Re: OpenGL causes hard system lockups

[Mike Brownlow]

Matthew Garman wrote:
> Some of these 3D screensavers will unexpectedly lock up my system---and I
> mean *completely* lock up my system, can't ctrl-alt-f[1-6] to switch to a
> text console, I can't kill X with ctrl-alt-backspace, etc.  Nothing works.
> 
> Obviously I don't like lockups like that, since I have to restart my
> computer without shutting down properly.
> 
> Anyone know what might be causing this?  Is it likely a hardware problem?
> Anyone had any experience with this kind of thing?

I have a G400 and I used to (long time ago) have that sort of thing
happen. It was more frequent with my TNT2ultra at the time. Lately I
haven't had problems (in fact, some speed improvements). I'm running
linux 2.4.10 and X 4.1.0.1.

Do you have the ability to telnet/ssh in? That's usually how I was able
to fix it cleanly...   but sometimes it wouldn't let you do that either.

-- 
Mike Brownlow   ><>  http://www.wsmake.org/~mike/
-
1024D/8AA6EAFD 3861 96B3 EEA2 285C BE23  F706 3E1E EBB2 8AA6 EAFD

Re: Sendmail not sending queue's out.

[Richard A Nelson]

On Mon, 1 Oct 2001, Mike Dresser wrote:

> Every couple days, i have to sendmail -q, beacuse /var/spool/mqueue has
> sometimes dozens of outbound unsent messages.  Once i do that, most of
> them get sent, leaving the ones that are unreachable/etc left to be sent.

what does mailq say about those messages (why are they queued)?
>
> My question is, why do I have to do this?  Shouldn't there be something to
> sendmail -q already?

There is, if sendmail is infact still running... AND is running in
queue delivery mode/

> I could just setup a cron job, but I'm wondering what's up with it.

Not enough information to know whats up, sorry...

> I'm running sendmail(8.9.3) out of potato.

Things will be much nicer when you go to woody... 8.12.0 (and 8.12.1,
due RSN) don't fork nearly as much, and have very nice queue management
facilities.

-- 
Rick Nelson
We are MicroSoft.  You will be assimilated.  Resistance is futile.
(Attributed to B.G., Gill Bates)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Voodoo3. DRI, X4.1.0 and Bus mastering <- Still no joy

[john]

Jason,

I can't get the bus mastering to work, and so cant get DRI to work (I think
that's the reason anyway).

If you trace the thread you'll see I've tried everything I can think of.

If I find time I'll build a new (actually old) box this weekend and see if I can
get the Voodoo to work - I suspect the MB is at fault here...

It's interesting that you say you have DRI working but not bus mastering. Can
you run setpci on the device and see if bit 3 of word 4 is set? If so my problem
isn't bus mastering at all.

Thanks,

John P Foster

Jason Healy wrote:

> At 1002217484s since epoch (10/03/01 22:44:44 -0400 UTC), john wrote:
> > Hi!
> >
> > I've tried both sets of advice, yet it still doesnt work.
>
> What, exactly, doesn't work?  No 3D, or no bus mastering?
>
> I have a Voodoo3, and I don't think I ever got bus mastering to work either.
> I thought that it was the reason why I couldn't get 3D to work, but it
> turned out that I had a few other problems.  I needed to install xlibmesa3,
> set my resolution to 1024x768 @ 16bpp, and get DRI working.  I kind of
> forgot about the bus mastering thing once it started working... =)
>
> Jason
>
> --
> Jason Healy| [EMAIL PROTECTED]
> LogN Systems   |   http://www.logn.net/
>
> --
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Woody printing problem

[Lars Jensen]

I just installed woody, but can't get printing to work. When I try to
print I get the error message that 


/dev/lp0 does not exist.

The same error happens if I do try to direct a file to the device
like this:

testfile > /dev/lp0


I have an HP Deskjet 540 hooked up to my Parallel port (LPT1 in
Windows).

What is going on?


Thanks,
Lars.

Re: SPAM WARNING: spammers use Debian lists for harvesting

[Johnny Ernst Nielsen]

>> I've been on the list, on and off, for roughly a year now and I have yet

>> to get a single bit of spam, other than the occasional one that is sent

>> to a list I am on.  None of them have been addressed directly to me.
>
>It wouldn't surprise me too much if the spammers filter or weight the
>addresses they see in various ways. It could be that a .edu address is
>less likely to be spammed heavily than a .com address at a
>consumer-oriented ISP. I'm just speculating, though.

Funny. As you see I have a .com address at a fairly consumer-oriented ISP.

I don't get annoying amounts of spam. 3-4 letters a year perhaps directly
addressed to me.
I wonder what the spammers really look for.

Cheers

Johnny :o)

smbmount without password

[Matthew Garman]

I've got some samba shares on my OpenBSD machine that I'd like to use as
drives on my Linux box.  The shares are setup such that I can connect to
them as guest (using no passowrd) with read/write access using smbclient.
However, when I try to mount them (using smbmount or mount -t smbfs), I
get the following error unless I specify a password:

session setup failed: ERRSRV - ERRbadpw (Bad password - name/password pair
in a Tree Connect or Session Setup are invalid.)
SMB connection failed

Am I missing something subtle here?

Thanks,
Matt

-- 
Matt Garman, [EMAIL PROTECTED]
"I'll tip my hat to the new constitution, Take a bow for the new revolution
 Smile and grin at the change all around, Pick up my guitar and play
 Just like yesterday, Then I'll get on my knees and pray..."
-- Pete Townshend/The Who, "Won't Get Fooled Again"

Re: SPAM WARNING: spammers use Debian lists for harvesting

[Robert Waldner]

(mantra: *please* do *not* Cc me on list-mail, I read all the lists I 
 post/mail to)

On Wed, 03 Oct 2001 18:47:12 +0200, martin f krafft writes:
>* Robert Waldner <[EMAIL PROTECTED]> [2001.10.03 18:26:21+0200]:
>> this.is.a.trap.graffl.net   TXT "this is just a spamtrap"
>> this.is.a.trap.graffl.net   A   127.255.255.255
>
>that's a splendid idea, but don't you want to consider 127.0.0.1
>instead. broadcast is kind of harsh...

There /will/ be people who are trying to send legitimate mail, which 
 could work with localhost, but mail over broadcast isn't implemented 
 AFAIK.

Also, consider the following:
- if the spammer uses an open relay[0], 127.0.0.1 would cause more 
 traffic as the box may try to do an entire SMTP-conversation, whereas 
 with broadcast it would be just a few unanswered packets
- 127.0.0.1 isn't that funny anymore, too much people use 
 [EMAIL PROTECTED]

0: most open relays aren't open on purpose, just run by clueless 
 ($exchange-) admins

cheers,
&rw
-- 
-- Writing everything for the dumbest user will
-- be insulting and irritating to everybody else.
--  - Andrew Suffield




pgpq5gIMe5n38.pgp
Description: PGP signature

Re: Mozilla is so slow! Problem with my upgrade to 2.2r3?

[Chris Cothrun]

Hi, 

I didn't see anyone else weigh in with quite my opinion so I 
thought I'd throw it in.

> What could be happening here?  I think I may be ready to buy a
> set of 2.2 disks because from a beginner's viewpoint it seems
> that successfully upgrading the OS to 2.2 was probably too
> much to expect.  It went very smoothly (I was very impressed
> with how easy it was!), but I did get some messages about what
> seemed to be housekeeping tasks that needed to be done as a
> result of the upgrade.  Some of them related to files that
> simply don't exist on my machine, so maybe something is
> seriously wrong.  

I would say that it went OK. maybe do some research, ask 
questions about those messages.

> Anyway, I thought I would post this question here just in case
> it's an obvious, easily fixed problem.   Otherwise, I'll just
> re-install from a new set of disks.  By the way, my computer
> is a Pentium 200 MHz, with 32 MB of RAM.  Is this too
> antiquated to run Mozilla, or perhaps Gnome or Galeon
> eventually???

Definately memory. Buy some more, it is relatively cheap 
right now. I've got a P200 with 96MB and it runs KDE and 
has all sorts of processes loaded (Zope, PosgreSQL, 
Apache, etc). Sure, it is a little slow and swaps lots 
whenever I start something big but it is perfectly useable. 
Galeon and Skipstone work great, Mozilla is slow but also 
works. With a simpler window manager, you're going to be 
much better off than I am with KDE.

Opera of course is highly reccomended but I see you've 
had bad luck with it. 




Chris

Re: SPAM WARNING: spammers use Debian lists for harvesting

[Antti Tolamo]

At 08:58 4.10.2001, Johnny Ernst Nielsen wrote:

>> I've been on the list, on and off, for roughly a year now and I have yet

>> to get a single bit of spam, other than the occasional one that is sent

>> to a list I am on.  None of them have been addressed directly to me.
>
>It wouldn't surprise me too much if the spammers filter or weight the
>addresses they see in various ways. It could be that a .edu address is
>less likely to be spammed heavily than a .com address at a
>consumer-oriented ISP. I'm just speculating, though.

Funny. As you see I have a .com address at a fairly consumer-oriented ISP.

I don't get annoying amounts of spam. 3-4 letters a year perhaps directly
addressed to me.
I wonder what the spammers really look for.


I think they look for easy targets and valids email
addresses.
Bigger, more public organization with easy way to gather
valid email  addresses and spammers strike. They don't
target ISP's as such I think. Another things, to
me it looks like they would just collect from certain
contexes(like from certain usenet groups) email addresses,
and if they know the organization  to which it belongs
and it fits to certain profile, they send. Propably
somebody makes real life lists of domain holders.
Spammes just match valid email addresses to it.
That's my guess anyway.

I for example have many years posted with valid
email addresses to usenet. Still, no spam. I suspect
my domain doesn't hold intrest and/or I post to wrong
places. Makes me wonder why people use those '_invalid_'
addresesses.

Now I get spam, but main reason is that I gave my email
addresses to few dubious web sites with questionable
content...effect was quick and easy to anticipate:)
But most of it, seems to come from one source.
edirectnetwork.

I've lost already 25 000 $ in
group lotto as I haven't clicked their adverts.
Stupid me.;)


Antti

Antti


My PGP public key:
http://linux.tola.org/~chicken/antti_pgp.txt

--
Sex, rags and rock'n roll!
--

Re: OT: forking so apache won't wait

[will trillich]

On Thu, Sep 27, 2001 at 04:28:01PM +0200, Erdmut Pfeifer wrote:
> On Thu, Sep 27, 2001 at 12:44:35AM -0500, will trillich wrote:
> > i know i've run across it before but when i WANT to stub my toe
> > on it, it's nowhere to be found: HOW can i have apache/web page
> > initiate a process and return quickly to generate a 'processing,
> > hold on' page while the process does its processing?
> >
> 
> > 
> > pointers? flames? any gdM you think i should RTF out of?
> 
> Hi Will,
> 
> ...as I've seen you being active on the mod_perl list, I would've
> assumed that you've already heard of "The Guide" ;)

duh! i've run my eyeballs over that website many a time, but i
didn't put two and two together for this oft-requested answer.
much thanks!

-- 
DEBIAN NEWBIE TIP #67 from Colin Watson <[EMAIL PROTECTED]>
:
Did you know MANPAGES ARE IN SEVERAL SECTIONS?  For example,
user commands are in section 2 of the manual, and system
administration items are in section 8; to request a particular
section via "man" include it before the item:
man 7 regex
(otherwise you'll probably see regex from section 3 instead.)
To see ALL pages with a particular name, try
man -a regex
every matching manpage (from whichever section) will be
presented, one-by-one.

Also see http://newbieDoc.sourceForge.net/ ...

Re: DHCP problems with cable modem

[Osamu Aoki]

On Wed, Oct 03, 2001 at 10:36:57PM -0500, Michael Heldebrant wrote:
> On Sat, 2001-09-29 at 19:56, Andrew Robertson wrote:
> > Hi.  I just installed Debian last night on my system,
> > and everything works great except the network.  I have
> > a cable modem, and the DHCP server requires that I
> > send in my hostname with my request.  So far, I have
> > tried (and failed at using):
  (Yep, you can set /etc/interfaces for DHCP and make sure your
ipchains/iptables are set corectly.)

Alternative is set up with FIXED IP.  Many cable service gives same IP
for same hostname (my cable) or same MAC address (some who rely on NIC
card) throgh DHCP.  To get current IP with Windows, use winipcfg command.

Heck, with this way, you can access your box from outside. ;-)

-- 
~\^o^/~~~ ~\^.^/~~~ ~\^*^/~~~ ~\^_^/~~~ ~\^+^/~~~ ~\^:^/~~~ ~\^v^/~~~ 
+  Osamu Aoki <[EMAIL PROTECTED]>, GnuPG-key: 1024D/D5DE453D  +
+  My debian quick-reference, http://www.aokiconsulting.com/quick/+

Re: Setting tab length universally

[will trillich]

On Tue, Sep 18, 2001 at 03:30:46PM +1000, Andrew Sione Taumoefolau wrote:
> Hi,
> 
> I was just wondering if there was some method of universally setting tab
> width. I find eight spaces a bit too much, and although I have found
> methods of changing it in some applications, it'd be nice if I could
> change it in one place and never have to worry about it again. Advice or
> pointers to some Ms to Fing R would be much appreciated :).

i wondered the same thing. i think the only approach would be to
establish a vt100-like setup string that plants console tabs
every so often...

that said, the rest of the world does expect tabs to be 1+mod8
(i.e. 1, 9, 17, 25, 33...) -- but if the focus of tabs is to
help indent code, then use one tab per level of indent, and let
the 8-char-tab purists watch their dislpays wrap to hell and
back. i've been using 4-per-tab a lot and it works well.

the trouble comes when you use nonstandard tab stops and then
try to align a column to the right of uneven text:

it  nicely  4-char-tab  : but this
lines   on  rxvt: is likely
up  my  window  : to suck

bad practice, mixing tabs and spaces. pick one and stay there.
(i vote heavily for tabs, as they're settable -- indent to
timbuktu and back if you like [despite what the purists say
about programs needing to stay no deeper than three levels] and
adjust your tab width to accomodate your display needs. your
recipients can do the same. and in vim you can include
'modelines' [see :help modeline] to pre-set nonstandard tabs for
you: "#!/usr/bin/perl   vim:ts=4" for example...)

-- 
DEBIAN NEWBIE TIP #19 from Dave Sherohman <[EMAIL PROTECTED]>
:
How do you determine WHICH NETWORK SERVICES ARE OPEN (active)?
Try "netstat -a | grep LISTEN". To see numeric values (instead
of the common names for services using a particular port) then
try "netstat -na" instead. For more info, look at "man netstat".
   Also try "lsof -i" as root. "man lsof" for details.
=Will Trillich <[EMAIL PROTECTED]>

Also see http://newbieDoc.sourceForge.net/ ...

Re: how to fight back against Michael Bramer's spam

[Glyn Millington]

Branden Robinson <[EMAIL PROTECTED]> writes:


> 1) apt-get install procmail
> 2) place all of grisu's spam as-is into its own folder; let's say you
>call it "BASTARD"
> 3) formail -s BASTARD < antispam
>

Haven't noticed the problem yet!!  But thanks for the recipe ;-)


Glyn
-- 

   **
   *  Here we are then...   *
   * http://members.tripod.co.uk/Christchurch2000uk *
   **

Re: Compiled kernel 2.4.9 does NOT boot

[Timeboy]

 - Re: Compiled kernel 2.4.9 does NOT boot -
 
 On Wednesday Oct 03 23:01 José Luis Rey wrote:
 
 > ** The problem is that after initrd is loaded kernel panics with:
 > **Kernel panic: VFS: unable to mount root fs on 03:03
 
 I had the same trouble after my first self compiled kernel. This
 happened cause i forget some things to set in the kernel configuration.
 Do you have enabled the following? This is needed to mount root fs.
 
 ---> Networking options
<*> Unix domain soccets
 
 ---> Character devices
<*> Unix98 PTY support
 
 Timo
 
 --
 Nothing is impossible!
 
 You only need to know the way and price.
 
  :-)
 

Re: SPAM WARNING: spammers use Debian lists for harvesting

[P Kirk]

Hi all,

Its my impression that there are only a few email harvesters out there
and that they don't work very hard.  Perhaps once you have a list with a
million names, its not worth wasting time building a new one.

For example, I found about the time of Clinton/Lewinski I was getting a
very well put together set of commentaries from a chap called Kristol.
Turns out he's a TV commentator who had bought the spam list.  Now for
the first time I could see proper headers because he didn't do any of
the usual munging stuff.  He had over 500 addresses with kirk in them
and was sending his stuff out in alphabetical blocks.  I create a
.forward rule that automatically drove all mail to that alias into a
spam file.  

That was over 3 years ago.  Since then I have used pknews all over
usenet and patrick in quite a few mailing lists.  Yet over half the 20
or so pieces of spam I get every day are to that old pkirk address which
hasn't been used publicly in 3 years.

Another example: its rare to get spam in the work inbox yet that address
has been blasted out on every PR release for over a year.  I think
spammers aim to get aol users or msn users in the hope that they are
less cynical.

I quite enjoy tinkering with my .forward to catch the poor devils.  Once
you get to the stage where only 1 or 2 gets through each week without
blocking people you know, its almost a hobby.

Perhaps I need to get out more :-(

Patrick

Re: directly installing a deb package

[Alex Hunsley]

Stephen Gran <[EMAIL PROTECTED]> wrote:

> Alex Hunsley wrote:
> >
> > Progress... I did apt-get install xaos from the net. It failed again, but
> > suggested I run "apt-get -f install", which I have done, and it seems to be
> > happily getting the needed dependencies. Is this the correct way to correct 
> > the
> > situation normally?
> >
> > thanks
> > alex
> >
> 
> How did it fail?  I guess what I'm trying to say is, can you post the
> failures here?  It'll give a better idea of what's happening.  the -f
> option works, but can cause some packages to break.  Write back and I'll
> see what can be done.
> Good luck,
> Steve

I'm afraid I didn't keep a copy of how apt-get failed, but I remember
the form.
The message looked like: 

Couldn't install A -
  dependency B wasn't installed.
  B depends on C which wasn't installed.

You say doing the -f things causes some packages to break -- hmm. The
software I installed does run now, but sometimes appears to hang, maybe
a symptom of this?

alex

NIS and initgroups

[Bristow Paul-BPB007]

Hi,
I've recently installed debian on a spare PC at work and I'm having 
real problems getting NIS to work properly with groups.  I'd appreciate any 
advice on the following problem.  
After installation I set up NIS using the nsswitch.conf file, the relevant 
lines are as follows.

passwd: files nis
shadow: files
group:  files nis

After doing this I was unable to log in as root!  Whenever I tried I got an 
error message
initgroups invalid argument
and was immediately thrown out.  However, I could log in as either a user local 
to the machine or as a user authenticated via nis without any problems and 
everything works properly!  

After booting from CD, mounting the partition and changing the group line to
group:  files #nis 
to comment out the nis part for groups I can log in fine as root or a local 
user but the users being authenticated via NIS now see their group as a number 
rather than a group name (due to disabling of nis group look up).

I've also tried using compat mode, i.e. nsswitch.conf is
passwd: compat
shadow: compat
group:  compat

and adding +:: to passwd and +::: to group but I get exactly the same 
problem.  I've also upgraded to the latest stable version of the nis debs, 
3.8.0 I think.

If it's any help, the nis server is an HP-UX machine and we also have a couple 
of Mandrake boxes which are using nis via the compat mode just fine so I'm 
pretty sure the problem is at the debian end.

Any thoughts greatly appreciated.
Regards,
Paul Bristow.

Re: directly installing a deb package

[Andrea]

On Wed, 2001-10-03 at 23:10, Alex Hunsley wrote:
> I've downloaded a .deb package from the net and want to install it. How do I
> tell apt-get that I'm giving it a direct file name rather than having it
> looking at the places in sources.list? The package is sitting in /tmp at the
> moment, and I've tried adding file:/tmp to sources.list, but still no luck - I
> think it's adding things like stable or main to the path.
> 

Normally dpkg -i package-name is enough, but if you want to use apt,
copy that package in /var/cacha/apt/archives (if you use sid, with
potato the location where apt keeps the package are), or try with
apt-get -o dir::cache::archives="/tmp/" install package-name (you need
to create a directory in temp, apt will tell the details you)...


Andrea

testing with 2.4.9 ppp apt-get upgrade errors

[Mark Robinson]

Preconfiguring packages ...
(Reading database ... 31851 files and directories currently installed.)
Preparing to replace ppp 2.4.1-4 (using .../ppp_2.4.1.uus-1_i386.deb) ...
Unpacking replacement ppp ...
depmod: Unexpected value (20) in
'/lib/modules/2.4.9-586tsc/kernel/drivers/ieee1394/sbp2.o' for
ieee1394_device_size
It is likely that the kernel structure has changed, if so then
you probably need a new version of modutils to handle this kernel.
Check linux/Documentation/Changes.
dpkg: warning - old post-removal script returned error exit status 255
dpkg - trying script from the new package instead ...
depmod: Unexpected value (20) in
'/lib/modules/2.4.9-586tsc/kernel/drivers/ieee1394/sbp2.o' for
ieee1394_device_size
It is likely that the kernel structure has changed, if so then
you probably need a new version of modutils to handle this kernel.
Check linux/Documentation/Changes.
dpkg: error processing /var/cache/apt/archives/ppp_2.4.1.uus-1_i386.deb
(--unpack):
 subprocess new post-removal script returned error exit status 255
depmod: Unexpected value (20) in
'/lib/modules/2.4.9-586tsc/kernel/drivers/ieee1394/sbp2.o' for
ieee1394_device_size
It is likely that the kernel structure has changed, if so then
you probably need a new version of modutils to handle this kernel.
Check linux/Documentation/Changes.
dpkg: error while cleaning up:
 subprocess post-removal script returned error exit status 255
Errors were encountered while processing:
 /var/cache/apt/archives/ppp_2.4.1.uus-1_i386.deb
E: Sub-process /usr/bin/dpkg returned an error code (1)

I'm not actually using ppp any more which is a good thing, but it won't let
me remove it either (package is too messed up to remove), and it appears the
problem is at least partly elsewhere.

Any pointers to a resolution appreciated.

Mark

aptitude basically unusable: what happened?

[Erik Steffl]

  I used to use aptitude for most of the package manging activities,
which mostly meant doing equivalent of apt-get update && apt-get
dist-upgrade.

  lately (few month) the aptitude became completely usuable:

  package screen: I only see few packages in package tree, e.g. the
Installed packages branch should have  a LOT of packages, however it has
only 10 packages.

  upgrade screen (screen you get into when you press 'g'): there should
be a list of new/held/removed/updated packages. however, there is
nothing there, the screen is empty.

  it looks like it got worse gradually, e.g. long ago it worked very
well, later on it started to put a huge number of packages on hold for
no apparent reason (apt-get wouldn't show them as being on hold,
pressing '+' changed the state to 'i' with no problems) and now it does
not work at all.

  I didn't see any bugs related to these problems, is it possible that
there is something wrong with my system/my using of aptitude or is
nobody using it?

  TIA

erik

Re: NIS and initgroups

[Miquel van Smoorenburg]

In article <[EMAIL PROTECTED]>,
Bristow Paul-BPB007  <[EMAIL PROTECTED]> wrote:
>Hi,
>   I've recently installed debian on a spare PC at work and I'm having
>real problems getting NIS to work properly with groups.  I'd appreciate
>any advice on the following problem.  
>After installation I set up NIS using the nsswitch.conf file, the
>relevant lines are as follows.
>
>passwd:files nis
>shadow:files
>group: files nis
>
>After doing this I was unable to log in as root!  Whenever I tried I got
>an error message
>initgroups invalid argument

Sounds like there is a (syntax?) error in the NIS groups map.
Use 'ypcat groups' and see if you can find something obvious.

Maybe you have groups with gids > 65535? That isn't supported by
Linux unless you install a 2.4 kernel and glibc >= 2.2.

Mike.
-- 
Move sig.

Re: directly installing a deb package

[Jussi Ekholm]

Alex Hunsley <[EMAIL PROTECTED]> wrote:

(Please write shorter lines - no more than 75 characters...)

> I've downloaded a .deb package from the net and want to install it. 
> How do I tell apt-get that I'm giving it a direct file name rather than 
> having it looking at the places in sources.list? The package is sitting 
> in /tmp at the moment, and I've tried adding file:/tmp to sources.list, 
> but still no luck - I think it's adding things like stable or main to 
> the path.

Use dpkg. 

$ dpkg -i /foo/bar.deb

-- 
Jussi Ekholm,   "Everything is so fine it could be
little, ill flower   don't let your mind take you in misery
[EMAIL PROTECTED]all the feelings you're not so much pleased
http://goa-head.org/ekhowl   they're just to take you to sweet harmony"

Re: downloading debian

[Wayne Topa]

csj([EMAIL PROTECTED]) is reported to have said:
> On Thu, 2001-10-04 at 03:49, Vineet Kumar wrote:
> > * kumar suresh ([EMAIL PROTECTED]) [011003 07:13]:
> > > Hello,
> > > 
> > > We tried to download debian cdimages as instructed by
> > > the latest psudo-image kit but from most of the sites
> > > we were unable to download the required files. With
> > > many sites, the make-psudo-imaage command terminated
> > > immediately and with someother sites we could donwload
> > > only upto 21MB or so. Thereafterwards,
> > > make-psudo-imaage command returned the prompt
> > > immediately.
> > > 
> > > So is there any site from where we can get latest
> > > debian? 
> > 
> > See www.bradmont.net (which, unfortunately, seems to be down at the
> > moment). I'll burn and mail you one for the cost of the disc + shipping,
> > mail me privately if interested. (potato discs only right now.)
> 
> Wouldn't it be a brilliant idea if someone came up with an Essential
> Debian ISO image, containing only the absolute minimum files necessary
> to have a working apt-get/dselect system, (e.g. the boot image,
> base*tgz, drivers.tgz)? Low-bandwidth folks can burn the mini-ISO
> (~30MB) image to CDRW or a mini CDR and just apt-get the .debs they
> really need (which for most people should be under 300MB)

That would be too good to be true.  From your lips to GOD's ear
or maybe a low bandwidth maintainer.
-- 
RAM disk is NOT an installation procedure.
___

DHCP PUMP trouble

[Eskild Wikkeling]

Hello,

I recently got a ADSL modem for babyxl (a dutch adsl provider that uses
dhcp to give me my ip) and am trying to get it to work.

in /etc/network/interfaces I have:
auto eth0 eth1

iface eth0 inet static


iface eth1 inet dhcp
  hostname MYNAME.demon.nl

and in /etc/pump.conf:
domainsearch "demon.nl"

retries 3
timeout 60

device eth1 {
}

when I run ifup eth1 I get the following error in /var/log/syslog:
Oct  4 12:52:41 cumulus pumpd[5098]: HOSTNAME: requesting
MYNAME.demon.nl
Oct  4 12:52:41 cumulus pumpd[5098]: PUMP: sending discover
Oct  4 12:52:41 cumulus pumpd[5098]: breq: opcode: 1
Oct  4 12:52:41 cumulus pumpd[5098]: breq: hw: 1
Oct  4 12:52:41 cumulus pumpd[5098]: breq: hwlength: 6
Oct  4 12:52:41 cumulus pumpd[5098]: breq: hopcount: 0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: xid: 0x45359f55
Oct  4 12:52:41 cumulus pumpd[5098]: breq: secs: 0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: flags: 0x
Oct  4 12:52:41 cumulus pumpd[5098]: breq: ciaddr: 0.0.0.0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: yiaddr: 0.0.0.0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: server_ip: 0.0.0.0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: bootp_gw_ip: 0.0.0.0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: hwaddr:
Oct  4 12:52:41 cumulus pumpd[5098]: breq: servername:
Oct  4 12:52:41 cumulus pumpd[5098]: breq: bootfile:
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: 0x63 0x53 0x82 0x63
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor:  53   1 0x01
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor:  12  17 0x73 0x74
0x72 0x6
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x74 0x75 0x73
0x2e
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x64 0x65 0x6d
0x6f
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x6e 0x2e 0x6e
0x6c
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x00
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: 0xff
Oct  4 12:52:41 cumulus kernel: Packet log: output DENY eth1 PROTO=17
192.168.0.
1:68 255.255.255.255:67 L=328 S=0x00 I=54009 F=0x T=64 (#6)

can someone tell me where things are going wrong?

thanks heaps,
Eskild

Re: SPAM WARNING: spammers use Debian lists for harvesting

[Bud Rogers]

On Thursday 04 October 2001 04:54 am, P Kirk wrote:

> Its my impression that there are only a few email harvesters out there
> and that they don't work very hard.  Perhaps once you have a list with a
> million names, its not worth wasting time building a new one.

I suspect there is some truth to that.

> That was over 3 years ago.  Since then I have used pknews all over
> usenet and patrick in quite a few mailing lists.  Yet over half the 20
> or so pieces of spam I get every day are to that old pkirk address which
> hasn't been used publicly in 3 years.

About half of the spam I get these days is addressed not to my real address 
but to about half a dozen addresses of the form
 
[EMAIL PROTECTED] 

always beginning with "m2" and ending with ".fsf" and not to my ISP's mail 
server but directly to my Linux box.  It took me a while to figure out they 
aren't email addresses at all.  They're usenet message ID's generated by 
gnus.  Gnus was my mail and news reader of choice for several years before I 
embraced KDE, but I haven't posted to usenet from gnus in a couple of years.

> Perhaps I need to get out more :-(

Perhaps we all do.  :-}

-- 
Bud Rogers <[EMAIL PROTECTED]>
They have awakened a sleeping giant and filled him with a terrible resolve.

DHCP PUMP trouble

[Eskild Wikkeling]

Hello,

I recently got a ADSL modem for babyxl (a dutch adsl provider that uses
dhcp to give out ip adresses) and am trying to get it to work.

in /etc/network/interfaces I have:
auto eth0 eth1

iface eth0 inet static


iface eth1 inet dhcp
  hostname MYNAME.demon.nl

and in /etc/pump.conf:
domainsearch "demon.nl"

retries 3
timeout 60

device eth1 {
}

when I run ifup eth1 I get the following message in /var/log/syslog:
Oct  4 12:52:41 cumulus pumpd[5098]: HOSTNAME: requesting
MYNAME.demon.nl
Oct  4 12:52:41 cumulus pumpd[5098]: PUMP: sending discover
Oct  4 12:52:41 cumulus pumpd[5098]: breq: opcode: 1
Oct  4 12:52:41 cumulus pumpd[5098]: breq: hw: 1
Oct  4 12:52:41 cumulus pumpd[5098]: breq: hwlength: 6
Oct  4 12:52:41 cumulus pumpd[5098]: breq: hopcount: 0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: xid: 0x45359f55
Oct  4 12:52:41 cumulus pumpd[5098]: breq: secs: 0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: flags: 0x
Oct  4 12:52:41 cumulus pumpd[5098]: breq: ciaddr: 0.0.0.0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: yiaddr: 0.0.0.0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: server_ip: 0.0.0.0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: bootp_gw_ip: 0.0.0.0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: hwaddr:
Oct  4 12:52:41 cumulus pumpd[5098]: breq: servername:
Oct  4 12:52:41 cumulus pumpd[5098]: breq: bootfile:
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: 0x63 0x53 0x82 0x63
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor:  53   1 0x01
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor:  12  17 0x73 0x74
0x72 0x6
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x74 0x75 0x73

0x2e
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x64 0x65 0x6d

0x6f
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x6e 0x2e 0x6e

0x6c
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x00
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: 0xff
Oct  4 12:52:41 cumulus kernel: Packet log: output DENY eth1 PROTO=17
192.168.0.
1:68 255.255.255.255:67 L=328 S=0x00 I=54009 F=0x T=64 (#6)

can someone tell me where things are going wrong?

thanks heaps,
Eskild

RE: NIS and initgroups

[Bristow Paul-BPB007]

Thanks but I don't think there's a problem with the group map as it's working 
with a lot of other machines (hp-ux/solaris/mandrake).  I did the "ypcat group" 
as you suggested on a working machine and it all looked normal.  When I did the 
same thing on the debian machine I got the same thing back, the group map from 
the nis server!  Given that my group line is "group: files" surely this 
shouldn't be happening?
I checked the gids as well and there all very low numbers.  Any suggestions?
Thanks,
Paul Bristow.


-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Sent: 04 October 2001 12:04
To: debian-user@lists.debian.org
Subject: Re: NIS and initgroups


In article <[EMAIL PROTECTED]>,
Bristow Paul-BPB007  <[EMAIL PROTECTED]> wrote:
>Hi,
>   I've recently installed debian on a spare PC at work and I'm having
>real problems getting NIS to work properly with groups.  I'd appreciate
>any advice on the following problem.  
>After installation I set up NIS using the nsswitch.conf file, the
>relevant lines are as follows.
>
>passwd:files nis
>shadow:files
>group: files nis
>
>After doing this I was unable to log in as root!  Whenever I tried I got
>an error message
>initgroups invalid argument

Sounds like there is a (syntax?) error in the NIS groups map.
Use 'ypcat groups' and see if you can find something obvious.

Maybe you have groups with gids > 65535? That isn't supported by
Linux unless you install a 2.4 kernel and glibc >= 2.2.

Mike.
-- 
Move sig.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: A challenge: 8,000 serial number labels

[Mario Olimpio de Menezes]

On Sun, 30 Sep 2001, Indraneel Majumdar wrote:

> Isn't there a label.sty in LaTeX ? I've used it for printing labels for
> posters a long time back using perl to generate the numbers. Have to
> look for the source though.

take a look in labels package; use texdoctk utility to browse the
non-standard document styles.

in my woody installation, the documentation is in:
/usr/share/doc/texmf/latex/styles/labels.dvi.gz


[]s,
Mario O.de Menezes"Many are the plans in a man's heart, but
IPEN-CNEN/SP is the Lord's purpose that prevails"
http://curiango.ipen.br/~mario Prov. 19.21

Upgradeing for largefile support?

[Stan Brown]

I have a machine that I installed Stable on a while back. The kernel is
2.2.19pre17.

What  do I need to do to upgrade this machine to have largefile ( >2G) file
size support?


-- 
Stan Brown [EMAIL PROTECTED]843-745-3154
Charleston SC.
-- 
Windows 98: n.
useless extension to a minor patch release for 32-bit extensions and
a graphical shell for a 16-bit patch to an 8-bit operating system
originally coded for a 4-bit microprocessor, written by a 2-bit 
company that can't stand for 1 bit of competition.
-
(c) 2000 Stan Brown.  Redistribution via the Microsoft Network is prohibited.

What's the device name for my tape drive?

[Stan Brown]

 I have a Debian stable machine with all IDE dirves. it also has an Adaptec
 AIC-7892 SCSi controler. This controler has one slave device, a HP C1537A
 4mm tape changer. It's at SCSI ID 5.

 What is it's device name? /dev/??

-- 
Stan Brown [EMAIL PROTECTED]843-745-3154
Charleston SC.
-- 
Windows 98: n.
useless extension to a minor patch release for 32-bit extensions and
a graphical shell for a 16-bit patch to an 8-bit operating system
originally coded for a 4-bit microprocessor, written by a 2-bit 
company that can't stand for 1 bit of competition.
-
(c) 2000 Stan Brown.  Redistribution via the Microsoft Network is prohibited.

How to generate a random number?

[Liu Tao]

I use random() in my program to generate a random number,
but every time when I restart my program I get the same number.
How can I get different random numbers each time?

--
Regards
Liu Tao

Q about autofs

[Bill Benedetto]

Good day.

Under HP-UX, SunOS, and IRIX, I could have an entry in my
/etc/auto.master file like this:

  /net -hosts

which would allow me to access any exported filesystem from a
remote system by accessing

  /net/REMOTE_MACHINE/EXPORT_FILESYSTEM

For example: /net/rem123/export/home

Unfortunately I haven't been able to find that same functionality
in autofs.  Is it there and I have just missed it (I hope I hope
I hope)?

Any guidance appreciated.

Thanks!

- Bill
+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
Bill Benedetto <[EMAIL PROTECTED]>The Goodyear Tire & Rubber Co.
I don't speak for Goodyear and they don't speak for me.  We're both happy.

Re: dual boot with Win 2000?

[Rino Mardo]

On Wed, Oct 03, 2001 at 10:23:37PM -0400, dman wrote:
> I too have set up GRUB to effortlessly dual-boot win2k and linux.  I
> breifly skimmed some of the other replies and saw suggestions
> reminiscent of the docs on linuxdoc.org :  do this with lilo, then do
> this with the windows loader, then finish doing this with lilo.
> While I don't want to condem lilo, I have never had any success with
> lilo -- either dual-booting win2k or simply booting linux on my
> previous machine (except via floppy).  With GRUB I have had no
> problems, and I find the configuration to be much easier.  You don't
> even have to touch window's boot setup to boot it from grub.
> 
i use debian2.2r0 and the stock lilo.  i don't have problems installing
or using lilo or linux with win2k.  only thing you have to remember is
install the jealous os first on the first partition, then linux on the
remaining partition, then install lilo on the mbr and let it control
dual booting between win/95/98/2k and linux.

-- 
"GUIs normally make it simple to accomplish simple actions and impossible
to accomplish complex actions."   --Doug Gwyn  (22/Jun/91 in comp.unix.wizards)


pgpoOoHmll9dF.pgp
Description: PGP signature

Re: downloading debian

[Alexander Wallace]

Well isnt the ReiserFS NetInst just that? I downloaded an iso and burned a
cd

On 4 Oct 2001, csj wrote:

> On Thu, 2001-10-04 at 03:49, Vineet Kumar wrote:
> > * kumar suresh ([EMAIL PROTECTED]) [011003 07:13]:
> > > Hello,
> > >
> > > We tried to download debian cdimages as instructed by
> > > the latest psudo-image kit but from most of the sites
> > > we were unable to download the required files. With
> > > many sites, the make-psudo-imaage command terminated
> > > immediately and with someother sites we could donwload
> > > only upto 21MB or so. Thereafterwards,
> > > make-psudo-imaage command returned the prompt
> > > immediately.
> > >
> > > So is there any site from where we can get latest
> > > debian?
> >
> > See www.bradmont.net (which, unfortunately, seems to be down at the
> > moment). I'll burn and mail you one for the cost of the disc + shipping,
> > mail me privately if interested. (potato discs only right now.)
>
> Wouldn't it be a brilliant idea if someone came up with an Essential
> Debian ISO image, containing only the absolute minimum files necessary
> to have a working apt-get/dselect system, (e.g. the boot image,
> base*tgz, drivers.tgz)? Low-bandwidth folks can burn the mini-ISO
> (~30MB) image to CDRW or a mini CDR and just apt-get the .debs they
> really need (which for most people should be under 300MB)
>
> --
> Neo:
> I don't know the future. I didn't come here to tell you how this is
> going to end. I came here to tell how it's going to begin. I'm going to
> hang up this phone, and then show these people what you don't want them
> to see. I'm going to show them a world without you.
>
>
> --
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>
>

test

[Marko Filipovic]

test

Re: esd esdctl?

[Robert L. Harris]

Yeah, esdctl was in esound-clients.

Now if I can get xmms, esd and vmware to play nicely together this will
be great.

Thus spake Vineet Kumar ([EMAIL PROTECTED]):

> * Robert L. Harris ([EMAIL PROTECTED]) [011003 10:32]:
> > 
> > 
> > Debian sid box running gnome.  I have esound installed and esound-common.
> > When I do a "ps -eaf | grep esd" it doesn't find anything.  I also can't 
> > find
> > which package contains esdctl which used to be the way to start and stop 
> > the esd daemon.
> 
> 
> 
> 
> > 
> > Anyone know which package?
> 
> /usr/bin/esdctl is part of esound-clients
> 
> I have esound, esound-common, and esound-clients, and It Works For Me.
> If just having those packages doesn't start esound under gnome
> automatically for you, check gnome's sound configuration, and ensure
> that the "start sound server" checkbox is selected. If it's still not
> working, ask again and I'll dig through my config files to try to find
> out where it's started from. (My initial searches returned nothing
> useful).
> 
> good times,
> 
> -- 
> Vineet   http://www.anti-dmca.org
> Unauthorized use of this .sig may constitute violation of US law.
> echo Qba\'g gernq ba zr\! |tr 'a-zA-Z' 'n-za-mN-ZA-M'





:wq!
---
Robert L. Harris|  Micros~1 :  
Senior System Engineer  |For when quality, reliability 
  at RnD Consulting |  and security just aren't
\_   that important!
DISCLAIMER:
  These are MY OPINIONS ALONE.  I speak for no-one else.
FYI:
 perl -e 'print $i=pack(c5,(41*2),sqrt(7056),(unpack(c,H)-2),oct(115),10);'

Re: resiserfs on woody install?

[Robert L. Harris]

Ok,
  where can I find info on ext3?  I don't see it in the kernel.  Can ext2 be
upgraded to ext3 once the machine is installed and I'm upgrading to 2.4.X?
I take it I have to patch the kernel for ext3 as I don't see it as an option 
in my 2.4.10 kernel config.

Thus spake A.R. (Tom) Peters ([EMAIL PROTECTED]):

> On Wed, 3 Oct 2001, Robert L. Harris wrote:
> 
> > I am rebuilding a box.  I have the woody floppies, I've gotten to the 
> > point of initializing the root filesystem.  Interestingly there is a
> > "mkreiserfs" available and it can make the filesystem, but doesn't seem
> > to be able to mount it.  I've tried to install the modules but it won't
> > install until root is mounted  
> > 
> > How do I mount the reiserfs filesystem so I can install on a reiserfs
> > cleanly?
> 
> Theoretically, this should work iff you have ReiserFS support built into
> your kernel.  Having it in a module is a pain, since you have to use a
> root floppy and|or an initrd which has the module tools and the modules
> and mount.
> 
> It has been advised to keep your root (and possibly /var) as ext2fs, so
> that use an ordinary kernel on a rescue disk (i.e. without ReiserFS) to
> repair your system.  If you want a journalling FS, you can easily put a
> journal on / and make it ext3fs.
> 
> I have no direct experience with your situation, because I also put my
> Reiserfs on logical volumes.  Theoretically it should be possible to put
> your root FS on a logical volume (again using an initrd and an init that
> keeps running while you chroot to your proper FS), but after 2 weeks I
> gave up.
> 
> --
> #>!$!%(@^%#%*(&([EMAIL 
> PROTECTED]@^$##*#@&(%)@**$!(&!^(#((#&%!)%*@)(&$($$%(@#)&*!^$)[EMAIL 
> PROTECTED]@)
> 
>   Tom "thriving on chaos" Peters
>   NL-1062 KD nr 149   tel.+31-204080204
>   Amsterdam   e-mail  [EMAIL PROTECTED]



:wq!
---
Robert L. Harris|  Micros~1 :  
Senior System Engineer  |For when quality, reliability 
  at RnD Consulting |  and security just aren't
\_   that important!
DISCLAIMER:
  These are MY OPINIONS ALONE.  I speak for no-one else.
FYI:
 perl -e 'print $i=pack(c5,(41*2),sqrt(7056),(unpack(c,H)-2),oct(115),10);'

Re: How to generate a random number?

[John Patton]

On Thu, Oct 04, 2001 at 09:21:34PM +0800, Liu Tao wrote:
> I use random() in my program to generate a random number,
> but every time when I restart my program I get the same number.
> How can I get different random numbers each time?

Look at srandom() to seed the generator. You can seed it with the
current time, or with a couple of bytes from /dev/urandom, and so
forth.

-- 
John Patton  [EMAIL PROTECTED]

"They couldn't hit an elephant at this dist..."
- General John B. Sedgwick Last words, 1864

Re: directly installing a deb package

[Colin Watson]

On Wed, Oct 03, 2001 at 10:51:15PM +0100, Alex Hunsley wrote:
> I've tried using dpkg as described, but there are problems:
> 
> pylori:~/debs# dpkg -i xaos_3_0-23_i386.deb
> Selecting previously deselected package xaos.
> (Reading database ... 16818 files and directories currently installed.)
> Unpacking xaos (from xaos_3_0-23_i386.deb) ...
> dpkg: dependency problems prevent configuration of xaos:
>  xaos depends on aalib1 (>= 1.2); however:
>   Package aalib1 is not installed.
>  xaos depends on libc6 (>= 2.2.2-2); however:
>   Version of libc6 on system is 2.1.3-18.

You're trying to install a package from testing/unstable on a stable
system. This isn't recommended; I'd suggest you build it from source
instead.

-- 
Colin Watson  [EMAIL PROTECTED]

RE: DHCP problems with cable modem

[David Priban]

> > > Hi.  I just installed Debian last night on my system,
> > > and everything works great except the network.  I have
> > > a cable modem, and the DHCP server requires that I
> > > send in my hostname with my request.  So far, I have
> > > tried (and failed at using):


I've got DHCP working on @home cable without a problem.
I'm usind dhcp-client with folowing in /etc/dhclient.conf:

   send host-name "whateveryournameis";
   supersede domain-name "yourdomainname";
   supersede domain-name-servers 127.0.0.1;

Supersede lines are optional. You can use it to override
some settings from DHCP server.
Interface configuration in /etc/network/interfaces is:

   auto eth1
   iface eth0 inet dhcp

Ifup command starts dhcp-client and the interface is configured
via dhclient-script according info recieved from DHCP server.

I hope this helps.

CheersDavid

Re: kernel-source-2.4.9-1 builds but does not boot

[Russell Coker]

On Thu, 4 Oct 2001 01:21, José Luis Rey wrote:
>make dep
>make
>make modules
>make modules_install
>make install

I strongly recommend that you install the package "kernel-package" and use 
make-kpkg to build the kernel.  This builds nice Debian packages of the 
kernel which make it a lot easier to manage and track which machines are 
running which versions.  Also the kernel package installation takes care of 
making the initrd.

>mkinitrd -o /boot/initrd-2.4.9-test /lib/modules/2.4.9-test

Hmm.  That doesn't look like the right command to get a fully functional 
initrd, also it puts all the kernel's modules into the initrd which isn't 
necessary (and if your kernel settings are like mine you'll get a 2M initrd 
image).

I've written some scripts to help manage this.  I've attached two scripts for 
inclusion in /etc/mkinitrd/scripts directory (make sure you don't run the 
devfs script in any other way), and the mkinitrd.conf file I use.  With that 
and the correct /etc/mkinitrd/modules file it should all work fine.

> and added a test entry in lilo.conf with:
>image=/vmlinuz.test
> label=Test
> read-only
> initrd=/boot/initrd-2.4.9-test
>
> The problem is that after initrd is loaded kernel panics with:
>Kernel panic: VFS: unable to mount root fs on 03:03
>
> I'm prety shure that my root file system is in /dev/hda3, so I ran rdev
> on two different kernel-images:

Well if your kernel is on /dev/hda3 then it is 03:03 so that part is OK.

> I think this had to do with the devfs thing, so I re-made the kernel
> without DEVFS support and ext2 in the kernel and it booted.
>
> Does anybody knows what happens with devfs, rdev or new kernel booting
> procedure ?, did I miss something from the docs ?

Firstly if you suspect a problem in a package then probably the best option 
is to contact the package maintainer (in the case of devfsd it's me).

Devfsd is necessary on all systems running devfs, and most systems won't 
complete the boot process without devfsd running.  However it won't affect 
the ability to mount root.

When you say "re-make the kernel without DEVFS support and ext2 in the 
kernel" are you saying that you make it with "ext2 in the kernel" or "without 
DEVFS support and ext2"?

I think that what you are doing with ext2 is what makes the difference...


I've CC'd this message to the debian-user list which is more appropriate for 
this discussion.  I've BCC'd debian-devel so people can see that it's moved 
to the user list.

Please CC me on all further discussion.

-- 
http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark
http://www.coker.com.au/postal/   Postal SMTP/POP benchmark
http://www.coker.com.au/projects.html Projects I am working on
http://www.coker.com.au/~russell/ My home page


copy-needed-modules
Description: copy-needed-modules


devfs
Description: devfs
# /etc/mkinitrd/mkinitrd.conf:
#  Configuration file for mkinitrd(8).  See mkinitrd.conf(5).
#
# This file is meant to be parsed as a shell script.

# What modules to install.
MODULES=none

# The length (in seconds) of the startup delay during which linuxrc may be
# interrupted.
DELAY=0

# If this is set to probe mkinitrd will try to figure out what's needed to
# mount the root file system.  This is equivalent to the old PROBE=on setting.
ROOT=probe

# This controls the permission of the resulting initrd image.
UMASK=022

# Turn this on to see the messages from mkcramfs.
VERBOSE=on

Re: What to choose

[Russell Coker]

On Wed, 3 Oct 2001 22:22, Teppo Hytönen wrote:
>   What comes to choosing between the two, it's personal preference that
> matters. I myself recommend Debian: I love it myself, and yes, apt-get is
> great. Then again, I haven't used Mandrake, but haven't heard a single
> positive comment about it, other than that it is easy to install; many say
> too easy, so that you can't configure things you might want even if you
> have enough skill to do it. I've heard that in many cases it doesn't work
> right.

I think you're being a bit hard on them.  I've heard positive reports about 
Mandrake, and the Mandrake developers I've talked to seem quite smart.

However I get the impression that Debian has more smart developers than any 
other distribution, and I think that upstream maintainers often have a 
similar opinion.  I recently received an email from an upstream author saying 
"I didn't know any distribution had included my code, I might have known 
Debian would do it first", I think that is an indication of the good opinion 
upstream authors have of Debian developers.

I think that you can compare distributions without using them.  If you know 
who develops a distribution, who pays them (if anyone), what their aims are, 
and how long they have been at it then you can get a good idea of what 
product they will develop and whether it will suit you.

Debian has a large team of people who work for fun (most of them get paid 
nothing for their work - many of them are doing Debian work instead of doing 
paid work).  The Debian developers are generally highly skilled by any 
standards of measurement.  The aim of the project is to develop the best 
possible OS according to the general aims of the FSF, where "best" is 
measured by the developers themselves.  The result is that there is a huge 
number of developers (any skilled person who has the time is welcome), a huge 
number of packages (any developer can add a new package at short notice 
without asking for permission), a good solid base, and a lot more work is 
needed on installation programs and documentation (no offense to the people 
who work on it - really they need more help from the rest of us).


So we can advocate Debian without mentioning other distributions.  Once they 
know what Debian is about they'll either like it or they won't.  Other 
distributions have much better installation routines, and may be more 
suitable for novices for that reason.

I've CC'd this message to debian-user as it really has nothing to do with 
laptops.

-- 
http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark
http://www.coker.com.au/postal/   Postal SMTP/POP benchmark
http://www.coker.com.au/projects.html Projects I am working on
http://www.coker.com.au/~russell/ My home page

Purging deb

[Jose Manuel Perez]

Hi all.
I've installed a package in my Potato but it crashed during
installation and I'd like to uninstall it. I've tried
dpkg --purge , with and without --force-??? option, but
I've not found the way to remove it. Allways I've got:

siul04:~# dpkg --purge lgtoclnt
(Leyendo la base de datos ... [reading database ...]
44862 ficheros y directorios instalados actualmente.) [4862 files and 
directories installed]
Desinstalando lgtoclnt ... [Deinstalling lgtoclnt ...]
rmdir: número de argumentos insuficiente [rmdir: insuficient number of 
arguments ]
Pruebe `rmdir --help' para más información. [Try ?? for more information]
dpkg: error al procesar lgtoclnt (--purge): [dpkg: error processing ...]
 el subproceso post-removal script devolvió el código de salida de error 1
[subprocess post-removal script returns error code 1]
Se encontraron errores al procesar: [errors where found procesing:]
 lgtoclnt

I have to say that this package was originally a rpm, and
I've converted it to deb. Someone knows how to remove that
package.

Thanks in advance,
-- 
--
José Manuel Pérez - Hispavista::Sistemas
P. Mikeletegi, 56 - 20009 Donostia - +34 943 448400
[EMAIL PROTECTED] - www.hispavista.com

Re: Purging deb

[Alexander Wallace]

try apt-get remove lgtoclnt --purge

On Thu, 4 Oct 2001, Jose Manuel Perez wrote:

> Hi all.
> I've installed a package in my Potato but it crashed during
> installation and I'd like to uninstall it. I've tried
> dpkg --purge , with and without --force-??? option, but
> I've not found the way to remove it. Allways I've got:
>
> siul04:~# dpkg --purge lgtoclnt
> (Leyendo la base de datos ... [reading database ...]
> 44862 ficheros y directorios instalados actualmente.) [4862 files and
> directories installed]
> Desinstalando lgtoclnt ... [Deinstalling lgtoclnt ...]
> rmdir: número de argumentos insuficiente [rmdir: insuficient number of
> arguments ]
> Pruebe `rmdir --help' para más información. [Try ?? for more information]
> dpkg: error al procesar lgtoclnt (--purge): [dpkg: error processing ...]
>  el subproceso post-removal script devolvió el código de salida de error 1
> [subprocess post-removal script returns error code 1]
> Se encontraron errores al procesar: [errors where found procesing:]
>  lgtoclnt
>
> I have to say that this package was originally a rpm, and
> I've converted it to deb. Someone knows how to remove that
> package.
>
> Thanks in advance,
> --
> --
> José Manuel Pérez - Hispavista::Sistemas
> P. Mikeletegi, 56 - 20009 Donostia - +34 943 448400
> [EMAIL PROTECTED] - www.hispavista.com
>
>
> --
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>
>

Re: Voodoo3. DRI, X4.1.0 and Bus mastering <- Still no joy

[Jason Healy]

At 1002229481s since epoch (10/04/01 02:04:41 -0400 UTC), john wrote:

> It's interesting that you say you have DRI working but not bus mastering. Can
> you run setpci on the device and see if bit 3 of word 4 is set? If so my 
> problem
> isn't bus mastering at all.

Here's my output from `glxinfo` (truncated).  As you can see, DRI is on.

name of display: :0.0
display: :0  screen: 0
direct rendering: Yes
server glx vendor string: SGI
server glx version string: 1.2
server glx extensions:
GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_EXT_import_context
client glx vendor string: SGI
client glx version string: 1.2
client glx extensions:
GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_EXT_import_context
GLX extensions:
GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_EXT_import_context
OpenGL vendor string: VA Linux Systems, Inc.
OpenGL renderer string: Mesa DRI Voodoo3 20010501 x86/MMX
OpenGL version string: 1.2 Mesa 3.4.2
OpenGL extensions:
GL_ARB_multitexture, GL_ARB_transpose_matrix, GL_EXT_abgr, 
GL_EXT_clip_volume_hint, GL_EXT_compiled_vertex_array, GL_EXT_histogram, 
GL_EXT_packed_pixels, GL_EXT_paletted_texture, GL_EXT_polygon_offset, 
GL_EXT_rescale_normal, GL_EXT_stencil_wrap, GL_EXT_texture3D, 
GL_EXT_texture_env_add, GL_EXT_texture_object, GL_EXT_texture_lod_bias, 
GL_EXT_vertex_array, GL_HP_occlusion_test, GL_MESA_window_pos, 
GL_MESA_resize_buffers, GL_NV_texgen_reflection, GL_PGI_misc_hints, 
GL_SGI_color_matrix, GL_SGI_color_table, GL_SGIS_pixel_texture, 
GL_SGIS_texture_edge_clamp, GL_SGIX_pixel_texture
glu version: 1.3
glu extensions:
GLU_EXT_nurbs_tessellator, GLU_EXT_object_space_tess


Here's the output of `lspci -vv` (just for my Voodoo3 card):

01:00.0 VGA compatible controller: 3Dfx Interactive, Inc. Voodoo 3 
(rev 01) (prog-if 00 [VGA])

   Subsystem: 3Dfx Interactive, Inc. Voodoo3 AGP

   Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- 
Stepping- SERR- FastB2B-

   Status: Cap+ 66Mhz+ UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- 
   SERR- 

I think that says that I don't have BusMastering on (That's what the little
minus sign "-" means, right?  I'm not a PCI expert, so you'll have to help
me on this one).  Let me know if I should run {ls,set}pci with different
options to give you better output.

This is a new motherboard, but even on my old crappy motherboard everything
worked the same, so I'm pretty sure that I have it working without bus
mastering.

Jason

--
Jason Healy| [EMAIL PROTECTED]
LogN Systems   |   http://www.logn.net/

Re: OpenGL causes hard system lockups

[Dave Sherohman]

On Wed, Oct 03, 2001 at 11:46:03PM -0500, Matthew Garman wrote:
> I bought a Matrox Millenium g450 over the summer.  I got it setup to do
> hardware accelerated 3d with XFree 4.x.
> 
> The only real 3d apps I've used are the "glxgears" test program, and the
> 3d screensavers that can be built with the xscreensaver package.
> 
> Some of these 3D screensavers will unexpectedly lock up my system---and I
> mean *completely* lock up my system, can't ctrl-alt-f[1-6] to switch to a
> text console, I can't kill X with ctrl-alt-backspace, etc.  Nothing works.

I doubt that your system is completely frozen.  More likely, it's
"just" the console.  If you have another box handy, you'll probably
be able to ssh in and do an orderly reboot that way.

Which brings me to what I've been wondering:

Once I've sshed into my G400/DRI box after the console has locked up,
is there any way to recover short of rebooting?  I suspect that it
should be OK if I can come up with a way to reset DRI or agpgart or
something, but I have yet to find any information on which something
to reset or how to do it.  (I know that the obvious solution of
killing and restarting X doesn't help...)

-- 
When we reduce our own liberties to stop terrorism, the terrorists
have already won. - reverius

Re: A simple question about wildcards with tar

[Yvonne Kelly]

-Original Message-
From:Walter Landry [EMAIL PROTECTED]
Sent:Wed, 03 Oct 2001 13:57:58 -0700
To:  debian-user@lists.debian.org
Subject: A simple question about wildcards with tar


Greetings,

I'm trying to make a backup with tar, but there are certain 
files that
I don't want to include in the backup.  Reading the info 
documentation
about tar, it says that I can use the --exclude=PATTERN 
option.  So if
I type

  tar -cvf backup.tar --exclude='*.fig' *

then it excludes all files that end with .fig.  However, if 
I also
want to exclude files that end with .fig.bak, then it
seems that

  tar -cvf backup.tar --exclude='*.fig*' *

should work.  But it doesn't.  Typing 

  tar -cvf backup.tar --exclude='*.fig*' --
exclude='*.fig' *

does work.  But it won't work for me in the general case, 
where I
won't know how many characters follow the pattern that I'm 
matching.
Does anyone know how to do this?

Thanks,
Walter Landry
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to debian-user-
[EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact 
[EMAIL PROTECTED]



How about:


tar -cvf backup.tar --exclude='*.fig*' --exclude='*.fig?*' *

If  matches exactly four characters, then ?* should 
match any string of at least one, am I right?

Y.Kelly


___
Visit http://www.visto.com.
Find out  how companies are linking mobile users to the 
enterprise with Visto.

IRC DCC through a firewall

[Andrew Pritchard]

I've got a Debian firewall setup, which is working ok. I can DCC receive files, 
but I can't DCC send. The ip_masq_irc module is installed on the firewall. If I 
try to send, it starts trying to send, the receiver gets the right IP address, 
and sends the acknowledgement, but the transfer never starts. Have I 
misconfiguered something or is this by design?

Cheers,

Andrew

"I do not agree with what you say,
but I will defend to the death your right to say it." 
Francois Marie Arouet Voltaire (1694-1778)

Re: IRC DCC through a firewall

[dman]

On Thu, Oct 04, 2001 at 03:58:21PM +0100, Andrew Pritchard wrote:
| I've got a Debian firewall setup, which is working ok. I can DCC
| receive files, but I can't DCC send. The ip_masq_irc module is
| installed on the firewall. If I try to send, it starts trying to
| send, the receiver gets the right IP address, and sends the
| acknowledgement, but the transfer never starts. Have I
| misconfiguered something or is this by design?

Are you using ipchains or iptables?  Also, what sort of setup does
your ISP have?

I tried to DCC send to someone once, we both have Frontier Lightning
Link and are on different masqueraded subnets.  The connection was
never established.  I was able to do a DCC chat with someone else who
has a decent ISP (no masq-ing) one time.

-D

Re: Voodoo3. DRI, X4.1.0 and Bus mastering <- Still no joy

[Rich Rudnick]

On Thu, 2001-10-04 at 07:31, Jason Healy wrote:
> At 1002229481s since epoch (10/04/01 02:04:41 -0400 UTC), john wrote:
> 
> > It's interesting that you say you have DRI working but not bus mastering. 
> > Can
> > you run setpci on the device and see if bit 3 of word 4 is set? If so my 
> > problem
> > isn't bus mastering at all.
> 
> 
> I think that says that I don't have BusMastering on (That's what the little
> minus sign "-" means, right?  I'm not a PCI expert, so you'll have to help
> me on this one).  Let me know if I should run {ls,set}pci with different
> options to give you better output.
> 

As well, here's an lspci report on my video: as you can see, the 3dfx
mentions nothing about bus mastering.

Bus  0, device  19, function  0:
VGA compatible controller: 3Dfx Interactive, Inc. Voodoo 3 (rev 1).
  IRQ 11.
  Non-prefetchable 32 bit memory at 0xfc00 [0xfdff].
  Prefetchable 32 bit memory at 0xf600 [0xf7ff].
  I/O at 0xe800 [0xe8ff].
Bus  1, device   0, function  0:
Display controller: ATI Technologies Inc 3D Rage Pro AGP 1X/2X (rev 92).
  Master Capable.  Latency=32.  Min Gnt=8.
  Non-prefetchable 32 bit memory at 0xf900 [0xf9ff].
  I/O at 0xd800 [0xd8ff].
  Non-prefetchable 32 bit memory at 0xfa8ff000 [0xfa8f].

Re: Purging deb

[Colin Watson]

On Thu, Oct 04, 2001 at 04:23:40PM +0200, Jose Manuel Perez wrote:
> I've installed a package in my Potato but it crashed during
> installation and I'd like to uninstall it. I've tried
> dpkg --purge , with and without --force-??? option, but
> I've not found the way to remove it. Allways I've got:
> 
> siul04:~# dpkg --purge lgtoclnt
> (Leyendo la base de datos ... [reading database ...]
> 44862 ficheros y directorios instalados actualmente.) [4862 files and 
> directories installed]
> Desinstalando lgtoclnt ... [Deinstalling lgtoclnt ...]
> rmdir: número de argumentos insuficiente [rmdir: insuficient number of 
> arguments ]
> Pruebe `rmdir --help' para más información. [Try ?? for more information]

The package is broken.

> I have to say that this package was originally a rpm, and
> I've converted it to deb.

Ah, that would explain it then ...

Edit /var/lib/dpkg/info/lgtoclnt.prerm and/or
/var/lib/dpkg/info/lgtoclnt.postrm, look for the rmdir call that's
failing, fix it, then try 'dpkg --purge lgtoclnt' again.

-- 
Colin Watson  [EMAIL PROTECTED]

Re: downloading debian

[Wayne Topa]

Alexander Wallace([EMAIL PROTECTED]) is reported to have said:
> Well isnt the ReiserFS NetInst just that? I downloaded an iso and burned a
> cd
> 

Wasn't aware of that one.  Thanks!  Now if he can get woody to
something of that size, I would be estatic!

Wayne
-- 
You know you've been spending too much time on the computer when your
friend misdates a check, and you suggest adding a "++" to fix it.
___

Re: Purging deb

[Jose Manuel Perez]

El Jue 04 Oct 2001 16:32, Alexander Wallace escribió:
> try apt-get remove lgtoclnt --purge
>

Thanks for your answer. I've finally purged the package
modifying postremoval script (lgtoclnt.postrm) at: /var/lib/dpkg/info/
to correct the error, well not at all, I've comented out the whole
script, but I don't mind because doing dpkg --purge  worked ok.

-- 
--
José Manuel Pérez - Hispavista::Sistemas
P. Mikeletegi, 56 - 20009 Donostia - +34 943 448400
[EMAIL PROTECTED] - www.hispavista.com

Re: dual boot with Win 2000?

[John]

On Thu, Oct 04, 2001 at 06:32:33PM +0800, Rino Mardo wrote:> 
> i use debian2.2r0 and the stock lilo.  i don't have problems installing
> or using lilo or linux with win2k.  only thing you have to remember is
> install the jealous os first on the first partition, then linux on the
> remaining partition, then install lilo on the mbr and let it control
> dual booting between win/95/98/2k and linux.
> 
I upgraded my son's dual booting box from win98 to win2k expecting the
upgrade to wipe out lilo on the mbr, but when rebooted, the lilo prompt
was still there.  
Win2k not as jealous/possesive of the mbr as win98?

John


-- 
-_-_-_-_-_-_-_-_-_-_-
 Using [Debian] Linux
_-_-_-_-_-_-_-_-_-_-_





pgpmmvTSqYiwx.pgp
Description: PGP signature

Re: How to generate a random number?

[Thomas Hallaran]

On Thu, 4 Oct 2001, Liu Tao wrote:
seed your random number generator like so:

srand(time(NULL));

before calling random.

> I use random() in my program to generate a random number,
> but every time when I restart my program I get the same number.
> How can I get different random numbers each time?
>
> --
> Regards
> Liu Tao
>
>
>

Re: OpenGL causes hard system lockups

[Andrew Agno]

Dave Sherohman writes:
 > Once I've sshed into my G400/DRI box after the console has locked up,
 > is there any way to recover short of rebooting?  I suspect that it
 > should be OK if I can come up with a way to reset DRI or agpgart or
 > something, but I have yet to find any information on which something
 > to reset or how to do it.  (I know that the obvious solution of
 > killing and restarting X doesn't help...)

I suppose you could try a few things:
- if you are using modules, unload agpgart and any other modules used
by your card, then reload them
- try running xf86config or SuperProbe (for those that it supports) to
see if they'll reset things for you
- try going to single user mode, then back to whatever runlevel you
were at; this seems least likely to do anything useful, but who knows

Andrew.

Re: DHCP PUMP trouble

[Michael Heldebrant]

On Thu, 2001-10-04 at 06:43, Eskild Wikkeling wrote:
> Hello,
> 
> I recently got a ADSL modem for babyxl (a dutch adsl provider that uses
> dhcp to give out ip adresses) and am trying to get it to work.
> 
> in /etc/network/interfaces I have:
> auto eth0 eth1
> 
> iface eth0 inet static
> 
> 
> iface eth1 inet dhcp
>   hostname MYNAME.demon.nl
> 
> and in /etc/pump.conf:
> domainsearch "demon.nl"
> 
> retries 3
> timeout 60
> 
> device eth1 {
> }
> 
> when I run ifup eth1 I get the following message in /var/log/syslog:
> Oct  4 12:52:41 cumulus pumpd[5098]: HOSTNAME: requesting
> MYNAME.demon.nl
> Oct  4 12:52:41 cumulus pumpd[5098]: PUMP: sending discover
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: opcode: 1
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: hw: 1
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: hwlength: 6
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: hopcount: 0
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: xid: 0x45359f55
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: secs: 0
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: flags: 0x
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: ciaddr: 0.0.0.0
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: yiaddr: 0.0.0.0
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: server_ip: 0.0.0.0
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: bootp_gw_ip: 0.0.0.0
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: hwaddr:
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: servername:
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: bootfile:
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: 0x63 0x53 0x82 0x63
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor:  53   1 0x01
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor:  12  17 0x73 0x74
> 0x72 0x6
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x74 0x75 0x73
> 
> 0x2e
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x64 0x65 0x6d
> 
> 0x6f
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x6e 0x2e 0x6e
> 
> 0x6c
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x00
> Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: 0xff
> Oct  4 12:52:41 cumulus kernel: Packet log: output DENY eth1 PROTO=17
> 192.168.0.
> 1:68 255.255.255.255:67 L=328 S=0x00 I=54009 F=0x T=64 (#6)
> 
> can someone tell me where things are going wrong?
> 
> thanks heaps,
> Eskild

Try echoing 1 into /proc/sys/net/ipv4/ip_dynaddr that may solve your
problem.

--mike

Re: IRC DCC through a firewall

[Andrew Pritchard]

Quoting dman <[EMAIL PROTECTED]>:
> On Thu, Oct 04, 2001 at 03:58:21PM +0100, Andrew Pritchard wrote:
> | I've got a Debian firewall setup, which is working ok. I can DCC
> | receive files, but I can't DCC send. The ip_masq_irc module is
> | installed on the firewall. If I try to send, it starts trying to
> | send, the receiver gets the right IP address, and sends the
> | acknowledgement, but the transfer never starts. Have I
> | misconfiguered something or is this by design?
> 
> Are you using ipchains or iptables?  Also, what sort of setup does
> your ISP have?
> 
> I tried to DCC send to someone once, we both have Frontier Lightning
> Link and are on different masqueraded subnets.  The connection was
> never established.  I was able to do a DCC chat with someone else who
> has a decent ISP (no masq-ing) one time.
> 
> -D

ipchains, fairly loose rules. It's a debian stable box, very very little 
running on the machine.

The ISP doesn't block anything as far as I am aware - it's a pretty good ISP, 
broadband and they aren't too fussy about people running webservers, etc off 
their connections. (No mail servers, and only 10 inbound connections allowed), 
and it is presented to us as an ethernet connection (Horrah - non of this "It's 
a consumer ISP therefore you MUST use USB!)

Andrew

Re: How to generate a random number?

[Jeffrey W. Baker]

On Thu, 4 Oct 2001, Thomas Hallaran wrote:

> On Thu, 4 Oct 2001, Liu Tao wrote:
> seed your random number generator like so:
>
> srand(time(NULL));
>
> before calling random.

That is an extraordinarily bad idea.  Any person will be able to guess the
sequence of random numbers simply by guessing the time at which your
program was started.

Use random data to seed the psuedo-random number generator.  Get that data
from /dev/random or /dev/urandom

-jwb

HELP: system hangs at boot

[Lars Jensen]

Whenever I boot up my system I get the following error message when the 
file system is being mounted:

ERROR: "cannot fsck root fs because fs is not mounted read-only"

and a little later the system hangs.

If I boot my system with the rw option like this:
boot: linux rw
the boot process appears normal, and the system doesn't hang, but I 
still get the above error. 

What suddenly caused this, I'm not sure, but does anyone know how to fix
it?

Thanks,
Lars.

%%%
Lars Jensen, Truckee Meadows Community College, Reno NV 89512-3999. 
Tel: 775.673.7113 E-mail: [EMAIL PROTECTED]

Failed Initialization of WD-7000 SCSI CARD

[Mike Towery]

I am using a Compaq ProSigna 300 with an Integrated
32bit FAST SCSI-2 Controller. I am getting the
following error message "Failed Initialization of
WD-7000 SCSI CARD" when I am booting from the rescue
disk.  I also have a Compaq CRD-254-V CDROM on ID 5 of
the SCSI.  It seems to recognize the CDROM but just
hangs when trying to boot from it with the Debian CD
is in it with a flashing "_".  This is a first time
installation of Debian on the PC.  Has anyone got any
suggestions?

Thanks,
Mike  

__
Do You Yahoo!?
NEW from Yahoo! GeoCities - quick and easy web site hosting, just $8.95/month.
http://geocities.yahoo.com/ps/info1

Re: IRC DCC through a firewall

[Fredrik Jagenheim]

On Thu, Oct 04, 2001 at 05:02:12PM +0100, Andrew Pritchard wrote:
> Quoting dman <[EMAIL PROTECTED]>:
> > I've got a Debian firewall setup, which is working ok. I can DCC
> > receive files, but I can't DCC send. The ip_masq_irc module is |
> > installed on the firewall. If I try to send, it starts trying to |
> > send, the receiver gets the right IP address, and sends the |
> > acknowledgement, but the transfer never starts. Have I |
> > misconfiguered something or is this by design?
> > 
> 
> ipchains, fairly loose rules. It's a debian stable box, very very
> little running on the machine.
> 

I'm a little unsure of the DCC protcol, but could it be that you never
see the ack?
That is, you send 'I have a file for you' over _normal_ IRC channel.
He receives this and then sends the ack to a different port on your
computer telling you 'fine, I accept that fine file you have for me,
let's use this socket-pair for the transfer'.
And as you're probably running NAT, your firewall won't know that the
port your friend is sending to should go to your IRC program, thus it
simply drops it, and you never see the ack, and the transfer doesn't
start.
As I said, I don't remember the IRC protocol, haven't been on there
for ages, but check your firewalls logging for what it drops...

Oh, you wanted to know the solution too?
Check if you can't either get a SOCKS-server running on the firewall
(I've tried it, and I can't make it work at all) or tell the IRC
program to use specific ports for DCC transfers and forward those
ports inward. For example, I've done this for accepting files through
ICQ (using iptables, but you get the idea):
iptables -t nat -A PREROUTING -i eth1 -p TCP --dport 6060 -j DNAT --to 
192.168.1.2

HTH,
//Fredde

Re: NIS and initgroups

[Miquel van Smoorenburg]

In article <[EMAIL PROTECTED]>,
Bristow Paul-BPB007  <[EMAIL PROTECTED]> wrote:
>Thanks but I don't think there's a problem with the group map as it's
>working with a lot of other machines (hp-ux/solaris/mandrake).  I did

But perhaps mandrake is using a 2.4 kernel with 32 bit uid/gid support?

>the "ypcat group" as you suggested on a working machine and it all
>looked normal.  When I did the same thing on the debian machine I got
>the same thing back, the group map from the nis server!  Given that my
>group line is "group: files" surely this shouldn't be happening?

Ofcourse that should happen. 'ypcat' doesn't look at /etc/passwd
or /etc/group or /etc/nsswitch.conf, it just transfers a NIS
map from a NIS server.

>I checked the gids as well and there all very low numbers.  Any suggestions?

Yep - login as root on a VT, then change /etc/nsswitch.conf to
use NIS. See under what pid another getty is running on a VT, and
strace it (something like strace -f -p put_pid_here -s512 -o login.trace).
Then login on that VT and let the error occur.

After that you should have a nice trace of what happened in
"login.trace". Possibly you can see what went wrong right away-
if not, send me the trace.

NOTE: your root password may be in the trace, edit it out or
trust me, but DO NOT send the trace to this list.

Or change the root password to 'fnord' just for the trace session.

Mike.
-- 
Move sig.

Re: Upgradeing for largefile support?

[Miquel van Smoorenburg]

In article <[EMAIL PROTECTED]>,
Stan Brown <[EMAIL PROTECTED]> wrote:
>I have a machine that I installed Stable on a while back. The kernel is
>2.2.19pre17.
>What  do I need to do to upgrade this machine to have largefile ( >2G) file
>size support?

You need to install a 2.4 kernel and the 'unstable' distribution.

Mike.
-- 
Move sig.

Fwd: Re: IRC DCC through a firewall

[Andrew Pritchard]

Quoting Fredrik Jagenheim <[EMAIL PROTECTED]>:

> On Thu, Oct 04, 2001 at 05:02:12PM +0100, Andrew Pritchard wrote:
> > Quoting dman <[EMAIL PROTECTED]>:
> > > I've got a Debian firewall setup, which is working ok. I can DCC
> > > receive files, but I can't DCC send. The ip_masq_irc module is |
> > > installed on the firewall. If I try to send, it starts trying to |
> > > send, the receiver gets the right IP address, and sends the |
> > > acknowledgement, but the transfer never starts. Have I |
> > > misconfiguered something or is this by design?
> > >
> >
> > ipchains, fairly loose rules. It's a debian stable box, very very
> > little running on the machine.
> >
>
> I'm a little unsure of the DCC protcol, but could it be that you never
> see the ack?
> That is, you send 'I have a file for you' over _normal_ IRC channel.
> He receives this and then sends the ack to a different port on your
> computer telling you 'fine, I accept that fine file you have for me,
> let's use this socket-pair for the transfer'.
> And as you're probably running NAT, your firewall won't know that the
> port your friend is sending to should go to your IRC program, thus it
> simply drops it, and you never see the ack, and the transfer doesn't
> start.
> As I said, I don't remember the IRC protocol, haven't been on there
> for ages, but check your firewalls logging for what it drops...
>
> Oh, you wanted to know the solution too?
> Check if you can't either get a SOCKS-server running on the firewall
> (I've tried it, and I can't make it work at all) or tell the IRC
> program to use specific ports for DCC transfers and forward those
> ports inward. For example, I've done this for accepting files through
> ICQ (using iptables, but you get the idea):
> iptables -t nat -A PREROUTING -i eth1 -p TCP --dport 6060 -j DNAT --to
> 192.168.1.2
>
> HTH,
> //Fredde

Yes that very much sounds like the problem - but isn't that supposed to be
handled by the ip_masq_irc module? I'm also using 2.2.19 kernel, so it's
IPchains. *sigh* don't really want to start using a socks proxy on the
firewall, but I will if I have to.

Andrew

"I do not agree with what you say,
but I will defend to the death your right to say it." 
Francois Marie Arouet Voltaire (1694-1778)

I nedd help with coccon !!

[debian]

Hi, I have been trying to install cocoon under jserv+apache but I don\'t see 
any 
progress with this 

I just have make all of what the How-To said... but I just don\'t get the 
http://127.0.0.1/Cocoont.xml  to work...

I am using woody.

Please help...

Re: fetchmail error

[Volker Schlecht]

* dman <[EMAIL PROTECTED]> [011002 22:44]:

> What are you using as the MDA in fetchmail?  I simply
> use '/usr/sbin/exim %T' and it delivers locally (via procmail).

I've used the same for a while, but that sometimes resulted in strange
and annoying delivery errors, that's why I switched to handling this
over SMTP.

> It
> appears that you are using fetchmail to perform and SMTP transfer for
> each message that it retrieves from the POP server and that is taking
> too long. 

It usually doesn't take noticeably longer than using the mda option
(that is, I as the user don't notice it).

I'm of course not sure that the problem is not with exim, but since my
fetchmail version has changed whereas exim stayed the same, I look in
that direction primarily. 

Also, while fetchmail was restarted when I re-connected to my ISP,
exim wasn't. (A restart of fetchmail alone did nothing to solve the
original problem, though)

> I have noticed, however, that sometimes when I send a
> message from mutt, it "hangs" for longer than I think it should while
> it says "sending message".  I imagine that is caused by something with
> exim acting a bit slower sometimes.  Maybe you are having a similar
> problem?

>From what I know, mutt invokes /usr/sbin/sendmail when sending a mail,
that should be unrelated to the SMTP connection that fetchmail
establishes?

regards,
Volker

-- 
In order to get a loan you must first prove you don't need it.

PHP Compilation error

[Paulo Henrique Baptista de Oliveira]

Hi all,

I'm trying to compile php4 on a potato machine from deb srcs to
install php interbase support.
The following error occurs when running debian/rules binary
Any hints?
Thanks, Paulo Henrique


#install the apache modules' files
for i in gd ldap mhash mysql pgsql snmp xml interbase ; do \
install -m 644 -o root -g root \
apache-build/ext/$i/.libs/$i.so \
debian/php4-$i/usr/lib/php4/apache/$i.so; \
done
install: apache-build/ext/ldap/.libs/ldap.so: Arquivo ou diretório não
encontrado
install: apache-build/ext/mhash/.libs/mhash.so: Arquivo ou diretório
não encontrado
install: apache-build/ext/mysql/.libs/mysql.so: Arquivo ou diretório
não encontrado
install: apache-build/ext/pgsql/.libs/pgsql.so: Arquivo ou diretório
não encontrado
install: apache-build/ext/snmp/.libs/snmp.so: Arquivo ou diretório não
encontrado
install: apache-build/ext/xml/.libs/xml.so: Arquivo ou diretório não
encontrado
install: apache-build/ext/interbase/.libs/interbase.so: Arquivo ou
diretório não encontrado
make: ** [install] Erro 1
condor:/usr/local/src/php4/php4-4.0.3pl1# 



-- 
Paulo Henrique B de Oliveira
Gerente de Operações - Linux Solutions -
http://www.linuxsolutions.com.br
O maior conteúdo de Linux em língua portuguesa - OLinux -
http://www.olinux.com.br
(21) 2526-7262 ramal 31

Characters can't be shown correctly on tty

[Liu Tao]

Hi
I typed "cat /dev/random" on tty1, and some strange characters scrolled ,
so I pressed Ctrl-c to terminate cat. But after that, characters can't be 
shown correctly on tty1, they are shown as strange characters.
Other ttys are normal. I tried "cat /dev/random" on Konsole, and it doesn't
have this problem.

Does that mean there are exploits  in ttys' driver code?

--
Regards
Liu Tao

X related

[Jeffrin Jose T .]

 Is  there any freely downloadable X window system administration guide
 available ? 

-- 
Jeffrin Jose T.
www.MSServices.org
GPG:1024D/F5726A1B

Re: setting a tty to login to remote computer via ssh

[Shriram Shrikumar]

--- Mike Alborn <[EMAIL PROTECTED]> wrote:
> init uses getty (or some variant - check /etc/inittab) to spawn a
> terminal, and getty spawns /bin/login to produce the prompt and handle
> user logins. getty can be instructed to spawn a different program with
> the -l option. Perhaps if you put an entry like this in /etc/inittab:
> 
> 8:23:respawn:/sbin/getty -l /usr/bin/ssh hostname 38400 tty8
> 
> where hostname is the remote host. According to the manpages, getty
> will pass the name from the login: prompt on to ssh (since /bin/login
> works in much the same way ssh does - for this purpose).

I tried that but getty just dies after getting the username and respawns asking
for it again. openvt [EMAIL PROTECTED] works fine though but that is not exactly
what I want. Is there any way of seeing what is going on ? Is there any way of
getting getty & ssh to display verbose messages ?

is there a way of logging onto the remote machine via the network other than
through ssh ?

any thoughts ?

Shri


__
Do You Yahoo!?
NEW from Yahoo! GeoCities - quick and easy web site hosting, just $8.95/month.
http://geocities.yahoo.com/ps/info1

LAPTOPS AND TWO DIFFERENT WORKPLACES (OT)

[D.]

Hi all,
  The system administrator at work is a subcontract
employee.  He is also the administrator at his
company.  The company that I work for also has
employe's from his company working here and they
travel back and forth between companies and use their
M$ laptops at both sites.  Here is the problem.. both
sites use M$ and my company  has the DHCP set and the
other company doesn't, so when these people go to
their company they have to go and set the nic card up.
  He asked me if I knew of a way to do this and I have
searched the web and came up dry... I think that a lot
of the people that are on this list are also very
knowledgable about M$ also.  Is there a way to do this
with out setting his company up to use DHCP?
Thanks,
  Appreciate all the help that I can pass on.
Don

__
Do You Yahoo!?
NEW from Yahoo! GeoCities - quick and easy web site hosting, just $8.95/month.
http://geocities.yahoo.com/ps/info1

Re: How to generate a random number?

[Alan Shutko]

"Jeffrey W. Baker" <[EMAIL PROTECTED]> writes:

> That is an extraordinarily bad idea.  Any person will be able to guess the
> sequence of random numbers simply by guessing the time at which your
> program was started.

And the impact of this depends on what the program is used for.  If
you're seeding the RNG for a crypto program, yes, this is bad.  If
you're seeding the RNG for a random sig generator, who cares?  Why use
up entropy for programs which don't need it?

-- 
Alan Shutko <[EMAIL PROTECTED]> - In a variety of flavors!
The grass is always greener on the other side of your sunglasses.

NIC problems

[Nils-Erik Svangard]

Hi!
I have a Realtek NIC, this is the output of cat /dev/pci
 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139 (rev 16).
  IRQ 10.
  Master Capable.  Latency=32.  Min Gnt=32.Max Lat=64.
  I/O at 0xcc00 [0xccff].
  Non-prefetchable 32 bit memory at 0xd581 [0xd58100ff].
When I try to ping a computer on the outside I get ping times rangeing
from 0.0 ms to 10ms
output of ping
64 bytes from 192.36.125.18: icmp_seq=201 ttl=248 time=54.3 ms
64 bytes from 192.36.125.18: icmp_seq=202 ttl=248 time=68.3 ms
64 bytes from 192.36.125.18: icmp_seq=203 ttl=248 time=85.2 ms
64 bytes from 192.36.125.18: icmp_seq=206 ttl=248 time=5309.9 ms
64 bytes from 192.36.125.18: icmp_seq=205 ttl=248 time=6313.9 ms
64 bytes from 192.36.125.18: icmp_seq=209 ttl=248 time=13611.6 ms
64 bytes from 192.36.125.18: icmp_seq=210 ttl=248 time=12617.1 ms
64 bytes from 192.36.125.18: icmp_seq=211 ttl=248 time=11618.0 ms
64 bytes from 192.36.125.18: icmp_seq=212 ttl=248 time=10656.8 ms
64 bytes from 192.36.125.18: icmp_seq=213 ttl=248 time=9661.6 ms
64 bytes from 192.36.125.18: icmp_seq=214 ttl=248 time=8667.3 ms
64 bytes from 192.36.125.18: icmp_seq=215 ttl=248 time=7667.9 ms
64 bytes from 192.36.125.18: icmp_seq=216 ttl=248 time=6672.2 ms

the drastic change in ping time is when I try to download something

if I dmesg I read this

NETDEV WATCHDOG: eth0: transmit timed out
eth0: Tx queue start entry 53  dirty entry 49.
eth0:  Tx descriptor 0 is 2000.
eth0:  Tx descriptor 1 is 2000. (queue head)
eth0:  Tx descriptor 2 is 2000.
eth0:  Tx descriptor 3 is 2000.
eth0: Setting half-duplex based on auto-negotiated partner ability .

I have 10Mbit line in.

Debian unstable
using kernel 2.4.4

I have no idea whats wrong
Can somone tell me whats wrong?

/nisse

Re: How to generate a random number?

[Jeffrey W. Baker]

On Thu, 4 Oct 2001, Alan Shutko wrote:

> "Jeffrey W. Baker" <[EMAIL PROTECTED]> writes:
>
> > That is an extraordinarily bad idea.  Any person will be able to guess the
> > sequence of random numbers simply by guessing the time at which your
> > program was started.
>
> And the impact of this depends on what the program is used for.  If
> you're seeding the RNG for a crypto program, yes, this is bad.  If
> you're seeding the RNG for a random sig generator, who cares?  Why use
> up entropy for programs which don't need it?

Right, but those conditions weren't explained in the original advice to
use srand(time()).  When someone asks "How do I generate a random number"
I think it's a lot safer to advise /dev/random than time().

In either case I would hope that critical software isn't being written by
people who don't even know how to generate random numbers.

-jwb

Re: What to choose

[Teppo Hytönen]

I admit that I might've been a little too hard on them, yes. But I've
heard much more good about Debian, and the examples about getting things
to work were true. But of course Mandrake can't be that bad, otherwise no
one would use it. It's just that I find Debian, both in abilities,
features and ideology, much more to my liking (based on what I've heard
about Mandrake and know about Debian).
This should be by no means be taken as an offense against Mandrake
developers, as I do not mean in that way. Just that I find Debian better
in most aspects. Consider this as an encouragement to Mandrake developers
to try to do the best they can : )
And as I mentioned, choosing a distribution really is a choice of personal
preference.
But as said, this really isn't laptop related anymore, so 'nuff said on my
part : )

Regards,
- Teppo


On Thu, 4 Oct 2001, Russell Coker wrote:

> On Wed, 3 Oct 2001 22:22, Teppo Hytönen wrote:
> >   What comes to choosing between the two, it's personal preference that
> > matters. I myself recommend Debian: I love it myself, and yes, apt-get is
> > great. Then again, I haven't used Mandrake, but haven't heard a single
> > positive comment about it, other than that it is easy to install; many say
> > too easy, so that you can't configure things you might want even if you
> > have enough skill to do it. I've heard that in many cases it doesn't work
> > right.
>
> I think you're being a bit hard on them.  I've heard positive reports about
> Mandrake, and the Mandrake developers I've talked to seem quite smart.
>
> However I get the impression that Debian has more smart developers than any
> other distribution, and I think that upstream maintainers often have a
> similar opinion.  I recently received an email from an upstream author saying
> "I didn't know any distribution had included my code, I might have known
> Debian would do it first", I think that is an indication of the good opinion
> upstream authors have of Debian developers.
>
> I think that you can compare distributions without using them.  If you know
> who develops a distribution, who pays them (if anyone), what their aims are,
> and how long they have been at it then you can get a good idea of what
> product they will develop and whether it will suit you.
>
> Debian has a large team of people who work for fun (most of them get paid
> nothing for their work - many of them are doing Debian work instead of doing
> paid work).  The Debian developers are generally highly skilled by any
> standards of measurement.  The aim of the project is to develop the best
> possible OS according to the general aims of the FSF, where "best" is
> measured by the developers themselves.  The result is that there is a huge
> number of developers (any skilled person who has the time is welcome), a huge
> number of packages (any developer can add a new package at short notice
> without asking for permission), a good solid base, and a lot more work is
> needed on installation programs and documentation (no offense to the people
> who work on it - really they need more help from the rest of us).
>
>
> So we can advocate Debian without mentioning other distributions.  Once they
> know what Debian is about they'll either like it or they won't.  Other
> distributions have much better installation routines, and may be more
> suitable for novices for that reason.
>
> I've CC'd this message to debian-user as it really has nothing to do with
> laptops.
>
> --
> http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark
> http://www.coker.com.au/postal/   Postal SMTP/POP benchmark
> http://www.coker.com.au/projects.html Projects I am working on
> http://www.coker.com.au/~russell/ My home page
>
>
> --
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>
>

ipmasqadm bug in potato 2.2r3 - 2.2.18 - i386

[Andreas Beham]

Sorry I dont know the package that covers ipmasqadm.
But on my system i386 Potato 2.2r3 kernel 2.2.18 the localhost cant open any
internetconnections anymore when I have defined some portforwarding rules
(or is it just the large number that causes the bug I dont know).

Here is the script that generates the rules ($EXTIP is my external ip
address):
# BEGIN ##
 /usr/sbin/ipmasqadm portfw -f

# DirectPlay Ports
#
port=2300
while [ $port -le 2400 ]
  do
   /usr/sbin/ipmasqadm portfw -a -P tcp -L $EXTIP $port -R 10.0.1.50
$port
   /usr/sbin/ipmasqadm portfw -a -P udp -L $EXTIP $port -R 10.0.1.50
$port
   port=$((port+1))
  done

# Suddenstrike Ports
#
 port=28800
 while [ $port -le 28900 ]
  do
   /usr/sbin/ipmasqadm portfw -a -P tcp -L $EXTIP $port -R 10.0.1.50
$port
   /usr/sbin/ipmasqadm portfw -a -P udp -L $EXTIP $port -R 10.0.1.50
$port
   port=$((port+1))
  done

# DirectPlay
#
 /usr/sbin/ipmasqadm portfw -a -P tcp -L $EXTIP 47624 -R 10.0.1.50 47624

# Operation Flashpoint
#
 /usr/sbin/ipmasqadm portfw -a -P tcp -L $EXTIP 2234 -R 10.0.1.50 2234
 /usr/sbin/ipmasqadm portfw -a -P udp -L $EXTIP 2234 -R 10.0.1.50 2234
### END ##

As soon as I flush the rules the internet connection from the localhost is
working again. From the network I can access the internet all the time. It
seems as if only the localhost is affected.
When I have the rules on, the connection looks like the following (netstat):

tcp 0 1 N698P029.adsl.high:2353 www.google.com:www SYN_SENT

It hangs at exactly that point. In lynx this is displayed as the message:
"Making http connection to www.google.com" and then nothing.

Ah yes I should also mention that dnslookup works though. So I can make
nslookup www.google.com and get the right answer. (with the rules enabled).
Though I cant open any tcp connection be it ftp or www.

If you would be so kind to report back directly to me too as I am not
subscribed to the mailing list.

Thanks,
Andreas

Re: Tulip driver unresolved symbols (UPDATE)

[D.]

Well in response to Does mine work the answer is No. 
I have tried for a few days to get this going and
finally figured out that I did not have a eth0 device
it was a sit0.  I tried and read everything that I
could and could not figure out a way to get the eth0. 
I tried to compile the driver like the instructions
told me, but came up with the error that I did not
have a modverdion.h. I copied tulp.o and pci-scan.o to
the appropiate folder though. Then I did a apt-cache
search for pcmcia and found out that I did not have
the right kernel and was missing the pcmcia packages. 
I installed those and still nothing.  At this point I
was really frustrated and figured I'd go about this
from a different perspective.  I installed a Distro
(Mandrake 8.0) that I knew that the tulip driver
worked in and was going to try and load Debian, but
leave the pcmcia stuff.. I'm not sure if I can really
do that, still investigating. (I saw something to this
effect in the last couple of days on the list)  I ran
across a site today that I think has the answer to How
do I get a eth0.  I think that if I do a ifconfig eth0
192.168.1.1 that might do the trick, This is a pure
guess though. 
  Have I given up on installing Debian on this Laptop,
No, The install went great its the Accton EN2242 nic
card thats giving me the problems, but I will get it.
  Any pointers from this rambling?
Thanks
Don
--- David Raleigh Arnold <[EMAIL PROTECTED]> wrote:
> I get tons every time I boot up, but it works fine. 
> Does yours work?
>  
>

> Information is not knowledge.   Belief is
> not truth.
> Indoctrination is not teaching.   Tradition is not
> evidence.
>  David Raleigh Arnold   [EMAIL PROTECTED]


__
Do You Yahoo!?
NEW from Yahoo! GeoCities - quick and easy web site hosting, just $8.95/month.
http://geocities.yahoo.com/ps/info1

Re: Characters can't be shown correctly on tty

[Colin Watson]

On Fri, Oct 05, 2001 at 01:11:20AM +0800, Liu Tao wrote:
> I typed "cat /dev/random" on tty1, and some strange characters scrolled ,
> so I pressed Ctrl-c to terminate cat. But after that, characters can't be 
> shown correctly on tty1, they are shown as strange characters.
> Other ttys are normal. I tried "cat /dev/random" on Konsole, and it doesn't
> have this problem.

Type 'reset' to reset your terminal.

Cheers,

-- 
Colin Watson  [EMAIL PROTECTED]

Re: dual boot with Win 2000?

[Sean Morgan]

On Wed, 03 Oct 2001 16:40:38 -0500
Kent West <[EMAIL PROTECTED]> wrote:

> I'm sitting in front of a W2K-Debian dual-boot system now, and I have a 
> laptop set up that way also. I don't remember any particular 
> tricks/problems, other than to install W2K first.

I just thought I'd post a working dual-boot lilo.conf...Just remember that
boot= must be set to either the MBR or the linux partition, if you attempt
to install to the Win2k partition Windows will fail to boot.

lba32
boot=/dev/hda
root=/dev/hda3
compact
install=/boot/boot.b
delay=20
map=/boot/map
image=/vmlinuz
label=Linux
read-only
other=/dev/hda1
   label=Win2k

Re: How to generate a random number?

[Liu Tao]

On Friday 05 October 2001 01:36, Jeffrey W. Baker wrote:
> On Thu, 4 Oct 2001, Alan Shutko wrote:
> > "Jeffrey W. Baker" <[EMAIL PROTECTED]> writes:
> > > That is an extraordinarily bad idea.  Any person will be able to guess
> > > the sequence of random numbers simply by guessing the time at which
> > > your program was started.
> >
> > And the impact of this depends on what the program is used for.  If
> > you're seeding the RNG for a crypto program, yes, this is bad.  If
> > you're seeding the RNG for a random sig generator, who cares?  Why use
> > up entropy for programs which don't need it?
>
> Right, but those conditions weren't explained in the original advice to
> use srand(time()).  When someone asks "How do I generate a random number"
> I think it's a lot safer to advise /dev/random than time().
>
> In either case I would hope that critical software isn't being written by
> people who don't even know how to generate random numbers.
>
> -jwb

Thanks for all replys.
I have known the methords to generate random numbers.
/dev/random is more safe , but time() is safe enough to me.
In fact  I  only need some different numbers for children's  mahts.

Liu Tao

Re: LAPTOPS AND TWO DIFFERENT WORKPLACES (OT)

[Dimitri Maziuk]

* D. ([EMAIL PROTECTED]) spake thusly:
> Hi all,
>   The system administrator at work is a subcontract
> employee.  He is also the administrator at his
> company.  The company that I work for also has
> employe's from his company working here and they
> travel back and forth between companies and use their
> M$ laptops at both sites.  Here is the problem.. both
> sites use M$ and my company  has the DHCP set and the
> other company doesn't, so when these people go to
> their company they have to go and set the nic card up.

*Blam* authomatic ATM machine.

>   He asked me if I knew of a way to do this and I have
> searched the web and came up dry... I think that a lot
> of the people that are on this list are also very
> knowledgable about M$ also.  Is there a way to do this ...

Yes. Hire a sysadmin.

...
> with out setting his company up to use DHCP?

No. The problem here is that lusers are by definition 
incapable of typing the right incantations into TCP/IP 
properties widget. Not to mention M$ lusers^Wprogrammers 
who are incapable of programming even remotely sane 
behaviour in their crapware. But then, a sysadmin would 
know that.

Dima
-- 
Surely there is a polite way to say FOAD.-- Shmuel Metz
"Go forth and multiply". -- Paul Martin

Re: DHCP PUMP trouble

[Antti Tolamo]

Eskild Wikkeling wrote:


Hello,

I recently got a ADSL modem for babyxl (a dutch adsl provider that uses
dhcp to give me my ip) and am trying to get it to work.

in /etc/network/interfaces I have:
auto eth0 eth1

iface eth0 inet static


iface eth1 inet dhcp
  hostname MYNAME.demon.nl

and in /etc/pump.conf:
domainsearch "demon.nl"

retries 3
timeout 60

device eth1 {
}

when I run ifup eth1 I get the following error in /var/log/syslog:
Oct  4 12:52:41 cumulus pumpd[5098]: HOSTNAME: requesting
MYNAME.demon.nl
Oct  4 12:52:41 cumulus pumpd[5098]: PUMP: sending discover
Oct  4 12:52:41 cumulus pumpd[5098]: breq: opcode: 1
Oct  4 12:52:41 cumulus pumpd[5098]: breq: hw: 1
Oct  4 12:52:41 cumulus pumpd[5098]: breq: hwlength: 6
Oct  4 12:52:41 cumulus pumpd[5098]: breq: hopcount: 0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: xid: 0x45359f55
Oct  4 12:52:41 cumulus pumpd[5098]: breq: secs: 0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: flags: 0x
Oct  4 12:52:41 cumulus pumpd[5098]: breq: ciaddr: 0.0.0.0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: yiaddr: 0.0.0.0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: server_ip: 0.0.0.0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: bootp_gw_ip: 0.0.0.0
Oct  4 12:52:41 cumulus pumpd[5098]: breq: hwaddr:
Oct  4 12:52:41 cumulus pumpd[5098]: breq: servername:
Oct  4 12:52:41 cumulus pumpd[5098]: breq: bootfile:
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: 0x63 0x53 0x82 0x63
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor:  53   1 0x01
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor:  12  17 0x73 0x74
0x72 0x6
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x74 0x75 0x73
0x2e
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x64 0x65 0x6d
0x6f
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x6e 0x2e 0x6e
0x6c
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: ++ 0x00
Oct  4 12:52:41 cumulus pumpd[5098]: breq: vendor: 0xff
Oct  4 12:52:41 cumulus kernel: Packet log: output DENY eth1 PROTO=17
192.168.0.
1:68 255.255.255.255:67 L=328 S=0x00 I=54009 F=0x T=64 (#6)

can someone tell me where things are going wrong?

thanks heaps,
Eskild



Dumb pump.  Pump has some problems with two networkinterfaces.
No, I don't use it. but after following few Usenet discussion in Finnish
usenet groups, problems seem to disappear when installing dhcp.

It works with some people, but all problems I've heard so far
concerning dhcp+ISP  have been by people who use pump.

Antti

Re: DHCP problems with cable modem

[Michael Fontenot]

Are you sure that dhcpcd isn't working?  I 
just installed
dhcpcd, and at first I thought it wasn't working, 
because
there was nothing about DHCP written to the console 
during 
boot.  But I eventually discovered that it was 
working fine...
it just doesn't print out its transmission and 
receptions anywhere.
Look in /etc/dhcpc, for a file ending in ".info" (I 
think).  If dhcpcd is
working (and has gotten a response back from the 
server), it
writes the received info in there.
 
    Mike Fontenot
    [EMAIL PROTECTED]
 

how to install debian originally from cd but copied to HD

[ramsubs]

i'm a linux newbie and playing around with 
debian to learn. i've installed and reinstalled it many times but the 
routine of swapping cds (from the 3 cd-set) is bothersome.
 
so i partitioned my HD increasing it by 3 
partitions with the idea of copying the 3 cds into the 3 partitions, and 
subsequently trying to install from the HD direct. (partitions are fat32; 
but i had no problems reading the boot floppies info from hda6)
 
the installation process asks for sources from 
CDROM, filesystem, ftp, http,.. I picked filesystem assuming i could refer 
to my local partitions.
 
i had to modify the /etc/apt/sources.list to 
read from the 3 new partitions (hda6, hda7, hda8) but i'm stumped. i get error 
messages. typically i used a line such as:
    deb 
file:/hda6/dists/potato/main/binary-i386/Packages
(the installer prompted this asking if i wanted to 
edit the file manually, and then displayed ae to edit the file)
 
i've checked man apt.get, apt.conf, sources.list 
but they don't provide help i need (at least to my little 
understanding).
 
hope someone can help me.
 
thanks in advance
 
ram
 
 

Re: dual boot with Win 2000?

[D.]

As a point of interest, When I used the same
configuration that you have for your Windows partation
"other=/dev/hda1" on a HP Pavilion 5450 my Windows Me
would not boot up.. It gave the error message that It
was missing the command.com file.  When I realized
what was wrong I changed it to "other=/dev/hda2" and
it booted right up. 
Don
--- Sean Morgan <[EMAIL PROTECTED]> wrote:
> On Wed, 03 Oct 2001 16:40:38 -0500
> Kent West <[EMAIL PROTECTED]> wrote:
> 
> > I'm sitting in front of a W2K-Debian dual-boot
> system now, and I have a 
> > laptop set up that way also. I don't remember any
> particular 
> > tricks/problems, other than to install W2K first.
> 
> I just thought I'd post a working dual-boot
> lilo.conf...Just remember that
> boot= must be set to either the MBR or the linux
> partition, if you attempt
> to install to the Win2k partition Windows will fail
> to boot.
> 
> lba32
> boot=/dev/hda
> root=/dev/hda3
> compact
> install=/boot/boot.b
> delay=20
> map=/boot/map
> image=/vmlinuz
> label=Linux
> read-only
> other=/dev/hda1
>label=Win2k
> 
> 
> -- 
> To UNSUBSCRIBE, email to
> [EMAIL PROTECTED] 
> with a subject of "unsubscribe". Trouble? Contact
> [EMAIL PROTECTED]
> 


__
Do You Yahoo!?
NEW from Yahoo! GeoCities - quick and easy web site hosting, just $8.95/month.
http://geocities.yahoo.com/ps/info1

Re: resiserfs on woody install?

[J.H.M. Dassen \(Ray\)]

On Thu, Oct 04, 2001 at 08:02:31 -0600, Robert L. Harris wrote:
>   where can I find info on ext3?  I don't see it in the kernel.

It's a separate patch at the moment, AFAIK. It is included in at least Alan
Cox' series of patches for 2.4 kernels (where you need to say [*] Prompt for
development and/or incomplete code/drivers to get [ ] Ext3 journalling file
system support under filesystems).

> Can ext2 be upgraded to ext3 once the machine is installed and I'm
> upgrading to 2.4.X?

Yes. As I understand it, you can even go back easily.

HTH,
Ray
-- 
POPULATION EXPLOSION  Unique in human experience, an event which happened 
yesterday but which everyone swears won't happen until tomorrow.  
- The Hipcrime Vocab by Chad C. Mulligan 

[OT] Re: LAPTOPS AND TWO DIFFERENT WORKPLACES (OT)

[Gary Hennigan]

[EMAIL PROTECTED] writes:
>   The system administrator at work is a subcontract
> employee.  He is also the administrator at his
> company.  The company that I work for also has
> employe's from his company working here and they
> travel back and forth between companies and use their
> M$ laptops at both sites.  Here is the problem.. both
> sites use M$ and my company  has the DHCP set and the
> other company doesn't, so when these people go to
> their company they have to go and set the nic card up.
>   He asked me if I knew of a way to do this and I have
> searched the web and came up dry... I think that a lot
> of the people that are on this list are also very
> knowledgable about M$ also.  Is there a way to do this
> with out setting his company up to use DHCP?
> Thanks,
>   Appreciate all the help that I can pass on.
> Don

Time to start using a new search engine...

Look at http://www.netswitcher.com. I use it for two static addresses,
but the docs mention DHCP so I assume it'll work with the scenario you
describe.

Gary

What am I doing wrong? (apt-get via http proxy)

[Stan Brown]

I have a Debian STABLE machien that I need to isntall at work behind a very
restricitive firewall. I have a Squid server runing thta can get through
the firewall (HTTP only ftp does not work). I have set up /etc/apt/apt.conf
like this:


// Pre-configure all packages before they are installed.
// (Automatically added by debconf.)
DPkg::Pre-Install-Pkgs {"/usr/sbin/dpkg-preconfigure --apt";};

  
  // HTTP method configuration
  http 
  {
Proxy "http://squid.ourcompany.com:3128";;
Timeout "120";
Pipeline-Depth "5";

// Cache Control. Note these do not work with Squid 2.0.2
No-Cache "false";
Max-Age "86400"; // 1 Day age on index files
No-Store "false";// Prevent the cache from storing archives
  };

But it does not work, and I don't see it trying to hit the proxy in the
squid access logs, whne I do a "Update Available".

What am I doing wrong?


-- 
Stan Brown [EMAIL PROTECTED]843-745-3154
Charleston SC.
-- 
Windows 98: n.
useless extension to a minor patch release for 32-bit extensions and
a graphical shell for a 16-bit patch to an 8-bit operating system
originally coded for a 4-bit microprocessor, written by a 2-bit 
company that can't stand for 1 bit of competition.
-
(c) 2000 Stan Brown.  Redistribution via the Microsoft Network is prohibited.

Re: resiserfs on woody install?

[Andrew Agno]

Robert L. Harris writes:
 >   where can I find info on ext3?  I don't see it in the kernel.  Can ext2 be
 > upgraded to ext3 once the machine is installed and I'm upgrading to 2.4.X?
 > I take it I have to patch the kernel for ext3 as I don't see it as an option 
 > in my 2.4.10 kernel config.

http://www.uow.edu.au/~andrewm/linux/ext3/

There you'll find the patch for 2.4.10 and the documentation.
Basically, grab the patch, cd to your sources, gunzip path/to/patch |
patch -p1, then do a make config (or menuconfig or whatever).

I have upgraded machines using ext2 to ext3 while they were running.
There is one important thing to note.  In 2.4.10, running tune2fs -j
/dev/hdXX then rebooting won't give you an ext3 partition.  You must
run tune2fs in a different kernel (like your existing one), because of 
some changes in the 2.4.10 kernel.

If you need to delete the .journal file, because you forgot to
create the journal in 2.4.9 (or whatever kernel) use:
"chattr -i .journal; rm .journal" BEFORE running "tune2fs -j"
(from
https://listman.redhat.com/pipermail/ext3-users/2001-September/001467.html)

Oh, and make sure e2fsprogs and mount are the correct versions.

Discussions about ext3 on 2.4.10 are in September's archives:
https://listman.redhat.com/pipermail/ext3-users/2001-September/thread.html

Andrew.

Re: how to install debian originally from cd but copied to HD

[Michael Heldebrant]

On Thu, 2001-10-04 at 14:12, ramsubs wrote:
> i'm a linux newbie and playing around with debian to learn. i've
installed and reinstalled it many times but the routine of swapping cds
(from the 3 cd-set) is bothersome.
> 
> so i partitioned my HD increasing it by 3 partitions with the idea of
copying the 3 cds into the 3 partitions, and subsequently trying to
install from the HD direct. (partitions are fat32; but i had no problems
reading the boot floppies info from hda6)
> 
> the installation process asks for sources from CDROM, filesystem, ftp,
http,.. I picked filesystem assuming i could refer to my local
partitions.
> 
> i had to modify the /etc/apt/sources.list to read from the 3 new
partitions (hda6, hda7, hda8) but i'm stumped. i get error messages.
typically i used a line such as:
> deb file:/hda6/dists/potato/main/binary-i386/Packages
> (the installer prompted this asking if i wanted to edit the file
manually, and then displayed ae to edit the file)
> 
> i've checked man apt.get, apt.conf, sources.list but they don't
provide help i need (at least to my little understanding).
> 
> hope someone can help me.
> 

Firstly please don't post to the list in html.  What I BELEIVE you are
going to need to mount each of these partitions as a file system let's
for the sake of ease create cdrom1 cdrom2 cdrom3 in /mnt.  I'm not sure
if apt is going to be able to understand the schizophrenia of the three
sources since I've never installed from cd copied to hd, or cd for that
matter.

So from a console (alt-f2 and enter) after you get the install running

cd /mnt

or if it isn't there just do it from the root system and adjust
accordingly, I haven't installed in a a while.

mkdir cdrom1
mkdir cdrom2
mkdir cdrom3
mount -t vfat /dev/hda6 /mnt/cdrom1
mount -t vfat /dev/hda7 /mnt/cdrom2
mount -t vfat /dev/hda8 /mnt/cdrom3

then in sources.list:

deb file:/mnt/cdrom1/dists/potato/main/binary-i386/Packages
deb file:/mnt/cdrom2/dists/potato/main/binary-i386/Packages
deb file:/mnt/cdrom3/dists/potato/main/binary-i386/Packages

That should work for you.

--mike

Re: Tulip driver unresolved symbols (UPDATE)

[David Raleigh Arnold]

Guess what?  If you have a netgear card, you probably want a ng_tulip
driver instead of a tulip driver.  I just found this out when I compiled
a new kernel.  Don't be mad, I didn't know either.  :-)
-- 
||/ Peace, understanding, health and happiness to all beings! ||
|-K ars sine scientia nihil[EMAIL PROTECTED] || 
||\ David Raleigh Arnold \\   Falls Church, Virginia \\   USA ||

eht0: Cared reports out of resouces

[Stan Brown]

I'm trying to install STABLE on a Compq Proliant server. It has a NIC that
dmesg reports as OEM i8257/i8258 10/100 Ethernet.

This machine will be an Amanda tapehost. When I start trying to run Amanda
(which loads tthe network up pretty good). I get a continous stream of
messages on the console about :eht0: Card reports out of resources"

It appears to work, but ifconfig reports many many collisions. I am using
this card on a 10M link.

Any sugestiosn? The machien is really unusable with all this going to the
screen.


-- 
Stan Brown [EMAIL PROTECTED]843-745-3154
Charleston SC.
-- 
Windows 98: n.
useless extension to a minor patch release for 32-bit extensions and
a graphical shell for a 16-bit patch to an 8-bit operating system
originally coded for a 4-bit microprocessor, written by a 2-bit 
company that can't stand for 1 bit of competition.
-
(c) 2000 Stan Brown.  Redistribution via the Microsoft Network is prohibited.

Re: How to generate a random number?

[Erik Steffl]

Alan Shutko wrote:
> 
> "Jeffrey W. Baker" <[EMAIL PROTECTED]> writes:
> 
> > That is an extraordinarily bad idea.  Any person will be able to guess the
> > sequence of random numbers simply by guessing the time at which your
> > program was started.
> 
> And the impact of this depends on what the program is used for.  If
> you're seeding the RNG for a crypto program, yes, this is bad.  If
> you're seeding the RNG for a random sig generator, who cares?  Why use
> up entropy for programs which don't need it?

  :-) considering that no matter what you do the entropy will increase I
don't think you have to be concerned about wasting it...

  but: using /dev/urandom can significantly slow down the program - e.g.
xdm's standard debian config specifies /dev/urandom and it takes a long
time (a minute or two) for xdm to start, replace it by /dev/mem and the
problem's gone...

erik

Postscript printer doesn't work...

[Michael Fontenot]

I recently installed debian 2.2r3, including the 
various "parport"
driver modules and the lp module.  I also 
istalled "lpr".  When
the lpr command didn't print, I installed 
apsfilter, but it failed to
print the test page.
 
I have a postscript parallel printer (Optra Ep), 
which is supposed
to make things easy.  I specified lp0, and all 
of the other questions
asked by apsfilter seemed straighforward, so I 
don't know why
lpr still doesn't print.  Any advice?  

 
    Mike Fontenot
    [EMAIL PROTECTED]
 

Building 2.4.9 kernel for a laptop

[Darryl L. Pierce]

I'm trying to build the 2.4.9 kernel for my laptop. Everything works except for 
the 
PCMCIA system. When I insert the card, /var/log/message outputs that it's 
noticed
the new hardware, but it doesn't identify the card (3COM 575BT) and there's no 
mention of eth0 in when I list interfaces with ifconfig.

Any help on building the kernel? My laptop is a Gateway Solo 9300cs with a 
Celeron
433mHz.

-- 
Darryl L. Pierce <[EMAIL PROTECTED]>
The Infobahn Offramp 
"Yes, Dougal, but you're not like most people..."


pgpmeavZ4Wu34.pgp
Description: PGP signature

requesting information

[Marc Weldon]

I am starting a new company, and I was curious to know what services and/or 
products you offer in the way of computer security.  Any information you can 
give me would be greatly appreciated.


Thank you,

Mark Weldon

_
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp

Re: Gnome + wmaker (problem switching workspaces)

[Antonio Arauzo]

"Karsten M. Self" wrote:
> > > > there is at least one window running on current workspace. Has anybody
> > > > experienced this? Is it any configuration problem?

> > > What happens if you open, then close, a window?

> > It doesn't matter. Whenever there is an open window on actual workspace
> > switching works. If there is no window open it doesn't.

> Weird.
> 
> I'd check open bugs, and file one if there isn't anything applicable.

Ok, thanks.

> I'd also prefer you kept list mail on-list.

Sorry, I forgot to change the "To:" field. 
Why doesn't debian lists set "Reply-To:" field to the list address?

Antonio

JDK on ftp.informatik.hu-berlin.de?

[Gary Hennigan]

I was getting the 1.3 JDK with the following line in my
/etc/apt/sources.list:

deb ftp://ftp.informatik.hu-berlin.de/pub/Java/Linux/debian woody non-free

but it seems that that directory has disappeared. Any idea what
happened to it? There's a version for potato, but the woody one is
gone.

Thanks,
Gary

Re: What's the device name for my tape drive?

[Michael Heldebrant]

On Thu, 2001-10-04 at 08:01, Stan Brown wrote:
>  I have a Debian stable machine with all IDE dirves. it also has an Adaptec
>  AIC-7892 SCSi controler. This controler has one slave device, a HP C1537A
>  4mm tape changer. It's at SCSI ID 5.
> 
>  What is it's device name? /dev/??
> 

as far as I know it should be st0

but, you're going to need st, scsi, and your low level driver support as
modules or compiled into the kernel.  I beleive you are going to need
something like (mtx - controls tape autochangers) to use the tape
changing aspects of the drive.

--mike