards which is supported in the 3c59x driver/module. -Ossama
EGCS 1.1 is out!
Hi all, Any idea when EGCS 1.1 will be packaged? I'd really like to have it. Otherwise, I'll just stick it in /usr/local myself. :) -Ossama ______ Ossama Othman <[EMAIL PROTECTED]>
Unrecognized binary format?
Hi, Oddly enough, I get an unrecognized binary format if I run gdb on a C++ program that I compiled. I get the same results if I run ldd on the program. However, I can run the program just fine. Any ideas why I would get unknown binary type errors from gdb and ldd? Thanks, -Ossama __ Ossama Othman <[EMAIL PROTECTED]>
Re: Unrecognized binary format?
Hi, > What does 'file
Re: Unrecognized binary format?
> file is a great tool. And I hate libtool. Slow, clunky, and odd. Not > to mention fun things like your problem (-: True, but it is nice not to have to figure out how to generate a shared library on different platforms. :) -Ossama
Re: EGCS 1.1 is out!
Hi, > Please can someone explain to me the difference between EGCS compilers > and non-EGCS ones? The little info given in dselect is not very > informative. EGCS is the Experimental GNU Compiler System. It is on a slightly faster development track then GCC 2.8.1, for example. The EGCS project was intended to speed up and improve GNU compiler development since development was very slow and seemed to be stagnant before gcc-2.8 was release. However, while the EGCS compilers tend to be very stable, they are still experimental. If you like, you can sort of think of the EGCS compilers as gcc/g++ 2.9.x, sort of :). G77 is also part of the EGCS suite, in addition to an implementation of the standard C++ library. Debian currently uses EGCS C++ and g77 as its default C++ and FORTRAN compilers since they tend to be better then the "standard" versions of the compilers. On the other hand, Debian still uses gcc 2.7.2.3 as its default C compiler since the 2.0.x Linux kernels make assumptions about the way the C compiler optimizes. These assumptions no longer work when using the EGCS C compiler, which is why the current 2.0.x kernels shouldn't be compiled with EGCS gcc (egcc package). Take a look at: http://egcs.cygnus.com for additional information. I hope that this helps. -Ossama
[off-topic] gethostbyname_r prototype
Hi, Sorry for the off-topic question but since I am using Debian, I figured why not. I am trying to find the man page for the netdb reentrant functions, e.g. gethostbyname_r. Unfortunately, they don't seem to be available in the man directories. I thought that I had all of the man pages I needed installed. Are there any that I may be missing, i.e. contain the man pages for the netdb reentrant functions? If they aren't available, I was hoping some could answer my question. According to /usr/include/netdb.h, the prototype for gethostbyname_r is: extern int gethostbyname_r __P ((__const char *__name, struct hostent *__result_buf, char *__buf, size_t __buflen, struct hostent **__result, int *__h_errnop)); where on Solaris, and most other platforms, it is: struct hostent *gethostbyname_r(const char *name, struct hostent *result, char *buffer, int buflen, int *h_errnop); Glibc2 adds an additional argument and returns int instead of struct hostent *. The additional argument is struct hostent **__result. Does this mean that I must dereference struct hostent **__result to get struct hostent *__result, which would then be analogous to the return value of the Solaris version "struct hostent *?" TIA, -Ossama __________ Ossama Othman <[EMAIL PROTECTED]>
Re: desktop calendar for linux
Try "ical." It is fairly useful. -Ossama
[off-topic] Best way to replace quotes with angle brackets
Hi all, What is the best way to replace quoted includes (#include "duh.h") with bracketed includes (#include )? I've got over one hundred files that use quoted includes and I would like to switch them over to bracketed includes. The package I am creating/maintaining uses the quoted includes and I would like to change them before I release the package. I was thinking of using sed but I am not at expert at using sed, and after several failed attempts sought your help :). Finding the include statements in the source isn't a problem via regex. The problem is changing the beginning and ending quotes to beginning and ending brackets. TIA, -Ossama __________ Ossama Othman <[EMAIL PROTECTED]>
Re: [off-topic] Best way to replace quotes with angle brackets
Cool! However, I have several "types" of includes, e.g.: #include "duh.h" # include "duh.h" #include /**/ "duh.h" # include /**/ "duh.h" # include ... etc. As such, I'd like to be able to do something like: #include "duh.h"-> #include # include /**/ "duh.h" -> # include /**/ ... etc. Is there a way to accomplish without making several passes? I'm not a Perl guru either so this is all very new to me. Sorry. Thanks again, -Ossama __ Ossama Othman <[EMAIL PROTECTED]>
Re: Voodoo Rush
Hi, > I have a friend interested in changing his machine's OS over to Debian, > the only problem that I can see is that he has a Intergraph Voodoo extreme > as his video card, and I can't find a X-server for it. I was simply > wondering if there is such an X-server for Debian. All Voodoo Rush cards, i.e. AT25/AT3D chipset, are supported by the XF86_SVGA server, including the Intergraph Intense 3D Voodoo. I had one running just fine under Debian. -Ossama
Savage3D X server?
Hi, Does anyone know if there is an X server that will support the Savage3D or if an X server/driver is in the works? Thanks, -Ossama __ Ossama Othman <[EMAIL PROTECTED]>
ACE packages available for testing
Hi, I am packaging ACE for Debian. I made some packages available for testing at: ftp://moc.tn.cornell.edu/pub/othman/aceconf/debian I'd appreciate any feedback. You may need to add -L/usr/X11R6/lib -lX11 -lXt -lpthread -ldl -lACE to the link line for your programs. Note these aren't "official" Debian packages yet since I haven't received my developer approval yet. Thanks, -Ossama __________ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: SCSI time out error
Did you enable SCSI tape drive support in the kernel? I'm not sure if the Debian kernel image has support for this compiled in or not. I'm using a Conner tape drive (DDS-2 and a couple of years old) with an Adapted 2940UW with no problem. -Ossama ______ Ossama Othman <[EMAIL PROTECTED]> --- PGP Keys --- Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Driver Module for 3C905
Hi, The 3c59x module supports both the Vortex and Boomerang series cards. The 3C905 is a Boomerang card so it is supported. :) The driver is does take advantage of 3C905 features meaning it isn't some cheapo driver that runs the 3C905 in some compatibility mode. I have the very same card installed on two machines and they work fine. -Ossama __ Ossama Othman <[EMAIL PROTECTED]> --- PGP Keys --- Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc On Tue, 12 May 1998, Paulo Almeida wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Hi everybody! > > I have a new 3Com 10/100 TX autosense PCI network card (3C905). I > would like to install it on my Debian box but I can't find the right > network module. Does anyone know where I can find the module or driver > source code I need? > > Thanks. > Paulo > > - - > Paulo Jorge de Almeida > Tel: +351 1 792 5500 > - - > -BEGIN PGP SIGNATURE- > Version: PGPfreeware 5.5.3i for non-commercial use <http://www.pgpi.com> > > iQA/AwUBNVgczxhjC44Onu5DEQL8YQCgh4/dOrGzWyXRTSpvzR1iof0by1kAoMFr > simc74rjg/gKbE3JHFdnz3CM > =jGzD > -END PGP SIGNATURE- > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: NFS and /etc/exports
I think that you are better off doing a: /etc/init.d/netstd_nfs reload instead of kill -HUPing the daemons individually. It would be equivalent, but I believe using the script in /etc/init.d is more desireable. -Ossama __ Ossama Othman <[EMAIL PROTECTED]> --- PGP Keys --- Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Help! Major crash in Hamm upgrade!
Hi, Use the Debian Rescue disk to boot your system but mount your existing root partition as the root instead of allowing the rescue disk to mount a ram disk as root. For example, when presented with the "boot:" prompt from the rescue disk type: linux root=/dev/hda1 Substitute whatever your root partition is. This causes your system to boot with the kernel on the rescue floppy but still uses your hard drive's root file system, meaning that you can easily reconfigure if you need to. Instead of manually configuring the lilo.conf file, try using liloconfig instead. Liloconfig is fairly interactive and it helps to make sure that your lilo.conf file is configured properly. -Ossama ______ Ossama Othman <[EMAIL PROTECTED]> --- PGP Keys --- Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: FTP site
On Mon, 1 Jun 1998, matthew.r.pavlovich.1 wrote: > I am still not able to log into the debian ftp server, seems it cannot > resolve my hostname. My machine has two dns entries, could that be a > problem? I think the Debian ftp server (ftp.debian.org) is down since I was able to connect to it a few days ago. Have you tried a mirror like: ftp.us.debian.org -Ossama -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Man errors?
Hi, What does all of the following mean and how can I correct the problem? # man afclient Updating index cache for path `/usr/man'. Wait...man: can't open /usr/man/man1/gnuserv.1: No such file or directory man: warning: /usr/man/man1/gnudoit.xemacs20.1.gz: bad symlink or ROFF `.so' request man: can't open /usr/man/man1/gnuserv.1: No such file or directory man: warning: /usr/man/man1/gnuclient.xemacs20.1.gz: bad symlink or ROFF `.so' request man: can't open /usr/man/man1/gnuserv.1: No such file or directory man: warning: /usr/man/man1/gnuattach.xemacs20.1.gz: bad symlink or ROFF `.so' request man: warning: /etc/alternatives/yaccman is a dangling symlink man: can't open /usr/man/man1/yacc.1: No such file or directory man: warning: /usr/man/man1/yacc.1.gz: bad symlink or ROFF `.so' request man: warning: /usr/man/man1/pstogif.xemacs20.1.gz is a dangling symlink man: can't open /usr/man/man1/pstogif.1: No such file or directory man: warning: /usr/man/man1/pstogif.1.gz: bad symlink or ROFF `.so' request man: warning: /etc/alternatives/tixwish.1.gz is a dangling symlink man: can't open /usr/man/man1/tixwish.1: No such file or directory man: warning: /usr/man/man1/tixwish.1.gz: bad symlink or ROFF `.so' request man: can't open /usr/man/man1/gnuserv.1: No such file or directory man: warning: /usr/man/man1/gnuclient.1.gz: bad symlink or ROFF `.so' request man: can't open /usr/man/man1/gnuserv.1: No such file or directory man: warning: /usr/man/man1/gnudoit.1.gz: bad symlink or ROFF `.so' request man: can't open /usr/man/man1/gnuserv.1: No such file or directory man: warning: /usr/man/man1/gnuattach.1.gz: bad symlink or ROFF `.so' request done. No manual entry for afclient I know that I have no man page for afclient. The above errors/warnings occur whenever I try to "man" a non-existent man page. Thanks, -Ossama -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: X11 & AGP Card
Hi, > Thank for the information. Does the AGP card improve the > graphics significantly and noticeable? Thanks! AGP is meant to significantly improve 3D performance, not 2D. There may be improvement in 2D performance but I don't think that it would be significant. I've got a Millenium II AGP. It's fast but I don't think that it being an AGP card helps all that much with its 2D stuff. I have no idea how much of an improvement there is for the 3D stuff since I don't have any 3D Linux applications that use 3D acceleration, unless Mesa takes advantage of the Millenium II's 3D abilities. This seems unlikely since my GL screen savers seem like they should run faster. -Ossama -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
afbackup problems
Hi guys,
I am still having problems with afbackup. This probably isn't the right
place to ask for help but I was hoping one of you might have already dealt
with this problem.
I am using my hamm system as the backup server. Two Solaris machines are
backup clients. The problem I am having is with remote starts. I can
start a backup to the backup server (hamm machine) from the Solaris backup
clients just run by running {full,incr}_backup but I can't remote start a
backup on the Solaris clients from the backup server. For example, I use
the following command on the server to start a remote backup:
afclient -X full_backup -h remote_host -k /etc/afbackup/cryptkey
but I get the following error on the server:
Server Error: device has not been opened for writing.
The server log says the following:
mt: /dev/nst0: Input/output error
However, if I "remote start" a backup on the localhost, i.e. the server
machine, as follows:
afclient -X full_backup -h localhost -k /etc/afbackup/cryptkey
the backup works fine. What's the deal? Any help would be greatly
appreciated.
Thanks,
-Ossama
__________
Ossama Othman <[EMAIL PROTECTED]>
--- PGP Keys ---
Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc
REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: KHOROS on Debian Problems
Hi, > So my question is: what is and where or how do I get libfl.a? libfl.a is a fast lexical analyzer library meaning that it should be part of the flex (Fast Lexical Analyzer) package. The function yywrap is in libfl.a I can't help you witht the other two problems you mentioned, sorry. -Ossama -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Tape drive "sense" errors
Hi, Does anyone know what the following means: Jun 12 05:47:54 squall kernel: st0: Error with sense data: Current error st09:00: sns = f0 4 Jun 12 05:47:54 squall kernel: ASC=44 ASCQ=ad Jun 12 05:47:54 squall kernel: Raw sense data:0xf0 0x00 0x04 0x00 0x00 0x00 0x3c 0x0a 0x00 0x00 0x00 0x00 0x44 0xad 0x00 0x2d Jun 12 05:47:54 squall kernel: st0: Error with sense data: Current error st09:00: sns = f0 4 Jun 12 05:47:54 squall kernel: ASC=44 ASCQ=ad Jun 12 05:47:54 squall kernel: Raw sense data:0xf0 0x00 0x04 0x00 0x00 0x00 0x01 0x0a 0x00 0x00 0x00 0x00 0x44 0xad 0x00 0x2d These probably occurred during my overnight backup. I am running kernel 2.0.34 with an onboard AIC7880 SCSI chipset. My tape drive is Conner 4mm DDS-2. TIA for any help or suggestions. -Ossama __ Ossama Othman <[EMAIL PROTECTED]> --- PGP Keys --- Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Tape drive "sense" errors
Great! Thanks! I was afraid something might be wrong. Although I can't get information from "mt" any more. I'll probably have to reboot. Thanks again. -Ossama __________ Ossama Othman <[EMAIL PROTECTED]> --- PGP Keys --- Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: XWindows
Hi, You need the xbase package. Dselect will tell you what else xbase depends on, so start off with that. Basically, you will need the xbase package, an X server package, X libraries, fonts and a window manager; all of which are in the X11 section. Have fun. :) -Ossama __ Ossama Othman <[EMAIL PROTECTED]> --- PGP Keys --- Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
GnuHoo a major BooHoo
Hi guys, For those of you who don't know about the GnuHoo web directory, keep on not knowing about it. :) GnuHoo is trying to recruit VOLUNTEER editors for the web directory. However, their copyright policy is downright anti-open-source. GnuHoo tries to make it appealing to become an editor but if you look at their FAQ all they are doing is blowing sunshine up your *ss. Editors basically contribute so that these people can make money off of them. GnuHoo says that they have a strict policy against copyright infringement but they later say that have the right to do anything they want to a contribution and that GnuHoo retains compilation copyrights on the directory. Why? so that can provide what they believe to be the best internet directory. Nice try, but who are they kidding. Sure, using the directory is free but that is all that is free. Hopefully the FSF and even Yahoo will come down on them for using the GNU name and "hoo," although admittedly the "hoo" part seems difficult to do anything about. I would like to thank some one on the Debiab-Devel list for pointing this out. I am sorry that forgot his name (perhaps James Tracey). Thanks again! Hopefully all of us who support Open Source can do something about this. -Ossama __________ Ossama Othman <[EMAIL PROTECTED]> -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
xterm-debian?
Hi, What's the difference between a standard xterm and a xterm-debian terminfo/termcap entry? It is annoying to see the following when I try to remote start an xterm on a non-Debian machine (via ssh): tcsh: No entry for terminal type "xterm-debian" tcsh: using dumb terminal settings. It's easy enough to do a "set term=xterm" but I really don't want to keep doing that. Why did Debian create their own terminfo/cap entry and why was such a change necessary? Thanks, -Ossama ______ Ossama Othman <[EMAIL PROTECTED]> --- PGP Keys --- Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Re[2]: Two 'gcc's? Yea or Nay?
Hi, > > $ ls -l /usr/bin/egcc /usr/bin/gcc > > -rwxr-xr-x 1 root root60320 Jun 17 03:04 /usr/bin/egcc* > > -rwxr-xr-x 1 root root49460 Jun 15 00:48 /usr/bin/gcc* > > Ok, I haven't posted a BIG DUMMY question in quite some time; I'm overdue! > > SO: what's the significance of the asterisks next to those two lines above? Heh, good question! :) The asterisk after the filename indicates that it is an executable file. Similarly a directory would have a "/" after it. This type of output is caused by doing a "ls -F." Debian sets up the following aliases: unalias ls alias ls 'ls-F -C' It seems that "ls-F" (no space after ls) is a shell built-in command. As stated above, shells that don't have this built-in command can simply use "ls -F" (with a space after ls) to produce the desired output. Incidentally, here is an excerpt from the ls man page: -F, --classify Append a character to each file name indicating the file type. For regular files that are executable, append a `*'. The file type indicators are `/' for directories, `@' for symbolic links, `|' for FIFOs, `=' for sockets, and nothing for regular files. -Ossama -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Once burned...
Hi, I can vouch for the Trackman Marble since I have the exact same thing. :) It works fine for me. All three buttons work in X Windows. If you are using the PS/2 mouse port, make sure you have PS/2 mouse support enabled in your kernel or installed as a module. The module for this psaux. From what I recall, Debian has this setup by default. I can't help you with the other hardware. Sorry. -Ossama -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: nvidia riva 128 - xf86config
Hi, > I`m useing a Diamond Viper V330 (Nvidia Riva 128) card but this card > isn`t listed... Try the SVGA driver. I believe support for the Riva 128 is in it. -Ossama -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Page counting / Print quota
Hi Have you tried pac, the printer accounting utility? Do a "man pac" do get some additional information. As far as I know, pac is part of the default Debian installation. Here is an excerpt from the pac man page: --- NAME pac - printer/plotter accounting information SYNOPSIS pac [-Pprinter] [-c] [-m] [-pprice] [-s] [-r] [name ...] DESCRIPTION Pac reads the printer/plotter accounting files, accumulating the number of pages (the usual case) or feet (for raster devices) of paper consumed by each user, and printing out how much each user consumed in pages or feet and dollars. -Ossama -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Did I do something wrong ?
Hi, Asking a question about fvwm95 probably isn't so bad since there is a Debian package for it. Not getting a response doesn't necessarily mean that you did something wrong. Some people are just more interested in questions or answers that are related to them in some way or just find interesting. In general, I delete about 90% of the e-mail I get from this list just based on the subject line before I even read any of them. Also, I generally don't get very many responses to my questions or posts to the mailing list, either. If you ask me there is a conspiracy out there where someone has setup a filter to send all of my posts to /dev/null. just kidding.. :) I wish I could answer your fvwm95 question but I deleted it before I read it. :) -Ossama ______ Ossama Othman <[EMAIL PROTECTED]> --- PGP Keys --- Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
NFS/Network locking available?
Hi, I am exporting some ext2 filesystems from my Debian box to a Solaris machine. While the export works, the Solaris automounter complains about network locking not being available on my Debian box. Is network locking supported in hamm? If so, how do I enable it? Thanks, -Ossama __ Ossama Othman <[EMAIL PROTECTED]> -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: CAP and linux
Hi, Try the netatalk package instead of CAP. I've been using it for over a month now and haven't had any problems. You will need to enable support Appletalk DDP in the kernel or as a module. -Ossama ______ Ossama Othman <[EMAIL PROTECTED]> -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: xcdroast and scsi
Hi, > Does I really need scsi support since my burner is IDE ? > Why this support is not detected by xcdroast (I have got no scsi card) ? I'm not sure if this will help, but have you tried enabling SCSI emulation in the kernel? Note that the SCSI drivers need to be enabled for the emulation to work, too. -Ossama -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Browser search
Hi, > I am having trouble finding browsers that use 128 bit encryption. I've > have found IE versions 3.0 and higher and Netscape versions 3.0 and > higher. Are these the only ones out there? What about AOL? If you > could shed light on this subject it would greatly be appreciated. Is this a Debian related question? :-) What's wrong with Netscape? -Ossama -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Debian & Std C++?
Hi, g++ 2.7.2.3 is fairly out of date with respect to C++ standards. Use egcs C++ 1.x instead. Egcs is the Experimental GNU Compiler System. It is somewhat equivalent to gcc/g++ 2.9x.x, sort of :). Anyway, the egcs version of g++ conforms much more closely to the latest C++ standard. However, the current "stable" version of egcs doesn't yet support namespaces, except for "namespace::std." If you can wait till August, when egcs 1.1 is supposed to be released then that version is supposed to support general namespaces. I'm not sure but it may also include the latest version of SGI's STL (3.1). Again, I am not sure about the latest version of SGI's STL. The new style C++ headers, for example, are also supported in the current version of egcs. Are using bo? Debian 2.0 (hamm) uses egcs c++ and FORTRAN by default. The old gcc 2.7.2.3 is still being used in hamm since there are kernel compile issues with gcc 2.8.x and egcs 1.x. By the way, this is not local to Debian. Any distribution that uses g++ will have these problems. -Ossama __________ Ossama Othman <[EMAIL PROTECTED]> --- PGP Keys --- Public: http://astrosun.tn.cornell.edu/staff/othman/OO_PUBLIC.asc REVOKED: http://astrosun.tn.cornell.edu/staff/othman/OO_REVOKED.asc -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: cd rom drives
Hi, > now.. here's the dumb newbie question.. apart from staying away from plug > and pray, is there any complication to installing a cd-rom drive and new > modem? do i have to invoke prayers to any particular UNIX gods? or do i > get to shove the new stuff into their proper bays and boot up? You should be able to use just about any "drop-in replacement" IDE ATAPI or SCSI (if you have SCSI) CD-ROM drive. Most modems should also work but stay away from Winmodems. -Ossama -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
xterm-debian and pine problem
Hi, I just added the xterm-debian terminfo file in the appropriate Solaris 2.6 location on our Solaris machine, in accordance with what /usr/doc/xbase/README.Debian says. When I start up an xterm on our Solaris machine remotely using SSH on our hamm system (Solaris xterm displayed on hamm system) and then run pine in that xterm I have problems with the colors. The colors look something along the lines of what dselect looks like in an xterm with a white background. However, the "highlight" bar that you would normally see when selecting an e-mail to read from the pine index is not at all visible. There is no change in the color of the text that is apparently "below" the highlight bar. I tryed reversing the video but the same problem of an "invisible" highlight bar occurs. If I do a "set term=xterm" in the very same xterm and then re-run pine the problem doesn't occur. Similarly, if I "slogin" into the Solaris machine and run pine I don't have the highlight problem, even though the terminal type is set to "xterm-debian." The problem only seems to occur if I start up a Solaris xterm via remote from my hamm system. Any ideas as to what the problem might be? I imagine that the same thing will happen with other curses based applications that use highlight bars. Is this a bug in the xterm-debian terminfo file/entry? Incidentally, I am running Pine 4.02. Thanks, -Ossama -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
imlib against gtk1.2?
Hi, I've gotten a number of requests for imlib packages that are linked against gtk/glib 1.2. The imlib packages linked against gtk/glib 1.2 have been ready for some time. However, they will not be released until the gnome/gtk/imlib coordination team feel that the new gnome/gtk/imlib packages are in a consistent state and are ready for release. Please bear with us. The new packages should be ready in a few days, and I believe that the wait will be worth it. I'm anxious to get them, too. :) Thanks, -Ossama ______ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
cron.daily e-mails for exim and suidmanager
Hi, I get the following e-mail from the cron.daily exim and suidmanager scripts: /etc/cron.daily/exim: cp: /var/log/exim/rejectlog: No such file or directory cp: /var/log/exim/paniclog: No such file or directory /etc/cron.daily/suidmanager: /usr/bin/elm PERMISSION MISMATCH: was root.mail 2755 changed to 4294967295.mail 2755 /bin/ping PERMISSION MISMATCH: was root.root 4755 changed to root.4294967295 4755 For exim, should I have those two files? Why does the suid manager think that root uid/gid is 4294967295? Here are the permissions for elm and ping: -rwxr-sr-x 1 root mail 424880 Mar 15 06:39 /usr/bin/elm* -rwsr-xr-x 1 root root14768 Mar 21 02:00 /bin/ping* I am not subscribed to the debian-user mailing list so please CC me a copy of any reponse. TIA, -Ossama __ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: TIOCSER?WILD Boot Error!
Hi, > Ever since I switched to the dev kernel, I've had the following > message on boot: > > TIOCSER?WILD ioctl obsolete, ignored > > I doesn't sound dangerous or anything, but a boot error just doesn't > look good. So does anyone know how I could fix this? Is it a bug, just > like the SIOCADDR thingy? It was my understanding that the SIOCADDR boot error was not a bug in the kernel since the dev kernel now sets up routing automatically and an explicit call to "route" is no longer needed. Perhaps your TIOCSER* problem is related. Check out the Documentation directory in your kernel source directory. It may contain files that explain the error/warning. -Ossama __________ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
how? m68k cross compile
Hi, I am trying to get the m68k cross compiler package to work. However, I have never cross-compiled anything before. I installed the gcc-m68k-linux and binutils-m68k packages. Do I need anthing else? I tried compiling a simple C++ program but the cross-compiler doesn't know where any of the header files are, nor can I find the m68k headers. What am I doing wrong. TIA, -Ossama __ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Checksums for compromise checks
Hi, My Debian 2.1 (unstable) system detected a dozen or so port scans, via the Debianized courtney package, coming from an older RedHat system that had the NFS server bug. The RedHat system in question was able to determine that ps and top were trojaned by using rpm to verify the ps and top md5 checksums. Does Debian have any such verification mechanism for individual files (i.e. not the entire deb)? It seems like it would be a useful feature to have available. -Ossama __ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: xconfig, menuconfig
Hi, There are no such packages. They are Makefile targets. Go into your Linux source directory and do "make xconfig" or "make menuconfig." Have fun! -Ossama __________ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: Warning on mounting partitions: automate e2fsck?
Hi > automatically runs e2fsck on the root partition. However, I mount > several other partitions automatically via fstab, and on those I will > after foo number of mounts get a warning that the drive should be > checked. However, it's a pain to switch to runlevel 1 and manually > check all those partitions. > > Why should I have to? Why can't the same process (init?) that > autochecks the root partition, just automatically check the others > instead of warning me? Does your /etc/fstab entry for those filesystems have the fsck-pass entry defined? For example, in my /etc/fstab file I have: # /etc/fstab: static file system information. # # /dev/sda1 / ext2defaults1 1 /dev/sda2 noneswapsw 0 0 /dev/sda3 noneswapsw 0 0 proc/proc procdefaults0 0 /dev/sda5 /usr ext2 rw 1 2 /dev/sda6 /tmp ext2 rw 1 2 /dev/sda7 /export/home/fuzzy ext2 rw 1 2 Notice that my last 3 filesystems are marked "pass 2" meaning that they will get checked second, i.e. after the root filesystem. From what I recall, those filesystems always get checked automatically for me. Make sure you don't have a zero "0" in the "pass" column of your filesystem(s). Swap and proc never get checked for obvious reasons. -Ossama
Re: non-free/libs package gets installed into libs
Hi, > When the packages was installed, you must have gotten an e-mail from > maor-installer saying: > > | If the override file requires editing, reply to this email. If your > | upload fixed reported bugs, you should close them now. > > I'm not sure which packages you're trying, but you should check > override files on the ftp sites (debian/indices/override.*.*.gz) first > and reply to the e-mail with notes what has changed. I didn't upload the packages to master by that time. The installation was local. I built the packages and then installed them as soon as I built them. I just sent the packages to master last night so I should find out if the problem persists once the package gets installed into Debian. Unfortunatley, I still can't figure out why the local installation went into libs instead of non-free/libs. -Ossama ______ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: xconfig, menuconfig
Hi Frederick, > It does not work, I still get the message "no target". I also cannot > locate any files with patterns "xconfig" or "menuconfig". > > I did install the kernel-package. > > I don't understand, what I'm doing wrong. "kernel-package" and "kernel-source" are two different things. You will need a kernel source package. The package "kernel-package" is used to create a custom Debian package of the kernel you configure with "make xconfig," for example. All of the config targets (config, xconfig and menuconfig) are normally in /usr/src/linux/Makefile. This is what I have in /usr/src/linux: COPYING Makefiledebian/ init/mm/ stamp-configure CREDITS README drivers/ ipc/ modules/ stamp-image Documentation/ Rules.make fs/ kernel/ net/ MAINTAINERS arch/ include/ lib/ scripts/ when you run "make xconfig" the "make" program will by default read the file "Makefile" and search for a target called "xconfig." If you don't have a Makefile then you will need to download a Linux kernel source package, preferrably from Debian. Note that you will need to have run X running and be logged in as root if you want to do an xconfig. The others config and menuconfig don't need X but also require you to be logged in as root. Here are some of the kernel-source packages available in the unstable distribution "potato": __ Opt develkernel-sourc 2.0.33-10 Linux kernel source. __ Opt develkernel-sourc 2.0.34-4Linux kernel source. __ Opt develkernel-sourc 2.0.35-3Linux kernel source. __ Opt develkernel-sourc 2.0.36-1Linux kernel source. __ Opt develkernel-sourc 2.1.125-1 Linux kernel source. Use the kernel source packages available in whatever Debian distribution you have. Let me/us know if you still have problems. -Ossama
weird vi exit behavior
Hi, For some months now I've been experiencing a weird problem with vi upon exit. After exiting vi, the shell prompt isn't where I left it prior to starting vi. It ends up a few lines above where it was prior to starting vi. Does anyone else have this problem? Any solutions or suggestions? I am using the unstable distribution. Thanks, -Ossama ______ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: Debian goes big business? [was: Re: Suggestion for RedHat (was: RH vs Debian)]
Hi, > An interesting question. The first step is (obviously) to convince enough > people. Especially the developers' we've been talking about. Surely they > have opinions? I'd be interested in such an entity. Count me in. -Ossama
Debian on Dell PowerEdge 6300 problems
Hi, Has anyone been able to install Debian on a Dell PowerEdge 6300 server? I am running into trouble during the scsi detection part of the boot procedure. I tried the aic7xxx=no_probe option but I still have problems. Apparently people have had success with RedHat but I have problems with RedHat, too. Any ideas? -Ossama __ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: boot disk question/suggestion
Hi again, > 2.0.x maxes out at 2^30-2^26 = 1006632960 bytes, or 960MB, of RAM. > > Thus, you'll wanna use "mem=960M". > > You can also adjust some headers (I forget which) to expand the kernel > memory / virtual memory split (it is adjustable, and it defaults to 1GB/3GB). Can the 2.1/2.2 kernels handle a gigabyte of memory? Also, I remember reading somewhere that the 2.1/2.2 kernels can handle swap partitions greater than 128MB. Is this also true? Sorry guys, I guess I should be asking this on debian-user. -Ossama
Re: boot disk question/suggestion
Hi Robert, I just wanted to thank you for your suggestion. It seems to be working! I am having problems getting the boot procedure find the root filesystem but at least I got this far! Any suggestions about the kernel being unable to find the root filesystem? Thanks, -Ossama > # dd if=resc1440.bin of=/dev/fd0 > # mount -t msdos /dev/fd0 /mnt > # cd /mnt > # rm linux > # cp /place/i/have/my/working/kernel linux > # ./rdev.sh > # cd / > # umount /mnt > > Yes, the rdev.sh script does require that you mount the disk on /mnt. > > Make sure your rescue disk contains ext2, msdos, ramdisk, initrd, and ELF > support. > > Happy booting.
Re: boot disk question/suggestion
Hi Robert, > > > Make sure your rescue disk contains ext2, msdos, ramdisk, initrd, and ELF > > > support. > > Oh, I forgot, you need minix fs support too. See if that helps. Indeed it did. :) I had to use the rescue disk on master in Incoming but with my custom kernel instead. It boots but it behaves strangely. First, it won't install with a gig of memory. I had to tell the kernel to use less otherwise it wouldn't find the root file system. Then when init dbootstrap menu comes up, it won't work. I had to go to the second console and start dbootstrap from there manually then it seemed like it was working. I'm in the middle of an install now. Wish me luck! Thanks for your help! -Ossama
boot floppies -- thanks!!
Hi, I wanted to thank Robert Woodcock and everyone who helped me (sorry, I don't recall their names :( ) for helping me figure out how to boot my Dell PowerEdge Server 6300. I am now the proud user (not owner, I wish! :) of a Quad-CPU Xeon Dell system with a gig (960MB with linux lim) of memory running Debian GNU/Linux 2.1 (potato). Now if I could just get the rest of my colleagues to switch to Debian. :p The system is _smooth_ and stable! Thanks again for the help and for a great OS! -Ossama __ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: Dell Poweredge 1300 can't boot kernel 2.0.36
Hi, On 15 Apr, Pete Templin wrote: > > I'm trying to set up a Poweredge 1300 (with integrated Adaptec 7890 Ultra2 > LVD SCSI controller) on a custom kernel v2.0.36 (added support for the > AIC-7xxx, of course), and I get the following error messages: [snip] > Any hints on getting it up and running? I really don't want to have to > put RedHat (Dell's buddy in the Linux game) on this box... You won't need to use RedHat. :-) I've documented how I installed Debian on my Dell PowerEdge Server 6300. The instructions are at: http://www.debian.org/~ossama/DELL6300 HTH, -Ossama (Sorry Adam. I haven't finished putting the instructions in a form suitable for inclusion in the Debian documentation. I'm working on it.) __________ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: Dell Poweredge 1300 can't boot kernel 2.0.36
On 16 Apr, Pete Templin wrote: > Although I can't stand the filesystem disorganization of RedHat, I almost > want to get my hands on a RedHat CD and try it on my server. I really > hope RedHat _can't_ boot it without a customized boot disk or something, > just to prove that they have a problem to solve. It'll probably work. :( You may still need to specify the amount of memory if they are still using an old kernel. Someone here tried a RedHat install on another Dell we have. It appeared to boot once we took care of the memory thing. Apparently RedHat doesn't load the SCSI drivers at boot time. Basically you specify which SCSI modules to load after their installation program has started, thus avoiding the boot time problems. > Anyway, thanks for the help. Still a 100% satisfied Debian "customer"... Indeed. Debian rules! :-) Thanks, -Ossama __________ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: Agggh! can't start new GNOME for slink
Hi, On 19 May, William R Pentney wrote: > Just downloaded the new GNOME from the slink staging area ... but when I > try to run it, all I get is the following error: > > GImLib ERROR: Cannot find file : /etc/im/im_palette.pal > gdk_imlib : Cannot find Palette. A Palette file is required for this mode > > What is the problem? Is there a package I need? You don't need a new package. The palette file was moved to `/etc/imlib/im_pallete.pal'. Check that your `~/.imrc' file reflects the new location. One of these days I should hack imlib so that it looks in the system dir if the user's `.imrc' is broken. If only I had the time. :( -Ossama -- Ossama Othman <[EMAIL PROTECTED]> Center for Distributed Object Computing, Washington University, St. Louis 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: Agggh! can't start new GNOME for slink
Hi Martin, On 19 May, Martin Bialasinski wrote: > Ossama, could you place a big note at the beginning of > /usr/doc/imlib_base/README.Debian ? I did. I just haven't uploaded a new set of packages yet since I've been waiting for a new imlib release. It doesn't seem like that is going to happen any time soon so I guess I'll do an upload. :-) -Ossama -- Ossama Othman <[EMAIL PROTECTED]> Center for Distributed Object Computing, Washington University, St. Louis 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: threads package for libc6
Hi, > pages, the functions may not be thread safe. However I think that if libc6 > in hamm supports POSIX threads, then the library functions in there must be > thread safe. Any ideas on this from you guys? I don't think it is safe to make this assumption on any system. All I can suggest is to make sure that _REENTRANT is defined and that you use the reentrant versions, if available, of a given function. A better place to ask about this issue would be one of the GNU newsgroups since Debian uses pretty much the same glibc2 that most glibc2 Linux distributions use. -Ossama
Re: mail failed, returning to sender
Somebody want to get rid of these failed mail messages? Thanks, -Ossama __ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Locale "C" in X missing?
Hi, After upgrading to the latest available XF86 Debian packages my programs no longer run properly. For example, one commercial package gives the following error: Error: X does not support locale C. Netscape complains about missing NLS stuff, and another program segmentation faults. These programs worked prior to the XF86 upgrade I did (which gave me a hard time). Incidentally, I also upgraded to Kernel 2.1.125 from Kernel 2.1.122 as well. Could the kernel be a problem, too? Thanks for any suggestions. -Ossama __ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
When will X be working properly?
Hi, Does anyone have idea when X will be functioning properly again? I've really hosed my X setup with the current packages. Perhaps I shouldn't have been so daring when I installed the current X packages. :) Thanks, -Ossama ______ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
stupid mistake: /usr/info/dir wiped out!
Hi, I did something stupid. I accidentally erased all but the *.gz files in /usr/info. Could someone please tell me how I can bring my /usr/info directory to a consistent state? I was intending to just erase the uncompressed info files but forgot about the 'dir' file. Thanks, -Ossama ______ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Where is makeinfo?
Hi, Is makeinfo available in the unstable tree yet? If so, what package is it in? I couldn't find one. Thanks, -Ossama __ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: linux operating system
Hi, > go to www.gnu.org to see the whole history of the thing. In a > nutshell: Richard M. Stallman had the idea to create a free unix > clone. Others liked the idea and helped him. Linux wrote the > first working kernel. The ESSENCE of the idea is computer code > that is totally open to examination and/or change. You've got your facts confused. Linux Torvald's created the Linux kernel. Richard Stallman founded the Free Software Foundation that is responsible for creating most of the GNU utilities that most, if not all, Linux distributions use. The Linux kernel basically runs an operating system that is comprised predominantly of GNU utilities, in most cases. Linus Torvald's did not create "GNU" nor did GNU, i.e. the Free Software Foundation, create Linux. Linux by itself is not an operating system. It is for this reason that Debian refers to its Linux distribution as: Debian GNU/Linux ^^^ ^ Here are some Linux related sites that might be of interest: www.linux.org www.linuxhq.com By the way, the FSF is working on their own kernel called the "Hurd." The Hurd will probably also run an operating system comprised mainly of GNU utilities, just like most Linux distributions. In fact, Debian is developing its own Hurd distribution. -Ossama ______ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Weird behavior with Tcsh
Hi, I came across some strange behavior with my tcsh and xterm on slink. Here is how I reproduced the behavior: Two xterms both running tcsh, starting out in home directory... xterm-1: mkdir -p test/me xterm-1: cd test/me xterm-1: pwd ---> /export/home/myhomedir/test/me xterm-2: mv test/me ~ xterm-1: pwd ---> /export/home/myhomedir/me The prompt in my "xterm-1" still shows "~/test/me." When I do an "ls" in xterm-1 I get no error. Shouldn't I get an error since the directory isn't supposed to exist anymore. Is this some behavior in tcsh that I wasn't aware of? In the past I used to get errors from "ls" of a non-existent directory if the directory was removed when I was in that directory, at least from what I recall. Am I missing something here? Thanks, -Ossama __________ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
Re: linux operating system
Hi Joseph, > I confess to my "Linux" typo when I meant Linus (Torvalds). By > "facts confused" are you referring to this typo? If not, I would > like to know to what. >From what I understood from your message, you were implying that Richard Stallman started Linux. Looking back at your message it seems that I may have misunderstood what you were saying. My apologies. -Ossama __________ Ossama Othman <[EMAIL PROTECTED]> 58 60 1A E8 7A 66 F4 44 74 9F 3C D4 EF BF 35 88 1024/8A04D15D 1998/08/26
