Re: suspend / hibernate
On Aug 8, 2008, at 12:40 AM, Dale wrote: 2008/8/6 Dale <[EMAIL PROTECTED]>: Hi all, I have had this problem for a little while now, when I come out of suspend / hibernation on on my Acer Aspire 5601AWLMi laptop I have very high system load. I am wondering if any one else has or noticed this problem. I have been informed it is a kernel problem and I am at a lost how to debug it. If am one can help with this problem it be very appreciated [.] After doing some research and testing I still have not been able to stop the initial system load when coming out of suspend / hibernate. The commands 'ps aux' and 'top' are to showing me what is causing the system load. from the proc(5) man page: /proc/loadavg The load average numbers give the number of jobs in the run queue (state R) or waiting for disk I/O (state D) averaged over 1, 5, and 15 minutes. They are the same as the load average numbers given by uptime(1) and other programs. While your machine is suspended, isn't every process waiting? So if your machine has 150 processes running and a normal load of 0, and the suspend/resume occupies the processor for 0.1 minutes = six seconds, the one-minute load average should jump up to 15? Even if no processes are waiting after the resume, the one-minute average won't go back down for ... a minute. Have you actually tried a benchmark, instead of looking at the system load? Something like $ for i in $(seq 10) ; do date ; time head -c 100 /dev/urandom | md5sum ; done 2>&1 | grep ^real $ sudo suspend ; for i in $(seq 10) ; do date ; time head -c 100 / dev/urandom | md5sum ; done 2>&1 | grep ^real This would give you an idea of how long soon after the suspend your machine gets back to its normal speed. The "load average" may not mean what it usually does here. I haven't followed the thread closely, apologies if this is duplicate information. Cheers, Rob -- Rob Mahurin Dept. of Physics & Astronomy University of Tennessee phone: 865 207 2594 Knoxville, TN 37996 email: [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Faster source of random numbers for blanking harddrives
On Thu, Oct 25, 2007 at 02:19:59PM +0800, Bob wrote:
> if you type
> dd if=/dev/urandom of=/dev/null count=10
> 10+0 records in
> 10+0 records out
> 5120 bytes (51 MB) copied, 14.0076 seconds, 3.7 MB/s
>
> whereas
> dd if=/dev/zero of=/dev/null count=10
> 10+0 records in
> 10+0 records out
> 5120 bytes (51 MB) copied, 0.0932449 seconds, 549 MB/s
>
> Is there a faster source or random or pseudo-random numbers because at
> 4MB/s it'll take a lng time to blank a 60GB drive let alone a 500GB job?
Most scientific libraries have fast, high-quality random number
generators. If you have libgsl0-dev installed, compile the attached
program with
$ make gsl_rng_info CFLAGS+="-lm -lgsl -lcblas -O6"
I get about 150 MB/s with the generator gfsr4:
$ time GSL_RNG_TYPE=gfsr4 ./gsl_random_spew spew | head -c 500m >
/dev/null
GSL_RNG_TYPE=gfsr4
generator type: gfsr4
seed = 0
first value = 2901276280
real 0m3.467s
user 0m2.612s
sys 0m0.672s
My computer has a different bus speed from yours, etc., so experiment.
I don't know whether this generator fills all 32 bits of each
generated number, or its period; those are in the GSL info manual
(from which I cribbed most of this program). If you care about those
sorts of details, you might follow another poster's advice and
consider an existing solution.
Nota bene: don't call this program without redirecting stdout :)
Rob
--
Rob Mahurin
Dept. of Physics & Astronomy
University of Tennessee phone: 865 207 2594
Knoxville, TN 37996email: [EMAIL PROTECTED]
#include
#include
gsl_rng * r; /* global generator */
#define CHUNKSIZE (1024*1024)
/*
* Calling this program with any arguments will spew infinite amounts
* of binary data to your terminal. Caveat emptor.
*/
int main (int argc, char* argv[])
{
const gsl_rng_type * T;
gsl_rng_env_setup();
T = gsl_rng_default;
r = gsl_rng_alloc (T);
fprintf (stderr,"generator type: %s\n", gsl_rng_name (r));
fprintf (stderr,"seed = %lu\n", gsl_rng_default_seed);
fprintf (stderr,"first value = %lu\n", gsl_rng_get (r));
if (argc==1) return 0;
else {
/* spew binary data to stdout */
unsigned long random_chunk[CHUNKSIZE];
int i =0;
while (1) {
for (i = 0; i
Re: HELP! can't become root
On Thu, Oct 04, 2007 at 08:06:56PM -0700, tom arnall wrote: > > > I got impatient with an aumix error and did 'chmod -R /dev' (and ctl-C'ed > > > out of it after ~3 min's.) now I can't become root. Some examples: [...] > i've been running the system for the last couple of days. the only visible > problems are: > > the font stinks for firefox. > i can't become root, not even in single-user mode, but i can 'sudo' all > i want. > > is it possible i could fix the two problems without having to reinstall? Can you "sudo chmod" to repair your permissions damage? I can send you an output from "find /dev -ls" if you don't have another machine to compare against. Rob -- Rob Mahurin Dept. of Physics & Astronomy University of Tennessee phone: 865 207 2594 Knoxville, TN 37996email: [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Random System Crashes
On Tue, Oct 02, 2007 at 10:33:27PM -0700, Kelly Clowers wrote: > > Lately (For the past month or so) I've been experiencing random system > > crashes. My system would just completely lock up and not respond to > > anything but a power-down (I've tried switching to TTY1, among other > > things, but nothing works). > > > > I don't even know where to start diagnosing this problem...I'm looking > > for help on where to start. ANY help would be appreciated. > > > > If you need any more information just ask, besides the fact that I'm > > running Debian Lenny with a teeny bit of Sid (amd64 as well). > > It may be this[1] xorg freezing problem that some people have been > having. Can do a remote login? If so does everything seem to work > alright in the new session? > > [1]http://lists.debian.org/debian-user/2007/09/msg01265.html I've been having this problem since upgrading to etch. The crashes happen once or twice a month, which is often enough to be really irritating but too infrequent for me to know how to debug. Restarting X remotely does seem to work, if the crash happens while my (laptop) is on a network where I can ssh in. Rob -- Rob Mahurin Dept. of Physics & Astronomy University of Tennessee phone: 865 207 2594 Knoxville, TN 37996email: [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Tool for document management
On Wed, Sep 26, 2007 at 10:39:55AM -0700, David Brodbeck wrote: > Looking at my copy of 'The Not So Short Introduction To LaTeX,' it's > not clear to me what document class I'd use [for a letter]. For some reason that book omits the LaTeX "letter" class. -- Rob Mahurin Dept. of Physics & Astronomy University of Tennessee phone: 865 207 2594 Knoxville, TN 37996email: [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: OT: Choice of OOo and LaTeX (Was: Tool for document management)
On Tue, Sep 25, 2007 at 10:11:31PM -0700, Steve Lamb wrote:
> Rob Mahurin wrote:
> > I know you've settled on OOo, but it's worth pointing out that TeX is
> > a simple language if you're writing a simple document. In particular
> > you are already writing valid plain TeX in your email. Copy the above
> > (without the >'s) into file.txt; change /'thinking'/ to {\it thinking}
> > and "saying" to ``saying''; type "pdftex file.txt" and "\end".
> > file.pdf looks like http://sns.phys.utk.edu/~mahurin/du/09-25.pdf,
> > which I think is what you're after.
>
> Uh, no. It's more than that. You're forgetting loading in the templates
> and the entire structure.
Sorry I wasn't clearer. I made the output linked above using /plain/ TeX;
the only \command was the italics. LaTeX is a set of templates and macros
for typesetting structured documents with TeX, which it sounds like you
don't need. When I was writing MLA-formatted papers as an undergraduate I
used plain TeX like this and was pretty happy with it.
Your other complaints, though, are all perfectly reasonable. It sounds
like you want to write your fiction using a word processor, not a
typesetting language. Great --- that's why the word processor was
invented, after all.
Let me see if I remember what you want:
1. revision control, including
- resurrect erased text
- merge changes from two computers
2. shallow learning curve, so you can focus on the writing
3. export to .doc that preserves italicization.
You're concerned (I think) about not being able to merge changes in
OpenOffice's data files using revision control, because those files
aren't straightforward text. Someone else mentioned Abiword, which
saves uncompressed XML; but there's metadata in there too, which might
not merge correctly. It looks like Abisource offering revision control
for collaborative writing, http://collaborate.abisource.com/faq/, but
that's probably not what you want either. These options give you #2 and
#3, maybe #1, or maybe a broken document after a certain level of
complexity is reached.
Many of the replies have been about TeX, its macro packages, etc. You
complain that gives you #1 at the expense of #2 and #3.
You mentioned you're not afraid of programming, so here's an idea.
You could just write in plain text, and use /italics/ the same you
have on this list. You said the publishers you've spoken to accept
plain text; that additional markup is easy enough to read.
If you /must/ send someone a .doc, you could write a Word macro (or a
macro in a program that produces Word files) to match and italicize text.
For that matter, a three-line perl (or whatever) script could
1. escape TeX's special characters, $%&#\{}^_~
2. replace / with "\it " (italicize) or "\rm " (roman) in alternation
3. run pdfTeX on the output
giving you something nice to print out. You should call the converter
SLIPTT, Steve Lamb's Italicized Plain Text Typesetter.
Don't want to print? Publisher can read plain text? You're all set.
The fact is that any document formatting specification is going to be in
SOME language, whether that language is embedded in the file format by
the word processor, marked up by the author, or whatever. Most of the
open-source revision control systems are for marked-up text, or code,
where the author can tell if the merged text is right or not. File
formats that warn "don't change this file manually," as AbiWord and
OpenOffice do, probably require specialized version control software
that's aware of those formats. The conclusion of this thread seems to
be that debian users are using source-controlled LaTeX markup, and that
there's not yet a good solution for source-controlled word processor
output. I'll be interested to know what you decide to do.
Good luck with your writing.
Rob
--
Rob Mahurin
Dept. of Physics & Astronomy
University of Tennessee phone: 865 207 2594
Knoxville, TN 37996email: [EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: OT: Choice of OOo and LaTeX (Was: Tool for document management)
On Tue, Sep 25, 2007 at 05:27:02PM -0700, Steve Lamb wrote:
> Good thing that what I'm writing is not at all complex. The two most
> complex things are italics and indent-first-line.
[...]
> Am I writing a book? Yes.
>
> Am I writing a technical book? No!
>
> I am writing fiction. I have no in-line graphics, complex font changes
> for examples, silly little icons to denote special sections, massive
> indention or the like. This is strictly line-after-line prose which
> could be done plain text except for the fact that I am making use of
> italics as a conscious style choice to reinforce when a character is
> /'thinking'/ something versus "saying" something.
I know you've settled on OOo, but it's worth pointing out that TeX is
a simple language if you're writing a simple document. In particular
you are already writing valid plain TeX in your email. Copy the above
(without the >'s) into file.txt; change /'thinking'/ to {\it thinking}
and "saying" to ``saying''; type "pdftex file.txt" and "\end".
file.pdf looks like http://sns.phys.utk.edu/~mahurin/du/09-25.pdf,
which I think is what you're after.
Good luck with your writing.
Rob
--
Rob Mahurin
Dept. of Physics & Astronomy
University of Tennessee phone: 865 207 2594
Knoxville, TN 37996email: [EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: mail messages with only html
On Fri, Sep 14, 2007 at 09:10:44AM +0100, Richard Lyons wrote: > On Fri, Sep 14, 2007 at 12:24:03AM +0100, Benjamin A'Lee wrote: > > In .muttrc: > > auto_view text/html > > > > In .mailcap: > > text/html; w3m -dump -T text/html '%s'; copiousoutput > > text/html; lynx -dump '%s'; copiousoutput # or something like that. or text/html; /usr/bin/elinks -force-html -dump -source '%s'; copiousoutput; description=HTML Text; nametemplate=%s.html which, among other things, gives you a numbered list of links. > Obviously a better method, yes. But when I reply, I will not get the > received text quoted ready for editing if the email is void. This isn't true for me: the quoted text sent to the editor for the reply includes all the auto_view'd attachments, including text, html, doc, pdf, ... with the [-- Autoview using ... --] header removed. Maybe I've tried a trick that you haven't, or you get more pathological messages than I do. Rob -- Rob Mahurin Dept. of Physics & Astronomy University of Tennessee phone: 865 207 2594 Knoxville, TN 37996email: [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
xfig: select many objects at once
Hi debian-user, I'm dealing with some auto-generated xfig images which have many thousands of overlapping objects all at the same depth. Were everything at a different depth, it would be easy to select things in groups by controlling what's shown and selecting in regions. How might I select many objects based on some other common attribute (like color)? The generating program (gnuplot) has some facility for setting depths, but it requires knowing some very Fortrannish 4-digit numbers that I'd rather not clutter my brain with. Xfig files are just text files full of lines of numbers; if there's no GUI way to do this I'm not afraid of, say, an awk script. Please CC to me, as I'm not subscribed to the list. Thanks! Rob -- Rob Mahurin Dept. of Physics & Astronomyphone: 865.974.8097 (sometimes) University of Tennessee fax:865.974.7843 Knoxville, TN 37996email: [EMAIL PROTECTED] -- Sorry, no fortune this time. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
woody on a Dell Inspiron 5100 laptop
Hello, I have installed Debian 3.0 on my Dell Inspiron 5100 laptop and have posted some notes at http://web.utk.edu/~rmahurin/inspiron5100. At this point essentially everything that I could want to use works, though it was something of a struggle, especially the networking hardware. If this is useful, great. If you have a question, CC to me as I'm no longer subscribed to this list. Rob -- The difference between the right word and the almost right word is the difference between lightning and the lightning bug. -- Mark Twain pgp0.pgp Description: PGP signature
Re: apt too big for its britches
On Mon, Jun 02, 2003 at 05:00:48PM +0200, VEGH Karoly wrote: > On Mon, Jun 02, 2003 at 05:00:57PM +0200, Rob Mahurin wrote: > > What, exactly, is running out of room here? I'm fine on memory and > > disk space. > > > > 10:43 $ sudo apt-get -u upgrade > > Reading Package Lists... Error! > > E: Dynamic MMap ran out of room > > E: Error occured while processing escputil (NewFileVer1) > > E: Problem with MergeList /var/lib/dpkg/status > > E: The package lists or status file could not be parsed or opened. > > AFAIK apt has a default that limits his memory usage. > fix for the problem: > > [EMAIL PROTECTED]:~$ cat /etc/apt/apt.conf.d/charlie > APT::Cache-Limit 8192000; > [EMAIL PROTECTED]:~$ > > (the filename can be just anything) > > hth That worked, you're my hero, have a nice day. Rob -- Meekness is uncommon patience in planning a worthwhile revenge. pgp0.pgp Description: PGP signature
apt too big for its britches
What, exactly, is running out of room here? I'm fine on memory and disk space. Please copy to me, I'm no longer on the list. Rob 10:43 $ sudo apt-get -u upgrade Reading Package Lists... Error! E: Dynamic MMap ran out of room E: Error occured while processing escputil (NewFileVer1) E: Problem with MergeList /var/lib/dpkg/status E: The package lists or status file could not be parsed or opened. 10:43 $ free -m total used free sharedbuffers cached Mem: 125122 3 38 35 46 -/+ buffers/cache: 40 84 Swap: 62 0 62 10:43 $ df -h FilesystemSize Used Avail Use% Mounted on /dev/hda5 57M 33M 21M 60% / /dev/hda7 484M 138M 321M 30% /var /dev/hdb6 1.3G 1.1G 211M 84% /usr /dev/hda1 484M 216M 243M 47% /usr/share/doc /dev/hda8 1.3G 1.1G 128M 90% /usr/local /dev/hdb5 486M 176M 285M 39% /usr/src /dev/hdb7 1.3G 1.1G 137M 90% /home /dev/hdb1 2.9G 2.0G 754M 74% /home/alphenglor/burn /dev/cdrom5.7M 5.7M 0 100% /cdrom -- HOST SYSTEM NOT RESPONDING, PROBABLY DOWN. DO YOU WANT TO WAIT? (Y/N) pgp0.pgp Description: PGP signature
Re: a quickie
On Wed, Sep 11, 2002 at 10:13:30PM -0600, Bob Proulx wrote: > Rob Mahurin <[EMAIL PROTECTED]> [2002-09-11 23:09:08 -0400]: > > Copy to me, I'm not on the list. Thanks, > > What's the perl command to convert a text file from DOS or unix > > newline format to the macintosh newline format? (CR->LF?) > > You know you could google for that information probably quicker. But > I am a sucker for those types of questions... Does it have need to be > perl? These are off of the top of my head and there may be mistakes > here. YMMV, etc., etc. > > MAC to UNIX: > > tr "\015" "\012" < uxfile > macfile Yup, that was it. I actually used "\r" and "\n", which are easier to remember than the numbers and work fine. I asked the list rather than searching on google because when I had time to read debian-user I always learned a lot from off-topic posts like this. Perhaps not the most efficient use of the internet but the internet is hardly optimized for efficiency. Thanks again, Rob -- Garbage In, Gospel Out msg02093/pgp0.pgp Description: PGP signature
a quickie
Hi all, What's the perl command to convert a text file from DOS or unix newline format to the macintosh newline format? (CR->LF?) Copy to me, I'm not on the list. Thanks, Rob -- Dammit Jim, I'm an actor, not a doctor. msg01899/pgp0.pgp Description: PGP signature
Re: [Samba] strange problem printing to windows box
On Wed, Aug 28, 2002 at 03:00:19PM +0200, Kovar Jan wrote: > Connect the printer to Windows 98. > Install drivers. > Test printing. > Unplug the printer and connect it to your linux box. > Setup samba to see the printer from Windows 98 on your samba server. > Go to printer settings in Windows 98 and change the port from LPT to > \\linux\lp Kovar, thanks for the reply. I had been trying to add a second printer on the windowsbo x rather than change the location of the first. Everything seems to work now: I can now print from either machine. However, this arrangement makes it _much_ slower to print from the windows machine, which is where 90% of the jobs will come from. So I'd like to move the printer to the windows box but I don't know how to debug this problem where pages sent from linux via SMB are truncated near the end. This seems a less common problem than simply having trouble getting printing to work at all, so I will re-post it to the lists. List folks please Cc: me as I'm not subscribed to either of them. Thanks again, Rob -- Amnesia used to be my favorite word, but then I forgot it. msg00193/pgp0.pgp Description: PGP signature
Re: Segfaults in seemingly unrelated programs
On Sat, Feb 02, 2002 at 08:52:36PM -0700, Rick Macdonald wrote: > On Sat, 2 Feb 2002 [EMAIL PROTECTED] wrote: > > > Due to a faulty fan, one CPU overheated and brought the system down. On > > restart, fsck indicated that some filesystem corruption occured. > > > > On startup, gdm would not start. After entering my username in the > > console, the login prompt came back without giving me the opportunity to > > enter my password. The logical next step, booting in single user mode. > > > > In single user mode, it quickly appeared that a few programs segfault. > > I had similar symptoms once. Segfaults and apparently corrupted disk > files. It turned out to be a bad memory SIMM. Try running memtest86 for > awhile (10 minutes to an hour or more; depends on how much memory you have > and how fast the cpu is). > > After I replaced the memory, I reinstalled all packages "in-place" > (declining any config files) to refresh any files that may have been bad. > Made me feel better, at least. I had the same thought, though I'm not sure how an overheated CPU fan could damage the memory. It seems more likely that the libraries Jim mentions are corrupted, and the one program that works is somehow "lucky". A memtest couldn't hurt, though. This is a randomly generated .sig, I swear. Rob -- You will be called upon to help a friend in trouble.
Re: What is a good, small, web browser?
On Sat, Feb 02, 2002 at 06:20:18PM -0800, Eric G. Miller wrote: > On Sat, 2 Feb 2002 17:58:06 -0800, "Karsten M. Self" > wrote: > > [snip] > > It's not for older/slower boxen though. I'd recommend *NO LESS* than a > > PII-233, and think you'll be happier with a PIII-600+ CPU. For memory, > > 128 MiB minimum, 256 strongly recommended. Particularly under intensive > > use (I can easily get over 100 tabs/windows open) it sucks both CPU and > > memory. But it does good things with both. > > Galeon does fine on this old 266 with 64MB of RAM. It certainly does > better than Mozilla proper or Konqueror (KDE > 1.x is just too memory > hungry). So, while 128 is probably nicer, I'd hesitate to say it was > the minimum. I'm also running Galeon quite happily on a P-200, albeit with 128 MB of RAM. The menus are a bit sluggish to display, but the browser as a whole is much better/faster/cheaper than anything else I've used. Rob -- If you are honest because honesty is the best policy, your honesty is corrupt.
Re: Wanda swam across my desktop ??
On Wed, Jan 30, 2002 at 12:14:35PM +0100, Tony Crawford wrote: > [EMAIL PROTECTED] wrote (on 30 Jan 2002 at 13:44): > > > On Tue, Jan 29, 2002 at 11:01:27AM -0600, Dimitri Maziuk wrote: > > > * [EMAIL PROTECTED] ([EMAIL PROTECTED]) spake thusly: > > > > ... gnome code bloat ... > > Like the man said, you have the > > source, go and fix it. > > > > Or not use it :) > > How about an option in the .deb installation script saying > something like "Would you like Gnome to provide a humorous > surprise on your desktop now and then? [y]/n" Or in the GNOME Configurator, which is a more appropriate place for it. Bug 131381 is tagged as wontfix, though, so nothing will happen for a while. Rob -- I know not how I came into this, shall I call it a dying life or a living death? -- St. Augustine
Re: moving from potato to woody (or sid)
On Sun, Jan 27, 2002 at 11:48:13PM -0600, [EMAIL PROTECTED] wrote: > I just wanted to know, as a "normal" user (as in, not completely > new, but no guru either), how hard would it be to move to woody, or > sid? > > Is it not that difficult? giving the fact that i have a fair knowledge of > "moving" around in console, editing files, do some tweaking, etc... I did a potato->woody the week of the 10th, no major problems. I asked on #debian first and was told to do upgrade apt-utils and debconf first, then a dist-upgrade would work. I had to manually go through the list of packages because I ran out of space on /var, but didn't have any problems besides that. Rob -- We are what we pretend to be. -- Kurt Vonnegut, Jr.
Re: home network, cant ping box 2
On Sun, Jan 27, 2002 at 05:53:09AM -0800, Jeff wrote: > Shawn Lamson, 2002-Jan-27 05:53 -0800: > > You have to be kiddin' me; i spent $25 on a 35' cable, and stood there > > while they made it and tested it... i figured that the cable from NIC > > to aDSL modem must be crossover too, since the long (35 footer) > > performed the same function when i hooked up either PC to the DSL with > > it! Are you SURE that the PC -> aDSL modem uses a straight thru cable? > > > > Thx, > > I don't have DSL, rather a cable modem, and I need a standard > (straight-thru) cable to go from modem to computer. And, I also > know that you need a cross-over cable to from computer to > computer. > > I hope your $25 cable can still be put to good use. :-) He could buy a crimper and make it a 34'11'' crossover cable :) Rob -- Marriage is a romance in which the hero dies in the first chapter.
Re: run scp in background?
On Sat, Jan 26, 2002 at 01:28:54PM -0800, Stonelx wrote: > Hi, > I tried this command: > > scp /etc/file remotehost:/etc/file & > > but when scp asks for the password, I'm actually > back on the command line. (thus the scp command fails) > Is there anyway around this? I would love > to be able to run scp in the background. > I looked at the man page, and see a flag -B , > but I don't quite get how that works. > I basically want to remote shell into my server, > run a scp command that may take serveral hours, but > be able to let the scp take place while I exit the shell > and go for lunch. ssh has the -f option that does what you want, but that doesn't seem to be enabled in scp. See bug #131090. Rob -- Fuch's Warning: If you actually look like your passport photo, you aren't well enough to travel.
Re: IPmasquerading
On Sat, Jan 26, 2002 at 12:40:22AM +0100, Pieter De Troyer wrote: > I'm setting up a full nat for the first time. > went through the howto and executed the example firewall rules. > Alas... the script first checks the availability of the concerned > kernel-modules. They are found, but I get errors returned that the > resource is busy... > > how do I solve this? What's the exact error message? I get something like that when a module is already loaded, but not exactly ... 21:14 tty2 $ /sbin/lsmod Module Size Used by [...] ide-scsi7508 0 [...] 21:15 tty2 $ sudo /sbin/insmod ide-scsi Password: Using /lib/modules/2.2.19/scsi/ide-scsi.o insmod: a module named ide-scsi already exists 21:15 tty2 $ Rob -- Oh yeah. Forgot about those. Getting senile, I guess... -- Larry Wall in <[EMAIL PROTECTED]>
Re: Debian, FHS & /floppy
On Tue, Jan 22, 2002 at 02:04:20PM -0800, Mark Ferlatte wrote: > True. But passing commandline args and editing fstab is still annoying > behavior... ideally, if I had a new disk and reboot, _nothing_ should > change with respect to my old disks. In this case, wouldn't you just put the new disk at the end of the chain (sd[last] rather than sd[first])? Rob -- Technological progress has merely provided us with more efficient means for going backwards. -- Aldous Huxley
Re: Printing problems
On Thu, Jan 17, 2002 at 03:11:51AM -0800, Jeff wrote: > Rob Mahurin, 2002-Jan-16 08:40 -0500: > > On Tue, Jan 15, 2002 at 07:38:04AM -0800, Jeff wrote: > > > Error: /invalidfont in findfont > > > Operand stack: > > > --nostringval-- > > > basefontdict > > > > > > So, I believe I have a font issue now. If you have some ideas on > > > this, I'd really appreciate them. I'm pretty lost...still. [..] > > I'm having this same problem (with an operand stack of > > F0 Times-Roman Font Times- > > > > after an upgrade this weekend from stable to testing. It seems to > > happen only from Netscape. [..] > I'm still fighting with it. It happens printing from any > application for me. > > Are you using lprng? magicfilter? What's an application that > works and what is it's print command? > > Sorry to bombard you with questions...I appreciate you help. No problem ... I offered. You guessed my setup: lprng with magicfilter. All the programs I use just call lp. Here's a dpkg -l on lprng, magicfilter, gs, and all of their Depends: and Suggests: (I think) ii lprng 3.8.2-1lpr/lpd printer spooling system ii debconf1.0.22 Debian configuration management system ii libc6 2.2.4-7GNU C Library: Shared libraries and Timezone ii magicfilter1.2-49 automatic printer filter. ii gs 6.51-5 The Ghostscript Postscript interpreter un enscript(no description available) ii bzip2 1.0.1-14 A high-quality block-sorting file compressor ii gzip 1.3.2-3The GNU compression utility. ii xpdf 0.93-4 Portable Document Format viewer for X11 ii libpaperg 1.1.2 Library for handling paper characteristics ii libgimpprint1 4.1.99-b1-1library necessary to use gimp-print programs ii libpng21.0.12-3 PNG library - runtime ii xlibs 4.1.0-11 X Window System client libraries ii zlib1g 1.1.3-18 compression library - runtime ii gs-common 0.3.1 Common files for different Ghostscript relea ii gsfonts6.0-2 Fonts for the ghostscript interpreter ii psfontmgr 0.10.1 PostScript font manager -- part of Defoma, D ii gs-pdfencrypt 1.0.4-6Provides support to view encrypted PDFs with After installing enscript and upgrading ghostview and gs-common, printing from netscape works: ii ghostview 1.5-27 a PostScript viewer for X11 ii gs-common 0.3.2 Common files for different Ghostscript relea ii enscript 1.6.2-4Converts ASCII text to Postscript, HTML, RTF Before I believe the problem was with converting html to postscript; html2ps produced output that crashed gv, that sort of thing. I was able to print plain text and .pdfs (produced by pdflatex, the one that worked yesterday is at http://web.utk.edu/~rmahurin/skool/syllabus.pdf) at least; I'm not sure what else I tested. I'm going to blame gs-common, since that was the only real dependency involved. Hope that helps. Let me know if I can test anything else for you. Rob -- Write yourself a threatening letter and pen a defiant reply.
Re: Printing problems
On Tue, Jan 15, 2002 at 07:38:04AM -0800, Jeff wrote: > Error: /invalidfont in findfont > Operand stack: > --nostringval-- > basefontdict > > So, I believe I have a font issue now. If you have some ideas on > this, I'd really appreciate them. I'm pretty lost...still. Hi Jeff, I'm having this same problem (with an operand stack of F0 Times-Roman Font Times- after an upgrade this weekend from stable to testing. It seems to happen only from Netscape. Don't know that that helps much, but it was certainly refreshing to see the same error message on your message and on my printer :) Let me know if I can test anything for you. Rob -- Revenge is a meal best served cold.
Re: SendMail died on SIGSEGV signal.
On Sun, Jun 10, 2001 at 05:01:30AM -0700, Account for Debian group mail wrote: > > Our SendMail server has been stopping with a SIGSEGV signal (SMTP-MAIL > died on a signal 11). This is something that has never happened before. We > upgraded about 2 months ago to potato and the server is running 8.9.3-23 > SendMail with kernel 2.2.18pre21. > > Anyone have an idea on what is causing this problem? We really need to get > it solved. Segfaults in gcc are often blamed on faulty RAM; when one of my RAM chips failed about a year ago, I got segfaults out of ordinarily stable programs like grep and ps. Look on the web for the "sig11 faq", which explains how to check for bad memory. Hope that helps. Rob -- Drew's Law of Highway Biology: The first bug to hit a clean windshield lands directly in front of your eyes.
Re: program response sluggish compared to Win98
On Sat, Jun 09, 2001 at 06:07:25PM +0200, Alex Suzuki wrote: > On Sat, Jun 09, 2001 at 08:02:40AM -0400, Rob Mahurin wrote: > > You are wrong --- WP8 was natively written for Linux. It's WP9, aka > > WP2000, that's run under wine, and wow is it horrible. > > Does anybody know how the Corel Office Suite (WP2000) compares to > StarOffice in terms of performance & compatbility to M$-Word format? > I have StarOffice installed here, and hey, I mean I've got a > 750 MhZ Duron processor, a quick harddrive and 512 Megs of RAM, > and I still wouldn't recommend it! I have a Pentium 200, with 40 MB RAM. StarOffice is too slow to be useable. WordPerfect was slightly less sluggish, but so unstable as to be useless; it frequently crashed for no good reason. In addition, it didn't understand the way that my windowmanager did focus (focus-follows-mouse in fvwm), so bringing up a dialog box caused everything to hang for as much as a minute while the focus shifted rapidly from the main window to the dialog box and back. Entirely unsatisfactory. It's possible that you might have a better experience on your faster machine, but a friend with dual-processor box installed from my CD and called me the next day to ask, "Is it _supposed_ to work this badly?" Just my experience. Didn't get to try and Word documents. Rob -- Paranoid Club meeting this Friday. Now ... just try to find out where!
SOLVED (was: [OT] windows networking issue)
On Thu, Jun 07, 2001 at 07:40:45AM -0500, ktb wrote: > > On Wed, Jun 06, 2001 at 05:02:53PM -0700, Mike Egglestone wrote: > > > what about duplexing? Is your nic set for half or full? > > There may be another way but I usually nab a utility program from the > manufacturer and run a diagnostic on the card. I've mostly worked with > older 3Coms. The utility fits on a bootable windows floppy so it > doesn't matter which operating system your running. On Fri, Jun 08, 2001 at 09:38:08AM +1000, Ian Perry wrote: > You've probably looked at this already...but... [...] > Last resort, try turning off Plug and Pray (Play) and setting the IRQ for > the card manually in the bios and try again. These two tidbits solved my problem, which was a mis-detected Ethernet card. I got the diagnostic tool from the 3com site (the "etherdisk", or something; it's now two floppies worth of self-extracting archive), and noticed that one contained drivers. I removed the driver I had, rebooted to disable plug&pray, and windows said, "You have new hardware, but I don't know what it is." Directing it to the drivers extracted from the disk revealed two that were compatible with the card, one PnP and one "legacy"; I chose the second. Ping works. Putty works. IP Masq works. This is what I wanted. Thanks to all who replied. This is a distressingly appropriate randomly generated signature. Rob -- If you find a solution and become attached to it, the solution may become your next problem.
Re: program response sluggish compared to Win98
On Sat, Jun 09, 2001 at 01:55:21AM -0500, Rob VanFleet wrote: > On Sat, Jun 09, 2001 at 01:10:01AM -0500, rich wrote: > > Howdy all, > > > > I've had a problem for a while in which my potato sytem seems to perform > > relatively sluggishly - most noticably on Wordperfect. Example: While > > WP8 under W98, I can hold the "page down" key for 5 seconds, and when I > > let go, the scroll immediately stops. Doing the same under linux, > > however, would result in a minute-long scroll through several dozen > > pages after the key was let go. Is there a keyboard buffer that I can > > adjust or something, or does the problem lie elsewhere, or is linux WP8 > > just not as well-designed? > > AFAIK, (and I could be wrong) WP8 for linux is basically the Windows > Wordperfect emulated through Wine, so naturally it would run more > sluggish than a native application on either OS. You are wrong --- WP8 was natively written for Linux. It's WP9, aka WP2000, that's run under wine, and wow is it horrible. However, because your name has a good ring to it, I forgive you for being wrong. Rob -- I'm going to Boston to see my doctor. He's a very sick man. -- Fred Allen
Re: Audio CDs [solved]
On Thu, Jun 07, 2001 at 01:05:41PM -0700, David Nusinow wrote: > It depends on which environment you're using. > KDE: kmix > Gnome: gmix > Anything else: xmix > I don't know of any console based ones, although I know they're out there. aumix? Rob -- Do you like "TENDER VITTLES"?
Re: [OT] windows networking issue
On Thu, Jun 07, 2001 at 09:18:44PM -0400, D-Man wrote: > On Thu, Jun 07, 2001 at 10:09:10AM -0400, Rob Mahurin wrote: > | > | Here's a tcpdump | grep gow with the static IP, watching gow boot up. > | [tcpdump] > > Apparently I don't have tcpdump on my system. Packet sniffers are quite useful to have around. > | > If you want more specifics on how I configured the doze systems I can > | > turn one on tomorrow and see what it says. > | > | This would be wonderful, thank you. > > On the Windo~1 98 box I have : > right-click on "Network Neighborhood" and click "Properties" > (can get here from control panel also) > > Under the Configuration tab I have : > Client for Microsoft Networks > Microsoft Family Logon > 10/100M PCI Fast Ethernet Adapter > TCP/IP -> 10/100M PCI Fast Ethernet Adapter > File and printer sharing to Microsoft Networks I lack the "family logon" and the file sharing; the former I've never seen before and the latter I'll enable when ping works. My ethernet card is detected as 3com EtherLink III ISA (3c509b-TPO) in PnP mode There's also some "Infrared PnP Serial Port (*PNP0510)" that is apparently part of the motherboard; I can't hardware disable it and if I delete it it reinstalls itself. It has a "Fast Infrared Protocol". > Properties for "Client for Microsoft Networks" : > "Quick Logon" selected under "Network Logon options", nothing > else selected. Mine has an unchecked "logon to Windows NT domain" box and the radio button for "Network Logon Options" is set to "Logon and restore network connections." This is probably what's sending those netbios requests on bootup. > Properties for "10/100M PCI Fast Ethernet Adapter" : > Driver Type : > Enhanced Mode 32bit and 16bit NDIS driver > (only choice) I have two "real mode (16 bit)" drivers, one which is NDIS (using it doesn't change anything that I can tell), and one which is ODI, which I wasn't able to use because the system asked me for a Novell disk I don't have. > Bindings : > TCP/IP -> 10/100M PCI Fast Ethernet Adapter Similar. > Advanced : > Early Tx Threshold : 10 > Link Speed / Duplex Mode : Auto Mode > Network Address : Not Present My only property is "Maximum Transmits", which is set to "Not Present." I can make it an arbitrary integer. Don't know what it does. > Properties for "TCP/IP -> 10/100M PCI Fast Ethernet Adapter" : > IP Address : > Specify an IP address (radio button selected) > IP Address : 192.168.0.3 > Subnet Mask : 255.255.255.0 Same, but 10.0.0.3. (255.255.255.0 is a subnet of the broader 255.0.0.0 permitted on the 10... private networks). > WINS Configuration : > Disable WINS Resolution Same. > Gateway : > 192.168.0.1 (my Debian box) 10.0.0.2 (bravo) > DNS Configuration : > Enable DNS > Host : Compaq (yeah I know, real original names) > Doman : > DNS Server Search Order : > 129.21.3.17 (the server from school, my ISP) Host:gow Domain: study DNS Server: 10.0.0.2, which will run bind > Bindings : > Client for Microsoft Networks > File and printer sharing to Microsoft Networks Haven't installed the second one yet; it wouldn't work without TCP/IP. > Identification : > (this is stuff for samba only, doesn't really matter what > names you fill in) For the record, I have: Host: gow Workgroup: WORKGROUP Description: Changing this hostname to "dude" (the name of the cat who says "gow") broke "ping gow", but not "ping 10.0.0.3", and didn't fix "ping 10.0.0.1", which still says "Request timed out." Adding the three local machines to c:\windows\hosts restored "ping gow" and made "ping peon" fail with "Request timed out" rather than "Unknown host peon." > If you need any more info/detail just let me know. I'd really like to see a tcpdump from your (or any working) network; I'm still curious about that "icmp router solicitation", which, after a 15-second power outage that deleted my previous fscking response, resolves to 08:46:57.211735 gow.study > ALL-ROUTERS.MCAST.NET: icmp: router solicitation Using tcpdump -n, it's the same 224.0.0.2 that I was wondering about earlier. > HTH, Very much, thank you. I have to go to work now, but I'll play with this again tomorrow. Rob -- The problem with any unwritten law is that you don't know where to go to erase it. -- Glaser and Way
multiple copies of mailing list messages
Has anyone else been receiving multiple copies of some messages (like those attached)? A glance at the headers shows that murphy.debian.org sent this one four times to myrealbox.com. Rob -- If ignorance is bliss, why aren't there more happy people? --- Begin Message --- Hello patrick, Sunday, May 27, 2001, 2:13:22 PM, you wrote: pq> --- Arian Novruzi <[EMAIL PROTECTED]> wrote: >> Hi, >> I'm trying to configure my ppp in my new Debian 2.2 (well, I think >> these >> developers guys can make more easy the installation ;-) >> My modem is a lucent one and I use the appropriate package to install >> it. >> I'm using the same script which worked perfectly in RH6.2. >> Running "ppp-on" gives these messages: >> >> /usr/sbin/pppd: The remote system is required to authenticate itself >> /usr/sbin/pppd: but I couldn't find any secret (password) for it to >> use to >> do so >> /usr/sbin/pppd" (none of the avalaible passwords would let it use an >> IP >> address) >> >> Thanks for any suggestion how to establish my ppp connection. >> Arian. hi all, Under woody, I use apt-get installing imp and horde. After installing,I find there isn't horde database in mysql. So imp cann't run normally. My operation steps are: apt-get install apache mysql-server php3 phplib apt-get install imp horde Under potato,I installed imp successfully with the above steps. Why not successfully under woody? -- Best regards, fgmailto:[EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] --- End Message --- --- Begin Message --- Hello patrick, Sunday, May 27, 2001, 2:13:22 PM, you wrote: pq> --- Arian Novruzi <[EMAIL PROTECTED]> wrote: >> Hi, >> I'm trying to configure my ppp in my new Debian 2.2 (well, I think >> these >> developers guys can make more easy the installation ;-) >> My modem is a lucent one and I use the appropriate package to install >> it. >> I'm using the same script which worked perfectly in RH6.2. >> Running "ppp-on" gives these messages: >> >> /usr/sbin/pppd: The remote system is required to authenticate itself >> /usr/sbin/pppd: but I couldn't find any secret (password) for it to >> use to >> do so >> /usr/sbin/pppd" (none of the avalaible passwords would let it use an >> IP >> address) >> >> Thanks for any suggestion how to establish my ppp connection. >> Arian. hi all, Under woody, I use apt-get installing imp and horde. After installing,I find there isn't horde database in mysql. So imp cann't run normally. My operation steps are: apt-get install apache mysql-server php3 phplib apt-get install imp horde Under potato,I installed imp successfully with the above steps. Why not successfully under woody? -- Best regards, fgmailto:[EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] --- End Message --- --- Begin Message --- Hello patrick, Sunday, May 27, 2001, 2:13:22 PM, you wrote: pq> --- Arian Novruzi <[EMAIL PROTECTED]> wrote: >> Hi, >> I'm trying to configure my ppp in my new Debian 2.2 (well, I think >> these >> developers guys can make more easy the installation ;-) >> My modem is a lucent one and I use the appropriate package to install >> it. >> I'm using the same script which worked perfectly in RH6.2. >> Running "ppp-on" gives these messages: >> >> /usr/sbin/pppd: The remote system is required to authenticate itself >> /usr/sbin/pppd: but I couldn't find any secret (password) for it to >> use to >> do so >> /usr/sbin/pppd" (none of the avalaible passwords would let it use an >> IP >> address) >> >> Thanks for any suggestion how to establish my ppp connection. >> Arian. hi all, Under woody, I use apt-get installing imp and horde. After installing,I find there isn't horde database in mysql. So imp cann't run normally. My operation steps are: apt-get install apache mysql-server php3 phplib apt-get install imp horde Under potato,I installed imp successfully with the above steps. Why not successfully under woody? -- Best regards, fgmailto:[EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] --- End Message --- --- Begin Message --- Hello patrick, Sunday, May 27, 2001, 2:13:22 PM, you wrote: pq> --- Arian Novruzi <[EMAIL PROTECTED]> wrote: >> Hi, >> I'm trying to configure my ppp in my new Debian 2.2 (well, I think >> these >> developers guys can make more easy the installation ;-) >> My modem is a lucent one and I use the appropriate package to install >> it. >> I'm using the same script which worked perfectly in RH6.2. >> Running "ppp-on" gives these messages: >> >> /usr/sbin/pppd: The remote system is required to authenticate itself >> /usr/sbin/pppd: but I couldn'
[OT] windows networking issue
Hello, Sorry for the non-linux question, but I haven't found an answer elsewhere. I have a functioning network of two Debian boxes, bravo (10.0.0.2) and peon (10.0.0.1). bravo is capable of IP routing, forwarding, and masquerading. peon has a fully-functioning TCP/IP connection with bravo, and when bravo dials in to the internet, peon's traffic is forwarded appropriately. I am trying to add to this network my fiancee's Win98 machine, named gow, with IP 10.0.0.3. gow is physically connected to the network through the same (Netgear 4-port) hub as the two Debian boxes; plugging in its cable illuminates the "connected" light on the port. Assigning a static IP, however, left gow unable to ping or be pinged. I set up dhcpd on bravo, and told gow to use that, and something interesting happened: gow correctly recives its IP information, sends 6 or 9 netbios broadcasts, and then stops sending traffic over the ethernet. Here is a "tcpdump | grep gow" : 18:58:04.943737 bravo.study.bootps > gow.study.bootpc: xid:0x721ddc08 Y:gow.study S:bravo.study [|bootp] [tos 0x10] 18:58:04.944643 arp who-has gow.study tell gow.study 18:58:05.848933 gow.study > 224.0.0.2: icmp: router solicitation 18:58:06.618817 gow.study.netbios-ns > 10.0.0.255.netbios-ns: [five more of these over the next second] 19:00:51.832071 bravo.study.bootps > gow.study.bootpc: xid:0xcf05ee00 Y:gow.study S:bravo.study [|bootp] [tos 0x10] 19:00:51.833017 arp who-has gow.study tell gow.study 19:00:52.825376 gow.study > 224.0.0.2: icmp: router solicitation 19:00:53.574929 gow.study.netbios-ns > 10.0.0.255.netbios-ns: [five more of these over the next second] However, after the boot is completed, I'm not able to ping or be pinged; no packet is sent to the hub (neither the light flashes nor does tcpdump show anything), and the error message on the windows box is "Request timed out." Furthermore, attempting (via winipcfg) to renew the DHCP lease gives 'Error: DHCP Server Unavailable: Renewing Adapter "*"'. Releasing the correct information obtained on boot and renewing again gives the same error, after which --- get this --- the IP assigned is 169.254.136.191! To summarize: 1. peon and bravo, two Debian boxes, are fully and correctly connected to each other. 2. gow, a win98 box on the same physical hub, correctly reads DHCP on boot and then stops sending packets to the hub. I believe I have eliminated: 1. DNS, since it doesn't work with IP numbers, either 2. Masquerading, since it's the local network that's broken 3. Faulty equipment, since a) the 'connected' light comes on when the card is plugged in b) gow correctly obtains its internet configuration on boot c) putting bravo's NIC in gow (though not the other way) exhibited the same behavior. My only possible lead is the line from tcpdump that says: 19:03:24.154942 gow.study > 224.0.0.2: icmp: router solicitation I don't know where this is configured, if it is configurable, but my local network is 10.0.0.0 (with netmask 255.0.0.0). Any advice, FMs to R, diagnostics to run, programs to install, etc. are more than welcome. I have been tooling around with this for weeks and I just want it to work. Thank you for any help. Rob -- You can rent this space for only $5 a week.
Re: [OT] windows networking issue
On Wed, Jun 06, 2001 at 05:02:53PM -0700, Mike Egglestone wrote: > Hi Rob... > I would guess its a hardware issue > Either the nic in the win9x box is flaky... or the cable or something > Keep troubleshooting... do you have an extra nic or cable...? Hmmm ... I took the card from one of the working linux boxes (bravo) and put it in the windows box, and had similar if not identical behavior. Also, the windows box does correctly get its DHCP information on boot; it just mysteriously stops using the interface after that. I'll try the switcheroo again in a few days, hoping that something easier gets suggested; the box doing my routing/gatewaying is headless, which makes this sort of thing hard. > what about duplexing? Is your nic set for half or full? How do I check this (on both operating systems)? A quick grep of my HOWTO directory doesn't reveal any information. Thanks so much for the reply, Rob -- "World domination. Fast" (By Linus Torvalds)
Re: [OT] windows networking issue
On Thu, Jun 07, 2001 at 02:46:21AM +0200, Debian User Jean-Baptiste Note wrote: > just by chance ... > did you put the masquerading box as the default gateway for the win98 machine > ? I've tried it with and without, although never while the masquerading box was connected to the wider internet (meaning that the gateway itself had no default gateway at the time). But since "ping" on the windows box doesn't seem to send a signal to the ethernet hub (it doesn't light the Rx light, and tcpdump doesn't show anything), I'm not sure that this is it. Thanks, Rob -- In West Union, Ohio, No married man can go flying without his spouse along at any time, unless he has been married for more than 12 months.
Re: [OT] windows networking issue
On Wed, Jun 06, 2001 at 10:48:45PM -0400, D-Man wrote: > On Wed, Jun 06, 2001 at 07:37:07PM -0400, Rob Mahurin wrote: > | Sorry for the non-linux question, but I haven't found an answer > | elsewhere. > | > | I have a functioning network of two Debian boxes, bravo (10.0.0.2) and > > Two Debian boxes ... sounds like linux to me Well, they aren't giving me any trouble. :) > | I am trying to add to this network my fiancee's Win98 machine, named > > Ooh, time for some "fun" ;-). "Fun" is relative. See following double-quoted paragraph. > | Any advice, FMs to R, diagnostics to run, programs to install, > | etc. are more than welcome. I have been tooling around with this for > | weeks and I just want it to work. Thank you for any help. > > I have here a network with 2 Potato boxen, 1 Win98 and 1 Win95 > working. I setup one Debian box with ipmasq. All NICs are statically > configured (no DHCP). On the 'doze box I set one of the Debian boxes > as the default gateway. I set a static IP and no WINS. I have a > D-Link switch physically connecting the boxes. This was what I tried first and I'd be happy to make it work; I have no particular attachement to DHCP. > Can the 'doze box ping itself? By localhost or 127.0.0.1? By it's > external IP address? Automatically assigning an IP via DHCP: "ping localhost": succesfully pings 127.0.0.1 "ping gow": succesfully pings 10.0.0.3 "ping 10.0.0.3":ditto "ping peon":"Unknown host peon." (DNS is disabled anyway) "ping 10.0.0.1":"Request timed out." Statically configured IP: "ping localhost": succesfully pings 127.0.0.1 "ping gow": succesfully pings 10.0.0.3 "ping peon":"Unknown host peon." (DNS is disabled anyway) "ping 10.0.0.1":"Request timed out." Here's a tcpdump | grep gow with the static IP, watching gow boot up. 09:53:40.859140 arp who-has gow.study tell gow.study 09:53:41.632729 gow.study.netbios-ns > 10.0.0.255.netbios-ns: [2 more of these] 09:53:41.967426 gow.study > 224.0.0.2: icmp: router solicitation 09:53:42.381266 gow.study.netbios-ns > 10.0.0.255.netbios-ns: [five more of these] and after 9:53:43, nothing. Do your winboxen perform this router solicitation? Is it also on this crazy address? > The biggest issue I had, once I had ipmasq installed, was the 'doze > boxes not cooperating with Samba. Oh, that's not good to hear. Maybe I won't need to set that up. > If you want more specifics on how I configured the doze systems I can > turn one on tomorrow and see what it says. This would be wonderful, thank you. > On the Debian systems I > use /etc/hosts to set names for the machines. 10:07 pts/0 $ grep study /etc/hosts 10.0.0.1peon.study peon 10.0.0.2bravo.study bravo 10.0.0.3gow.study gow Thanks for the help. Rob -- Remember though that THERE IS NO GENERAL RULE FOR CONVERTING A LIST INTO A SCALAR. -- Larry Wall in the perl man page
Re: [OT] windows networking issue
On Wed, Jun 06, 2001 at 09:15:02PM -0600, John Galt wrote: > > Maybe he's got two HURD boxen? > Nope, it's a windows question. Rob -- Dr. Livingston? Dr. Livingston I. Presume?
Re: [users] My Dilemma w/ ssh
On Fri, Jun 01, 2001 at 11:47:05PM -0500, will trillich wrote: > it took me a while to figure out that "ssh" contains "sshd" > whereas, say, "telnet" is the client and "telnetd" is the > server. with "ssh" the package wonk put all pieces into the one > package, so you get both client and server in one swell foop. > > probably a good reason for it, tho i dunno what it is. I think it's because telnet is useful for non-telnet purposes, while ssh is only set up for ssh. For example, you can 06:54 $ telnet localhost 25 Trying 127.0.0.1... Connected to localhost. Escape character is '^]'. 220 peon ESMTP Exim 3.12 #1 Sat, 02 Jun 2001 06:54:18 -0400 HELO 501 syntactically invalid HELO argument(s) and, if you know SMTP, send yourself a message to debug your mail server, but ssh -p 25 localhost tries to authenticate itself and just sits there. This seems to me a much better reason to split client and server than "I want to telnet to other machines, but I don't want anyone to telnet to me." Rob -- Santa's elves are just a bunch of subordinate Clauses.
Re: Currency of packages
On Fri, Jun 01, 2001 at 09:32:53AM -0700, Jaye Inabnit ke6sls wrote: > > where stable rules. Maintenance then isn't 'tinker & reboot', Shouldn't this be 'tinker ; reboot' ? If you use an ampersand your tinkering will be backgrounded and get interrupted by the reboot process. Rob -- The goys have proven the following theorem... -- Physicist John von Neumann, at the start of a classroom lecture.
Re: abbreviations for non-native english speakers
On Fri, Jun 01, 2001 at 11:15:10PM -0500, will trillich wrote: > On Thu, May 31, 2001 at 05:55:37PM -0500, Andrew D Dixon wrote: > > will trillich wrote: > > > > > > BYCMUYJTCP > > > (but you can alaways make up your own, just to confuse people.) > > > > And you can alaways make up words too :0) > > and on a bad day, even leave a or two out. :( [ shoulda been bycAmuyOjtcp ] IRTYMDFJ (in response to your message dated first june), leaving a couple out doesn't hurt. Rob -- Only that in you which is me can hear what I'm saying. -- Baba Ram Dass
Re: apt-get --print-uris in NoLocking mode
On Thu, May 31, 2001 at 10:59:53PM -0400, Rob Mahurin wrote: > On Wed, May 30, 2001 at 03:42:23PM -0500, will trillich wrote: > > On Sun, May 20, 2001 at 01:56:23AM +0200, Santi B?jar wrote: > > >I try to get a list of all the .deb to be installed (with > > >--print-uris), but I have to do it like root. I thought that if I add > > > a "-o Debug::NoLocking=yes" would make it possible to run it like > > > user. But I get this: > > > > > > bash$ apt-get --print-uris -o Debug::NoLocking=yes upgrade > > > E: Unable to write to /var/cache/apt/ > > > E: The package lists or status file could not be parsed or opened. Hmmm, wait a minute: 10:37 $ apt-get --print-uris -o Debug::NoLocking=yes install proftpd Reading Package Lists... Done Building Dependency Tree... Done The following NEW packages will be installed: proftpd 0 packages upgraded, 1 newly installed, 0 to remove and 0 not upgraded. Need to get 357kB of archives. After unpacking 897kB will be used. 'ftp://ftp-linux.cc.gatech.edu/pub/linux/distributions/debian/dists/potato/main/binary-i386/net/proftpd_1.2.0pre12 10:38 $ apt-get --print-uris -o Debug::NoLocking=yes install proftpd | grep ^\' | cut -d\' -f 2 | xargs wget --10:42:43-- ftp://ftp-linux.cc.gatech.edu:21/pub/linux/distributions/debian/dists/potato/main/binary-i386/net/pb => `proftpd_1.2.0pre10-2.0potato1.deb' Connecting to ftp-linux.cc.gatech.edu:21... Why can I do this and Santi can't? (My /var/cache/apt doesn't conatain a lock file, but with NoLocking=yes, that shouldn't matter.) Rob -- No small art is it to sleep: it is necessary for that purpose to keep awake all day. -- Nietzsche
Re: Currency of packages
Hi Jenny W, Please set your mailer to wrap long lines at about 70 characters. On Thu, May 31, 2001 at 09:44:46AM -0700, jennyw wrote: > I started using Debian because folks suggested it was easy to > maintain (e.g. apt-get). I really like apt -- it certainly makes > installing and downloading things a lot easier than other > distributions. However ... I'm not (yet) a heavy Linux user, and I > find that a lot of times I want software that's more recent than the > stuff in the packages. This usually means that I end up downloading > stuff and compiling new software. > How do other people deal with this? Do you install the new software > you need manually and then when the package gets released install it > again with apt-get (probably not practical for frequently updated > software)? Do you just use apt-get by default and download and > compile when you need to (and then never use apt for those > packages)? I put deb-src lines to unstable in my sources.list: 10:02 $ grep ^deb-src /etc/apt/sources.list | grep -v cdrom deb-src ftp://ftp-linux.cc.gatech.edu/pub/linux/distributions/debian unstable main contrib non-free deb-src http://non-us.debian.org/debian-non-US unstable/non-US main contrib non-free and, if there's a newer version in sid that hasn't made it to potato yet, I can do user$ apt-get -b source package [get coffee here] root# dpkg --install package_version.deb Much easier than getting tarballs off the net, and you have all the advantages of the dpkg system. > Also, I'm currently using potato because it's labeled as > stable. Would I be making a bad mistake by moving to Woody to get > some of the more recent stuff? What about mixing stable and > testing/unstable distributions? Is this generally unwise? I get really sad when my computer gets b0rked, so I run stable. If the risk of occasional breakages doesn't bother you (and I hear they're pretty rare), run testing. > It just seems like it's hard to use Debian without downloading > stuff. Some of the software I'm installing seems like it would be > pretty common: > > Cyrus IMAP (2.x -- current version is 1.5.19; there > is no later version in testing or unstable) Then you should file a "new upstream version available" bug, if there's not one already. > Postfix (20010228 or > higher -- current version is 19991231; there is a later version in > testing) Then you can apt-get -b source as described, or just grab the newer .deb from http://packages.debian.org/postfix . Unstable seems to have 20010502. > I'd like to hear how Debian old-timers work with this. Thanks! Jen Hope this helps. Rob -- It is contrary to reasoning to say that there is a vacuum or space in which there is absolutely nothing. -- Descartes
Re: First Debian Install Question
On Wed, May 30, 2001 at 08:48:59PM -0400, Bill Witherspoon wrote: > Can anyone tell me how to re-run the X setup proggy that > you get through dselect? I think it's "XF86Setup", which is an absolutely stupid name. Rob -- hubub, hubub, HUBUB, hubub, hubub, hubub, HUBUB, hubub, hubub, hubub.
Re: apt-get --print-uris in NoLocking mode
On Wed, May 30, 2001 at 03:42:23PM -0500, will trillich wrote: > On Sun, May 20, 2001 at 01:56:23AM +0200, Santi B?jar wrote: > >I try to get a list of all the .deb to be installed (with > >--print-uris), but I have to do it like root. I thought that if I add > > a "-o Debug::NoLocking=yes" would make it possible to run it like > > user. But I get this: > > > > bash$ apt-get --print-uris -o Debug::NoLocking=yes upgrade > > E: Unable to write to /var/cache/apt/ > > E: The package lists or status file could not be parsed or opened. > > installing a package is something that'll affect anything from a > tiny pocket of a remote corner of a dusty drawer in the far > reaches of your system -- to upgrading your kernel. > > you wouldn't want gomer or floyd or opie to be able to do that, > now, would you? gotta be root to do root-level things. Ahh, but Santi's not trying to install, I don't think, but just get the list of uris (perhaps to pipe to wget); --print-uris is a glorified^Wenhanced --no-act. I don't much care if gomer knows my download queue. I think you should be able to run this command without root privileges. Imagine (guessing on the output format and syntax here): apt-get --print-uris upgrade | tr " " "\n" | grep ^http | \ xargs wget && dpkg -i *.deb && rm -rf *.deb versus apt-get --print-uris upgrade | tr " " "\n" | grep ^http | \ xargs wget && dpkg -i *.deb && rm -rf * .deb ^^^ Sure, I wouldn't do either of these things, but I can imagine reasons to do something similar, and I'm such a putz that I'd much rather have my script execute without any privileges. I don't see why you need to be root to get a URL listing. Rob -- Q: Heard about the who couldn't spell? A: He spent the night in a warehouse.
Re: The use of source cds.
On Wed, May 30, 2001 at 09:02:38PM -0600, Simmons-Davis wrote: > Dear Debian Group, > > I would like to know exactly what purpose the source cds serve and who would > need to use them. I will only be using Debian for things like word > processing, email, the internet, etcetera and wouldn't be doing any > development or performing any major alterations to the debian distribution. > My question then is do you think that I have any need for purchasing the > source cds along with the main cds? > > Thank you for your time, > Ry The license for most of the programs released with Debian (the GNU General Public License, or GPL) is designed to make sure that you have the freedom to distribute copies of free software (and charge for this service if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs; and that you know you can do these things. This is why there are source CDs. You don't need them if you're not going to use them, but the software licenses stipulate that Debian must make them available. Rob -- Diplomacy is about surviving until the next century. Politics is about surviving until Friday afternoon. -- Sir Humphrey Appleby
Re: small school: replacements for MS Word and Excel
On Wed, May 30, 2001 at 09:27:43PM -0400, Alan Shutko wrote: > "Steve R. Hastings" <[EMAIL PROTECTED]> writes: > > > Corel is selling the whole WordPerfect Office suite, which includes > > WordPerfect and Quattro Pro. > > Note that last I checked, there's no commitment to ongoing support of > this product, there was almost no support of WP8, and getting fixes [blah blah blah] > > In other words, if you want WPO2k/Linux, you'll be on your own. In addition, the WPO2k for linux is, in my experience, not very good. It's essentially the windows version and a WINE environment, and crashed so frequently that I could hardly use it. > -- > Alan Shutko <[EMAIL PROTECTED]> - In a variety of flavors! > Siskel and Ebert gave this tagline two thumbs up. The random .sig generator strikes again! Rob -- I put up my thumb... and it blotted out the planet Earth. -- Neil Armstrong
Re: Can't find ssh
On Mon, May 28, 2001 at 08:54:46PM -0600, ray p wrote: > Also I don't know > how hard or easy to bring the version of SSH from Woody to Potato > but that would give you SSH2. It's very easy, just add an unstable (for sid, or testing for woody) deb-src line as already mentioned in this thread and do apt-get -b source ssh Get a cup of coffee and it will be done; install the .deb file it creates. You may need one or two library packages; if they're not in stable either, you can install them the same way. Rob -- To write good code is a worthy challenge, and a source of civilized delight. -- stolen and paraphrased from William Safire
cd audio problems under 2.2.19?
Hello, Under kernel 2.2.16 I can play CD audio normally; however, under 2.2.19 I get 14:58 tty1 $ cdir unknown cd - 64:30 in 6 tracks 10:11.62 1 15:17.00 2 9:59.08 3 8:19.55 4 3:39.12 5 17:01.40 6 14:58 tty1 $ cdplay sr0: CDROM (ioctl) reports ILLEGAL REQUEST. sr0: CDROM (ioctl) reports ILLEGAL REQUEST. cdplay: ioctl cdromplaytrkind 14:58 tty1 $ cdinfo no-status and nothing happens. Is this a known issue or have I done something dumb in my compile? The kernel is stock 2.2.19, patched (without errors) from the working 2.2.16 source with the kernel.org patches and configured with 'make oldconfig'. I did not use the Debian kernel packages. Thanks for any help. Rob -- Let me put it this way: today is going to be a learning experience.
Re: downgrading with apt
On Sun, May 27, 2001 at 06:50:50PM -0500, John Patton wrote: > Is there a way to force apt-get to install downgraded > packages? One or two packages are no problem... using dpkg > works just fine with them. But how could you downgrade > your system from testing to stable, for example? Is it > even possible? AFAIK, apt wont even download packages if > it thinks that they are older than what's registered. If I > could at least get it to do that, then I could have dpkg > install them over the newer versions. The apt in potato won't do this, but the apt in woody can be coaxed into it. Rob -- Are you still an ALCOHOLIC?
Re: I've been getting scanned...
On Sat, May 26, 2001 at 02:39:46PM -0400, Paul Wright wrote: > Hi all, > > Someone's been port-scanning me, checking only some high ports. Here are > my relevant log entries: > > > May 26 13:39:30 j001 ippl: port 37397 connection attempt from 216.136.179.238 > May 26 13:43:03 j001 ippl: port 37404 connection attempt from 216.136.179.238 > May 26 13:43:06 j001 ippl: port 37404 connection attempt from 216.136.179.238 > May 26 13:45:55 j001 ippl: port 37406 connection attempt from 216.136.179.238 > May 26 13:45:58 j001 ippl: port 37406 connection attempt from 216.136.179.238 > May 26 13:47:10 j001 ippl: port 37408 connection attempt from 216.136.179.238 > May 26 13:49:30 j001 ippl: port 37412 connection attempt from 216.136.179.238 > > Does anyone know what they may be looking for in that range? Looks to me like an FTP site you connected to, but the port numbers are a little high. > Does anyone know of a good reference for info (vulnerabilities sorted by > port, service, etc)? Add a deb-src line to unstable to your sources.list: 09:22 tty2 $ tail -2 /etc/apt/sources.list deb-src http://non-us.debian.org/debian-non-US unstable/non-US main contrib non-free and do apt-get -b source portsentry. Install the .deb it produces. You might also want logcheck. > Thanks in advance for any help / advice. Hope this helps. Good luck in your job search. Rob -- Be nice to people on the way up, because you'll meet them on your way down. -- Wilson Mizner
Re: Firing on all CPUs
On Mon, May 14, 2001 at 01:27:03PM -0700, Osamu Aoki wrote: > I can not tell program name by memory but there are one cpu load program > in X (in potato) which display SMP load properly. It's the one with all > information shown as horizontal bar graph with red/green/yellow for each > cpu load/memory/... something. Go look for it in X memu. It's xosview. Rob -- "The geeks shall inherit the earth." -- Karl Lehenbauer
Re: apt-get offline
On Wed, May 16, 2001 at 10:34:20AM +0200, Joerg Johannes wrote: > Sorry, I did not express myself clear enough. This sed thing changes > every line in the "wget-script" script. This, I did easily with emacs > (while sed is much more elegant). My problem is: I do not want to do > hundreds of > mv package1_1x3a.deb package1_1%3a.deb > mv package2_1x3a.deb package2_1%3a.deb and so on > > Any idea how to do this automatically? Use mmv. Rob -- I tell ya, I was an ugly kid. I was so ugly that my dad kept the kid's picture that came with the wallet he bought. -- Rodney Dangerfield
Re: Emacs sends mail, but doesn't show-up in exim.
On Mon, May 07, 2001 at 01:58:45PM -0700, Robert Cymbala wrote: > What are the "D"s next to some addresses but not others in this mailq > output: > 0m 2.3K 14wrw8-0003EP-00 <[EMAIL PROTECTED]> > D [EMAIL PROTECTED] > [EMAIL PROTECTED] > [...] > D [EMAIL PROTECTED] > [...] > > In my last posting, Hot49erfan was a recipient that would get into > exim's queue, then disappear. It has a "D". > > What's D? The exim manpage, in the section about the -bp option, says, The recipients of the message (taken from the enve- lope, not the headers) are displayed on subsequent lines. Those addresses to which the message has already been delivered are marked with the letter D. If an original address gets expanded into sev- eral addresses via an alias or forward file, the original is displayed with a 'D' only when deliver- ies for all of its child addresses are completed. I once had an interrupted exim delivery like the one you described freeze delivery on all the messages to certain addresses. Quite mysterious. I think I solved it by removing the half-delivered message with exim -Mrm. On Sat, May 12, 2001 at 01:53:11PM -0700, Robert Cymbala wrote: > > I honestly think that they are disappearing. Have made the trip from > LA to NYC, and right now I'm at my friend's place. Just now I > plugged-in my laptop. The D-Link card is _not_ connected to the phone > line. Command "mailq" showed nothing in queue. As non-root user, I > typed 'echo "TEST" | mail [EMAIL PROTECTED]' and did not receive any > error messages. Immediately, I switched to tty1 and did "mailq" as > root. It reported: > 0m 312 14yfxa-7u-00 <[EMAIL PROTECTED]> > [EMAIL PROTECTED] > Then, about two seconds later, I repeated "mailq" and it didn't report > anything! It disappeared, rather than was sent, since the only wire > going into laptop is the power cord. Note that there's no " D " to > the left of address; in another posting > ( http://lists.debian.org/debian-user-0105/msg01047.html ) there was a > " D " next to that particular email address. Yes, only having one wire does seem to indicate that the message has not successfully been delivered. The reason there's no "D" in that mailq output is that, at the time you ran mailq, the message had not been delivered to that address. To see a "D", you would send a single message to multiple recipients; the partially delivered message would show "D"s indicating which addresses had already been processed and which were still waiting to be delivered. > >If you get another reproducable one, try this: > > > >echo "don't go away" | [EMAIL PROTECTED] ; > >tail -f /var/spool/exim/msglog/$(mailq |head -1 | cut -b11-26 ) ; > > > >which will get the first message ID out of mailq and follow that > >logfile. > > It's reproducable. I sent a message to the vanishing address > (N.B. with "mail" before the vanishing address), followed immediately > by said tail command. Within one second of sending email, that tail > command gave: > $ tail -f /var/spool/exim/msglog/$(mailq |head -1 | cut -b11-26) > tail: /var/spool/exim/msglog/14ygCG-AN-00: No such file or directory > tail: no files remaining > After a second or two, the same command gave: > $ tail -f /var/spool/exim/msglog/$(mailq |head -1 | cut -b11-26) > tail: /var/spool/exim/msglog/: Is a directory > tail: /var/spool/exim/msglog/: cannot follow end of non-regular file > tail: `/var/spool/exim/msglog/' has been replaced with a non-regular file; > cannot follow end of non-regular file > tail: no files remaining > > Woe is me... I don't know what to do next! to get it to work again. > I'm also coming to grips with the fact that, historically, I don't > know which messages were sent and which messages disappeared (don't > know when this 'behavior' began). OK, here's what that means. "mailq", as you know, outputs a list of messages waiting to be delivered: 11:22 # mailq 14m 2.7K 14zJxH-0006OB-00 <[EMAIL PROTECTED]> [EMAIL PROTECTED] 14m 5.8K 14zJxi-0006OB-00 <[EMAIL PROTECTED]> [EMAIL PROTECTED] 11:22 # "head -1" keeps only the first line of the standard input and throws the rest away: 11:23 # mailq | head -1 16m 2.7K 14zJxH-0006OB-00 <[EMAIL PROTECTED]> 11:24 # "cut" selects part of the line: 11:24 # mailq | head -1 | cut -b11-26 14zJxH-0006OB-00 11:25 # The bash idiom $(cmd) substitutes the output of "cmd" at that point in the command line: 11:25 # msgid=$(mailq | head -1 | cut -b11-26) ; echo $msgid 14zJxH-0006OB-00 11:27 # ls 1* ; touch $(mailq | head -1 | cut -b11-26) ; ls 1* ls: 1*: No such file or directory -rw-rw-r--1 root root0 May 14 11:27 14zJxH-0006OB-00 11:27 # So "tail -f /var/spool/exim/msglog/$(mailq |head -1 | cut -b11-26)" means, "follow the end of the file /var/spool/exim/ms
Re: GDM Broken: the fix
Hello, On Wed, May 09, 2001 at 06:14:55PM +0100, John Mylchreest wrote: > This communication is private and confidential and > is intended solely for the use of the named > addressee(s). Access, copying or re-use of the > information in it by any other person is not > authorised by OCRA. If you are not the intended > recipient please notify the sender immediately by > return e-mail to [EMAIL PROTECTED] When replying to I recieved this email even though I was not in the recipient list. (So did everyone else subscribed to debian-user.) Rob -- where am I and what am I doing in this handbasket?
Re: X question
On Tue, May 08, 2001 at 08:50:53PM -0700, Steve Cooper wrote: > The only issue I have with the X behavior is that I can't > figure out to efficiently perform a replacement paste. E.g. > If you want to paste a URL into a browser in Windoze you'd > select the current URL and paste over it. > > In X once you select the text you want to replace it > becomes the new copied text. I had some success using > gclipper, because it allowed me to access a stack of copied > text items. Is there a better way? I paste the new text at the beginning of the line, then hit Ctrl-K to erase the old text. Rob -- Backed up the system lately?
Re: Setiathome Packet Caching
On Tue, May 08, 2001 at 12:07:09PM -0500, Benjamin Pharr wrote: > I'm looking for a program to cache Setiathome packets, both incoming and > outgoing. Does anyone know of any such program for Linux? If so, is it > packaged for Debian? Thanks! There are some ways to do this on the Setiathome site; look for "using setiathome with a seldom-connected computer." Rob -- "If a computer can't directly address all the RAM you can use, it's just a toy." -- anonymous comp.sys.amiga posting, non-sequitir
Re: [users] Re: vim tip-o-rama
On Mon, May 07, 2001 at 06:39:30PM -0400, Brian Nelson wrote: > Well, damn, let's just throw them all out there: > > Eight Megabytes And Constantly Swapping [...] > Emacs Makers Are Crazy Sickos > > Did I miss any? Gracious, no: u've eliminated most all conceivable selections. Rob -- Old programmers never die, they just become managers.
Re: Installing Debian over a Suse system
On Mon, May 07, 2001 at 11:33:22PM +0200, [EMAIL PROTECTED] wrote: > I have installed Suse 7.0, but want to shift to Debian 2.2. I have read > installation manuals and FAQs regarding installing Debian, but have not > found anything on this item. There was a guy a couple years ago who upgraded a Red Hat system to Debian without rebooting it (!). He wrote down what he did and put it on a web page somewhere; maybe that will give you some ideas. I think the simplest way to change distributions will be to (a) back up /home and all of your tweaks, (b) do a clean install, formatting the drive and everything, and (c) restore your /home and put your tweaked software in /usr/local. Hope that helps. Rob -- If the meanings of "true" and "false" were switched, then this sentence would not be false.
Re: getting procmail to pipe to a gzipped file
On Tue, May 08, 2001 at 03:05:02PM +1000, [EMAIL PROTECTED] wrote: > I have these variables set at the top of my .procmailrc > PATH=/usr/bin:/usr/local/bin > MAILDIR=$HOME/mail # all mailboxes are in mail/ > LOGFILE=/dev/null > SHELL=/bin/sh I have /bin in my path. 02:47 $ which date /bin/date 02:48 $ which gzip /bin/gzip This is your problem. Rob -- You're not an alcoholic unless you go to the meetings.
Re: getting procmail to pipe to a gzipped file
On Tue, May 08, 2001 at 01:25:37PM +1000, [EMAIL PROTECTED] wrote: > On Mon, May 07, 2001 at 10:20:51PM -0400, Rob Mahurin wrote: > > Try "| gzip -c >> testing.gz". > > That still seems to not work when I sent a mail with Subject: gzip > And I know the condition works since if I remove the gzip pipe and just > place a filename there then it does append to it OK. Hmmm ... worked for me here, but I'm not sure what your definition of "working" is. When I used that syntax I got a file called Mail/testing.gz, which gunzip transformed into a normal-looking mbox file. mutt, however, said (rightly) that the compressed file was not a mailbox. For what it's worth, I have procmail v3.13.1 1999/04/05 and Mutt 1.2.5i. > > How will you read mail from the compressed box? > > I was going to use zless. That should work shouldn't it? Well, only as well as "less /var/mail/$USER". That doesn't seem to me like a very effective way to read a mailing list archive with thousands of threaded messages; I have trouble when I can see all the headers. I can't think of a better way, though. Chalk it up as a wishlist bug for mutt 2.0. Rob -- Being owned by someone used to be called slavery -- now it's called commitment.
Re: Question for gurus of mutt / scripting
On Tue, May 08, 2001 at 11:54:19AM +1000, [EMAIL PROTECTED] wrote: > It would be really nice to be able to keep the debian-user list a little > smaller and up to date on my system. How could I make mutt move old > threads that have not had a reply in say 5 days into a gzipped file? > This way I could have an archive of everything but my debian-user file > will remain up to date. > > Maybe it would be easier to write a script to do this? I am sure most > of you on this list have the same problem with information overload > especially on active lists. How do you guys handle this? I've halfway figured this out, but not automated it yet. I have the following lines in a file: 22:08 $ cat ~/muttcommands :push "D~r>2w\nU~f debian-user-request\n*" :set wrap_search=no :push /~D\n^U :set wrap_search=yes 22:11 $ To mark old messages for deletion, I highlight the first two in one xterm, and paste into mutt. To undelete the recent threads, I highlight the third line and middle-click until I'm back at my subscription message from way far back ago. I haven't figured out how to automate this because I don't know if it's possible to get mutt to conditionally undelete a thread. Hope that helps. Rob -- Shit Happens.
Re: getting procmail to pipe to a gzipped file
On Tue, May 08, 2001 at 11:37:19AM +1000, [EMAIL PROTECTED] wrote: > I thought I had this worked out, but it doesn't seem to work quite how > it should. Bascially I want to have mail with a certain subject piped > into a .gz file in my ~/mail directory. > > Here is what I had in .procmailrc: [...] > # testing of gzip pipe > :0 bw: > * ^Subject:.*gzip > | gzip >>testing.gz 22:08 $ gzip --help 2>&1 | grep stdout -c --stdout write on standard output, keep original files unchanged Try "| gzip -c >> testing.gz". How will you read mail from the compressed box? Rob -- "I'd love to go out with you, but I'm doing door-to-door collecting for static cling."
Re: Emacs sends mail, but doesn't show-up in exim.
On Mon, May 07, 2001 at 12:01:02PM -0700, Robert Cymbala wrote: > > I'm still seeing a problem. Have eliminated Emacs, by sending test > messages with: > $ echo "TEST" | mail [EMAIL PROTECTED] > > This time I've noticed that an email message will appear in mailq > output immediately after using mail, AND some messages just disappear > for some reason. > > Here's a test script: [snip] > > Here is the report: [one message evaporates] > > Where did message 2 disappear to? I've run your script as a normal user and did not see this behavior: all messages were visible both in the $RPT file and in the output of mailq. > This is what's under /var/spool/exim: This is the same as what I have, except that my undelivered mail is still in exim/input. Are these messages disappearing, or are they being delivered? > Reading the man page, it says that per-message log files are > generated, and that per-message log files are deleted after message is > sent. I've seen such messages before. Is there a log of messages > that have been successfully sent, or a place that tracks what is > happening to the Hot49erfan message after it is in exim's queue? This I don't know. > The problem doesn't seem to follow particular email addresses. The > other day, email to sam was disappearing. Today email to sam stays in > queue (waiting for PPP). Today email to Hot49erfan is disappearing > (reproducibly, at that). If you get another reproducable one, try this: echo "don't go away" | [EMAIL PROTECTED] ; tail -f /var/spool/exim/msglog/$(mailq |head -1 | cut -b11-26 ) ; which will get the first message ID out of mailq and follow that logfile. > Thing is: I'm driving to N.Y. and will be on the road for a few days. > But I wanted to post more information before leaving. Thanks again! Enjoy your trip. Rob -- If it's working, the diagnostics say it's fine. If it's not working, the diagnostics say it's fine. - A proposed addition to rules for realtime programming
Re: Cloning your Debian system
Hello, Viktor. I'm a little mystified as to why you CC'd this to me, but here goes. On Mon, May 07, 2001 at 10:00:54PM +0100, Viktor Lakics wrote: > I got a crazy idea: if you want to make 100 % sure that you do not > brake your production system, why don't you have an exact copy of > that system before you do an upgrade (or just use that system for > trying out things, which you cannot afford on your system). > > I got a spare 2 Gig partition on my system, and I want to give it a > try. I just do not know how to start this? O.K. I could probably do > it with hard disk image program, but then what configuration files > do I have to edit (exept /etc/fstab) after my root partition moves > from /dev/hda8 to /dev/hda7... > > Anyone have done it already? All comments and possible ways to do it > are welcome! I would you rsync(1) to copy the relevant parts of the system over (say, to $HOME/system2) and chroot(8) to switch to the subsystem. Rob -- Comparing information and knowledge is like asking whether the fatness of a pig is more or less green than the designated hitter rule." -- David Guaspari
Re: install screensaver with setuid root
On Mon, May 07, 2001 at 04:15:36PM +0200, Alex Suzuki wrote: > On Mon, May 07, 2001 at 04:12:51AM -0400, Rob Mahurin wrote: > > You should probably file a bug on the package, if it's not useable > > without introducing a security risk. Perhaps someone more familiar > > with the code could seperate out the part that needs special > > privileges into a seperate setuid program. > > The problem is that the ping sensor is actually a special feature, you > can use the screensaver if it's not setuid root, you just can't access > that feature. This is the purpose of "wishlist" bugs. See bug#96677. Rob -- Kent's Heuristic: Look for it first where you'd most like to find it.
Re: Gnome Packages annoyances
On Mon, May 07, 2001 at 01:47:32PM +0200, Preben Randhol wrote: > Preben Randhol <[EMAIL PROTECTED]> wrote on 07/05/2001 (11:50) : > > Thanks, I'll try it. > > # apt-get install reportbug [...] > Sorry, but the following packages have unmet dependencies: > reportbug: Depends: python-newt but it is not going to be installed > E: Sorry, broken packages > > :-( Try apt-get -f install reportbug. Note that the reportbug in potato predates some changes to the BTS (specifically, some new allowed types of bugs) and last I tried to install it, it didn't work. However, I was able to apt-get -b source the reportbug from unstable with no problems. Rob -- Any program which runs right is obsolete.
Re: Gnome Packages annoyances
On Mon, May 07, 2001 at 11:43:08AM +1000, Brian May wrote: > > "Karsten" == Karsten M Self writes: > > Karsten> Place your response *after* quoted material. > > Why? Makes it read like a conversation. "Question? Answer!" makes more sense to read than "Answer! Question?" > I like to see the new message at the top, instead of having to scroll > down to access it. Especially in this case, since you can tell what > the poster is saying without reading the quoted message. > > I thought it would make it easier for you too, since you don't have to > read the entire message to find out if it is relevant or not. Ahh, but if your mailer colors quoted text, you know what to skim and what to read carefully. I far prefer to see the context before the response; otherwise I forget what the response is to and it doesn't make sense. Rob -- First Law of Socio-Genetics: Celibacy is not hereditary.
Re: install screensaver with setuid root
On Sat, May 05, 2001 at 08:33:38PM +0200, Alex Suzuki wrote: > On Sat, May 05, 2001 at 12:42:36PM -0500, John Hasler wrote: > > Even if the application itself is free of buffer overflows and such you > > would be granting root access to wads of unaudited and buggy library code. > > yeah, I guess you're right... too bad I love that screensaver :) You should probably file a bug on the package, if it's not useable without introducing a security risk. Perhaps someone more familiar with the code could seperate out the part that needs special privileges into a seperate setuid program. Rob -- Subtlety is the art of saying what you think and getting out of the way before it is understood.
Re: attempted windows install, linux won't boot
On Fri, May 04, 2001 at 07:20:52PM -0700, Forrest English wrote: > > 1. i tried to install win98 to play games > 2. it decided it wouldn't get through the setup. locked 4 times. > 3. so, i put in my debian cd > 3. rescue root=/dev/hda5 > 4. it boots up, all good... > 5. login, re-run lilo > 6. lilo seems to work fine. > 7. i reboot > 8. I9990305 is all that apears where i expect to see: > MBR > LILO: > > 9. so i use the rescue disk again. i try loging in and doing fdisk /mbr, > it reports, unable to open mbr. uh, greeet ^^ I think (could be wrong) that "fdisk /mbr" is the syntax for the DOS version (which is why it works in the next step). You would get this message unless you had a hard disk device file named "mbr" in your / directory. Perhaps you should try "fdisk /dev/hda." > 10. reboot using the win98 cd to dos.fdisk /mbr. it doesn't complain > 11. do the reboot, rescue, lilo thing again. > 12. stare at I9990305 again. Rob -- "It's a summons." "What's a summons?" "It means summon's in trouble." -- Rocky and Bullwinkle
Re: How to find uptime stats after a reboot
On Fri, May 04, 2001 at 04:55:11PM -0500, Kent West wrote: > I just had a power loss that brought down a Potato server. Is there any > way to find out now what the uptime was prior to the unannounced shutdown? I've installed "ud": 00:30 $ ud -d - Uptime for peon - Now : 97 day(s), 10:48:07 running Linux 2.2.16 One : 97 day(s), 10:47:51 running Linux 2.2.16, ended Sat May 5 00:31:18 2001 Two : 91 day(s), 02:47:39 running Linux 2.2.16, ended Sun Dec 10 17:24:13 2000 Three: 61 day(s), 21:21:36 running Linux 2.2.14, ended Tue Jul 18 19:57:18 2000 although that just keeps the three highest, don't know where/if to find data on the most recent. Rob -- Horner's Five Thumb Postulate: Experience varies directly with equipment ruined.
Re: get number of pages of a PostScript file
On Tue, May 01, 2001 at 03:04:41PM +0200, Iwan Mouwen wrote: > * Rob Mahurin <[EMAIL PROTECTED]> [010428 23:09]: > > 10:48 $ psselect -p _1 zeeman.ps > /dev/null > > [8] Wrote 1 pages, 86473 bytes > > Shouldn't that be > psselect -p - file.ps > /dev/null > instead? > > [EMAIL PROTECTED]:~/tmp$ psselect -p _1 report.ps >/dev/null > [11] Wrote 1 pages, 68447 bytes > [EMAIL PROTECTED]:~/tmp$ psselect -p - report.ps >/dev/null > [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] Wrote 11 pages, 100834 bytes That also works; the question was how to find out the number of pages in the document, though, so I only care about the last number. You could even do 16:28 $ psselect -p _1 xray.ps 2>&1 > /dev/null | tr "[]" "\n\n" | grep ^[0-9] 8 although I'm sure there's a more elegant way. Rob -- "Text processing has made it possible to right-justify any idea, even one which cannot be justified on any other grounds." -- J. Finnegan, USC.
Re: Dselect chown: root.root: invalid group
On Sun, Apr 29, 2001 at 08:04:18PM +, Victor wrote: > Any suggestion? > I'm reproposing my previous message. > By the way, when it all happened I was either root or sued from a user login. > Vittorio > -- Forwarded Message -- > Subject: Dselect chown: root.root: invalid group > Date: Fri, 27 Apr 2001 12:21:53 + > From: Victor <[EMAIL PROTECTED]> > To: debian-user@lists.debian.org > > > Trying to unpack devscripts by means of dselect I've got the following error [...] > Setting up devscripts (2.5.8.1) ... > chown: root.root: invalid group > dpkg: error processing devscripts (--configure): > subprocess post-installation script returned error exit status 1 > Errors were encountered while processing: > devscripts Do you have a root group? 22:33 $ grep root /etc/group root:x:0: Can you perform this operation manually? 22:35 # su alphenglor 22:36 $ touch somefile 22:36 $ ll somefile -rw-r--r--1 alphengl alphengl0 Apr 29 22:36 somefile 22:36 $ exit 22:36 # chown root.root somefile 22:36 # ls -l somefile -rw-r--r--1 root root0 Apr 29 22:36 somefile What is the ownership of /etc/ or /bin/, which are ordinarily root.root? 22:37 $ ll -d /etc/ /bin/ drwxr-xr-x2 root root 2048 Jan 30 01:52 /bin/ drwxr-xr-x 88 root root 6144 Apr 29 13:16 /etc/ Hope that gets you started. Rob -- Rule of Defactualization: Information deteriorates upward through bureaucracies.
Re: get number of pages of a PostScript file
On Fri, Apr 27, 2001 at 02:41:04PM +0200, Alexander Steinert wrote: > What's the fastest way to do this on the command line? I do: 10:48 $ psselect -p _1 zeeman.ps > /dev/null [8] Wrote 1 pages, 86473 bytes Rob -- It seems to make an auto driver mad if he misses you.
Re: Shutting down as a user
On Sat, Apr 28, 2001 at 12:19:26AM +1000, Mark wrote: > > What is the "KeyboardSignal"? Can you change it? Would that be a good idea? 10:41 $ dumpkeys | grep KeyboardSignal alt keycode 103 = KeyboardSignal See dumpkeys(1), loadkeys(1), and keymaps(5). Rob -- The story you are about to hear is true. Only the names have been changed to protect the innocent.
Re: page up/dn keys unbound
On Thu, Apr 26, 2001 at 08:21:38PM +0200, Matthieu Paindavoine wrote: > I have noticed today that my page up/dn keys are not responding... > In order to backtrace this problem, I would like to know if there exist > a utility which echoes the codemap received by the keyboard driver, or > any keyboard troubleshooting prgm for that matter. showkey does this on the console; I use xkeycaps for X. Rob -- While it may be true that a watched pot never boils, the one you don't keep an eye on can make an awful mess of your stove. -- Edward Stevenson
Re: mutt's default colors (Potato)
On Tue, Apr 24, 2001 at 06:15:51PM -0500, will trillich wrote: > what *I* would like to know is -- is there a way to hilite > messages in a newsgroup based on a score of some sort? how? Yup, sure is. Interactively, I did :score "~f [EMAIL PROTECTED]" 100 :color index white default "~n 100-" Surely more elaborate schemes are possible. Rob -- Another Armenia, Belgium ... the weak innocents who always seem to be located on a natural invasion route. -- Kirk, "Errand of Mercy", stardate 3198.4
Re: GUI Email program
On Thu, Apr 26, 2001 at 07:05:53PM -0500, will trillich wrote: > On Thu, Apr 26, 2001 at 07:59:41AM -0400, Rob Mahurin wrote: > > -*-Mutt: =debian-user [Msgs:6445 New:39 Inc:2 > > 22M]---(threads/date)--(92%)--- > > c'mon, now you KNOW you'll never go back and look at 6000 old > messages. what gives? I've just recently re-subscribed to the list and I'm waiting until I finish finals to re-implement auto-deletion. Probably the mailbox size will stabilize in the 5-10 MB range. The challenge was, "mutt's not very good with mailboxes with thousands of messages"; apparently _somebody_ has this situation. > > -- > > The beauty of a pun is in the "Oy!" of the beholder. > > people have been burnt at the stake for less. >:/ 09:15 $ echo People have been burned at the stake for puns like yours | fortune I have not yet begun to byte! Rob -- A triangle which has an angle of 135 degrees is called an obscene triangle.
Re: Timeout for shell script
On Thu, Apr 26, 2001 at 04:34:25PM +0200, Andre Berger wrote: > * Keith O'Connell <[EMAIL PROTECTED]>, 2001-04-26 16:29 +0200: > > > I have a shell script in /etc/ppp/ip-up.d/ that synchronizes the system > > > clock with time servers via rdate. Sometimes the servers can't be > > > reached, and the rest of my scripts are blocked. Is there a way to say: > > > "You have at most 10 seconds to complete this command, or fail"? > > > > Have a look at ntp - you may be attempting to solve the wrong problem! > > Thanks... but still I'd like to know if it's possible to use a shell > script with a timeout, in general. In the script, instead of directly calling rdate, call 'exec rdate'. rdate will run and the script will exit. Note that you can't do this if you need to do some processing in your script after rdate runs. Rob -- They are called computers simply because computation is the only significant job that has so far been given to them.
Re: Shutting down as a user
On Fri, Apr 27, 2001 at 08:16:06AM -, [EMAIL PROTECTED] wrote: > > Using debian on a stand-alone laptop I usually work as an ordinary user and > find somewhat awkward the fact that I have to su in order to shutdown the PC. > Is there a way to power my PC off as a user? Use sudo. Someone posted recently that they had created users "reboot" and "halt" with shells "/bin/reboot" and "/bin/halt" to address this, but it didn't quite work like they expected; check the archives. Rob -- Go not to the elves for counsel, for they will say both yes and no. -- J.R.R. Tolkien
Re: using Xfig with LaTeX
On Wed, Apr 25, 2001 at 11:57:15PM -0400, MaD dUCK wrote:
> beating around my thesis, i am trying several methods of including
> Xfig drawings in LaTeX. currently, i am using fig2dev to create .eps
> file, which I \psfig into the document. nevertheless, while the result
> is fine, mpage seems unable to handle the documents and corrupts the
> .ps file right around the included picture (which is rotated 90
> degrees by mpage). furthermore, latex2html can't extract a single of
> these pictures and just reports errors, leaving .png files of zero
> length everywhere.
>
> so i am interested to hear how you guys embed .fig files in latex with
> the ability to scale them to a certain width. and yes, it should be a
> batch process to be run from Makefile.
I do this by creating exporting to .eps from Xfig's file menu (it'll
even prompt you if you've not saved yet, so it's two-click). I
include use a style which contains the following:
\usepackage{graphicx}
[...]
\newcommand{\smepsfigure}[4][htbp]{
\begin{figure}[#1]
\begin{center}
\begin{minipage}{0.4\textwidth}
{\includegraphics[width=0.9\textwidth]{#2}}
\end{minipage}
\caption{#3}
\label{#4}
\end{center}
\end{figure}
}
I've also got a \medepsfigure and an \epsfigure with larger
minipages. (I forget why I changed the size of the minipage instead of
the width= line.) So now I can do
\epsfigure[placement]{file.eps}{This is the caption for this figure}{fig:label}
or copy what I've already written if I need something different.
I use psnup from the psutils package instead of mpage and have had no
problems with it. I've not tried latex2html.
Hope that helps,
Rob
--
About the time we think we can make ends meet, somebody moves the ends.
-- Herbert Hoover
Re: cron: nth of month?
On Wed, Apr 25, 2001 at 03:58:28AM -0700, Karsten M. Self wrote: > I've got a job I'd like to run once a month, on a set day of the week, > say, the first Sunday of the month. Suggestions as to how to do this > with cron? > > My understanding is that I can specify day of month, or weekday, but if > both are specified the condition is treated as an OR (if first week of > month OR Sunday, then do...). > > E.g.: > > 48 2 1 * 7 do-stuff > > ...would run on the first of the month, and each Sunday. Yup, that's how I read it, too. > I could set a job to run either each Sunday, or each of the first seven > days of the month, and test the appropriate other condition. This is the solution which seems the prettiest to me; something like 0 0 1-7 * * [ "$(date +%a)" == "Sun" ] && command -args Rob -- One can search the brain with a microscope and not find the mind, and can search the stars with a telescope and not find God. -- J. Gustav White
Re: Problem with fetchmail.
On Thu, Apr 26, 2001 at 12:06:07AM +0200, Alberto Garc?a wrote: > I am not able to download my email with fetchmail. I try to download being > root and I configured it with fetchmailconf. It just picks the lists of > emails, but fails to download them. What could be the problem. > Thanks in advance. Please post your .fetchmailrc and the output of "fetchmail --verbose" to the list. Rob -- It is illegal to drive more than two thousand sheep down Hollywood Boulevard at one time.
Re: GUI Email program
On Wed, Apr 25, 2001 at 03:56:45PM -0700, Ross Boylan wrote: > Since mutt is based on the file system, groups with lots of mail tend > to overwhelm it--opening, closing, and sorting the boxes gets > sluggish. Eh? -*-Mutt: =debian-user [Msgs:6445 New:39 Inc:2 22M]---(threads/date)--(92%)--- Rob -- The beauty of a pun is in the "Oy!" of the beholder.
Re: Getting package from "testing" for potato
On Wed, Apr 25, 2001 at 06:15:08AM -0400, Harry Henry Gebel wrote: > On Wed, Apr 25, 2001 at 04:15:09PM +1000, Mark wrote: > > > I can't seem to figure out what the build-depends are. I tried doing > > "apt-cache show hotplug", but it can't find it because I only have deb-src > > pointing to testing. (The rest is pointing to stable and Adrian Bunk's > > stuff). > > Use the command > > apt-get build-dep hotplug > > to fulfill hotplugs's build dependencies. Sometimes this will give you an > error that it can't find one (or more) of the packages needed to fill the > build dependancies; if this is the case build them from source first (using > `apt-get build-dep` and `fakeroot apt-get -b source`). > Erm ... apt-get build-dep isn't supported by the apt in potato (0.3.19). Rob -- There are worse things than PerlASP comes to mind
Re: mutt's default colors (Potato)
On Tue, Apr 24, 2001 at 04:02:11PM -0700, Steven E. Harris wrote: > Where are these default colors defined in the Potato mutt package? I 19:55 $ ls -l /etc/Muttrc -rw-r--r--1 root root 4183 Oct 29 07:13 /etc/Muttrc 19:55 $ grep color /etc/Muttrc # colors color hdrdefault cyan default color quoted green default ... Rob -- Work expands to fill the time available. -- Cyril Northcote Parkinson, "The Economist", 1955
Re: OT: command line vs. cron
On Tue, Apr 24, 2001 at 05:59:20PM -0500, Stephen E. Hargrove wrote: > > i've run into a problem i've never seen before. when i execute > > /usr/bin/tvguide.pl > /www/htdocs/tvguide.inc > > from the commandline, it works perfectly. however, when i put > > 0,30 * * * * /usr/bin/tvguide.pl > /www/htdocs/tvguide.inc > > into root's crontab, tvguide.inc is created with nothing in it. Try it without the redirection and see if cron mails you anything. You might even do 0,30 * * * * echo 'starting' && /usr/bin/tvguide.pl && echo 'finished' so that you're sure to get some output. Perhaps the script depends on the value of some environment variable which is set in your shell but not in your crontab. Rob -- Jesuit priests are DATING CAREER DIPLOMATS!!
Re: EMAIL PROCESSORS WANTED IMMEDIATLY
On Tue, Apr 24, 2001 at 09:02:53AM -0700, paul wrote: > EMAIL PROCESSING COMPANY LOOKING FOR EMAIL PROCESSORS IMMEDIATELY, > TO SUSTAIN EXPLOSIVE GROWTH. EARN $5,000- $10,000.00 AND > MORE MONTHLY. NO EXPERIENCE NECESSARY. SEND AN EMAIL TO > [EMAIL PROTECTED] WITH "EMAIL PROCESSOR" IN THE > SUBJECT LINE. WE WANT SERIOUS INQUIRIES FROM PEOPLE WHO > WANT TO MAKE SERIOUS MONEY! > > To remove from this list just insert remove in the > subject line. > Thank you for your time Paul, You should look into procmail. It's an excellent email processor and available for much less than $5,000. Rob -- No animal should ever jump on the dining room furniture unless absolutely certain he can hold his own in conversation. -- Fran Lebowitz
Re: How to remove messages from mbox based on relative date?
On Mon, Apr 23, 2001 at 06:05:07PM -0700, Mike Fedyk wrote: > Hi, > > I have a .procmailrc and am filtering, but now I don't want to have to > delete my messages when they get old in certain mbox files. > > Can I run procmail with another conf file and have it send messages with a > "delivered" date older than N days? Is procmail even the right tool, I'm > not so sure... You can run procmail on existing mailbox if you use formail to split the box into messages, for example, mv mbox mbox.safe && cat mbox.safe | formail -ds procmail You probably don't want to filter into and out of the same mailbox, though; might produce an infinite loop or erase all your mails. I'm not sure how you do 'older than' in procmail. You can also use mutt's folder hooks. For example, I have # # Tag "old" messages in mailinglists for "expiring" them. # Simply press "d", after entering one of the folders, iff mutt asks # "tag-" # folder-hook =debian-user$ 'push T~r>2w\n\;' # 2 weeks in my .muttrc, so when I switch to =debian-user and there are old messages I see 'tag-' at the bottom of the screen and hit 'd' to delete everything that's old. Note that with 'push' mutt can do just about whatever you want. Hope that helps. Rob -- You are here: *** *** * *** * *** * But you're not all there.
Re: GUI Email program
On Mon, Apr 23, 2001 at 05:06:26AM -0500, Rob Zietlow wrote: > Hi there, I am looking for suggestions of a GUI email program (sorry i > just like them better). That's what I thought, until I tried mutt. Rob -- Cold, adj.: When the politicians walk around with their hands in their own pockets.
Re: why does ps->eps conversion reduce line thickness?
On Mon, Apr 23, 2001 at 01:08:30PM +1000, Mark Mackenzie wrote: > I do ps->eps conversions using: > > echo -n '\004' | gs -q -dNOPAUSE -sDEVICE=epswrite \ > -sOutputFile=box.eps box.ps >/dev/null > > using box.ps below. When box.ps is printed, it comes out as a 1mm thick > 1" box (quite dark). When box.eps is printed, the lines are very thin > (perhaps .2mm) and faint. [...] > Using ps2epsi doesn't seem to help either. > I am using gs_5.10-6_i386.deb I get this same effect using "convert box.ps box.eps", which, like ps2epsi, calls gs. Perhaps there is a 'line width' option somewhere which you're not setting explicitly and which is different for ps and eps files. Good luck. Rob -- Prediction is very difficult, especially of the future. -- Niels Bohr
Re: HOWTO remove a previous wildcard file list?
On Sun, Apr 22, 2001 at 10:15:40PM -0400, Mark Hurley wrote: > > Anyone know of a method to easily solve this ... > > I *sometimes* list the files before deleting them: > > ls A*.pdf > > Ensuring I have only listed the ones I wish to delete, I then enter: > > rm A*.pdf > > Great, but anyone know of an easier why than the following? > > 1) Type the second command followed by the selected mask? > 2) Go back one in bash "history", [home] to beginning of line replace > "ls" with "rm". I would either do $ ^ls^rm or $ rm !!:$ or $ rm Pressing repeatedly cycles through the last word in each command in your history. Rob -- In case of atomic attack, the federal ruling against prayer in schools will be temporarily canceled.
Re: upgrading with dselect
On Sun, Apr 22, 2001 at 10:12:12AM +0200, Hans wrote: > I think you shouldn't mix apt-get with dselect, but I'd like to hear some > comments on this. Or maybe use deselect to make choices and use upt-get > dselect-upgrade, but I haven't tried if this works. Erm ... dselect and apt-get are both front-ends to dpkg. dselect changes a package's desired status (which you can also do by saying "echo [install|deinstall|hold|...] | dpkg --set-selections" ); "apt-get dselect-upgrade" tries to make your system match the state described by "dpkg --get-selections". You can choose whatever packages you want with either, if you're careful, and you always have the option to "dpkg --install file.deb". Rob -- I've known him as a man, as an adolescent and as a child -- sometimes on the same day.
Re: Backups and the distribution
On Fri, Apr 20, 2001 at 12:59:08PM -0700, Alex Swavely wrote: > What file(s) would I need to backup to get a complete snapshot of all > installed packages from the distribution I'm using in order to be able to > recover a system to it's previous state? My nightly backup includes /home /root /etc /usr/local /usr/src/linux/.config the output of dpkg --get-selections (unedited) Last year I had a SIMM fail and didn't diagnose it for a while; I got some corruption on my /var and /home paritions and decided to do a full reinstall. I didn't re-format my /usr/local partition, so I didn't need that backup. When I got to the "install packages" part of the install, after the reboot, I exited dselect without doing anything and said # dpkg --set-selections < /mnt/backup/dpkg.selections # apt-get dselect-upgrade As I needed to configure things, I copied their information from my /etc backup and voila! things were as before. All in all, it was quite a smooth recovery. Rob -- You will inherit some money or a small piece of land.
Re: user=halt=halt, user=reboot=halt?
On Thu, Apr 19, 2001 at 04:12:58PM -0700, Lang Hurst wrote: > I added users to halt and reboot, respectively, by making the formers login > shell /sbin/halt and the latter's /sbin/reboot. Neat idea; if I rebooted more often I might try it. > If I login as the halt user, the system shuts down and powers off. Great. > When I login as the reboot user, the system shuts down and powers off. Not > what I wanted. I assume these commands work as expected from the command line? Rob -- QOTD: "Like this rose, our love will wilt and die."
Re: Unsubscribing issues (was Re: )
On Fri, Apr 20, 2001 at 01:14:45PM -0700, Karsten M. Self wrote: > on Fri, Apr 20, 2001 at 08:07:30AM -0700, JC Portlock ([EMAIL PROTECTED]) > wrote: > > On Friday 20 April 2001 02:03 am, Felicity wrote: > > > unsubscribe > > > Try again with "unsubscribe" in the subject line. > > Nope. > > Standard unsubscription: > > Send a message 'unsubscribe' to [EMAIL PROTECTED] > > Or, as a GNU/Linux command: > > $ echo 'unsubscribe' | mail [EMAIL PROTECTED] > > ...which is what the list tag used to say, but it was considered too > confusing. Subject line doesn't matter. Funny, the tag now says >-- >To UNSUBSCRIBE, email to [EMAIL PROTECTED] >with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] which reminds me that the old confusing one was mail -s 'unsubscribe' [EMAIL PROTECTED] < /dev/null Rob -- Let's remind ourselves that last year's fresh idea is today's cliche. -- Austen Briggs
Re: applying _only_ security updates
On Thu, Apr 19, 2001 at 03:32:20PM +1000, Price, Tim wrote: > Regardless, > > Is it possible to do? > > I ask because I am running a potato system with many packages from woody, > and some security packages take longer to get into testing then they do to > get into stable. I've heard that the more recent versions of apt have a feature called "pinning," wherewith you can request certain packages to come only from certain sources, or some such magic. Perhaps that would help you. Rob -- ... the flaw that makes perfection perfect.
Re: logcheck
On Wed, Apr 18, 2001 at 04:34:05PM -0700, [EMAIL PROTECTED] wrote: > i have logcheck installed on a few systems. i cleared out most > of the things generating the reports but..it still emails me > every hour and the only contents of the email are the log entries > of it sending the previous email(messages about root sending > email to me using postfix). any way to get rid of those so only > emails that contain something useful are generated? being emailed > by a program about activities it performs isnt ideal for me :) I've added the log messages associated with e-mail to /etc/logcheck/logcheck.ignore. Rob -- "My life is a soap opera, but who has the rights?" -- MadameX
Re: Shell prompts (was Re: floppy permissions)
On Wed, Apr 18, 2001 at 12:48:55AM -0700, Karsten M. Self wrote:
> > It's a shame that you've solved your problem; now I don't have an
> > excuse to tell you that this is a great root prompt.
>
> Not half bad. My own approach is slightly more subtle, but reasonably
> effective:
Here's mine, while we're comparing; I put "source $HOME/.prompt &&
prompt" in by .bashrc.
Rob
--
... this must be what it's like to be a COLLEGE GRADUATE!!
#!/bin/bash
black () { echo -en "\e[0;30m" ; }
red () { echo -en "\e[0;31m" ; }
green () { echo -en "\e[0;32m" ; }
brown () { echo -en "\e[0;33m" ; }
blue() { echo -en "\e[0;34m" ; }
purple () { echo -en "\e[0;35m" ; }
cyan() { echo -en "\e[0;36m" ; }
gray() { echo -en "\e[0;37m" ; }
dgray () { echo -en "\e[1;30m" ; }
bred() { echo -en "\e[1;31m" ; }
bgreen () { echo -en "\e[1;32m" ; }
bbrown () { echo -en "\e[1;33m" ; }
bblue () { echo -en "\e[1;34m" ; }
pink() { echo -en "\e[1;35m" ; }
bcyan () { echo -en "\e[1;36m" ; }
white () { echo -en "\e[1;37m" ; }
norm() { echo -en "\e[0m" ; }
prompt () {
local BLACK="\[\e[0;30m\]"
local RED="\[\e[0;31m\]"
local GREEN="\[\e[0;32m\]"
local BROWN="\[\e[0;33m\]"
local BLUE="\[\e[0;34m\]"
local PURPLE="\[\e[0;35m\]"
local CYAN="\[\e[0;36m\]"
local GRAY="\[\e[0;37m\]"
local DGRAY="\[\e[1;30m\]"
local BRED="\[\e[1;31m\]"
local BGREEN="\[\e[1;32m\]"
local BBROWN="\[\e[1;33m\]"
local BBLUE="\[\e[1;34m\]"
local PINK="\[\e[1;35m\]"
local BCYAN="\[\e[1;36m\]"
local WHITE="\[\e[1;37m\]"
local NORM="\[\e[0m\]"
local TIME="$PURPLE$(date +%H:%M)$NORM "
[ "$(whoami)" == "root" ] && local DOLLAR="$WHITE\\$\[\]${NORM} " \
|| local DOLLAR="$GREEN\\$\[\]${NORM} "
# this will let me put a history number in ... I've decided I don't like it
#local HIST="$RED\\!${NORM} "
if [ $PWD == $HOME ] ;
then
local DIR=""
local TBAR="[EMAIL PROTECTED]"
else
local DIR="$BROWN[\w]$NORM\n"
local TBAR="[EMAIL PROTECTED]:\w"
fi
[ "$TITLEBAR" == "" ] || local TBAR=$TITLEBAR
local TBAR="\[\e]0;$TBAR\007\]"
case $HOSTNAME in
peon) local HOST="" ;;
*) local HOST="$BLUE\h$NORM " ;;
esac
local TTY="$(tty | cut -d/ -f3)"
if [ "$TTY" == "pts" ]
then
local TTY="$TTY/$(tty | cut -d/ -f4)"
fi
local TTY="$CYAN$TTY$NORM "
case $TERM in
xterm*) local TBAR=$TBAR
local TTY="";;
dtterm*)local TBAR=$TBAR
local TTY="";;
*) local TBAR=""
local TTY=$TTY ;;
esac
export PS1="$TBAR$DIR$TIME$HOST$TTY$DOLLAR"
export PROMPT_COMMAND='prompt'
} && prompt
Re: any women here?
On Wed, Apr 18, 2001 at 03:05:32PM -0600, Joe 'Zonker' Brockmeier wrote: > On Wed, 18 Apr 2001, will trillich wrote: > > > Not that there aren't some men out there with real lives, of course. > > [EMAIL PROTECTED]:~# apt-get install real-life > Reading Package Lists... Done > Building Dependency Tree... Done > E: Couldn't find package real-life > > Hmmm. Nope didn't work. Maybe if I upgrade to Sid... I don't know that I'd be so thrilled about an unstable real life. Rob -- No problem is insoluble. -- Dr. Janet Wallace, "The Deadly Years", stardate 3479.4
Re: lists.debian.org: FAQ / tip-of-the-day
On Tue, Apr 17, 2001 at 11:19:11PM -0500, will trillich wrote: > On Fri, Apr 13, 2001 at 02:01:24PM -0400, Rob Mahurin wrote: > > On Fri, Apr 13, 2001 at 09:05:43AM +0100, Glyn Millington wrote: > > > What might help is a daily post to the list with a subject line > > > "NEW TO LIST? READ THIS!!" [...] > daily is waaay too often. there's plenty of traffic on the list > already... YET a periodic re-post is not a bad (or original) > idea. many newsgroups publish their faq, sometimes multi-part > faq's, monthly or quarterly. Truth. > > I imagine the following format: > > > > --- begin hypothetical message --- > > Subject: Frequently Asked Questions on debian-user > > nice start so far. add it to http://newbieDoc.sourceForge.net? You must have left town for the weekend, Will; I expected you to ask me that days ago. I'll look at it. > redundancy in mentioning a document source is NOT a bad thing. > HOWTOs can be mentioned in fifteen different places on this FAQ > document, and it won't hurt a thing. Good point. By the way, you should read the Linux HOWTOs. > don't let anyone put a damper on this effort. it's important! > (and when you're done -- we prefer sgml, docbook style, but we'll > accept html -- add it to our itty-bitty library at newbiedoc over > at http://sourceforge.net/projects/newbiedoc/ !) Thanks for the encouragement. I'm keeping working drafts at http://web.utk.edu/~rmahurin/debian-user-faq/faq.html ; s/faq.html// for a little front matter. Rob -- QOTD: "A university faculty is 500 egotists with a common parking problem."
