Re: OT: regexp spam filter
the following regexp *should work: 217\.78\.(6(4|5|6|7|8|9)|7(0|1|2|3|4|5|6|7|8|9))\.[0-9]+ it's not really obscure, but it's not really all that clever either. =) i'm sure there are more compact solutions out there! -alan Gary Turner wrote (Mon, 20 May 2002 16:11:00 -0500 ): |I am not practiced in the use of regexp's, and need some help. I want |to block a Nigerian domain (I wonder why?) whose IP block is 217.78.64.0 |- 217.78.79.255. For unknown reasons (ignorance?), my various |incantations and curses have failed the test. | |If the cognoscenti among you would post examples of regexps that will |work, I will be thankful. And if yours is a more obscure or esoteric |example, please explain the manipulation so that I can maybe figure out |the next magic spell without leaning on the list. | |tnx, |-- |gt |Yes I fear I am living beyond my mental means--Nash | | |-- |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]
eterm broken in latest woody upgrade?
ever since doing 'apt-get update' then 'apt-get upgrade' last night, my eterms have been broken. when i start one, the window is mapped to the screen, but my shell prompt doesn't appear. about a minute after i start the eterm, the shell prompt finally appears and all is well. to try to debug this, i ran Eterm in gdb, interrupted while the window was hung, and spit out the frames on the stack. here's what i see: (gdb) where #0 0x4032d82e in select () from /lib/libc.so.6 #1 0x4023bfdc in _XlcPublicMethods () from /usr/X11R6/lib/libX11.so.6 #2 0x401973ba in _XRead () from /usr/X11R6/lib/libX11.so.6 #3 0x40197dc3 in _XReply () from /usr/X11R6/lib/libX11.so.6 #4 0x40180e24 in XListFontsWithInfo () from /usr/X11R6/lib/libX11.so.6 #5 0x401f12c2 in _Xutf8DefaultDrawImageString () from /usr/X11R6/lib/libX11.so.6 #6 0x401f16f2 in _Xutf8DefaultDrawImageString () from /usr/X11R6/lib/libX11.so.6 #7 0x401f1be1 in _Xutf8DefaultDrawImageString () from /usr/X11R6/lib/libX11.so.6 #8 0x401f215b in _Xutf8DefaultDrawImageString () from /usr/X11R6/lib/libX11.so.6 #9 0x401f25ad in destroy_fontdata () from /usr/X11R6/lib/libX11.so.6 #10 0x401ae125 in XCreateOC () from /usr/X11R6/lib/libX11.so.6 #11 0x401ad655 in XCreateFontSet () from /usr/X11R6/lib/libX11.so.6 #12 0x4002e4c3 in create_fontset () from /usr/lib/libEterm-0.9.2.so #13 0x4002e66f in init_locale () from /usr/lib/libEterm-0.9.2.so #14 0x4002faf7 in init_command () from /usr/lib/libEterm-0.9.2.so #15 0x40062381 in eterm_bootstrap () from /usr/lib/libEterm-0.9.2.so #16 0x08048584 in eterm_bootstrap () #17 0x4027c17f in __libc_start_main () from /lib/libc.so.6 it seems something is going on with fonts and such...does anyone know what might be going on? thanks... -alan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
testing + some unstable packages
i've got an installation of testing on my laptop, but i wanted to upgrade a few packages (notably galeon) to the unstable version. i'd rather not just point apt to the unstable distribution, as i'm pretty happy with the way testing is working out for me. in order to do this, i'm manually using dpkg to install packages that i download from the unstable distribution. however, for this to work, i need to upgrade a bunch of other packages on which these few packages depend. is there a way to determine the set of packages which must be installed (from the unstable distribution) before i go about installing the packages i want? thanks for any ideas you might have! -alan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
error installing emacs?
i just did an update for my potato box, and the only error that didn't correct itself with several Install passes was emacs. here's the error, as far as i can tell: install/dpkg-dev: Byte-compiling for emacs20 cp: debian-changelog-mode.el: No such file or directory emacs-install: /usr/lib/emacsen-common/packages/install/emacsen-common-install emacs20 failed at /usr/lib/emacsen-common/emacs-install line 28. dpkg: error processing emacs20 (--configure): subprocess post-installation script returned error exit status 29 does anyone know what's going on? any ideas on how to fix it? thanks! -alan
installing on an ibm tp-i1452
hi everyone- i'm trying to install slink onto an ibm thinkpad i1452, and i'm having some problems when it comes to configuring pcmcia support. i allow all the defaults to stand (i82365 chipset, no special options), but the next thing it does (presumably probing for the pcmcia hardware?) locks the machine hard. has anyone run into this before? does anyone know of parameters i need to pass? if so, i'd appreciate some pointers to get around this rather substantial bump in the road... thanks! -alan
smbfs question...
hi everyone- i have a linux box and a win98 box on a network, with the windows box exporting a directory via SMB. i use smbmount to mount the share, which works fine. the problem arises when i try the following operation: 9:27 [28] alsu (alsu):tmp/salishan99/CVS - ls -l Entries* -rwxr-xr-x 1 alsu alsu0 Apr 17 09:26 Entries* -rwxr-xr-x 1 alsu alsu 308 Apr 17 09:26 Entries.Backup* -rwxr-xr-x 1 alsu alsu 320 Apr 17 09:26 Entries.Log* 9:27 [29] alsu (alsu):tmp/salishan99/CVS - mv -f Entries.Backup Entries mv: cannot move `Entries.Backup' to `Entries': Permission denied basically, mv doesn't want to unlink the Entries file to rename Entries.Backup to Entries. this is essentially the operation that CVS wants to perform when updating a working directory. has anyone encountered this problem and know of a solution? i could be way off here, but it seems like a problem in the implmentation of smbfs. anyway, thanks! -alan
Re: smbfs question...
R. Brock Lynn wrote (Sat, 17 Apr 1999 12:16:27 -0500 ): | |Well here's a simple thing to check: | |Do you have write permission on the SMB share? In windows you can have two |passwords... one for read and one for write. | Brock- thanks for the idea. actually, i'm quite sure i have the correct permissions to do this, as i can manually delete the file with /bin/rm. Shaleh suggested to me that this is a basic limitation in windows filesystems. that is to say, if you go into the explorer or a command shell on a win box and try something analogous, you get an error. so, now i have a question for the list. i guess i can accept this behavior, but now i'm curious: what does /bin/mv do when the target file already exists (and isn't a directory). it must do something other than unlink the target file and relink the source file to that name, right? otherwise, it seems like it should work. thanks, everyone. -alan
switching virtual consoles...
Hi everyone- we have a box that doesn't run X. i'd like to configure it to switch to a specific virtual console after some amount of idle time on the other virtual consoles. is there a solution out there? thanks. -alan
xserver fixated on tty7?
i probably missed something as i upgraded to slink today, but is the tty that the xserver uses somehow changed to be in some config file (rather than taking the first unused one)? i had a couple extra virtual consoles which i used to use on tty7 and tty8. after the upgrade, the xserver periodically got hosed (no response to keyboard input/mouse clicks) and when i was able to switch over to virtual consoles, i saw that the one sitting on tty7 was hosed. i modified inittab to remove the two extra getty's, and now everything's back to normal. anyone have any ideas? thanks. -alan
testing video ram?
does anyone know of a a utility to test memory on the video card? kinda like a memtest for the video memory? thanks! -alan
Re: testing video ram?
[EMAIL PROTECTED] wrote (Thu, 18 Feb 1999 13:54:37 EST ): | | does anyone know of a a utility to test memory on the video card? | kinda like a memtest for the video memory? thanks! | | |You can see how much video memory you have with SuperProbe - if you have X |installed, you have SuperProbe. | i was actually looking for something that would test whether or not the memory is good, sort of like a stress test for the video memory. any ideas? thanks. -alan
audio devices...
i just did a fresh install of slink, and for some reason, the audio devices (most notably /dev/audio and /dev/dsp) were not created. was this done by design? anyway, i created the necessary devices by doing '(cd /dev ; ./MAKEDEV audio)' as root. the devices were created, but the ownership of the devices were root.root, rather than root.audio, as i'm used to seeing. is this right? thanks everyone... -alan
pcmcia card...
Hi everyone- i'm having some problems with pcmcia on a laptop here. i'm trying to get it to recognize a 3com 3c575bt cardbus ethernet card, and the stock pcmcia modules aren't cutting it. so, i'm installing hamm from scratch. then, i install new libc6 (v2.0.7u) so that i can install v3.0.6 pcmcia-cs and pcmcia-modules (modules corresponding to the 2.0.34 kernel). this actually works pretty well, and at this point, the card is usable with the kernel that comes in the kernel-image package. now i want to recompile the kernel for apm, ps/2 mouse, etc., so i: - install kernel-source (still version 2.0.34) - reconfigure kernel for apm, ps/2, and pcmcia - compile and install new kernel image. when i reboot with the newly compiled kernel, the 3c575_cb module doesn't want to load. the relevant errors are: pcmcia/3c575_cb.o: unresolved symbol netif_rx_R6a97aa56 pcmcia/3c575_cb.o: unresolved symbol unregister_netdev_R84fb480b pcmcia/3c575_cb.o: unresolved symbol init_etherdev_Rc8833d67 pcmcia/3c575_cb.o: unresolved symbol dev_kfree_skb_R61d50ea2 pcmcia/3c575_cb.o: unresolved symbol eth_type_trans_R81c3bb43 pcmcia/3c575_cb.o: unresolved symbol dev_alloc_skb_R36d7a14b are there kernel configuration options that i've missed in order to get these symbols defined? do i basically need to install pcmcia-source and recompile modules myself? thanks in advance... -alan
laptop internal modem...
hi all- i'm trying to set up the internal modem on a Winbook laptop. it's configured to be on COM2 under the BIOS. as described by the Serial HOWTO, i'm using kermit to test the modem, and this is the result: [/home/alsu] C-Kermitset line /dev/ttyS1 Sorry, can't open connection: /dev/ttyS1 my suspicion is that this is a winmodem...is this the expected behavior of kermit on a winmodem? thanks... -alan
pcmcia cardmgr doing the right thing after resuming?
Hi- i have a laptop here that does not see the pcmcia cards after resuming. basically, suspending the machine goes fine, and after resuming, i can do most everything. however, the network is not there. bouncing the card manager with '/etc/init.d/pcmcia restart' gets things back to normal. what am i missing here? is there a script that gets run on a resume where i can do this? thanks...and be gentle; this is my first attempt at a laptop installation. =) -alan
weird x error?
this probably isn't a debian specific error, but i think it might have something to do with how i have the system set up. i'm logging in remotely to an SGI IRIX machine and attempting to run an X application. it fails with something like the following error: Error initializing colors for colormap i'm running in 32 bit color depth, and no color intensive processes are being displayed. (don't tell me that sgi has invented 64-bit color...right?) what does this mean? thanks. -alan
Re: As user: swap capslock/cntrl?
Anders Hammarquist wrote (Fri, 28 Aug 1998 19:45:20 +0200 ): | |I think you can swap them in the console as well by using loadkeys. I'm |afraid I can't offer you any help as to how to go about doing it though. | so i'll shamelessly use this as a segue to a question i've had for a while. i've actually accomplished this by running 'loadkeys -d' after changing /usr/src/linux/drivers/char/defkeymap.map. the following lines: keycode 29 = Control keycode 58 = Caps_Lock were changed to: keycode 29 = Caps_Lock keycode 58 = Control since this works so well, i decided to try to roll my own kernel with the key table in that file. so i run 'loadkeys -d -m' to generate defkeymap.c and recompile the kernel. for some reason the key bindings don't take. am i doing this correctly? has anyone else accomplished such a feat? thanks. -alan
Re: getting fqdn from gethostname on a hamm box
Martin Schulze wrote (Wed, 26 Aug 1998 02:12:58 +0200 ): | Martin Schulze wrote (Wed, 26 Aug 1998 01:04:36 +0200 ): | |If you want to write portable programs then you use gethostname() to | |determine the hostname, check for a '.' and if it's not included you | |use gethostbyname() to get the fqdn. For an example, take a look at | |the syslogd source. | |This doesn't work on Solaris? Why, you should tell us why it fails. | ok, i've found one case where neither gethostname() nor the method Joey descibed worked in getting the fqdn. on that particular machine, i found that the h_name field still had the simple name, but that h_aliases[0] contained the fqdn. so, it's in there somewhere, it's just not always in the h_name field of the struct hostent. now my question is this: how does one tell how many aliases are in the h_aliases vector? do i just keep examining the elements of the vector until one of them is null? thanks everyone... -alan
getting fqdn from gethostname on a hamm box
when i call gethostname() on my hamm, it returns only the name of the machine, rather than a fqdn. i distinctly remember that when i was running bo, i got a fqdn. does anyone know how i can make it return the fqdn? or is there another function i should be using? thanks. (btw, getdomainname() returns (none), which suggests to me that something else is going on here. am i right?) -alan
dependencies in dselect...
is there anyway to permanently override the suggests form of dependency in dselect? i have a system on which i want xbase installed, but not an xserver. every time i run dselect, it insists on changing the status of xserver-vga16, xfntbase75, and xfntbase100 (i think those are the ones) to install. i've put these packages on hold, but that doesn't seem to help, it just goes ahead and resets the status. what's the right way to do this, other than remembering to hit Q instead of just the Return key... thanks. -alan
simple server?
i'm pretty confused about some weird behavior i'm seeing with the
socket routines on a couple hamm machines. i've written a simple dumb
server to demonstrate the behavior, and i'll attach it below.
basically, it opens a socket and tries to bind it to a specific port.
the wierd thing is that bind doesn't care if i try to bind to a
privileged port (numbered less than 1024). the same code compiled
and run on a solaris box results in an error generated by the call to
bind() -- with errno set to EACCES, i believe.
the weirdest thing is that the socket doesn't seem to be actually
bound when i run it under linux, regardless of whether or not the port
i try to use is privileged or not. i extended the code to actually
bind to port 23, listen on it, and go into an accept loop. the code
runs to the accept loop without errors, but telnet connections still
seem to go to the right place (i.e. telnetd).
can anyone spot what i'm doing wrong? anyone care to try this on
their hamm box? thanks.
-alan
#include stdio.h
#include sys/types.h
#include sys/socket.h
#include netinet/in.h
#include netdb.h
#define DS_DEFAULTPORT
#define USAGE Usage: %s [port]\n
int
main(int argc, char **argv)
{
int serverSock;
struct sockaddr_in serverSockName;
int port;
struct protoent *proto;
if (argc 2) {
fprintf(stderr, USAGE, argv[0]);
exit(1);
}
if (argc == 2) {
if (sscanf(argv[1], %d, port) != 1) {
fprintf(stderr, USAGE, argv[0]);
exit(1);
}
}
else {
port = DS_DEFAULTPORT;
}
fprintf(stderr, using port %d\n, port);
/* 6 == protocol no. for tcp */
serverSock = socket(AF_INET, SOCK_STREAM, 6);
if (serverSock == -1) {
perror(socket);
exit(1);
}
serverSockName.sin_family = AF_INET;
serverSockName.sin_addr.s_addr = INADDR_ANY;
serverSockName.sin_port = port;
if (bind(serverSock, (struct sockaddr *)serverSockName,
sizeof (struct sockaddr)) != 0) {
perror(bind);
exit(1);
}
fprintf(stderr, socket bound\n);
return(0);
}
kernel image size...
just out of curiousity, if i do 'make clean ; make zImage' in /usr/src/linux repeatedly and compare the size of /usr/src/linux/arch/i386/boot/zImage, should i expect to get the same size image? i'm currently burning in a computer by sticking it in such a loop, and i'm saving the images, just to tax the filesystem a bit. this is a partial listing of the images i've saved: -rw-r--r-- 1 root root 45 Aug 8 23:25 zImage.26 -rw-r--r-- 1 root root 45 Aug 8 23:31 zImage.27 -rw-r--r-- 1 root root 45 Aug 8 23:37 zImage.28 -rw-r--r-- 1 root root 44 Aug 8 23:43 zImage.29 -rw-r--r-- 1 root root 45 Aug 8 21:05 zImage.3 -rw-r--r-- 1 root root 44 Aug 8 23:49 zImage.30 -rw-r--r-- 1 root root 45 Aug 8 23:55 zImage.31 -rw-r--r-- 1 root root 45 Aug 9 00:01 zImage.32 -rw-r--r-- 1 root root 48 Aug 9 00:07 zImage.33 -rw-r--r-- 1 root root 411120 Aug 9 00:14 zImage.34 -rw-r--r-- 1 root root 411121 Aug 9 00:20 zImage.35 -rw-r--r-- 1 root root 47 Aug 9 00:26 zImage.36 -rw-r--r-- 1 root root 411122 Aug 9 00:32 zImage.37 -rw-r--r-- 1 root root 49 Aug 9 00:38 zImage.38 -rw-r--r-- 1 root root 411121 Aug 9 00:44 zImage.39 -rw-r--r-- 1 root root 46 Aug 8 21:11 zImage.4 -rw-r--r-- 1 root root 411121 Aug 9 00:50 zImage.40 -rw-r--r-- 1 root root 46 Aug 8 21:17 zImage.5 -rw-r--r-- 1 root root 46 Aug 8 21:23 zImage.6 the file sizes are pretty close, but the're not identical. of course, i haven't actually tried any of these, but i have no reason to suspect that they won't work. am i right, or should i conclude that this machine has problems? thanks. -alan
Re: XDM / KDM problem...
Well, that is a bit extreme. don't worry about shutting down the X server. when the tty on which the X server is running is active, you can switch to other tty's by using Control-LeftAlt-Fx, where x is the number of a different tty (presumably a text console). in response to the original question, if you still don't want xdm to start when the machine is booted, simply edit /etc/X11/config and change the line which says 'start-xdm' to 'no-start-xdm' (or just comment the darn thing out). finally, i have no idea how to do change to kdm. sorry. -alan Zaphod Beeblebrox wrote (Sun, 02 Aug 1998 10:28:53 PDT ): | |This may be kind of extreme, but I can get out of XWindows (xdm) |using the Ctr-Alt-BackSpace sequence that forces a reset. Then, at |some critical point during the reset, you can use the LeftAlt-RightArrow |sequence to switch to a different tty. | |Debian 1.3 user | |Xwindow... is there a way to get out of Xwindow if XDM is turned |on? | | | |__ |Get Your Private, Free Email at http://www.hotmail.com | | |-- |Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev /null | -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
log file rotation in hamm...
I just installed hamm, and I noticed that the log file rotation in
/etc/cron.{daily,weekly}/sysklogd now uses a facility called
syslogd-listfiles. does anyone know how to change the periodicity
with which logs are rotated? most notably, i want to rotate syslog on
a weekly, not daily basis, but it seems that there's no way to make
syslogd-listfiles to do the right thing without munging
/etc/syslog.conf in unspeakable ways.
anyone got a solution? thanks...
-alan
--
Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: log file rotation in hamm...
Nathan E Norman wrote (Thu, 30 Jul 1998 12:51:59 -0500 (CDT) ):
|On Thu, 30 Jul 1998, Alan Su wrote:
|
|: I just installed hamm, and I noticed that the log file rotation in
|: /etc/cron.{daily,weekly}/sysklogd now uses a facility called
|: syslogd-listfiles. does anyone know how to change the periodicity
|: with which logs are rotated? most notably, i want to rotate syslog on
|: a weekly, not daily basis, but it seems that there's no way to make
|: syslogd-listfiles to do the right thing without munging
|: /etc/syslog.conf in unspeakable ways.
|:
|: anyone got a solution? thanks...
|
|Comment out the following lines in /etc/cron.daily/sysklogd
|
|for LOG in `syslogd-listfiles`
|do
| if [ -f $LOG ]; then
| savelog -g adm -m 640 -u root -c 7 $LOG /dev/null
| fi
|done
|
|(Try running `syslogd-listfiles' - you'll notice that /var/log/syslog
|is the output)
|
Yea, i tried this first to see what it did, and you're right it spits
out /var/log/syslog. my question was basically: how do you modify the
behavior of syslogd-listfiles?
|If you don't want ANY logs rotated on a daily basis, remove
|/etc/cron.daily/sysklogd, or add an exit 0 to the beginning of that
|file.
|
Well, this is only half the solution; if i simply exit the script
without doing anything, i don't get the daily rotations. however, i
don't get the weekly ones either (which is what i want).
-alan
--
Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: log file rotation in hamm...
Shaleh wrote (Thu, 30 Jul 1998 14:00:56 -0400 ): |Alan Su wrote: | Well, this is only half the solution; if i simply exit the script | without doing anything, i don't get the daily rotations. however, i | don't get the weekly ones either (which is what i want). | | -alan | |So move the script from daily to weekly. Then it gets run when you want |it to be. i could do that. it just feels a little strange since 'syslogd-listfiles' is supposed to output the list of logs to rotate daily, and 'syslogd-listfiles --weekly' is supposed to do the same for those to be rotated weekly. moving the script doesn't seem like it should be the Debian Way. how would i go about making the ppp log file rotate daily (for example)? this isn't going to be possible using syslogd-listfiles, as far as i can tell. anyway, this isn't that big a deal to me. if this isn't possible with syslogd-listfiles, i'll just install my old bo scripts... -alan -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: log file rotation in hamm...
OK, both Norman and Shaleh didn't understand my question. i can only conclude that i'm an idiot. =) here's a second shot: these are the output of 'syslogd-listfiles' and 'syslogd-listfiles --weekly': alsu# syslogd-listfiles /var/log/syslog alsu# syslogd-listfiles --weekly /var/log/messages /var/log/mail.warn /var/log/mail.err /var/log/mail.info /var/log/uucp.log /var/log/lpr.log /var/log/user.log /var/log/ppp.log /var/log/kern.log /var/log/auth.log /var/log/mail.log /var/log/daemon.log /var/log/debug /etc/cron.daily calls 'syslogd-listfiles' and concludes that /var/log/syslog should be rotated daily. /etc/cron.weekly calls 'syslogd-listfiles --weekly' and rotates all the other logs on a weekly basis. the change i want to make is to rotate /var/log/syslog on a *weekly* basis, rather than a daily basis. as Shaleh points out, this particular modification can be accomplished by simply moving /etc/cron.daily/sysklogd to /etc/cron.weekly/sysklogd2 or something. i claim this isn't the Debian Way, but rather a hack. for example, how would one make /var/log/messages rotate daily rather than weekly? i would like a way for 'syslogd-listfiles' and 'syslogd-listfiles --weekly' to output something like the following: alsu# syslogd-listfiles /var/log/syslog /var/log/messages alsu# syslogd-listfiles --weekly /var/log/mail.warn /var/log/mail.err /var/log/mail.info /var/log/uucp.log /var/log/lpr.log /var/log/user.log /var/log/ppp.log /var/log/kern.log /var/log/auth.log /var/log/mail.log /var/log/daemon.log /var/log/debug it seems to me that there should be a way to do this, but currently it requires an act of god, or at least some really gross changes to /etc/syslog.conf which will probably change the way stuff gets logged. (i admit, i don't really understand what facilities and priorities are, which is what syslogd-listfiles uses to decide when things need to be logged.) looked at it another way, my question could be, why doesn't syslogd-listfiles have it's own config file which describes directly when things should be rotated, rather than inferring that from /etc/syslog.conf? does this make sense? -alan -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: log file rotation in hamm...
Shaleh wrote (Thu, 30 Jul 1998 16:58:18 -0400 ): |Have you tried man on syslogd-listfiles?? It should give you a |pointer. If not, contact the packages maintainer and ask him. He/she |will know the correct way to do this. If the script needs editing or a |conf file he can either make his own or ask for help. If this is not |resolved, I will look into it when I get home. | ok, let's put this to rest: - YES, i've read the man page. - YES, it does give me a pointer. the pointer essentially says that syslogd-listfiles decides when a log should be rotated based on its entry in /etc/syslogd.conf. this was made clear very early on in the man page (and in my first email today). - to do what i want to do requires that log file rotation periodicity be decided independently of /etc/syslog.conf entry. it's entirely possible that (a) what i'm trying to do is supported by syslogd-listfiles, but i don't know what i'm doing; (b) what i'm trying to do is too stupid for words; or (c) it's a reasonable thing to want, but syslogd-listfiles does not support it. i believe (c) is true. =) seriously though, i'll wait a little while to see what the debian-user community comes up with. personally, i trust that if there is a coorect way to do this, someone on debian-user knows it. if not, then i'll file an enhancement request with the sysklogd package. -alan -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
fresh hamm installation...
i'm doing a fresh install of hamm, and i'm just wondering: what happened to the ftp method of installation? my choices were floppy, cd-rom or hard drive, but no option to do an ftp install. basically, i'm forced to do a floppy install since all i have on the system is win98, and in their great wisdom, microsoft has not provided a way to make a vanilla FAT partition. oh well...just curious. -alan -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
installing without a floppy?
Hi- i'm kicking windows off my disk, and i'd like to put hamm in it's place. currently, i'm dual booting between windows and bo, but i'd like to keep it available, just so i can compare and fiddle with both of them. My question is: what's the best way to do this? i've considered: - somehow installing hamm off one of the existing ext2 partitions - somehow installing hamm from a vfat partition - repartitioning the old disk and essentially backing up all the necessary partitions from my bo system to the old disk, then running an upgrade the catch here is that my floppy drive is completely out of commission. can this be done? most of the instructions i've seen involve copying the disk images to floppy and booting from them... thanks! -alan -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: installing without a floppy?
ok, if i may be flippant for a moment, i had a thought about this. the installation instructions have this to say about installing from a DOS partition: 1.Get the following files from your nearest Debian ftp mirror and put them into a directory on your DOS partition: resc1440.bin, drv1440.bin, base2_0.tgz, root.bin, linux, install.bat and loadlin.exe. 2.Execute install.bat from that directory in DOS. 3.skip down to Installation . install.bat simply calls loadlin to load the linux kernel with the magic incantations to set up the installtion RAM disk and such. what if i did the following: 1. Put resc1440.bin, drv1440.bin, base2_0.tgz, root.bin, and linux into a directory /hamm-install 2. add a new entry into /etc/lilo.conf with: image=/hamm-install/linux initrd=/hamm-install/root.bin root=/dev/ram 3. continue with the installation? will this work? (the goal being to install hamm on this system with no working floppy drive, while preserving the existing bo installation.) thanks for your help! -alan Alan Su wrote (Wed, 22 Jul 1998 08:51:37 -0700 ): | |i'm kicking windows off my disk, and i'd like to put hamm in it's |place. currently, i'm dual booting between windows and bo, but i'd |like to keep it available, just so i can compare and fiddle with both |of them. | -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: What's a good video card?
Shaleh wrote (Tue, 07 Jul 1998 21:59:32 -0400 ): |Well on Alan Cox's web page for TV in Linux he has a pretty GIF that has |the red circle w/ the slash on an ATi logo and specifically states that |until ATi gives out specs their all-in-wonder card will not do anything |more than X. SO I refuse to support ATi. And would ask all others to |do the same. A capitalist society is the purest form of democracy. You |vote with your money. I refuse to gve ATi any of mine. | This is all well and good, but the fact remains that your original statement (ATI does NOT support Linux or Xfree in ANY way) was wrong. This isn't really a black and white issue, and while ATI may not be the most open-computing-friendly company, I really think there are many more worthy targets for such boycotts. As far as the original question goes, I'm running XFree86 on my [EMAIL PROTECTED] and it works great. If you're still wary about installing hamm and opt for bo instead, you may want to upgrade to a later version of the XFree86 Mach64 server (at least I did when I installed it...the version being 3.3.2 I think). -alan -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
obscure kernel error
Does anyone know what the following error (extracted from /var/log/syslog) means: :May 28 13:46:18 alsu kernel: Warning: kfree_skb passed an skb still on a list (f :rom 017ff8f8). :May 28 13:46:18 alsu kernel: bounds: :May 28 13:46:18 alsu kernel: CPU:0 :May 28 13:46:18 alsu kerneld: error: exit: Identifier removed thanks! -alan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
kernel error?
After my machine was completely hung this morning, I did a hard reset, fsck'd the filesystems, and found the following in /var/log/syslog: :May 20 02:10:42 alsu kernel: Warning: kfree_skb passed an skb still on a list (from 01d5ccf0). :May 20 02:10:42 alsu kernel: general protection: 4674 :May 20 02:10:42 alsu kernel: CPU:0 does anyone know what actually happened here? This message appeared in the log about 6.5 hours before the machine actually hung, I think. I don't know if it's related to the crash, but it did look suspicious. thanks... -alan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
faking the hardware address?
My linux box is connected to a campus network, and i'd like to provide access to that network to a machine connected over a serial/modem line via PPP. i *think* the normal way to do this would be: - get an IP assigned for the PPP box - compile the linux kernel with IP forwarding and have it route packets to and from the PPP box - make the gateway that my linux box uses accept packets from the new IP with my ethernet card's hardware address and through my ethernet port the first two I can do, but the third one may be problematic, as I'm not an administrator for the gateway or anything. i have been told that i can have multiple machines connect through my port (using a hub), but it expects each IP to have a unique hardware address. My question is: can I fake this? all i think i need is to have the linux box use a different hardware address for packets using the second IP, right? thanks in advance for any random thoughts or ideas... -alan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: why debian?
Ian Stuart wrote (Wed, 15 Apr 1998 15:14:33 + ): | On Wed, 15 Apr 1998, Ian Stuart wrote: | | RedHat is easier to install initially (it's only 2 floppies as opposed | to Debians 6+), however the updating system in debian (dselect) is _far_ | superior.. | |I feel I must clarify my position here - I've done ally my installs from |FTP sites. | |It's one of the advantages of working for an Academic site | |In this situation, RedHat seems to offer an easier install... | I have to jump in here. If you're installing from FTP sites, you only need 2 floppies maximum: the rescue/boot disk and the drivers disk (as Isabelle noted). I've done all my installs (granted only two =)) from an academic site, and I've never used more than two disks. Perhaps you didn't realize that the base system could be installed over FTP? There may be advantages to Red Hat, but I don't think they are in the installation process... -alan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Setting up Anon FTP?
Richardson,Anthony wrote (Wed, 25 Feb 1998 20:01:00 -0500 ): | |I'm trying to set up an anonymous ftp server. Everything seems to be |working except |ls or dir. I copied /bin/ls to /home/ftp/bin/ls and set up permissions |as described in the |ftpd man page. When I type ls however I get: | | 150 Opening ASCII mode data connection for '/bin/ls'. | 226 Transfer complete. | |and then nothing. | |nlist works. | |Any ideas? Thanks | Well, and idea, but I'm not sure if it's right or if this is going to help. I think that the problem is that /home/ftp/bin/ls depends on libc.so.?. Since anonymous ftp sessions run chroot'd to the ~ftp directory, it can't see the shared library. I tried making a /home/ftp/lib directory and putting a copy of libc.so.5 there, but that didn't help. Is my/our only recourse to compile a statically-linked version of ls? -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Setting up Anon FTP?
Ossama Othman wrote (Thu, 26 Feb 1998 12:31:56 -0500 (EST) ): |Why are you trying to chroot to /bin/ls? The ftpd daemon automatically |does a chroot when someone logs in as anonymous or ftp. Here is an |excerpt from the ftpd man page. Did you follow what it says? | Tony was trying to diagnose the problem, the same problem I'm having. Namely, ls depends on libc (at least) and simply copying the library to the ~ftp/lib directory doesn't work. The man page mentions nothing about this, leading me to believe that it assumes that ls is statically linked. -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Setting up Anon FTP?
Ossama Othman wrote (Thu, 26 Feb 1998 14:03:54 -0500 (EST) ): | |Ah, I see. However, what is the specific problem? I assume Tony ran ldd |on /bin/ls and copied over the necessary shared libraries. What is wrong |or isn't happening now? Well, I can't speak to his situation, but currently, when I do 'ls' from the ftp prompt, this is the result: ftp ls 200 PORT command successful. 150 Opening ASCII mode data connection for '/bin/ls'. 226 Transfer complete. ftp No output...I think ls is failing, and the only reason I can think why is because of the shared library failing to link. Tony made a good point though: there are a lot of files that are required to do dynamic linking, and these are all out of the scope of a chroot'd file system. Heck, a chroot'd process won't even see /lib/ld.so, right? So, do we need to copy /lib/ld.so, /etc/ld.so.*, and a bunch of other stuff to the ~ftp area? |In my experience, the steps listed in the man page are all I had to do |(except for the character devices). | Hmmm...does your man page describe the process for a dynamically linked ls? In my man page, it just says copy /bin/ls to ~ftp/bin/ls and it should work. Clearly, this isn't the case... Thanks. -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Problems with java
Steve Witt wrote (Thu, 29 Jan 1998 19:27:58 -0800 (PST) ): | |[/usr/home/witt/devel/java] $ java HelloWorldApp.class |Can't find class HelloWorldApp.class | | |I've checked the environment variables CLASSPATH (I don't have it set, so |it should be set by the java scripts) and can't think of anything else. |Anyone have any ideas?? | I haven't used jdk 1.1 yet, but I suspect that even if the CLASSPATH variable is being set in the java wrappers, it's not including '.' (the current directory). In jdk 1.0.2, the default was to not include the '.' directory. You might want to just set the variable explicitly to include '.' or just hack the java wrapper. hope that helps... -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Xsession (help)
Greg Green wrote (Sat, 24 Jan 1998 19:12:45 -0700 ): |Hi, |I am having a very frustrating time with my .xsession. I can login just |fine if I don't have a .xsession of my own, but if I copy the |/etc/X11/Xsession to my $HOME/.xsession, my console just hangs. After |this hanging, I have to reboot. | |Here is what my Xsession looks like (the one I copy to $HOME/.xsession) |#! /bin/sh |# |[...] |startup=$HOME/.xsession This is the first thing to notice...$startup is the file to which you have copied the global Xsession file. |[...] |if [ -x $startup ] grep -q ^allow-user-xsession /etc/X11/config |then | exec $startup |else |[...] Here's your problem. It's going into an infinite loop, reading your $HOME/.xsession file. Why do you want to copy the global Xsession file to $HOME/.xsession? -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
inconsistency between server and config file?
All this talk about setting up the X server got me wondering. So, I checked it out, and it seems that there's some kind of inconsistency on my machine: :11:31 [1] alsu (alsu):/home/alsu - ls -l /usr/X11R6/bin/X :lrwxrwxrwx 1 root root 26 Nov 11 23:31 /usr/X11R6/bin/X - /usr/X11R6/bin/XF86_Mach64* :11:31 [2] alsu (alsu):/home/alsu - cat /etc/X11/Xserver :/usr/X11R6/bin/XF86_SVGA :Console : :The first line in this file is the full pathname of the default X server. :The second line shows who is allowed to run the X server: :RootOnly :Console (anyone whose controlling tty is on the console) :Anybody : 'ps' shows that X (and therefore, XF86_Mach64) is running. So, what is it that the /etc/X11/Xserver file does? Thanks! -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Need help in X Windows installation
Daniel Martin at cush wrote (Wed, 10 Dec 1997 22:34:52 -0500 ): |Hamish Moffatt [EMAIL PROTECTED] writes: | | The window manager should always be last too. Specifically, the last | command should not end in , but it's most useful if that's the window | manager. You could make it xclock or something, but then you'd | have to kill the clock somehow to logout. | |Depends - most window managers will send a message to all active X |clients when they exit that causes them to shut down. | I don't think this is right...I've fiddled a lot with window managers, and I switch them ``mid-flight'' quite a bit. (Since I have an xterm as the final exec'd command, killing my window manager doesn't end my x session.) If what you're saying is true, every time I switch window managers, all my windows would die, effectively ending the session. Needless to say, this doesn't happen. -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Need help in X Windows installation
Hamish Moffatt wrote (Thu, 11 Dec 1997 18:36:55 +1100 ): |On Wed, Dec 10, 1997 at 08:00:30PM -0800, Alan Su wrote: | Daniel Martin at cush wrote (Wed, 10 Dec 1997 22:34:52 -0500 ): | |Hamish Moffatt [EMAIL PROTECTED] writes: | | | | The window manager should always be last too. Specifically, the last | | command should not end in , but it's most useful if that's the window | | manager. You could make it xclock or something, but then you'd | | have to kill the clock somehow to logout. | | | |Depends - most window managers will send a message to all active X | |clients when they exit that causes them to shut down. | | I don't think this is right...I've fiddled a lot with window managers, | and I switch them ``mid-flight'' quite a bit. (Since I have an xterm | as the final exec'd command, killing my window manager doesn't end my | x session.) If what you're saying is true, every time I switch window | managers, all my windows would die, effectively ending the session. | Needless to say, this doesn't happen. | |The window manager will replace itself with the new one, |(using an exec() call, presumably). So the same command in |your .xsession/.xinitrc is still running. | This is only true if you use the window manager facility to restart. I usually kill the window manager with (kill pid), temporarily leaving me window manager-less. Then, I start another window manager from a shell I have open. I suppose I could configure my window manager to exec a different window manager, but that's too much effort. or i'm too lazy. (probably the latter.) -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: xdm....HELP!!!!!!???????
Brian V Bonini wrote (Thu, 11 Dec 1997 14:28:59 -0500 ): |What I need to do is somehow stop xdm at boot up (I'm booting Linux from a |floppy) so I can get access to the console as root and delete the |.xsession file that resides in /root. Please HELP |Thanks |-Brian, ([EMAIL PROTECTED]) | Why don't you just log in with your user account and su to or even login as root? You can also wait for the X server to start up and then hit Ctrl-alt-f1 to get to a console... If there's some reason why you don't want to do this, you can try to boot into single user mode. This would probably require you to have LILO installed on your floppy. If you do, at the LILO prompt, hit Shift, and a boot: prompt should appear. type the image name you want to boot (probably linux, hit tab to get a list), followed by -single. So, the whole boot line will look something like: LILO boot: linux -single Once in single user mode, just go to /root and make the fix. Alternatively, if you still have the Debian rescue/install disk, stick that in there, and boot. Once the install starts, you can use alt-f2 to switch to the virtual console which runs ash (a striped down shell). From there, mount the appropriate file system in /mnt or something and modify the file. Then remove the install/rescue disk, stick in your own, and you should be fine. -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: staroffice libc5
This is odd...the level of civility on this list is astounding. On any other list I've been on, if someone were to write this, they would immediately be ``punished''. this is almost disconcerting...=) Mike wrote (Thu, 11 Dec 1997 15:49:27 -0600 ): |take me off this list! | Mike, and anyone else who would want to unsubscribe from this cool list: |-- |TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to |[EMAIL PROTECTED] . |Trouble? e-mail to [EMAIL PROTECTED] . | This message is included in every mail which comes to you from the list...including the ones to which you have replied. Just follow those directions, and if they don't work, send mail to the administrator email, which is also specified in the message. In the future, you would do well to keep mail that you get when you first sign on to any mailing list. it usually has information about how to unsubscribe. I think you'll find nearly every other mailing list in the known world to be less forgiving that this one... I hope I haven't violated some unspoken convention for this list... I'm not usually the one that ends up doing this... -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: alt-F7 reassignment
Aaron Brick wrote (Wed, 10 Dec 1997 10:21:29 -0500 (EST) ): | |How can I set the ttys corresponding to ralt-Fx? /etc/inittab says nothing |about which alt key is in use. | The point is you don't have to. By simply duplicating the getty/agetty lines in /etc/inittab, new virtual consoles will be associated with the lalt-fX first, and when those are exhausted, with the ralt-fX keystrokes. So, with 12 virtual consoles, lalt-f1 through lalt-f12 would correspond to your 12 virtual consoles, and ralt-f1 would get you to your x server. -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Need help in X Windows installation
Sten Anderson wrote (10 Dec 1997 16:32:12 +0100 ): |Jens B. Jorgensen [EMAIL PROTECTED] writes: | | These messages do not indicate a fatal error, only that you haven't | installed these X extensions. If you're using your own .xinitrc, | are you sure that the last command uses 'exec' rather than just | running a program? This would cause the X server to immediately exit. | |It is NOT necessary for the last command to be started with exec. In |fact, exec should only be used on the window manager, and only if that |is the last command in .xinitrc. The problem is more likely the use of |'s. Every program started in .xinitrc should have appended EXCEPT |the window manager. | Is there something wrong with using exec on an xterm, rather than a window manager? I'm currently doing that which makes the duration of an x session depend on the xterm rather than the window manager (which is what I prefer), and I haven't run into any problems. Am I doing something inherently bad? -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: alt-F7 reassignment
Aaron- When X starts, it will always locate itself on the next available virtual terminal. Since the default is to have six, X gets tty7. If you end up having 12 tty's as login consoles, X will (presumably) get the 13th, ttyd. I have no idea how you would map alt-esc to switch back to X. However, you can cycle through the tty's by using alt-right_arrow and alt-left_arrow. So, the key sequence alt-f1/alt-left_arrow (or I suppose alt-f12/alt-right_arrow) would get you back to X. This begs the question: why would you need 12 login consoles? =) hope that helps... -alan Aaron Brick wrote (Wed, 10 Dec 1997 01:32:49 -0500 (EST) ): | |I see that in /etc/inittab I can set which ttys correspond to alt-fkeys; I |would like to set all 12 of them to correspond to terminals. However, I |see no reference in that file to the action associated with alt-F7, which |brings up the x window interface. how can i set this action to another |key (like alt-esc)? | -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
two random questions...
Hi everyone- I have a couple unrelated questions to which I'm hoping someone might have answers. First, I've installed APS for my HP DeskJet 682C. Everything works wonderfully for PS files, but now when I pipe plain text to the printer (via lpr), I get output that looks as if it's been rotated to landscape format and is printed with a smaller font than I'm used to seeing for plain text output. (It's almost as if it's been enscript'ed with the -2r arguments.) It's only very slightly annoying, but I'm just curious as to why this is happening. Secondly, I'm using the ctwm window manager, and I want to have the pointer warp to newly mapped windows. For example, if I'm using Netscape to view a secure document, Netscape puts up a modal warning dialog box. I'd like to be able to just hit the Enter key on my keyboard rather than having to navigate the mouse over to the dialog box. Does anyone know how to do this? Thanks! -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: two random questions...
Alan Su wrote (Sun, 07 Dec 1997 16:32:47 -0800 ): | |First, I've installed APS for my HP DeskJet 682C. Everything works |wonderfully for PS files, but now when I pipe plain text to the |printer (via lpr), I get output that looks as if it's been rotated to |landscape format and is printed with a smaller font than I'm used to |seeing for plain text output. (It's almost as if it's been |enscript'ed with the -2r arguments.) It's only very slightly |annoying, but I'm just curious as to why this is happening. Hmmm...sorry for responding to my own message, but I was able to finally dig through the /etc/apsfilterrc configuration file to figure out why this was happening. In case anyone using APS is wondering, take a look at the section of /etc/apsfilterrc where the FEATURE environment variable is defined. Again, sorry for the noise... -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
installing 1.3 and Intel EtherExpress Pro/10 (ISA)
Hi everyone- I hope this is the right list...I'm having a problem getting Debian Linux to recognize my Ethernet adapter, the Intel EtherExpress Pro/10 (the ISA version). So far, I've done the following: - disabled PnP for the card - set the interrupt to IRQ 10 - set the I/O address to 0x240 - installed Debian 1.3 - passed the following paramaters to the eepro driver (during installation): irq=10 io=0x240 During the driver installation, it claims to find the card. When I boot up the system, however, nothing seems to be working. ifconfig reports the correct interrupt and i/o address (as well as the ip address, netmask, and such), but there's no activity. Any suggestions are welcome...thanks! -alan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
