Re: Anyone using apt-listchanges?

[Graham Williams]

Received Fri 10 Oct 2003  4:27am +1000 from [EMAIL PROTECTED]:
> I'd like to look at the change log of a package before installing it. 
> I installed apt-listchanges, but it seems like I have to download the 
> .deb first.  Is that true?

I sometimes like to check the changelog first. Perhaps this will help:

wajig news 

will get the latest update of the pkg's changelog from
packages.qa.debian.org.

The next release of wajig (1.0.1) will support (it's already
implemented):

wajig changelog 

which will retrieve the whole changelog from
http://people.debian.org/~noel

Then if the package looks okay, you can:

wajig install 

Regards,
Graham


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

/dev/hda gone missing with kernel 2.4.19 when CMD680 detected?

[Graham Williams]

I'm installing Debian 3.0 on a new machine with a 80GB IDE recognised
as /dev/hda. I've upgraded to kernel-image-2.4.19-686 as we've added
another IDE card to the machine (a CMD680). Now when it boots (using
Grub) the boot starts okay and then finds the CMD680 (presumably
before it finds the original IDE) and starts assigning its 3 hard
disks to /dev/hda, /dev/hdc, and /dev/hdd! Then I presume it's trying
to mount /dev/hda, presumably expecting this to be the original
/dev/hda that I installed Debian on.  It doesn't seem to get to
recognising the original IDE drive any more.  It simply kernel panics
and stops. Can still boot into the older 2.4.18 kernel (which doesn't
recognise the CMD680 IDE card)

Any ideas on what I can do here to get out of this pickle?

Thanks,
Graham


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

Re: ASUS P4B533 Motherboard

[Graham Williams]

Received Sun 10 Nov 2002  3:53am +1100 from Bill Moseley:

> Does anyone have experience installing Debian on a ASUS P4B533 based system
> that can offer any comments?  I'm not finding much on-line discussion of
> this motherboard.

Hi Bill,

I'm at this moment in the throws of installing Debain 3.0 on an ASUS
P4B533 based machine. So I would love to hear of any insights you
get. 

My current state of play is that Debian GNU/Linux 3.0 installed just
fine (booted with bf24 to use the 2.4.18 kernel) but the onboard
network card is not recognised. 

I have an additional IDE (CMD680) in the machine but when moving to
kernel 2.4.19, which supports this card, it becomes /dev/hda and my
original /dev/hda seems to be lost and I can't get past trying to mount
root partition.

Regards,
-- 
[EMAIL PROTECTED]   Tel: (+61 2) 6216 7042
Enterprise Data Mining Fax: (+61 2) 6216 7111
CSIRO AustraliaGPO  Box 664  Canberra
  ACT2601  Australia


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

Re: ASUS P4B533 Motherboard

[Graham Williams]

Received Sun 10 Nov 2002  8:08am +1100 from Bill Moseley:

> At 07:22 AM 11/10/02 +1100, Graham Williams wrote:
> > ... Debain 3.0 on an ASUS P4B533 based machine. ...

> > ... Debian GNU/Linux 3.0 installed just fine (booted with bf24 to
> > use the 2.4.18 kernel) but the onboard network card is not
> > recognised.

> ... Is there a specific MB model number you have?  e.g. something
> that specifies what (lan, video, audio) is included?

I'll check on Monday - without the NIC I can't connect to it at
present.

> > ... additional IDE (CMD680) ... kernel 2.4.19, which supports
> > this card, it becomes /dev/hda and my original /dev/hda seems to
> > be lost and I can't get past trying to mount root partition.

> Were you able to boot fine with the on-board IDE (without the IDE card)?

Sure. Without the extra IDE Debian installed and now runs just fine
(so far) with kernel 2.4.18.

> What are you running for video and audio?

We've had X running but I'll need to check the machine directly (on
Monday) to get the specs.  As I recall the AGP was recognised
automatically.

Regards,
Graham


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

Re: ASUS P4B533 Motherboard

[Graham Williams]

Received Sun 10 Nov 2002 10:53am +1100 from Graham Williams:
> Received Sun 10 Nov 2002  8:08am +1100 from Bill Moseley:
> 
> > At 07:22 AM 11/10/02 +1100, Graham Williams wrote:
> > > ... Debain 3.0 on an ASUS P4B533 based machine. ...
> 
> > > ... Debian GNU/Linux 3.0 installed just fine (booted with bf24 to
> > > use the 2.4.18 kernel) but the onboard network card is not
> > > recognised.
> 
> > ... Is there a specific MB model number you have?  e.g. something
> > that specifies what (lan, video, audio) is included?
> 
> I'll check on Monday - without the NIC I can't connect to it at
> present.

Unfortunately lspci comes up with everything unknown! No help
there. I guess the motherboard is too new (it's a P4B533-VM (and
that's essentially all I know).

Regards,
Graham


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

Re: ASUS P4B533 Motherboard

[Graham Williams]

Received Wed 13 Nov 2002  3:53am +1100 from Mike Dresser:

> On Sun, 10 Nov 2002, Graham Williams wrote:

> >
> > I'm at this moment in the throws of installing Debain 3.0 on an ASUS
> > P4B533-VM based machine.
> > ...

> I've got ... 4 of these boards in Debian machines here, with one that is
> the P4B533-E.
> 
> 2.4.19 doesn't support the onboard ethernet because of a new PCI ID by
> Intel.  2.4.20 has the added PCI ID added, and as well supports the onboard RAID
> just fine(at least as a regular controller)  So you could hunt a 2.4.20
> kernel, or use 2.4.20 from kernel sources.

Thanks for the information.  I grabbed source for kernel 2.4.19,
patched it to 2.4.20-pre11, compiled it and installed the resulting
Debian package and everything is working wonderfully, except XFree
4.2.1 which doesn't support the video card (using the vesa driver
works, just). Fixed in Xfree86 4.3 according to the
documentation. Some information: The audio is i810_audio, video will
be i810, and NIC is eepro100. I now have 8 IDE hard drive slots on 2
extra IDE CMD680 cards in addition to the hard drive on the onboard
IDE chip!

Regards,
Graham


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

Gnome runs out of sockets without sound?

[Graham Williams]

I've tried on and off to figure this one out or to find some other
posting with similar problems, to no avail.

Debian 2.1 on a Notebook running xserver-neomagic (using XFree86 3.3.2
as comes with Debian 2.1 rather than the newer 3.3.3) and supplied
kernel 2.0.36. I installed gnome 1.0.2 through the collection of
helper packages as suggested on the gnome home page.

All works great, of course, except Gnome seems to be trying to open
some audio device (I haven't compiled any kernels and I don't think my
Sound Blaster Pro compatible card is supported in the kernel I have?),
failing, then gets into a loop trying to open sockets, eventually
failing by hitting the maximum open file limit?

This tends to happen after I've already got one or two gnome
applications already running (like two gnome-terminals---the third one
fails as described above).

Any ideas?

Here's a bit more detail:

Start first gnome-terminal from menu, no problem.  Start next
gnome-terminal from first gnome-terminal, no problem.  Start a third
gnome-terminal (or any gnome application, like gnomecc) and I get:

from .xsession-errors I see lots of:

 Trying 8.192Khz, 8bit mono.
 /dev/dsp: No such file or directory
 Audio device open for 8.192Khz, mono, 8bit failed
 Trying 8Khz, 8bit mono.
 /dev/dsp: No such file or directory
 Sound device inadequate for Esound. Fatal.

and then

Unable to create socket
_X11TransSocketOpen: socket() failed for local
_X11TransSocketOpenCOTSClient: Unable to open socket for local
_X11TransOpen: transport open failed for local/inca:0
Gtk-WARNING **: cannot open display: :0

And then I start to get "Too many files open" errors for gnome and
other applications.

Seems like after the first 2 or 3 gnome applications, any subsequent
ones I try to start fail. Other applications (e.g., xterm) seem just fine.

Thanks, in advance, for any help.

Cheers,
Graham

Re: Curriculum Vitae in LaTeX

[Graham Williams]

A great resource for finding packages in LaTeX is
http://www.ctan.org/CTANfind.html.

A new package called currvita has just been released.

Cheers,
Graham

A travelogue installing Debian 2.1 on a laptop

[Graham Williams]

I was updating my page on installing Debian GNU/Linux on the Dell
Latitude CPi D300XT when it occurred to me that a lot of the material
there might be of interest in general. It's a log of how to install
Debian on this laptop (and maybe installing Debian in general) and
aims to be fairly descriptive.

In case anyone is interested in it

   http://www.dit.csiro.au/~gjw/linux/dlcpid300xt.html


Warning: I've been a silent user of Debian for many years but would
not claim to be a Debian expert

Of course, updates are welcome!

Cheers,
Graham

Re: Accessing RealPlayer broadcasts from Broadcast.com

[Graham Williams]

I have Debian GNU/Linux (Slink) with a 2.2.12 kernel on a Dell
Latitude CPi D300XT which has a Crystal Semiconductor CS4237B Sound
Blaster Pro compatible sound card.

On boot I get the message:

Sound initialization started
 at 0x220 irq 5 dma 0,5
This sound card may not be fully Sound Blaster Pro compatible.
In many cases there is another way to configure OSS so that
it works properly with OSS (for example in 16 bit mode).
Please ignore this message if you _really_ have a SB Pro.
Sound initialization complete

So I've probably not configured things properly. With RealAudio (using
rvplayer) and the Myth2 demo I don't get proper sound, but instead
hear a horrible screaching sound.  MP3 sounds okay.

While the sound card is a CS4237B there is only a CS4232 available in
the kernerl configuration.  I tried various parameters for this but
all failed (sometimes I got very faint sounds, sometimes sound repeated
3 times, and always a regular click).  Instead I compiled the kernel
for Sound Blaster Pro with the following CONFIG options:

CONFIG_SOUND=y
CONFIG_SOUND_OSS=y
CONFIG_SOUND_SB=y
CONFIG_SB_BASE=220
CONFIG_SB_IRQ=5
CONFIG_SB_DMA=0
CONFIG_SB_DMA2=5
CONFIG_SB_MPU_BASE=0
CONFIG_SB_MPU_IRQ=-1
CONFIG_SOUND_ADLIB=y

Is there something I haven't configured here?  How do I get 16 bit
sound? I've had a read through the HOWTO and perhaps I need to install
ALSA? Is that the only answer?

Cheers,
Graham

Slink -> Potato: Problem with ppp/chat/pcmcia/modem?

[Graham Williams]

I just used "apt-get dist-upgrade" to move from a slink to a potato
distribution on my Laptop (Dell Latitude CPi D300Xt). Had a minor
problem with configuring some packages because of perl.  I manually
"dpkg --install perl*" and then got apt-get going again. Almost
succeeded, failing in the end on make-doc (which I'll ignore for now).

My problem is that "pon" (as in ppp) is now failing and I've spent a
good few hours trying to figure out why.  I'm using PCMCIA. I've
recompiled the kernel 2.2.12 and pcmcia-sources, just as I used to
with slink.

After some odd module problems (unresolved symbols in parport.o and
couldn't find module char-major-108, problems that I did not get under
slink, even though I started out using the same kernel .config file)
the kernel/module things seem okay.

Issuing "pon" or pppd directly, I get (in the plog file)

chat[...]: Can't get terminal parameters: Input/output error.

I believe pppd is working okay, since if I mimic the modem on my
interactive tty to get the ATZ stuff, I can respond "OK" to get ATDT
etc.

The pcmcia modem card seems to be fine, since ejecting and inserting
it generates various beeps and the syslog file records the appropriate
IRQ (3) and identifies it as a Serial or Modem card.  Similarly,
"cardinfo" reports the card.

If I reboot under WindowsNT the modem works just fine.  The modem was
working just fine under Slink until I did the upgrade. Installing
pcmcia-cs package suggested restarting the daemon, so I did, and since
then have had this problem.

I'm running out of ideas and things to read.  Any suggestions? I wish
I had redirected the output from apt-get to some place so I could now
read it.  Is this information stored anywhere, so I might check for
other messages that I missed along the way (the apt-get download took
some 24 hours over the modem!).

Many thanks in advance.

Cheers,
Graham

Re: Slink -> Potato: Problem with ppp/chat/pcmcia/modem?

[Graham Williams]

Thanks for the reply John.  Below are details from my log files.
pointers to what I should be reading or exploring further are most
welcome.

>>>>> Graham Williams wrote to debian-user on 22 Nov 1999 08:02:22 +1100:

Graham> I just used "apt-get dist-upgrade" to move from a slink to
Graham> a potato ... "pon" (as in pppd) is now failing ...
Graham> chat[...]: Can't get terminal parameters: Input/output error.

These are the messages I get:


Nov 22 21:00:33 inca pppd[254]: pppd 2.3.10 started by root, uid 0
Nov 22 21:00:34 inca chat[255]: Can't get terminal parameters: Input/output 
error
Nov 22 21:00:34 inca pppd[254]: Connect script failed
Nov 22 21:00:35 inca pppd[254]: Exit.


Below is a successful connection when I had slink installed a few days
ago, followed by its termination.  During this particular session I
downloaded and installed potato and then tried to reconnect.


Nov 18 19:17:47 inca pppd[2947]: pppd 2.3.5 started by XXX, uid 1000
Nov 18 19:17:48 inca chat[2948]: abort on (BUSY)
Nov 18 19:17:48 inca chat[2948]: abort on (NO CARRIER)
Nov 18 19:17:48 inca chat[2948]: abort on (VOICE)
Nov 18 19:17:48 inca chat[2948]: abort on (NO DIALTONE)
Nov 18 19:17:48 inca chat[2948]: abort on (NO ANSWER)
Nov 18 19:17:48 inca chat[2948]: send (ATZ^M)
Nov 18 19:17:49 inca chat[2948]: expect (OK)
Nov 18 19:17:49 inca chat[2948]: ATZ^M^M
Nov 18 19:17:49 inca chat[2948]: OK
Nov 18 19:17:49 inca chat[2948]:  -- got it 
Nov 18 19:17:49 inca chat[2948]: send (ATDT^M)
Nov 18 19:17:49 inca chat[2948]: expect (CONNECT)
Nov 18 19:17:49 inca chat[2948]: ^M
Nov 18 19:18:08 inca chat[2948]: ATDT^M^M
Nov 18 19:18:08 inca chat[2948]: CONNECT
Nov 18 19:18:08 inca chat[2948]:  -- got it 
Nov 18 19:18:08 inca chat[2948]: send (\d)
Nov 18 19:18:09 inca pppd[2947]: Serial connection established.
Nov 18 19:18:10 inca pppd[2947]: Using interface ppp0
Nov 18 19:18:10 inca pppd[2947]: Connect: ppp0 <--> /dev/modem
Nov 18 19:18:18 inca pppd[2947]: Remote message: 
Nov 18 19:18:18 inca pppd[2947]: Cannot determine ethernet address for proxy ARP
Nov 18 19:18:18 inca pppd[2947]: local  IP address XXX.XX.X.XX
Nov 18 19:18:18 inca pppd[2947]: remote IP address XXX.XX.X.X
Nov 19 19:29:55 inca pppd[2947]: Terminating on signal 15.
Nov 19 19:29:55 inca pppd[2947]: Connection terminated.
Nov 19 19:29:55 inca pppd[2947]: Exit.
Nov 19 20:17:13 inca pppd[29861]: pppd 2.3.10 started by root, uid 0
Nov 19 20:17:14 inca chat[29862]: Can't get terminal parameters: Input/output 
error
Nov 19 20:17:14 inca pppd[29861]: Connect script failed
Nov 19 20:17:15 inca pppd[29861]: Exit.



The following is from syslog for my current boot up. Some odd things
are:

The kernel ppp version is 2.3.7, but pppd is 2.3.10.  Will
this cause a problem? Previously I was using pppd version
2.3.5 (from slink).

I don't know why I get the message about modprobe not being
able to find module char-major-108?


Nov 22 21:00:15 inca syslogd 1.3-3#33: restart.
Nov 22 21:00:15 inca kernel: klogd 1.3-3#33, log source = /proc/kmsg started.
Nov 22 21:00:15 inca kernel: Inspecting /boot/System.map-2.2.12
Nov 22 21:00:16 inca kernel: Loaded 8033 symbols from /boot/System.map-2.2.12.
Nov 22 21:00:16 inca kernel: Symbols match kernel version 2.2.12.
Nov 22 21:00:16 inca kernel: Loaded 12 symbols from 1 module.
Nov 22 21:00:16 inca kernel: Linux version 2.2.12 ([EMAIL PROTECTED]) (gcc 
version 2.95.2 19991109 (Debian GNU/Linux)) #1 SMP Sat Nov 20 21:47:44 EST 1999
Nov 22 21:00:16 inca kernel: mapped APIC to e000 (00273000)
Nov 22 21:00:16 inca kernel: mapped IOAPIC to d000 (00274000)
Nov 22 21:00:16 inca kernel: Detected 300686087 Hz processor.
Nov 22 21:00:16 inca kernel: Console: colour VGA+ 80x25
Nov 22 21:00:16 inca kernel: Calibrating delay loop... 299.83 BogoMIPS
Nov 22 21:00:16 inca kernel: Memory: 127824k/131008k available (1112k kernel 
code, 424k reserved, 1588k data, 60k init)
Nov 22 21:00:16 inca kernel: Checking 386/387 coupling... OK, FPU using 
exception 16 error reporting.
Nov 22 21:00:16 inca kernel: Checking 'hlt' instruction... OK.
Nov 22 21:00:16 inca kernel: POSIX conformance testing by UNIFIX
Nov 22 21:00:16 inca kernel: per-CPU timeslice cutoff: 100.22 usecs.
Nov 22 21:00:16 inca kernel: CPU0: Intel Pentium II (Deschutes) stepping 02
Nov 22 21:00:16 inca kernel: SMP motherboard not detected. Using dummy APIC 
emulation.
Nov 22 21:00:16 inca kernel: PCI: PCI BIOS revision 2.10 entry at 0xfbcde
Nov 22 21:00:16 inca kernel: PCI: Using configuration ty

Re: Boot disk for kernel-image-2.2.13 not working

[Graham Williams]

I've also just started moving to Potato and was trying the 2.2.12
(kernel-image-2.2.12_2.2.12-4.deb) and 2.2.13
(kernel-image-2.2.13_2.2.13-1.deb) kernels, creating a boot floppy,
and getting exactly the same problem, with the same message over and
over again:

 0400
 AX: 0212
 BX: 7000
 CX: 5001
 DX: 

I'd also welcome any suggestions as to what the problem is?

Cheers,
Graham

> Bart Szyszka wrote to  on 28 Nov 1999 12:32:20 +1100:

Bart> Hi, I'm trying to upgrade my kernel and have downloaded some
Bart> of the kernel... packages, including the kernel-image-2.2.13
Bart> one. My problem is that during its setup, when it asks me to
Bart> create a boot disk and makes one, the disk doesn't seem to
Bart> work. When I reboot and put it in, all it says is 'Loading'
Bart> (no Syslinux, no version numbers, nothing) followed by a few
Bart> lines of dots, and then this gets repeated over and over:
Bart> AX: 0212 BX: 7000 CX: 5001 DX: .

Bart> Any ideas? Is there a package I can download that will let
Bart> me make another boot disk? I prefer booting to Linux by just
Bart> putting the floppy in and turning the computer on instead of
Bart> needing to wait for that right moment for LILO to come up (I
Bart> dual-boot with Windows and unfortunately Win98 needs to be
Bart> the default). BTW, I did use a different disk so I still
Bart> have the disk that'll boot with my older kernel (2.0.38), so
Bart> I still have access to Debian (potato).

Re: questions about slink to potato upgrade

[Graham Williams]

Hi Dan,

I went that path also (dist-upgrade to get potato) and had some of the
same problems.  

The most serious for me was that my pcmcia modem card stopped working
somewhere along the way. After recompiling kernels and pcmcia and
different options etc I gave up.  I reinstalled (slink) from scratch
and have proceeded step-by-step in moving to potato. Compiling the
kernel-source-2.2.13 and pcmcia-modules-2.2.13 (and versions 2.2.12)
from potato results in a non-functioning pcmcia.  Previoulsy I
compiled the kernel/pcmcia sources (2.2.12) from slink (under slink)
with no problem. I have yet to see if I were to grab the kernel/pcmcia
sources from slink whether it makes any difference.

I trust someone can shed some light on this? 

Cheers,
Graham

Re: Boot disk for kernel-image-2.2.13 not working

[Graham Williams]

>>>>> John Carline wrote to Graham Williams on 29 Nov 1999 17:02:13 +1100:

John> Hi folks,

John> Every time this has happened to me, it was either a bad
John> floppy or a bad floppy drive.  On one occasion, the
John> connector to the drive became slightly loose and all the
John> disks I created with "make zdisk or make bzdisk"
John> failed. They gave basicly the same indidatons that your
John> getting. The strange part is that the disks using syslinux
John> still worked. Still don't understan why?

John> Assuming that all your hardware is good, you *might* be able
John> to create a new boot disk by going to "/usr/src/linux" and
John> doing a "make zdisk" or "make bzdisk".


Thanks John.

I'm not sure I'm seeing hardware problems.  The odd thing is that when
I build my own kernel, to build my own kernel-image package, and
install that to make a boot floppy, all works okay (except my pcmcia
stuff, but that's another story). No problems with the boot floppy.

Very odd.

Cheers,
Graham




John> Graham Williams wrote:
>> I've also just started moving to Potato and was trying the
>> 2.2.12 (kernel-image-2.2.12_2.2.12-4.deb) and 2.2.13
>> (kernel-image-2.2.13_2.2.13-1.deb) kernels, creating a boot
>> floppy, and getting exactly the same problem, with the same
>> message over and over again:
>> 
>> 0400 AX: 0212 BX: 7000 CX: 5001 DX: 
>> 
>> I'd also welcome any suggestions as to what the problem is?
>> 
>> Cheers, Graham
>> 

>>  >>>>> Bart Szyszka wrote to on 28 Nov 1999 12:32:20 +1100:
>> 
Bart> Hi, I'm trying to upgrade my kernel and have downloaded some
Bart> of the kernel... packages, including the kernel-image-2.2.13
Bart> one. My problem is that during its setup, when it asks me to
Bart> create a boot disk and makes one, the disk doesn't seem to
Bart> work. When I reboot and put it in, all it says is 'Loading'
Bart> (no Syslinux, no version numbers, nothing) followed by a few
Bart> lines of dots, and then this gets repeated over and over:
Bart> AX: 0212 BX: 7000 CX: 5001 DX: .
>>
Bart> Any ideas? Is there a package I can download that will let
Bart> me make another boot disk? I prefer booting to Linux by just
Bart> putting the floppy in and turning the computer on instead of
Bart> needing to wait for that right moment for LILO to come up (I
Bart> dual-boot with Windows and unfortunately Win98 needs to be
Bart> the default). BTW, I did use a different disk so I still
Bart> have the disk that'll boot with my older kernel (2.0.38), so
Bart> I still have access to Debian (potato).
>> 
>> --
>> Unsubscribe?  mail -s unsubscribe
>> [EMAIL PROTECTED] < /dev/null

John> --

John> Powered by the Penguin

xserver 30 seconds startup delay with usb?

[Graham Williams]

Any ideas on why starting X or changing virtual terminals to the X
window (C-A-F7) might pause for 30 seconds with a blank screen or at
best the background and window outlines showing. Seems like it is
timing out waiting for something? Perhaps associated with usb or a usb
mouse?

This is occurring on a Dell OptiPlex GX270 with an Intel 82865G
Integrated Graphics Device (i810), currently up to date with unstable
but with the xserver-xfree86 from experimental (for support of the
chipset) 4.3.0-0pre1v4.

Regards,
Graham


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

Re: LP -> CD

[Graham Williams]

Received Sat 06 Dec 2003 10:32pm +1100 from Luis Fernando Llana D?az:
> Hi all,
>   I have several LP's that I would like to transfer to CD's. I have already 
> connected a LP player to my sound card and I cad heard it. I am searching 
> programs to:
> 1. Record the LP in the hard disk.
> 2. Split the sound files in their tracks.
> 

I put together some notes about how I do this at

  http://togaware.com/linux/survivor/Recording.html

and particularly "Using Gramofile" and "Process"

Regards,
Graham


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

Re: Debian and Dell?

[Graham Williams]

Received Fri 17 Dec 2004  1:41am +1100 from Jon Dowland:
> Writing this using Debian/sarge kernel 2.4.27-1-686 on a Dell Optiplex
> GX270 with a 17" Dell TFT (not sure of model #).
> 
> So far no problems although I think I faffed around with XF86Config
> quite a bit.

Likewise, running Debian sid on Dell Optiplex GX270 with TFT. Very
nice set up. Only problem was in being mislead by Dell about the audio
chip that is sold as a Sound Blaster Live! sound card but is NOT - it
is not a real SB Live! but some cut down version that requires
software (MSWindows) support :-(

My experience with installing is at 
   
http://www.togaware.com/linux/survivor/Dell_Optiplex.shtml

Regards,
Graham


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

Re: using cd-rw

[Graham Williams]

Received Fri 17 Dec 2004  3:15pm +1100 from jack kinnon:
> Hi,
> 
>  Is there a 'cd-rw howto' for Debian? Or one that is
> applicable to Debian too?

Some information at

 http://www.togaware.com/linux/survivor/CD_Reading.shtml

Hope it might help.

Regards,
Graham


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

Re: tg3 driver

[Graham Williams]

Received Wed 11 May 2005  7:41pm +1000 from Amira Youssef:
> Hello,
>
> I'm looking for tg3 driver source or bin (kernel 2.4.18-1-686-smp)
>
> can any one advise where I can find them?

Perhaps http://survivor.sarovar.org/Tigon3_Network.html might help.

Regards,
Graham


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

Re: wajig broken in unstable?

[Graham Williams]

Received Wed 29 Jun 2005  7:39pm +1000 from Anthony Campbell:
> I found that wajig disappeared here after an update to unstable a couple
> of days ago and attempts to reintall it are greeted with the message
> that it has unmet dependencies and I should file a bug report. However,
> bug reporting seems to be failing on Debian at present.
> 
> Not sure what is happening here.

Hi Anthony. Any more specific details? What is it dependent on that is
failing? I presume you might be using sarge (stable) and have upgraded
to unstable? Did you use wajig to do this? 

Regards,
Graham


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

Re: wajig broken in unstable?

[Graham Williams]

Received Wed 29 Jun 2005  8:43pm +1000 from Clive Menzies:
> On (29/06/05 11:09), Anthony Campbell wrote:
> > To: debian-user@lists.debian.org
> > From: Anthony Campbell <[EMAIL PROTECTED]>
> > Date: Wed, 29 Jun 2005 11:09:18 +0100
> > Subject: Re: wajig broken in unstable?
> > 
> > On 29 Jun 2005, Graham Williams wrote:
> > > Received Wed 29 Jun 2005  7:39pm +1000 from Anthony Campbell:
> > > > I found that wajig disappeared here after an update to unstable a couple
> > > > of days ago and attempts to reintall it are greeted with the message
> > > > that it has unmet dependencies and I should file a bug report. However,
> > > > bug reporting seems to be failing on Debian at present.
> > > > 
> > > > Not sure what is happening here.
> > > 
> > > Hi Anthony. Any more specific details? What is it dependent on that is
> > > failing? I presume you might be using sarge (stable) and have upgraded
> > > to unstable? Did you use wajig to do this? 
> > > 
> > > Regards,
> > > Graham
> > > 
> > 
> > No, I've been following unstable for many months. In fact, I discovered
> > that synaptic and aptitude had also been removed, but I've reinstalled
> > those without problems. I've been using wajig routinely for some time.
> > 
> > The message I get now is:
> > 
> > 
> > Reading package lists... Done
> > Building dependency tree... Done
> > Some packages could not be installed. This may mean that you have
> > requested an impossible situation or if you are using the unstable
> > distribution that some required packages have not yet been created
> > or been moved out of Incoming.
> > 
> > Since you only requested a single operation it is extremely likely that
> > the package is simply not installable and a bug report against
> > that package should be filed.
> > The following information may help to resolve the situation:
> > 
> > The following packages have unmet dependencies:
> >   wajig: Depends: python-apt but it is not going to be installed
> > E: Broken packages
> 
> I'm coming in late on this.  Aptitude threatened to remove itself here
> (sid) during the last couple of upgrades because apt and apt-utils were
> being upgraded but aptitude is dependant on the currently installed
> version; so I held apt and apt-utils.
> 
> I suspect that you're seeing this behaviour because you've upgraded
> these to 0.6.38 but synaptic/aptitude depend on 0.5.28.6.
> 
> You could downgrade these two packages manually to rectify the problem
> perhaps?
> 
> Regards
> 
> Clive

Thanks for the information.  Looks like the apt upgrade is breaking a
few dependencies.  I am getting the same problem now! I'll have a look
to see what needs to be done for wajig. As suggested, downgrading apt
and apt-util and then putting them on hold might allow wajig to be
reinstalled.

Regards,
Graham


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

Re: wajig broken in unstable?

[Graham Williams]

Received Wed 29 Jun 2005  7:39pm +1000 from Anthony Campbell:
> I found that wajig disappeared here after an update to unstable a couple
> of days ago and attempts to reintall it are greeted with the message
> that it has unmet dependencies and I should file a bug report. However,
> bug reporting seems to be failing on Debian at present.
> 
> Not sure what is happening here.
> 
> AC

I believe some of the apt induced issues have been resolved - at least
it is no longer attempting to remove wajig. Should be able to
reinstall.

Regards,
Graham


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

Info on kernel-image packages?

[Graham Williams]

Susan G Kleinmann wrote  on 03 Aug 1996 20:14:45 +1000:

Susan> ...

Susan> Yes, you need a kernel with special support for the mouse
Susan> compiled in.  Reading the 'config-?' files in
Susan> buzz-fixed/disks-i386/special-kernels it appears that all
Susan> of the special kernels there have support for the PS/2
Susan> mouse incorporated as a module.  Several other types of bus
Susan> mice are supported as well.

Apart from doing a diff on these config-? files, is there any
documentation on what each version of the kernel-image contains.  Is
it also correct to assume that the -0 one is the default that ships
with Debian?

Thanks

Re: Installed from floppies, now what?

[Graham Williams]

I personally find dselect very confusing.  I install all packages
using just dpkg (dselect is a front-end to dpkg). It is actually very
easy.  The only thing that is more difficult is in manually dealing
with dependencies between packages.

The simplest thing to do is to identify any packages you want, e.g.,
minicom, ppp, etc. Download your selected packages from any Debian
archive (best to get them from the stable distribution).  For the two
I mention above, this would be:

stable/binary-i386/comm/minicom_1.75-1.deb
stable/binary-i386/net/ppp_2.2.0f-2.deb

Once you have them on your Linux box you should check what other
packages they depend on:

$ dpkg --info minicom_1.75-1.deb

This lists various things about the package.  Look for a line that
begins with "Depends: "

You can do a "dpkg --list" to see if you have all of the required
packages.

If so, then simply install the new package:

$ dpkg --install minicom_1.75-1.deb

This actually unpacks the package and then configures the package.

If you don't like the package you can simply remove it with:

$ dpkg --purge minicom

It removes all files that were installed and undoes appropriate
configurations.

Hope this helps.

(My Linux box is home at present so I'm working from memory but I've
done this often enough to be pretty sure it's correct.  The "--help"
option of dpkg lists all available options.)

--
[EMAIL PROTECTED],--_|\   Tel: (+61 6) 216 7042
Senior Research Scientist, Data Mining /  \  Fax: (+61 6) 216 7112
CSIRO Div of Info Tech, Aust Nat Univ  \_.--_*/  GPO Box 664  Canberra
http://www.dit.csiro.au/~gjw>   v   ACT  2601   Australia

Re: catch 22?

[Graham Williams]

Jim Pick wrote on 08 Aug 1996 04:44:21 +1000:

wb2oyc> <...> In an effort to get a working 2.0.0 kernel with the
wb2oyc> proper options <...> results in errors during the prerem
wb2oyc> or postrem scripts for both the source and image package.

Jim> I had the same problem with a virgin Debian 1.1 installation.
Jim> I investigated and found that the kernel-image/kernel-source
Jim> postinst and prerm scripts reference "#! /bin/perl".
Jim> However, on my virgin Debian 1.1 installation, there was no
Jim> symlink from /bin/perl to /usr/bin/perl!  I manually created
Jim> the symlink and that fixed the problem.

This is Bug#3951 I think.  I had the same problem.

Cheers,
Graham

Re: Backspace key doesn't work

[Graham Williams]

> Dave Sherohman wrote to Colin Watson on 13 Nov 2000 12:55:36 +1100:

Dave> Different person, same problem.  It _was_ working yesterday,
Dave> then I upgraded a bunch of (woody) packages and backspace
Dave> stopped working.  The Windows key also stopped being usable
Dave> as a meta.

Ditto.  Also, and at the same time, under one user's account (and only
one) on the same machine the left mouse button click with X stopped
working?  I wish I knew where to look to fix the mouse problem.  A
temporary fix that worked for me for the backspace problem was:

  xmodmap -e 'keycode 22 = BackSpace'

Cheers,
Graham

XFree 4.0.1 upgrade problem: Mouse now a big blob?

[Graham Williams]

I've just upgraded my "unstable" installation to use XFree 4.0.1
(previously XFree 3.3.6) on a neomagic-based laptop. I run Helix GNOME
and gdm with two displays active by default. Soon after installing I
was moving from one X display to the other (Ctl-Alt-F8) and suddenly
the mouse changed from the usual pointer to a roughly 1 inch square
blob!!  It's now basically a white square with black vertical
lines through it a bit like a bar code. Any ideas what could have
happened and how to fix?  I've rebooted, I've hunted around for image
masks, I've rapidly swapped between displays, I've run out of
ideas.

Cheers,
Graham

Re: apt-get new helixcode gnome

[Graham Williams]

My system also started life as slink, and has been potato for the past
few weeks. When I tried Helix Gnome I also got some error like yours
about a package (as I recall it was some library rather than grip). I
simply changed potato to woody in my sources.list (thought I would
take the leap) and things have worked just fine since.  (I have not
dist-upgrade'd or even simply upgrade'd to woody,  instead planning
to install packages as I see fit.)

I did have another problem though. I did:

  apt-get install task-gnome-helix

and left it going unattended. When i go back what was left on the
screen looked okay, but for some reason not all of the Helix gnome
packages were obtained.  I ended up with a mix of new and old gnome
packages (things like panel and gnome-core had not been updated).  I
then grepped the package names that Helix supply on their ftp site and
did:

 apt-get install 

This got everything else I needed (I did not install the dev packages)
and now I'm happily running a very nice looking Helix Gnome.

This dose not specifically sound like your problem, but it is a
"success" installing Helix Gnome. 

gmc icons end up with title bars on Helix Gnome?

[Graham Williams]

I've installed Helix Gnome on a Debian "Potato moving to Woody"
machine.  It works great, and the extra bit of effort that Helix has
put into the presentation is really worth it (IMHO).

I've installed a number of user accounts starting from scratch for
each one of them (removing previous .gnome etc directories - I'm sure
it wasn't necessary but I wanted to start fresh).

All have moved from enlightenment to sawfish.

Just one of them has odd behaviour that I'm trying to understand.
With a number of workspaces, their gmc icons (Home directory,
Slashdot, etc) remain only on the first workspace.  But worse than
this, the icons sometimes get a title bar with buttons!!  Looking
very ugly.  Often happens when using the Control Center and also it
seems under other circumstances that I can't quite get a grip on yet.

This user has the cooper sawfish theme and the seaGTK gtk theme, but a
little experimenting isn't conclusive about whether it is the themes
that aftects this?

I do not (yet) see this behaviour on other accounts.

Is this something controlled by the themes, and any ideas where I look
to "fix" this.

Many thanks,
Graham

Finding old potato packages (downgrading libpng2)

[Graham Williams]

I need to downgrade libpng2 from the current 1.0.5 in potato to the
older 1.0.3 (which I believe will fix a problem with pdflatex in
tetex-bin - Debian bug #49834).

Where do I find old version of deb packages?

Cheers,
Graham

pcmcia/pppd problem in potato

[Graham Williams]

To recap: I'm somewhere along the way of moving from slink to potato.
After recompiling the kernel and pcmcia modules "pon" stopped working
with the log indicating from pppd:

  Can't get terminal parameters: Input/output error

Doing a

  setserial /dev/ttyS1 irq 0

sometimes fixes this, but turning polling off reduces the throughput
dramatically. Sometimes doing:

  setserial /dev/ttyS1 irq 3

fixes things up again (the 3 came from hunting through the messages
log and guessing the association between irq 3 and pcmcia).  Sometimes
I need to reboot to get these setserial commands to have an effect
(i.e., to get "pon" working again).

I'm getting by, but I'd sure like to understand what's going on, and
fix it?

Cheers,
Graham

>>>>> Dan Christensen wrote to Graham Williams on 30 Nov 1999 02:57:31 +1100:

Dan> ...  One thing that makes my modem work again is to type:
Dan>   setserial /dev/ttyS1 irq 0
Dan> ...

dpkg update of ALSA and ESound and GNOME

[Graham Williams]

I've mostly updated a slink installation on a Dell Latitude CPi D300XT
to potato.

In an attempt to get sound working I've compiled ALSA (alsa-source)
using make-kpkg (kernel-package) together with kernel-source-2.2.13
and pcmcia-source.

I want to remove libesd0 and replace it with libesd-alsa0 which
provides libesd0 (but also "Conflicts:" with libesd0?). I want to do
this to get sound working again within GOME/Enlightenment. (I have
other problems with ALSA at present, with no CD sound and can't play
sound files, but that's another story for now).

But 
dpkg --remove libesd0 
wants to remove all of the installed gnome et al. packages.  So 
dpkg --ignore dependency libesd0 --remove libesd0
works, then 
dpkg --install libesd-alsa0_0.2.15-12.deb
works, but the dependencies seem to want libesd0 installed again:
apt-get -f install
installs libesd0 0.2.15-12 and removes libesd-alsa0.

Have I misunderstood dpkg/apt-get or is there something wrong here?
How do I replace a library like this in-place?


Cheers,
Graham

Re: how to find a package that provides ____._____

[Graham Williams]

> I need libXt.so.6 and libX11.so.6 to run Wordperfect how do
> I find out which .deb package provides these?

If you want to find where some file lives then you can ask at
http://packages.debian.org/.

You probably want to apt-get install xlibs.

I have a script that searches from the command line:

  $ wajig whichpkg libXt.so.6
  usr/X11R6/lib/debug/libXt.so.6devel/xlibs-dbg
  usr/X11R6/lib/debug/libXt.so.6.0  devel/xlibs-dbg
  usr/X11R6/lib/libXt.so.6  libs/xlibs
  usr/X11R6/lib/libXt.so.6.0libs/xlibs
  usr/lib/libc5-compat/libXt.so.6   oldlibs/xlib6
  usr/lib/libc5-compat/libXt.so.6.0 oldlibs/xlib6
  $

Underneath this simply queries the web server:

 $ wget 
http://packages.debian.org/cgi-bin/search_contents.pl\?word=libXt.so.6\&case=insensitive\&version=unstable\&arch=i386\&directories=yes

Regards,
Graham

NIS and No buffer space available

[Graham Williams]

I'm trying to set up NIS on a "unstable" Debian machine. I believe
I've set up the config files appropriately.

$ apt-get install nis
...
Starting NIS services: ypserv yppasswdd ypxfrd ypbind

But it then sits there waiting?

I exit from this and try 

$ ypbind -d
...
Cannot register service: RPC: Unable to send; errno = No buffer space available

Any ideas what might be going on or where to start investigating?
(I've had the "No buffer space available" problem in other places
also?)

Regards,
Graham

wajig update

[Graham Williams]

Thanks for all the comments I've received re my "wajig" script.

I've put the current version at http://velox.act.cmis.csiro.au/~gjw/wajig

This is a script that simply collects together some typical package
management operations into one place.

A sample iteration for me is:

  # wajig update  (to update the list of available pkgs)

  # wajig new (to list new (to Debian) pkgs)

  Package Available
  ===-
  bugzilla  1.13+cvs20010521-2  
  bugzilla-doc  1.13+cvs20010521-2  
  gutenbook 0.1.10-1
  wing-data 0.7-6   
  xmms-infinity 0.2-1   
  yale  1.0-1   

  # wajig newupgrades (to list Debian pkgs updated on archive)

  Package Available   Installed
  ===-===-
  base-files2.2.9   2.2.8   
  blt   2.4u-4  2.4u-2  
  blt-common2.4u-4  2.4u-2  
  blt-dev   2.4u-4  2.4u-2  
  debhelper 3.0.36  3.0.18  
  fetchmail 5.8.7-1 5.8.1-5 
  fetchmailconf 5.8.7-1 5.8.1-5 
  finger0.17-3  0.17-1  
  gnomeicu  0.96.1-30.96.1-2
  libnspr4  2:0.9.1-3   M18-3   
  libpspell20.11.2-2.2  0.11.2-2.1  
  python2-mysqldb   0.9.0-1 0.3.5-2 

  # wajig install(to install the named package)
  
  # wajig whatis galeon

  galeon 0.11.0-1.1  (763k)
Mozilla based web browser with GNOME look and feel

  # wajig whichpkg bin/domainname

  bin/domainnamenet/nis

  # wajig install nis

and so on.

Re: apt-get/dselect-problem

[Graham Williams]

On 12/07/01 22:38 +0200, Martin Bretschneider wrote:
[...] 
> Hm, I yust wanna update few debs day by day. If I understand it the right
> way, I can make an 'apt-get update' and then install my wanted deb with
> "apt-get [deb]. Right?
[...]

# apt-get install emacs20

will get emacs20 deb and required dependencies from where you've told
apt to look (in /etc/apt/sources.list) and install them all.

I also like to do an update regularly and see what's new (new packages
and new upgrades). Then choose which I want to install and upgrade.
I've used dselect, deity, and aptitude, and they're okay but I like to
feel a bit more in control of what is happening.  So I wrote a script
(which I call wajig and make available at
http://edm.act.cmis.csiro.au/debian/wajig) that keeps track of changes
since I last did an update (the script simply captures some of the
tricks for managing packages I've observed over the years). I
regularly do something like:

# wajig update  (just apt-get update)
[...]

# wajig new
Package Available
===-
3dwm-geoclient  0.2.4-2 
3dwm-server 0.2.4-2 
3dwm-texclient  0.2.4-2 
3dwm-vncclient  0.2.4-2 
abbrowser   4:2.1.1-4   
gphoto2 2.0beta1-4  
gphoto2-dev 2.0beta1-4  
jython  2.1-alpha1-1
jython-doc  2.1-alpha1-1
[...]

# wajig whatis jython
jython-doc 2.1-alpha1-1  (269.3k)
Jython documentation including API docs
jython 2.1-alpha1-1  (751.8k)
Python seamlessly integrated with Java

# wajig install jython
[...]

# wajig newupgrades
Package Available   Installed
===-===-
adduser 3.393.33
auctex  10.0g-3 10.0d-1 
base-files  2.2.10  2.2.4   
bonobo  1.0.7-1 0.37-5  
cdrecord4:1.10-13:1.9-1 
debconf 0.9.77  0.9.31  
emacs20 20.7-9  20.7-3  
eperl   2.2.14-32.2.14-0.6  
[...]

Regards,
Graham

Re: apt-get dselect upgrade problems

[Graham Williams]

Received Fri 13 Jul 2001  9:16pm +1000 from Joost Kooij:
> On Thu, Jul 12, 2001 at 05:58:19PM -0400, John Covici wrote:
> > But this doesn't solve my problem -- why am I getting errors which say
> > no such file or directory -- its looking for a file which isn't there

Sorry not to have any useful insights but I have over the past week or
so been noticing exactly the same behaviour on occasion.  My apt-get
sources.list points to ftp.debian.org (amongst others) unstable.

Quite a few times I've "apt-get update".  I then check to see which
packages have been upgraded wrt my installation, then go to install
them with apt-get install and the deb files are not there. Checking
the actual ftp archive and sure enough they are not there. If I come
back some hours later and try the install again, it's okay.  Seems
like a lag somehow between the available packages list and the actual
packages appearing.

Regards,
Graham

fetchmail fails to RETR?

[Graham Williams]

I've set up fetchmail to retireve via POP3 on one machine
successfully. Same set up (as far as I can determine) on another
machine fails.  Using other ways of retrieiving POP3 mail on this
other machine works just fine.

The POP3 with fetchmail fails once it tries to download the messages:


fetchmail: POP3< +OK Mailbox open, 3 messages
fetchmail: selecting or re-polling default folder
fetchmail: POP3> STAT
fetchmail: POP3< +OK 3 27521
fetchmail: POP3> LAST
fetchmail: POP3< +OK 0
3 messages for  at  (27521 octets).
fetchmail: POP3> LIST
fetchmail: POP3< +OK Mailbox scan listing follows
fetchmail: POP3< 1 12237
fetchmail: POP3< 2 1710
fetchmail: POP3< 3 13574
fetchmail: POP3< .
fetchmail: POP3> RETR 1
fetchmail: POP3< +OK 12237 octets
reading message 1 of 3 (12237 octets)
About to rewrite Return-Path: 
Rewritten version is Return-Path: 

About to rewrite From: 
Rewritten version is From: 

About to rewrite To: 
Rewritten version is To: X



It now just sits there?

On the machine where it works I get exactly as above but then


fetchmail: forwarding to localhost
fetchmail: SMTP> MAIL FROM: SIZE=1710
fetchmail: SMTP< 250  is syntactically correct
fetchmail: SMTP> RCPT TO:
fetchmail: SMTP< 250  is syntactically correct
fetchmail: SMTP> DATA


and seems to be just fine.

Any ideas how to hunt this down?

I'm using exim, same version and config files on both machines. But I
don't even seem to get to "forwarding to localhost".  If I tell
fetchmail to use mda procmail (on this otherwise failing machine) it
does get the message from the POP server?

ii  exim   3.22-4 Exim Mailer
ii  fetchmail  5.8.6-1POP3, APOP, IMAP mail
gatherer/forwarder

Thanks,
Graham

Re: Wajig toupgrade reports stuff that wajig upgrade doesn't upgrade!

[Graham Williams]

Yes, that's an error in my comparison of versions.  I'll fix it soon
(I think there's a dpkg-compare-version script somewhere otherwise
I'll grab to code from dpkg or apt). Thanks for giving wajig a try.
All comments/suggestions most welcome.

Received Fri 20 Jul 2001 10:35pm +1000 from Colin Watson:
> [EMAIL PROTECTED] wrote:
> >I now use Graham Williams wajig shell script as a front end to apt, 
> >mostly because of the extra facilities it provides an easy interface 
> >to.
> >  ...
> >e2fsprogs   1.18-3  1.18-3.0potato1
> 
> That looks like a bug in his version comparison; 1.18-3 is older than
> 1.18-3.0potato1.
> 

Re: Wajig toupgrade reports stuff that wajig upgrade doesn't upgrade!

[Graham Williams]

> Received Fri 20 Jul 2001 10:35pm +1000 from Colin Watson:
> > [EMAIL PROTECTED] wrote:
> > >I now use Graham Williams wajig shell script as a front end to apt, 
> > >mostly because of the extra facilities it provides an easy interface 
> > >to.
> > >  ...
> > >e2fsprogs   1.18-3  1.18-3.0potato1
> > 
> > That looks like a bug in his version comparison; 1.18-3 is older than
> > 1.18-3.0potato1.

Received Sat 21 Jul 2001 10:08am +1000 from Graham Williams:
> Yes, that's an error in my comparison of versions.  I'll fix it soon

I've fixed WAJIG.

WAJIG's idea of what was updated was based on versions in the Debian
archive being different to those installed (rather than being newer
than those installed). This is mostly okay as one would normally
expect newer versions to appear in the archive.  Occasionally you
might install a newer version yourself from a different Debian
archive, or you may move your default archive from unstable to stable,
for example. Thus the behaviour you were noticing (I think).

I now check the version numbers as well (using "dpkg
--compare-versions") to only print those that are actually newer (this
does slow down this part of the script).

This new version also adds new commands:

 # wajig bug   Review bugs associated with package
 # wajig purge Remove a package and all its config files
 # wajig start  Equivalent to /etc/init.d/ start
 # wajig stop   Equivalent to /etc/init.d/ stop
 # wajig restartEquivalent to /etc/init.d/ restart

Available from: http://edm.act.cmis.csiro.au/debian/wajig

Regards,
Graham

Driver in boot floppy for install?

[Graham Williams]

I'm installing Debian from CD-ROM (2.2r2 since that's what I have
lying around).

The hard disk is a RAID array (Adaptec 2100s).  I've the driver for it
from adaptec for kernel 2.2.19 (and the CD-ROM has 2.2.18pre21 kernel
but the driver seems okay?). 

During a clean install I need to load the driver from floppy right
after the keyboard config (the driver is dpt_i2o.o). I can then
partition and initialise the hard disk.  

But what now??  I end up being asked to create a boot floppy (it
skips asking me to install on the MBR with lilo for some reason).  I
create the boot floppy and reboot but end up with a kernel panic (with
mention of block-major-8 and unable to mount root fs)

No doubt this is because it can not find any hard disks because the
driver (dpt_i2o) is not available - it's still on it's original floppy
:-)

What is the process from here to get this driver somehow on the boot
floppy so I can reboot the machine?

Thanks,
Graham

Adaptec RAID 2100s new install of Debian - please help!

[Graham Williams]

I'm installing Debian from CD-ROM (2.2).

The hard disk is a RAID array (Adaptec 2100s).  I've the driver for it
from adaptec for kernel 2.2.19 (and the CD-ROM has 2.2.18pre21 kernel
but the driver seems okay?). 

During a clean install I need to load the driver from floppy right
after the keyboard config (the driver is dpt_i2o.o). I can then
partition and initialise the hard disk.  This works just fine.

But what now??  

I end up being asked to create a boot floppy (it skips asking me to
install on the MBR with lilo for some reason).  I create the boot
floppy and reboot but end up with a kernel panic (with mention of
block-major-8 and unable to mount root fs)

No doubt this is because it can not find any hard disks because the
driver (dpt_i2o) is not available - it's still on it's original floppy
:-)

What is the process from here to get this driver somehow on the boot
floppy so I can reboot the machine?

Another attempt tried copying the driver to the right place:
  Clean install
  Load dpt_i2o.o from floppy early in install
  Partition and initialise hard disk   
  Install drivers and base from CD-ROM
  In a console window
  Copy dpt_i2o.o to /lib/modules/2.2.18pre21/scsi/ 
  Do depmod -a
  Back to the install
  Create a boot floppy
  Reboot

The initial part of the reboot goes extremely slow.  This is the part
where it prints Loading Linux...  This takes a minute or more per
dot!

Eventually VFS Kernel Panic.  Cant mount root fs.  There's also a
message about block-major-8.

Thanks,
Graham

Modules suddenly missing

[Graham Williams]

About a week ago I updated a Debian box, running a custom kernel
2.4.16, with a dist-upgrade in unstable (it's been running unstable
for a long time, and this just brought it up to date).

On the next reboot (a few days later) things started breaking! In
particular nfs, audio, and cdrom and cdrw access.  I think I've
tracked the problems to the appropriate kernel modules not being
loaded? Why would a dist-upgrade cause this? I can add nfs, and nfsd
to /etc/modules and NFS is working again. I can add i810_audio and
audio is up agian. Also isofs, ide-cd, ide-probe-disk, and ide-disk
get CDROM going again.  I've failed so far to get the CD-RW working
again:

# mount /cdrw
mount: /dev/scd0 is not a valid block device

My questions: 

   What could have changed to the modules stuff to cause these to no
   longer be loaded on boot?

   Any ideas on why my ide-scsi CD-RW set up would be failing.  A
   scanbus does not find the ide-scsi CD-RW.

Thanks
Graham

==
Debian GNU/Linux Guide at http://edm.act.cmis.csiro.au/debian/book/
==

Re: Package of a file in debian?

[Graham Williams]

Received Sun 03 Mar 2002  7:02am +1100 from Xeno Campanoli:
> Is there a similar command in apt or dpkg to this thing with rpm that
> allows you to figure out the package of a file?  That really would have
> helped me recently.

As others have noted, "dpkg -S" works for installed packages and the
web page works for all packages.

My "wajig" package (available from Debian unstable) queries the web
page from the command line:

$ wajig whichpkg libcrypto
File Path   Package
===-=
usr/lib/erlang/lib/crypto-1.1.2/priv/lib/elibcrypto.so  non-us/erlang
usr/lib/libcrypto.sonon-us/libssl-dev
usr/lib/libcrypto.so.0  non-us/libssl09
usr/lib/libcrypto.so.0.9.4  non-us/libssl09
usr/lib/libcrypto.so.0.9.5  non-us/libssl095a
usr/lib/libcrypto.so.0.9.6  non-us/libssl0.9.6

Regards,
Graham

Re: geforce2 'rapid' screen save

[Graham Williams]

I've the NVIDIA driver on two machines, one a TNT2 M64 the other a
GeForce 256.

On the TNT M64 machine, when switching virtual screens using
Ctl-Alt-F8 (e.g., running two X servers) the screen goes blank as
usual, but then stays that way permanently! The whole machine is
locked, keyboard and all. The only solution seems to be a power cycle.

This has also happened when running some of the GL screen savers in
xscreensaver.

Using the "nv" driver from XFree86 I have never had this
problem. Normally when switch virtual screens the current screen goes
blank, the monitor momentarily looses the signal, then the new screen
comes to life! This seems to normal behaviour.

On the GeForce machine I only run a single X server and have not seen
this problem, although exiting from the evas_test program from the
evas-demo package crashed the X server.

Cheers,
Graham

ipchains, log message on console, and ppp networking?

[Graham Williams]

I'm get the following Packet log each time I'm connected to my ISP,
every 4 seconds

Packet log: input DENY ppp0 PROTO=88 152.83.4.1:65535 224.0.0.10:65535
L=60 S=0xC0 I=0 F=0x T=2 (#9)

Any help explaining what this is?  (I've read through the IPCHAINS
HOWTO and have searched quite a bit. I'm close but I don't know what
PROTO=88 is nor what the significance of port 65535 is nor why I
should be getting these messages every 4 seconds, nor why the
destination is 224.0.0.10 which has nothing to do with my host -
152.83.4.1 is the ISP.)

These messages are displayed on the virutal console that is currently
active (tty1 to tty6) and quickly fill the screen. Why would it be
placing the messages there instead of /var/log/kern.log?

I also run ipmasq with another machine connected to my main machine
(main machine doing the PPP). This works but when I exit from ppp it
takes down the local network (on eth0).  How do I stop this happening?

Thanks, in advance, for any help.

Cheers,
Graham

NFS kernel daemon: nfsdnfssvc: Invalid argument

[Graham Williams]

I've just "apt-get dist-upgrade" a SS2.  I'm running unstable and when
then nfs-kernel-server was being upgraded it gave the message:

Starting NFS kernel daemon: nfsdnfssvc: Invalid argument
 mountd.

Until now NFS has been working just fine.  I'm running kernel 2.2.19.
(NFS is now not working - remote hosts can not access this host.)

Any ideas?

Regards,
Graham


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

NFS kernel daemon: nfsdnfssvc: Invalid argument

[Graham Williams]

I've just "apt-get dist-upgrade" a couple of machines (one sparc one
i386).  I'm running unstable and when nfs-kernel-server was being
upgraded it gave the message:

Starting NFS kernel daemon: nfsdnfssvc: Invalid argument
 mountd.

Until now NFS has been working just fine.  I'm running kernel 2.2.19. 
(NFS is now not working - remote hosts can not access this host.)

Changing to nfs-user-server only partially works.  Some exports are
working just fine while others are giving me:

mount: wrong fs type, bad option, bad superblock on hostname:/dmnt/vlx,
   or too many mounted file systems

and yet other "identical" mounts work just fine..

Any ideas?

Regards,
Graham

wajig: Simplified command line admin for Debian

[Graham Williams]

Wajig is a simplified command line interface to many of the typical
Debian administration tasks, including package management and
configuration, and daemon control. It is essentially a collection of
"tricks of the Debian trade" packaged up into one python
program. Underneath it uses apt-get, dpkg, apt-cache, wget, and much
more.  On the surface it is based around a collection of commands that
hopefully make some sense.

Wajig can be installed by adding the following to
/etc/apt/sources.list (not yet an official Debian package):

deb http://edm.act.cmis.csiro.au/debian ./

and then "apt-get update" and "apt-get install wajig".

Wajig is designed to run as a normal user, using "sudo" as required,
but can also run as root without any extra setup. Using "sudo"
requires a little setting up as described with the doc command "wajig
doc".

Try "wajig help" for a list of common commands and "wajig -v help" for
a complete list of available commands. Examples include:

  wajig update   (= apt-get update)
  wajig install less (= apt-get install less)
  wajig new  (list new packages since last update)
  wajig newupgrades  (list packages upgraded since last update)
  wajig updatealts editor(update the default "editor")
  wajig restart apache   (restart the apache daemon)
  wajig listfiles less   (list the files supplied by the "less" pkg)
  wajig whichpkg stdio.h (what package supplies this header file)
  wajig whatis rats  (one line description of the pakacge "rats")
  wajig orphans  (list libraries not required by other pkgs)

and many more (see below for current complete list).

Wajig has evolved over many years and there's a small band of
users. It has some of the same aims as the feta package and I thought
to wrap the extra wajig features into feta, but a number of users
suggested that wajig should stay.  So it was rewritten recently from
being a shell script to a python program. It is available under the
GPL.

$ wajig -v help
All wajig commands:

 autoclean  Remove all superseded downloaded deb files
 bugCheck for reported bugs in package using the Debian Bug Tacker
 build  Retrieve/unpack sources and build .deb for the named packages
 dist-upgrade   Upgrade to a new distribution (installed and new rqd packages)
 clean  Remove all downloaded deb files
 describe   Provide a description of package (-v and -vv for more detail)
 docEquivalent to help with -verbose=2
 editsourcesEdit the sources.list file which locates Debian archives
 findfile   Search for a file within installed pacakges
 force  Install package even if it overwrites files from other pkgs
 help   Print documentation (detail depends on --verbose)
 hold   Place listed packages on hold so they are not upgraded
 init   Initialise or reset the wajig archive files
 installInstall (or upgrade) one or more packages or a .deb file
 integrity  Check the integrity of installed packages (through checksums)
 lastupdate Identify when an update was last performed
 list   List the status of every known package
 listalts   List the objects that can have alternatives configured
 listfiles  List the files that are supplied by the named package
 listnames  List all known packages or those containing supplied string
 newList packages that became available since last update
 newupgradesList packages newly available for upgrading
 orphansList libraries not required by any installed package
 purge  Remove one or more packages and configuration files
 reconfigureReconfigure the named installed packages
 reinstall  Reinstall each of the named packages
 remove Remove one or more packages (see also purge)
 repackage  Generate a .deb file for an installed package
 reset  Initialise or reset the wajig archive files
 search Search for packages containing listed words
 source Retrieve and unpack sources for the named packages
 status Show the version and available version of packages
 toupgrade  List packages with newer versions available for upgrading
 update Update the list of downloadable packages
 updatealts Update the default alternative for things like x-window-manager
 upgradeUpgrade all of the installed packages
 whatis For each package named obtain a one line description
 whichpkg   Find the package that supplies the given command or file 

Command line options:

 -h|--help  Print usage message.
 -q|--quiet Do system commands everything quietly.
 -t|--teaching  Trace the sequence of commands performed.
 -v|--verbose=n Increase (or set) the level of verbosity (to n).

Run `wajig -vv help' for documentation.

Regards,
Graham Williams

Re: latex packages

[Graham Williams]

Received Thu 24 Jan 2002 12:12pm +1100 from martin f krafft:
> also sprach nori heikkinen <[EMAIL PROTECTED]> [2002.01.24.0203 +0100]:
> > copying the list because i don't know, and i certainly didn't write it
> > myself.  i just have it, and i find that explicitly setting the
> > margins gives much better results -- i hate those huge top margins,
> > too.  anyone know where to get extra LaTeX packages?
> 
> the geometry package is in Debian's 'tetex-extra' package.
> 
> and check [1] for more TeX packages, which Debian doesn't provide...
> 
>   1. http://www.ctan.org/

The geometry package is often the right way to go. Also, have a look
at

  /usr/share/doc/texmf/help/Catalogue/catalogue.html

for a list of many LaTeX packages.

Regards,
Graham

Re: latex packages

[Graham Williams]

Received Thu 24 Jan 2002  3:18pm +1100 from Brian P. Flaherty:
> I would check the catalouge on-line at www.ctan.org, under 'Graham
> Williams's catalogue.'  I would assume a package installed version
> could get old pretty quickly.  But, obviously it is better than
> nothing if you don't have constant web access.  As for which Debian
> package has it, I would guess tetex-doc, but I am not positive.
> 
> Brian

Yes, tetex-doc, viz

$ wajig whichpkg catalogue.html
File Path   Package
===-=
usr/share/doc/texmf/help/Catalogue/catalogue.html   tex/tetex-doc
usr/share/doc/texmf/help/Catalogue/entries/catalogue.html   tex/tetex-doc

Nothin' like pushing one's borrow here :-) Woops.

Regards,
Graham

gnome-panel not sharing /tmp/languages?

[Graham Williams]

I've a Debian "unstable" installation, up-to-date.

If more than a single user is using gnome on the console, then the
latter users can not start the gnome panel (as of
gnome-panel_1.4.0.3-1). 

After some hunting around I think the problem might be that
/usr/bin/panel seems to create /tmp/languages owned by the first user.
When another user starts up Gnome and panel the panel application is
unable to start up the panel because the file /tmp/languages already
exists and is owned by someone else. This file does not go away,
either, when the first user exits. 

Temporary solution is for the user to remove /tmp/languages each time
they exit, although this does not help multiple concurrent Gnome
sessions.

Any ideas if there is something in my set up that causes this? 

Seems like older versions of gnome-panel do not have anything to do
with /tmp/languages?

Cheers,
Graham

cdrecord can't get mmap on /dev/zero

[Graham Williams]

Using cdrecord 1.10a18-2 on an up to date Debian unstable installation
with kernel 2.2.18 (self compiled) trying to write to CD writer:

 cdrecord: Invalid argument. Cannot get mmap for 4198400 Bytes on
 /dev/zero

Any ideas what the problem is and a solution?

Many Thanks,
Graham

Re: Proposal: Weekly Debian user FAQ.

[Graham Williams]

I've just caught up on some of the discussion on debian-doc re a
Debian FAQ.

For my own purposes I generally grab tidbits of information from
debian-user and debian-devel etc, plus my own experiences in
installing debian, and record them in a document that I also happen to
convert to HTML.

It's nothing fancy nor complete, but perhaps there are bits there that
others may find useful.  I also regularly update it and am
transferring some older documents I have into this same format. (Over
the years I've maintained documents on installing Debian on particular
Dell laptops and have received a lot of positive and constructive
feedback.)

There are two documents available from:

  http://edm.act.cmis.csiro.au/debian/debtopics/
  http://edm.act.cmis.csiro.au/debian/debinstall/

There are PDF versions available:

  http://edm.act.cmis.csiro.au/debian/debtopics.pdf
  http://edm.act.cmis.csiro.au/debian/debinstall.

Regards,
Graham Williams

Wanted: apt-changes tool

[Graham Williams]

Is there such a thing as a tool for debian that looks at a current
installation and compares it to what's available (via whatever
sources.list is pointing to) and summarises the changes.  

For example, list all installed packages, their current version
number, and the available version number.

Something like:

# apt-changes
Package Available   Installed
===
adduser 3.363.34 
alsa-base   0.9+0beta4-30.9+0beta3-1
auctex  10.0g-2 10.0d-1 
binutils2.11.90.0.7-2   2.11.90.0.1-1 


Then, I would like to give the name of a particular package and it
will summarise changes made to the package from the version installed
to the version available. This information is perhaps cached somewhere
on the net since it goes in debian-devel-changes mailing list?

# apt-changes adduser

Changes: 
 adduser (3.36) unstable; urgency=medium
 .
   * The "I hope it resolves NIS issues" release
   * make -C /var/yp is called when invalidating nscd cache. Thanks to
 Juergen Fischer for testing. Closes: #22412
   * Included example adduser.local script from John Zaitseff.
 Closes: #94245

Pointers to where such information lives might be useful if such a
tool does not exist but could be written.

Regards,
Graham Williams

List new packages after an apt-get update

[Graham Williams]

After doing an apt-get update there are two types of packages I'm
particularly interested in: newly available packages (new to Debain
since the last time I did an apt-get update) and packages I have
already installed but that have been updated since the last time I did
an apt-get update. (I.e., ignore packages that have been updated but
not since the previous time I did an apt-get update)

Are there any simple command line methods for getting these two lists
(or do I need to persist in learning what seems to me to be the quite
un-intuitive interfaces in deity and aptitude).

Many thanks in advance,
Graham

wajig: a simple command line interface to Debian package mgmt

[Graham Williams]

I've been juggling the use of various Debian Package Management tools
over the years (including the likes of dselect, deity, deity-gtk,
aptitutde, apt-get, dpkg, gnome-apt, apt-cache, and so on).

I've written a shell script front-end which I call "wajig" to do some
of the common tasks I do in managing various Debian installations. I
include the script below in case others might find it useful or others
might give me wisdom on how to do some of these better. I would not
yet vouch for it's accuracy (in terms of identifying new and updated
packages) 

WARNING: I suggest you review the code carefully to be sure for
yourself that I don't include any trojan horses. Currently I assume
the script is being run as root. Thus, someone with less scruples than
myself :-) could mislead you into running code as root that trashes
your system, or worse.  Nonetheless, I've used this for quite a while
now and I've never trashed my system.

In brief I use it as follows:

# wajig update  (update available list: apt-get update)
# wajig new (list new packages to Debian)
# wajig newupgrade  (list upgradeable pkgs since update)
# wajig toupgrade   (list all upgradeable pkgs)

# wajig install xmms(install the xmms package)
# wajig remove xmms

# wajig listfiles xmms  (list files supplied by installed pkg)

# wajig whichpkg ppmmake(search Debian packages web page for
 packages containing ppmmake. This one
 lets me avoid going to the web page
 to find information each time)
  
and many more, including upgrade, dist-upgrade, reconfigure,
repackage, source, clean, autoclean, status, describe, search.

Code is provided as is without warranty.

Finally, I've been using Debian for many years but am not an
officially registered Debian developer.

#!/bin/bash
#
# WAJIG: Debian Package Management Front End
#
# Remembering all the different commands to get different
# information about different aspects of Debian package management and 
# then using other commands to install and remove packages 
# was all becoming too much for me.  
# 
# Swaping between dselect, deity, deity-gtk, aptitutde, apt-get, dpkg, 
# gnome-apt, apt-cache, and so on was interesting but cumbersome.
#
# This script simply collects together what I have learnt about 
# various commands!  Clearly I have yet to learn all there is.
# Also, this script is a bit of a hack, and many things could surely be 
# done better. I find it useful and in case others do I release it.
#
# Suggestions are most welcome.  Email [EMAIL PROTECTED]
#
# Currently it only runs as root which needs to be rectified. 
# Perhaps using fakeroot or sudo when (and only when) it is required.
#
# Copyright (c) 2001 Graham Williams, Kayon Toga.
#
# Code is provided as is without warranty.
#
# Released under the GNU General Public License.

self=$(basename $0)

init_dir=/root/.wajig
scratch_dir=${init_dir}

avail_now=${init_dir}/pkg.available
avail_tmp=${init_dir}/pkg.available.tmp
avail_prev=${init_dir}/pkg.available.prev

install_now=${init_dir}/pkg.installed

function usage ()
{
echo "Usage: ${self} [-h|--help]  []"
echo ""
echo "   init   Initialise the internal archives"
echo "   reset  Reset the internal archives"
echo ""
echo "   update Update the list of available packages"
echo "   upgradeUpgrade all installed pacakges"
echo "   dist-upgrade   Upgrade to a new distribution"
echo "   install  Install package from net"
echo "   installdebInstall package from a .deb file"
echo "   force "
echo "   reconfigure "
echo "   repackage Generate .deb from installed package"
echo "   remove "
echo "   source   Retrieve package source from net"
echo ""
echo "   clean  Remove all cached apt-get files"
echo "   autoclean  Remove old cached apt-get files"
echo ""
echo "   status []"
echo "   version []"
echo "   availableList versions of given packages"
echo "   describe "
echo "   description "
echo "   fulldescription "
echo "   orphans []"
echo ""
echo "   list"
echo "   listnames []"
echo "   listfiles "
echo "

Re: Anything similar to Powerpoint?

[Graham Williams]

> Matthias Richter wrote to debian-user on 09 Jun 2001 23:30:41 +1000:

>> Is there anything for Linux that provides at least some of the
>> facilities of Powerpoint?

Matthias> I've just completed a presentation using pdflatex
Matthias> (texpower, hyperref, ifmslides) \ldots so if you're
Matthias> already very familiar with (pdf)latex you might also use
Matthias> this cross-plattform way of presenting things nicely ;-)

In a similar vane I use the LaTeX "prosper" package which does a very
good job of bringing together texpower, hyperref, ifmslides, in a
simple to use and familiar (LaTeX) environment and provides
plug-n-play glitz just like powerpoint, all presented using PDF.

http://prosper.sourceforge.net/

Regards,
Graham

Re: Combining PDF documents

[Graham Williams]

Received Thu 14 Oct 2004  7:14am +1000 from Martin Dickopp:
> Michael Satterwhite <[EMAIL PROTECTED]> writes:
> 
> > In Adobe Acrobat, there is the ability to combine PDF documents, inserting 
> > document 2 at a given location within document one. Actually, all I really 
> > need is to take a document and append it to another.
> >
> > Does anyone know of a Linux tool for this.
> 
> There's a tool "pdfjoin" (in the "pdfjam" package) for this purpose.
> I've never used it myself, but according to the documentation, it can
> do this.

Works very well.  I use it regularly to combine and manipulate PDF
pages. 

Regards,
Graham


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

Re: mozilla 1.3.1 java plugin

[Graham Williams]

Received Thu 26 Jun 2003 11:03am +1000 from Tom Allison:
> OK, I'm confused.
> 
> I have a javaplugin_oji.so in whatever directory I'm supposed to have it in 
> according to the mozilla dev website.
> 
> Nothing works.  I get that stupid busted puzzle piece.

Have a read of

 http://togaware.com/linux/survivor/Mozilla_Java.html 

Perhaps it will help.


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

Ouch... kernel-image-2.4.21 - now all I get is LIL at boot

[Graham Williams]

Installed kernel-image-2.4.21-1-686. Updated lilo.conf to use this as
default. Ran lilo.  Now all I get when I reboot is "LIL" and it then
hangs.

What's the next step.. no boot floppy for this machine :-(

This is an uptodate "unstable" installation.

Regards,
Graham


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

DVD+RW gone missing on moving to kernel 2.6.6

[Graham Williams]

My "_NEC DVD+RW ND-2100AD, ATAPI CD/DVD-ROM drive" that is easily
identified under 2.4.25 (as /dev/hdc) goes missing under my 2.6.6
kernel. dmesg has no hint at all of the DVD. This is my only IDE
device (using a SATA hard drive).

Is this a matter of finding the right module? The machine is sid,
up-to-date.

Regards,
Graham


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

D-I to SATA (hda) but move to 2.6 kernel (sda) fails

[Graham Williams]

The current beta 4 debian-installer will install to SATA okay, but
treats the disk as IDE (/dev/hda). In moving to kernel 2.6.6 from sid
this uses SCSI (/dev/sda) and the reboot fails. Is there a migration
path one needs to follow?

Regards,
Graham


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

Re: cdrom does not un-mounts

[Graham Williams]

Received Sat 22 May 2004  3:04pm +1000 from J.S.Sahambi:
> I am using Debian/sid. Some times when I mount a cdrom and try to 
> unmount it immediately, it gives the following error on the terminal:
> 
> umount: /cdrom: device is busy
> umount: /cdrom: device is busy
> 
> But the fact is the cdrom is not being used by any of my terminals or 
> programs (as I have not used the terminal or any program to browse or 
> use the cdrom). The only way I have found to umount the cdrom in this 
> case is to logout (or kill the xserver with ctrl-alt-backspace) and 
> login are root or same user and unmount. Then the cdrom unmounts with 
> out any problem.

Not sure if this might be your problem, but I used to have the same,
and found it was caused by the "fam" daemon. I used to kill it with
"wajig stop fam" (or more directly as "/etc/init.d/fam stop"). Then I
was able to unmount the cdrom.

Regards,
Graham


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

Re: D-I to SATA (hda) but move to 2.6 kernel (sda) fails

[Graham Williams]

Received Sat 22 May 2004  1:53pm +1000 from David Purton:
> On Sat, May 22, 2004 at 08:59:14AM +1000, Graham Williams wrote:
> > The current beta 4 debian-installer will install to SATA okay, but
> > treats the disk as IDE (/dev/hda). In moving to kernel 2.6.6 from sid
> > this uses SCSI (/dev/sda) and the reboot fails. Is there a migration
> > path one needs to follow?
> > 
> 
> Not an answer, but a me too ;)
> 
> So when you figure out how to do this, it would be great to post your
> method to the list

I'm installing on a Dell (Optiplex GX270) so I decided to install with
Scott Kveton's Dell Debian Install disk instead (from
http://wiki.osuosl.org/display/LNX/Debian+on+Dell+Servers) which has
the libata compiled in with the 2.4.25 kernel, thus the SATA is
/dev/sda. So the upgrade to kernel 2.6.6 did not present any problem.
I put some rough notes on
http://survivor.sarovar.org/Dell_Optiplex.html.

Haven't had any response regarding migrating from /dev/hda to
/dev/sda for the SATA.

Regards


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

Re: DVD+RW gone missing on moving to kernel 2.6.6

[Graham Williams]

Received Sat 22 May 2004  7:17pm +1000 from Jonathan Matthews:
> Graham Williams had the gall to say:
> > My "_NEC DVD+RW ND-2100AD, ATAPI CD/DVD-ROM drive" that is easily
> > identified under 2.4.25 (as /dev/hdc) goes missing under my 2.6.6
> > kernel. dmesg has no hint at all of the DVD. This is my only IDE
> > device (using a SATA hard drive).
> > 
> > Is this a matter of finding the right module? The machine is sid,
> > up-to-date.
> 
> Just hit this problem myself.  Here's what I did - let me know if it 
> works:
> 
> 1) Remove any "hdc=ide-scsi" lines from /etc/lilo.conf (grub? No 
> idea...) and then run "lilo" as root.
> 
> 2) # modprobe ide-cd
> 
> 3) $ cdrecord dev=ATA -scanbus
> 
> ... and use the x,y,z triple that cdrecord reports, as in
> 
> $ cdrecord dev=ATA:x,y,z 

Thanks for this and other suggestions.  I had to also modprobe
ide-generic, after a bit of trial and error.  All works now (and I've
somewhat documented it at
http://survivor.sarovar.org/Dell_Optiplex.html.)

Regards,
Graham


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

Re: here's how I installed Sun's Java, comments are welcome

[Graham Williams]

Received Thu 03 Jun 2004  2:37am +1000 from David Baron:
> On Friday 14 May 2004 12:57, [EMAIL PROTECTED] 
> wrote:
> > If I remember correctly, all the Java stuff is installed in one place
> > (e.g. /usr/local/java/j2sdk1.4.2_04), so it is easy to maintain. You
> > can however create a deb package from the JSDK or JRE, either be
> > downloading the RPM and using alien, or by converting the installer:
> >
> > 
> 
> I did not have any luck with that package converter. I used alien on the RPM 
> and installed Sun Java 1.5.0. I places it in /usr/java which is OK.
> 
> Attempting the run Java, I get that it cannot find the main objects.
> 
> I followed instructions with the "virtual packages" which set up pointers to 
> Java and it man pages. Still cannot run it.

My log of installing Java is at: 

  http://www.togaware.com/linux/survivor/Install_Sun.shtml

Regards,
Graham


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

Re: apt-get: purging a package that has already been removed?

[Graham Williams]

Received Tue 02 Mar 2004  9:08pm +1100 from Adam Funk:
> Sorry if this is a stupid question, but...
> 
> Suppose foo has already been removed without the --purge option, and I
> later want to purge it.
> 
> # apt-get --purge remove foo
> gives the error that foo is not installed and therefore takes no action.
> 
> I can work around this with
> # apt-get install foo
> # apt-get --purge remove foo
> but that seems silly.  Is there a "correct" way to purge foo?

I do it with:

# wajig purge foo 


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

Re: what app includes 'startx' command

[Graham Williams]

Received Wed 03 Mar 2004 10:39am +1100 from MJ Inabnit:
> Greetings:
> 
> I'm testing the latest debian "Sarge" installer.  It's improving, but . . .  What 
> application includes the "startx" command?  I realize I can install 
> xdm/gdm/whateverDM, but I like keeping things "lite" on the older computers.  
> Anyway, I have x configured, I bet it'd display Fluxbox great, if I could only 
> launch x :)

Another alternative way to find out is:

$ wajig whichpkg startx
File Path   Package
===-=
/usr/X11R6/bin/startx   xbase-clients


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

Re: automating apt-get

[Graham Williams]

Received Fri 06 Feb 2004  5:18am +1100 from Deryk Barker:
> I think I'm on the verge of persuading our technical people that we
> should be using debian in our labs rather than RH.
> 
> One thing that would tip the scales is if it were possible to
> *completely* automate the upgrade process.
> 
> For instance: our techies would do an update followed by an upgrade on
> a central/test machine to ensure everything was OK. They would then
> like to be able to run the same commands on several dozen lab machine
> *without* any manual intervention (e.g. not having to answer those
> questions about  opackage configuration etc).
> 
> Is this feasible/easy? Is there a better way of mirroring a debian
> system to many others?


I think it is. I use wajig (which I wrote but is available from
debian) for something similar. On the lab machines you can configure
debconf not to ask questions and then run wajig daily-upgrade (perhaps
as a cron job if you wanted it fully automated, otherwise as a script
run manually from the central host), making sure the lab machines only
reference the central machine (i.e., the central machine is the only
one listed in /etc/apt/sources.list) which you can set up as a debian
archive of just the packages you've downloaded to that central
machine.

Regards,
Graham


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

Re: To dselect or aptitude, that is the question

[Graham Williams]

On Wed, 2004-04-21 at 06:37, Alex Malinovich wrote:
> On Tue, 2004-04-20 at 15:08, Florian Ernst wrote:
[...]
> I prefer dselect for the one task that I use it for, and that's
> dist-upgrades with the dependency resolution screen. The rest of the
> time I use wajig. The only thing that I can't do that I wish I could
> (And someone please fill me in if you know of a way) is remember which
> packages got brought in for installation of package 'a'. So when I
> remove package 'a' it'll remove it and all of the packages that were
> brought in with it.

Hi Alex,

I think there's something like this in wajig:  purge-depend. This will
purge a package and those it depend on and not required by others. Not
exactly what you were asking, but close?

Regards,
Graham



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

Re: Where is growisofs in Sarge

[Graham Williams]

On Wed, 2004-04-21 at 05:31, Florian Ernst wrote:
> Hello Paul!
> 
> On Tue, Apr 20, 2004 at 12:21:10PM -0600, Paul E Condon wrote:
> > I've moved to Sarge in anticipation of it's soon becoming
> > 'stable'. But I can't find the program growisofs, which I have come to
> > depend on to do backups of DVD+RW. What is the situation in re. a
> > Debian package that contains this program? And, if the program is
> > deprecated, or in some trouble, what is the recommended alternative
> > for writing DVDs?
> 
> Seems you missed the fine package search on the Debian homepage ;)
> 
> It's in dvd+rw-tools.

Or wajig 

$ wajig whichpkg growisofs
File Path Package
=-=
 
usr/bin/growisofs utils/dvd+rw-tools

Regards,
Graham


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

Re: Oracle install

[Graham Williams]

On Wed, 2004-04-21 at 16:32, Alex Malinovich wrote:
> On Tue, 2004-04-20 at 20:57, Cage wrote:
> > I got my hands on a copy of Oracle 8i. I'm mostly trying it out for fun. 
> >   But from what I've read it's a bear to install. I'm up to the 
> > challenge. Been getting bored lately with no new projects. Anyone 
> > install this bugger in Debian Woody and what to look out for. I found a 
> > few links on Linuxquestions.org and am looking them over but also 
> > looking for insight from y'all.
> 
> 8i is rather out of date. I'd suggest downloading either 9i or 10g from
> Oracle and trying them out. I wrote a guide last year for installing 9i
> on Sid. You can find it at:
> 
> http://www.the-love-shack.net/oracle-on-sid.shtml
> 
> I'm currently working with someone on trying to get 10g to install
> properly on a Woody system. If/when we get it taken care of, I'll add
> that to the 9i page.

And a 10g install guide is at:

http://survivor.sarovar.org/Oracle_10g.html

Regards,
Graham


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

Re: edit pdf's

[Graham Williams]

Received Wed 12 May 2004  4:14am +1000 from Antonio Rodriguez:
> On Tue, May 11, 2004 at 01:01:16PM -0400, Matt Price wrote:
[...]
> Well, if you have scanned all the pages in about the same position,
> and you can establish reassonably well the coordinates of the crop,
> you can write a script that does all the work in one step (containing
> all the inner steps).

I can't find your original post, so I might be off topic, but I use
LaTeX and the pdfpages package to combine multiple pdf documents, crop
off page headers and footers, and add my own page numbering and
running page titles, and to generate a new, single PDF.  Works rather
nicely. But if you are talking about editing the text, then this isn't
the answer.

Regards,
Graham


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

Re: dell poweredge 2200 scsi installation problem

[Graham Williams]

On Fri, 2004-05-14 at 07:09, Stu Woodbridge wrote:
> I have a dell poweredge 2200 and a problem with installing debian.
> I am using a net install from cd, the installation process does not
> detect my hard drive and asks for 
> essential modules to be loaded from floppy. I have also tried from 
> disk1 of the full set (which I downloaded and burned)

Did you see  http://oregonstate.edu/~kveton/debian/ which includes a
Debian install CD for many Dell PowerEdges (but I note 2200 is not
listed there)?

Regards,
Graham


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

Re: dell poweredge 2200 scsi installation problem

[Graham Williams]

Received Fri 14 May 2004  8:14am +1000 from Stu Woodbridge:

> that did the trick thanks

Excellent.  Would be worth reporting this to Scott Kveton so he might
add the 2200 to his list of machines his Debian Install disk is known
to work on.

Regards,
Graham

Received Fri 14 May 2004  8:14am +1000 from Stu Woodbridge:

> - Original Message - 
> From: "Graham Williams" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Friday, May 14, 2004 9:25 AM
> Subject: Re: dell poweredge 2200 scsi installation problem
> 
> 
> > On Fri, 2004-05-14 at 07:09, Stu Woodbridge wrote:
> > > I have a dell poweredge 2200 and a problem with installing debian.
> > > I am using a net install from cd, the installation process does not
> > > detect my hard drive and asks for
> > > essential modules to be loaded from floppy. I have also tried from
> > > disk1 of the full set (which I downloaded and burned)
> >
> > Did you see  http://oregonstate.edu/~kveton/debian/ which includes a
> > Debian install CD for many Dell PowerEdges (but I note 2200 is not
> > listed there)?
> >
> > Regards,
> > Graham


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

Re: libXrandr.so.2 not installed

[Graham Williams]

On Sun, 2004-05-02 at 05:36, Bruce Miller wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
> 
> On May 1, 2004 15:07, Juan Noguera wrote:
> > Hello,
> >
> > When trying to run the mysqlcc, the system complains:
> >
> > "Error while loading shared libraries: libXrandr.so.2 cannot open
> > shared object file: No such file or directory"
> >
> > I've been searching around in the web but I nothing I have read works
> > for me. The thing is that there is no libXrandr.so.2 (or anything
> > with a similar name) in my system. XFree86 seems to be running
> > reasonably well, but this library is not there.
> >
> > Can someone tell me how to get  libXrandr.so.2  installed?
> 
> dpkg -S libXrandr.so.2
> apt-cache show libxrandr2
> apt-get install libxrandr2

or if you prefer not remembering the right command for the right job:

 wajig whichpkg libXrandr.so.2
 wajig detail libxrandr2
 wajig install libxrandr2



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

Re: dist-upgrade problem

[Graham Williams]

Received Sun 08 Jul 2007  7:23pm +1000 from Jonathan Kaye:
> Zach wrote:
> 
> > On 7/8/07, Jonathan Kaye <[EMAIL PROTECTED]> wrote:
> >>
> >> Hi Zach,
> >> It might be a better idea to run aptitude (or apt-get) upgrade rather
> >> than dist-upgrade as a matter of routine. I think you will avoid these
> >> kinds of problems this way.
> >> Cheers,
> >> Jonathan
> > 
> > Hi Jonathan,
> > 
> > Ok, I'm curious though 1) why it wants to install all the texlive-*
> > packages when I don't even have texlive installed and 2) how i could
> > force it to not install these during the dist-upgrade?
> > 
> > Zach
> I think you can mark them as "hold" as in aptitude hold texlive.
> I still don't know why you want to do a dist-upgrade rather than a simple
> upgrade as a matter of routine. Have you tried doing just aptitude upgrade
> and see if it still wants to install texlive? I would use dist-upgrade if I
> were going from Etch to Lenny for example. dist-upgrade is more agressive
> than upgrade about what it chooses to upgrade. Try reading the manual for a
> better description than mine.
> Cheers,
> Jonathan 

tetex no longer exists (except historically) and to automatically and seamlessly
transition people to the new TeX (texlive) the tetex packages are
probably empty packages depending on texlive. Thus apt-get is trying
to upgrade you.

Unless you have a specific reason to stay with tetex, might be best to
just go with the flow. My observation is that the transition, left to
apt-get, works just fine.

Regards,
Graham


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

etch to lenny upgrade - X apps no longer see keystrokes?

[Graham Williams]

Have just upgraded a fairly vanilla etch install on a Dell Precision
690 (AMD64) with an nvidia graphics chip. All seemed to proceed well,
but on reboot and starting up GDM, most key presses result in the
screen resolution changing - I can't login!

After quite a bit of research and attempts to determine what is going
on, I have run out of ideas! Ctrl-Alt-f1, etc, do not function.  The
simplest way I've figured out to log on is through single user
mode. Keyboard works just fine there. Booting into a Red Hat partition
is also just fine.

I've created a .xinitrc which only runs xev so I can see what keys it
is seeing. When I run with video driver as "nv" (in
/etc/X11/xorg.conf) xev is not seeing any keyboard activity. Changing
to "vga" at least I can see that xev gets the keystrokes (but the
screen is not usable). Changing to "vesa" exhibits the same behaviour
as "nv" - that is, no keys reported by xev, and any key press seems to
change the screen resolution.

With the "nv" driver (xserver-xorg-video-nv) I can seeverything on the
screen. Mouse and menus work.  Ctl-Alt-Backspace works (to terminate
X11). But most other keys simple cause this screen resolution change.

Any ideas?

Thanks,
Graham


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: etch to lenny upgrade - X apps no longer see keystrokes?

[Graham Williams]

Received Thu 09 Apr 2009  9:12pm +1000 from Thorny:
> On Thu, 09 Apr 2009 18:10:41 +1000, Graham Williams posted:
> 
> > Have just upgraded a fairly vanilla etch install on a Dell Precision 690
> > (AMD64) with an nvidia graphics chip. All seemed to proceed well, but on
> > reboot and starting up GDM, most key presses result in the screen
> > resolution changing - I can't login!
> > 
> > After quite a bit of research and attempts to determine what is going on,
> > I have run out of ideas! Ctrl-Alt-f1, etc, do not function.  The simplest
> > way I've figured out to log on is through single user mode. Keyboard works
> > just fine there. Booting into a Red Hat partition is also just fine.
> > 
> > I've created a .xinitrc which only runs xev so I can see what keys it is
> > seeing. When I run with video driver as "nv" (in /etc/X11/xorg.conf) xev
> > is not seeing any keyboard activity. Changing to "vga" at least I can see
> > that xev gets the keystrokes (but the screen is not usable). Changing to
> > "vesa" exhibits the same behaviour as "nv" - that is, no keys reported by
> > xev, and any key press seems to change the screen resolution.
> > 
> > With the "nv" driver (xserver-xorg-video-nv) I can seeverything on the
> > screen. Mouse and menus work.  Ctl-Alt-Backspace works (to terminate X11).
> > But most other keys simple cause this screen resolution change.
> > 
> > Any ideas?
> 
> When you upgraded from etch to lenny did you follow the release notes for
> upgrading? 
> http://www.debian.org/releases/stable/releasenotes
> If not, have a look now and see if anything you did might have caused
> trouble, and then determine if there is any way you can back out
> gracefully and redo things.

Thanks Thorny. Yes I did follow the release notes in upgrading and
have been trawling through the upgrade-lenny.script file and my wajig
log for clues. Trying to purge various X and friends and reinstalling
(and trying to stay with stable rather than testing or sid because
this is a test upgrade for a bunch of servers deployed in
production). no luck yet.

Regards,
Graham


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: etch to lenny upgrade - X apps no longer see keystrokes?

[Graham Williams]

Received Fri 10 Apr 2009  6:31am +1000 from Florian Kulzer:

Thanks for the suggestions Florian. More details below.

> On Thu, Apr 09, 2009 at 18:10:41 +1000, Graham Williams wrote:
> > Have just upgraded
> 
> To what? Lenny, Squeeze, or Sid?

>From etch to lenny, as per Subject.

> >a fairly vanilla etch install on a Dell Precision
> > 690 (AMD64) with an nvidia graphics chip.
> 
> Which chipset is it? Post the output of
> 
> lspci -nn | grep -Ei 'vga|graphic|display'

07:00.0 VGA compatible controller [0300]: nVidia Corporation NV43GL [Quadro FX 
550] [10de:014d] (rev a2)

> >   All seemed to proceed well,
> > but on reboot and starting up GDM, most key presses result in the
> > screen resolution changing - I can't login!
> > 
> > After quite a bit of research and attempts to determine what is going
> > on, I have run out of ideas! Ctrl-Alt-f1, etc, do not function.  The
> > simplest way I've figured out to log on is through single user
> > mode. Keyboard works just fine there. Booting into a Red Hat partition
> > is also just fine.
> 
> Post the output of these three commands:
> 
> awk '/Section.*"InputDevice"/,/EndSection/' /etc/X11/xorg.conf

Section "InputDevice"
Identifier  "Generic Keyboard"
Driver  "kbd"
Option  "XkbRules"  "xorg"
Option  "XkbModel"  "pc104"
Option  "XkbLayout" "us"
EndSection
Section "InputDevice"
Identifier  "Configured Mouse"
Driver  "mouse"
EndSection

> grep -E '^\((EE|WW)\)' /var/log/Xorg.0.log

(WW) The directory "/usr/X11R6/lib/X11/fonts/misc" does not exist.
(WW) The directory "/usr/share/fonts/X11/cyrillic" does not exist.
(WW) The directory "/usr/X11R6/lib/X11/fonts/cyrillic" does not exist.
(WW) The directory "/usr/X11R6/lib/X11/fonts/100dpi/" does not exist.
(WW) The directory "/usr/X11R6/lib/X11/fonts/75dpi/" does not exist.
(WW) The directory "/usr/X11R6/lib/X11/fonts/Type1" does not exist.
(WW) The directory "/usr/X11R6/lib/X11/fonts/100dpi" does not exist.
(WW) The directory "/usr/X11R6/lib/X11/fonts/75dpi" does not exist.
(WW) The directory "/var/lib/defoma/x-ttcidfont-conf.d/dirs/TrueType" does not 
exist.
(WW) Open ACPI failed (/var/run/acpid.socket) (No such file or directory)
(WW) NV(0): Failed to set up write-combining range (0xc800,0x800)

> grep -Ei 'keyboard' /var/log/Xorg.0.log

(**) |-->Input Device "Generic Keyboard"
(II) Initializing built-in extension XKEYBOARD
(**) Option "CoreKeyboard"
(**) Generic Keyboard: always reports core events
(**) Generic Keyboard: Protocol: standard
(**) Generic Keyboard: XkbRules: "xorg"
(**) Generic Keyboard: XkbModel: "pc104"
(**) Generic Keyboard: XkbLayout: "us"
(**) Generic Keyboard: CustomKeycodes disabled
(II) evaluating device (Generic Keyboard)
(II) XINPUT: Adding extended input device "Generic Keyboard" (type: KEYBOARD)

> 
> (You can do this in single user mode because the Xorg log from the
>  previous start of X should still be present as long as GDM has not been
>  restarted.)
> 
> -- 
> Regards,| http://users.icfo.es/Florian.Kulzer
>   Florian   |

Regards,
Graham


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: etch to lenny upgrade - X apps no longer see keystrokes?

[Graham Williams]

Received Fri 10 Apr 2009 11:30pm +1000 from Thorny:
> On Fri, 10 Apr 2009 06:44:42 +1000, Graham Williams posted:
> 
> > Received Thu 09 Apr 2009  9:12pm +1000 from Thorny:
> >> On Thu, 09 Apr 2009 18:10:41 +1000, Graham Williams posted:
> >> 
> >> > Have just upgraded a fairly vanilla etch install on a Dell Precision
> >> > 690 (AMD64) with an nvidia graphics chip. All seemed to proceed well,
> >> > but on reboot and starting up GDM, most key presses result in the
> >> > screen resolution changing - I can't login!
> >> > 
> >> > After quite a bit of research and attempts to determine what is going
> >> > on, I have run out of ideas! Ctrl-Alt-f1, etc, do not function.  The
> >> > simplest way I've figured out to log on is through single user mode.
> >> > Keyboard works just fine there. Booting into a Red Hat partition is
> >> > also just fine.
> >> > 
> >> > I've created a .xinitrc which only runs xev so I can see what keys it
> >> > is seeing. When I run with video driver as "nv" (in
> >> > /etc/X11/xorg.conf) xev is not seeing any keyboard activity. Changing
> >> > to "vga" at least I can see that xev gets the keystrokes (but the
> >> > screen is not usable). Changing to "vesa" exhibits the same behaviour
> >> > as "nv" - that is, no keys reported by xev, and any key press seems to
> >> > change the screen resolution.
> >> > 
> >> > With the "nv" driver (xserver-xorg-video-nv) I can seeverything on the
> >> > screen. Mouse and menus work.  Ctl-Alt-Backspace works (to terminate
> >> > X11). But most other keys simple cause this screen resolution change.
> >> > 
> >> > Any ideas?
> >> 
> >> When you upgraded from etch to lenny did you follow the release notes
> >> for upgrading?
> >> http://www.debian.org/releases/stable/releasenotes If not, have a look
> >> now and see if anything you did might have caused trouble, and then
> >> determine if there is any way you can back out gracefully and redo
> >> things.
> > 
> > Thanks Thorny. Yes I did follow the release notes in upgrading and have
> > been trawling through the upgrade-lenny.script file and my wajig log for
> > clues. Trying to purge various X and friends and reinstalling (and trying
> > to stay with stable rather than testing or sid because this is a test
> > upgrade for a bunch of servers deployed in production). no luck yet.
> > 
> 
> "Trying to stay with stable rather than..." Were you trying to do a
> dist-upgrade with "testing" and "unstable" repositories in your sources
> list? You would probably be better advised to switch to codename, lenny in
> your sources list and/or not have testing or unstable available. Perhaps I
> misunderstood what you wrote but you may now have a mixed system which
> might not be trivial to recover from. Are you sure you followed the
> release notes correctly.

There is no testing/unstable. Install done from DVD (5.0.0) whilst
server was standalone. The only entries in sources.list are the 5
DVDs. I followed the release notes carefully, and used the "script"
command to record the process.

Thanks,
Graham


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: etch to lenny upgrade - X apps no longer see keystrokes?

[Graham Williams]

Received Wed 15 Apr 2009  5:32am +1000 from Florian Kulzer:
> On Tue, Apr 14, 2009 at 09:21:11 +1000, Graham Williams wrote:
> > Received Fri 10 Apr 2009  6:31am +1000 from Florian Kulzer:
> > > On Thu, Apr 09, 2009 at 18:10:41 +1000, Graham Williams wrote:
> > > > Have just upgraded
> > > 
> > > To what? Lenny, Squeeze, or Sid?
> > 
> > >From etch to lenny, as per Subject.
> 
> I tend to forget to check the subject again once I start to compose a
> reply in vim.
>  
> > > >a fairly vanilla etch install on a Dell Precision
> > > > 690 (AMD64) with an nvidia graphics chip.
> > > 
> > > Which chipset is it? Post the output of
> > > 
> > > lspci -nn | grep -Ei 'vga|graphic|display'
> > 
> > 07:00.0 VGA compatible controller [0300]: nVidia Corporation NV43GL [Quadro 
> > FX 550] [10de:014d] (rev a2)
> 
> OK, that is nothing exotic and should work well with the nv driver.
> 
> > > >   All seemed to proceed well,
> > > > but on reboot and starting up GDM, most key presses result in the
> > > > screen resolution changing - I can't login!
> > > > 
> > > > After quite a bit of research and attempts to determine what is going
> > > > on, I have run out of ideas! Ctrl-Alt-f1, etc, do not function.  The
> > > > simplest way I've figured out to log on is through single user
> > > > mode. Keyboard works just fine there. Booting into a Red Hat partition
> > > > is also just fine.
> > > 
> > > Post the output of these three commands:
> > > 
> > > awk '/Section.*"InputDevice"/,/EndSection/' /etc/X11/xorg.conf
> > 
> > Section "InputDevice"
> > Identifier  "Generic Keyboard"
> > Driver  "kbd"
> > Option  "XkbRules"  "xorg"
> > Option  "XkbModel"  "pc104"
> > Option  "XkbLayout" "us"
> > EndSection
> 
> That looks OK to me.
> 
> [ snip: only harmless warnings in Xorg.0.log as far as I can tell ]
> 
> > > grep -Ei 'keyboard' /var/log/Xorg.0.log
> > 
> > (**) |-->Input Device "Generic Keyboard"
> > (II) Initializing built-in extension XKEYBOARD
> > (**) Option "CoreKeyboard"
> > (**) Generic Keyboard: always reports core events
> > (**) Generic Keyboard: Protocol: standard
> > (**) Generic Keyboard: XkbRules: "xorg"
> > (**) Generic Keyboard: XkbModel: "pc104"
> > (**) Generic Keyboard: XkbLayout: "us"
> > (**) Generic Keyboard: CustomKeycodes disabled
> > (II) evaluating device (Generic Keyboard)
> > (II) XINPUT: Adding extended input device "Generic Keyboard" (type: 
> > KEYBOARD)
> 
> Hmm, no real clues so far. I would like to see the status of certain
> packages on your system. Please post the output of:
> 
> dpkg -l udev {,lib}hal\* {,lib}dbus\* xserver-xorg\* libx11\* xkb\* | awk 
> '/ii/{print$2,$3}'

This produces:

dbus 1.2.1-5
dbus-x11 1.2.1-5
hal 0.5.11-8
hal-info 20080508+git20080601-1
libdbus-1-3 1.2.1-5
libdbus-glib-1-2 0.76-1
libhal-storage1 0.5.11-8
libhal1 0.5.11-8
libx11-6 2:1.1.5-2
libx11-data 2:1.1.5-2
libx11-dev 2:1.1.5-2
udev 0.125-7
xkb-data 1.3-2
xserver-xorg 1:7.3+18
xserver-xorg-core 2:1.4.2-10
xserver-xorg-input-all 1:7.3+18
xserver-xorg-input-evdev 1:2.0.8-1
xserver-xorg-input-kbd 1:1.3.1-1
xserver-xorg-input-mouse 1:1.3.0-1
xserver-xorg-input-synaptics 0.14.7~git20070706-3
xserver-xorg-input-wacom 0.7.9.3-2
xserver-xorg-video-all 1:7.3+18
xserver-xorg-video-apm 1:1.2.0-1
xserver-xorg-video-ark 1:0.7.0-1
xserver-xorg-video-ati 1:6.9.0-1+lenny4
xserver-xorg-video-chips 1:1.2.0-1
xserver-xorg-video-cirrus 1:1.2.1-1.lenny1
xserver-xorg-video-cyrix 1:1.1.0-8
xserver-xorg-video-dummy 1:0.3.0-1
xserver-xorg-video-fbdev 1:0.4.0-1
xserver-xorg-video-glint 1:1.2.1-1
xserver-xorg-video-i128 1:1.3.0-1
xserver-xorg-video-intel 2:2.3.2-2+lenny6
xserver-xorg-video-mach64 6.8.0-1
xserver-xorg-video-mga 1:1.4.9.dfsg-1
xserver-xorg-video-neomagic 1:1.2.1-1
xserver-xorg-video-nv 1:2.1.10-1
xserver-xorg-video-openchrome 1:0.2.902+svn579-4
xserver-xorg-video-r128 6.8.0-1
xserver-xorg-video-radeon 1:6.9.0-1+lenny4
xserver-xorg-video-radeonhd 1.2.1-2
xserver-xorg-video-rendition 1:4.2.0.dfsg.1-2
xserver-xorg-video-s3 1:0.6.0-1
xserver-xorg-video-s3virge 1:1.10.1-1
xserver-xorg-video-savage 1:2.2.1-2
xserver-xorg-video-siliconmotion 1:1.6.0-1
xserver-xorg-video-sis 1:0.10.0-1
xserver-xorg-video-sisusb 1:0.9.0-1
xser

Re: etch to lenny upgrade - X apps no longer see keystrokes?

[Graham Williams]

Thanks again Florian.

Received Sat 18 Apr 2009  3:34am +1000 from Florian Kulzer:
> On Fri, Apr 17, 2009 at 14:46:58 +1000, Graham Williams wrote:
> > Received Wed 15 Apr 2009  5:32am +1000 from Florian Kulzer:
> > > On Tue, Apr 14, 2009 at 09:21:11 +1000, Graham Williams wrote:
> > > > Received Fri 10 Apr 2009  6:31am +1000 from Florian Kulzer:
> > > > > On Thu, Apr 09, 2009 at 18:10:41 +1000, Graham Williams wrote:
> > > > > > Have just upgraded
[...]
> > > > >From etch to lenny, as per Subject.
[...]
> > > > 07:00.0 VGA compatible controller [0300]: nVidia Corporation NV43GL 
> > > > [Quadro FX 550] [10de:014d] (rev a2)
[...]
> > > dpkg -l udev {,lib}hal\* {,lib}dbus\* xserver-xorg\* libx11\* xkb\* | awk 
> > > '/ii/{print$2,$3}'
> 
> [ output edited ]
> 
> > udev 0.125-7
> 
> You should upgrade udev to version 0.125-7+lenny1 (security.debian.org).
> 
> > xserver-xorg-core 2:1.4.2-10
> 
> Rmadison tells me that the current version of this package for Lenny is
> 2:1.4.2-10.lenny1. I would try to upgrade to that. 
> 
> Other than those two packages, I did not see anything unusual your list.

I have upgrade both. No change in the X/kbd behaviour

> > > Another thing to check is which processes are using files in
> > > /dev/input/.  Ideally, this check should be done after X has started.
> > > Using CTRL-ALT-Fn does not work for you, but you could use a simple
> > > ~/.xinitrc that runs "sudo chvt 1" in an xterm, which would return you
> > > to the text terminal. (You have to configure your system to allow your
> > > user to run sudo with this command without password.) Then I would like
> > > to the output of:
> > > 
> > > lsof /dev/input/*

I am getting:

COMMANDPID USER   FD   TYPE DEVICE SIZE NODE NAME
hald-addo 4357 root4r   CHR  13,69  5819 /dev/input/event5
hald-addo 4357 root5w   CHR  13,68  5772 /dev/input/event4
hald-addo 4357 root6r   CHR  13,67  5654 /dev/input/event3
hald-addo 4357 root7r   CHR  13,66  5620 /dev/input/event2

> Another thing to check is if certain processes are running:
> 
> ps -ef | grep -E 'X|hal|dbus|udev'

root  1370 1  0 13:38 ?00:00:00 udevd --daemon
103   2862 1  0 13:38 ?00:00:00 /usr/bin/dbus-daemon --system
root  3775  3638  0 13:39 ?00:00:00 /sbin/dhclient -1 -lf 
/var/lib/dhcp3/dhclient.eth1.leases -pf /var/run/dhclient.eth1.pid -q -e 
dhc_dbus=31 -d eth1
105   4336 1  0 13:44 ?00:00:00 /usr/sbin/hald
root  4337  4336  0 13:44 ?00:00:00 hald-runner
root  4357  4337  0 13:44 ?00:00:00 hald-addon-input: Listening on 
/dev/input/event5 /dev/input/event4 /dev/input/event3 /dev/input/event2
105   4362  4337  0 13:44 ?00:00:00 hald-addon-acpi: listening on 
acpid socket /var/run/acpid.socket
root  4368  4337  0 13:44 ?00:00:00 hald-addon-storage: polling 
/dev/hda (every 2 sec)
anet  4421  4404  0 13:44 tty2 00:00:00 xinit /home/anet/.xinitrc -- 
/etc/X11/xinit/xserverrc :0 -auth /tmp/serverauth.sVdbIjaCrQ
root  4422  4421  0 13:44 tty7 00:00:00 /usr/bin/X11/X -nolisten tcp
root  4439  3960  0 13:45 tty1 00:00:00 grep -E X|hal|dbus|udev

> > Note that I am booting single user mode to do this, so it is logging
> > in as root and running startx as root. I modified the .xinitrc to chvt
> > 1 and to then xterm. I ran lsof on the console.
> 
> I think it would be better to do further tests in the normal runlevel.
> I would temporarily uninstall or at least disable [xkg]dm to allow you
> to log in at the normal tty prompt.

Yes - now in user land.

> > I've been playing with xev, looks like it is getting KeyRelease events
> > but not the KeyPress events for the keys that actually result in the
> > screen resolution being reset. Without knowing how hal and the kbd
> > device works, it is almost as if X is capturing these KeyPress events
> > and not passing them on, instead treating them as a screen resolution
> > change shortcut.
> 
> I think the capturing would be normal for real resolution-change key
> combinations, but we have to figure out why your system misidentifies
> other key press events as this combination.

Regards,
Graham


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: etch to lenny upgrade - X apps no longer see keystrokes?

[Graham Williams]

Received Tue 21 Apr 2009  7:57am +1000 from Florian Kulzer:
> On Mon, Apr 20, 2009 at 13:49:53 +1000, Graham Williams wrote:
> > Received Sat 18 Apr 2009  3:34am +1000 from Florian Kulzer:
> > > On Fri, Apr 17, 2009 at 14:46:58 +1000, Graham Williams wrote:
[...]
> There should be some input devices opened by Xorg. Please run this
> command instead (as root with X started):
> 
> lsof $(find /dev/input/)

Thanks. Starting in normal multi-user mode, having removed gdm,
logging in as a normal user, starting X ("startx") with a .xinint
consisting of "chvt 1" and "xterm", login in as root, run the above
"lsof" gives:

COMMANDPID USER   FD   TYPE DEVICE SIZE NODE NAME
hald-addo 3640 root4r   CHR  13,69  5677 /dev/input/event5
hald-addo 3640 root5w   CHR  13,68  5671 /dev/input/event4
hald-addo 3640 root6r   CHR  13,67  5567 /dev/input/by-path/../event3
hald-addo 3640 root7r   CHR  13,66  5589 /dev/input/by-id/../event2

> 
> > > Another thing to check is if certain processes are running:
> > > 
> > > ps -ef | grep -E 'X|hal|dbus|udev'
> 
> [ snip: all normal, except that dbus is not running ]
> 
> Did you run this in single user mode, or did you deliberately kill dbus?

Multi user mode. I think dbus is running? Isn't it the dbus-daemon?

root  1369 1  0 08:38 ?00:00:00 udevd --daemon
103   2832 1  0 08:38 ?00:00:00 /usr/bin/dbus-daemon --system
105   3619 1  0 08:39 ?00:00:00 /usr/sbin/hald
root  3620  3619  0 08:39 ?00:00:00 hald-runner
root  3640  3620  0 08:39 ?00:00:00 hald-addon-input: Listening on 
/dev/input/event5 /dev/input/event4 /dev/input/event3 /dev/input/event2
105   3648  3620  0 08:39 ?00:00:00 hald-addon-acpi: listening on 
acpid socket /var/run/acpid.socket
root  3651  3620  0 08:39 ?00:00:00 hald-addon-storage: polling 
/dev/hda (every 2 sec)
root  3745  3608  0 08:39 ?00:00:00 /sbin/dhclient -1 -lf 
/var/lib/dhcp3/dhclient.eth1.leases -pf /var/run/dhclient.eth1.pid -q -e 
dhc_dbus=31 -d eth1
anet  4419  4402  0 08:44 tty3 00:00:00 xinit /home/anet/.xinitrc -- 
/etc/X11/xinit/xserverrc :0 -auth /tmp/serverauth.LvCwEgYUrc
root  4420  4419  0 08:44 tty7 00:00:00 /usr/bin/X11/X -nolisten tcp
root  4488  4446  0 08:47 tty1 00:00:00 grep -E X|hal|dbus|udev

[...]
> > Yes - now in user land.
> 
> Does "now" encompass all the output that you posted in your last
> message?

Yes.

> Does the keyboard work in Xorg if you boot from a Debian Lenny live CD
> (http://debian-live.alioth.debian.org/)? 

Running debian 5.0.0 live amd64 gnome ISO image, all seems to be
working just fine.

Thanks again for your help Florian. It is particularly useful, though
we haven't solved it yet.

Regards,
Graham


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: etch to lenny upgrade - X apps no longer see keystrokes?

[Graham Williams]

> ldconfig -pNX | grep /local/

Bingo!

Our servers have XWin32 LIVE libraries installed (A X11 server for
MS/Windows, but which installs some libraries on the Linux box). The
install of XWin32 placed links in /usr/local/lib. I removed those
links and X is now functioning

Thanks for that pointer, and for persisting in helping to resolve
this. It is much appreciated.

Regards,
Graham


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

XFree 4.1 on ATI Rage 128 Ultra Pro II

[Graham Williams]

Running Debian unstable up to date installation, Kernel 2.4.18.

Trying to get XFree86 on an ATI Rage 128 Ultra Pro II.

Works just fine on a RedHat XFree86 4.1 installation. But can't get it
going with XFree86 4.1 on Debian. Seems that it is not supported under
4.1 but is on 4.2? Not sure why it it would work on RedHat.

What are my options?

Thanks,
Graham


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

Debconf problem: DbDriver "config": /var/cache/debconf/config.dat is locked by another process

[Graham Williams]

I've got myself stuck somehow with upgrading some packages (am current
with "unstable"). After a recent apt-get dist-upgrade I now have:

debconf: DbDriver "config": /var/cache/debconf/config.dat is locked by
another process

and the various messages about 

  Package debconf is not configured yet.

Any ideas what I've done to get myself into such a pickle? And how to
get out of it.

Regards,
Graham


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

Re: Debconf problem: DbDriver "config": /var/cache/debconf/config.dat is locked by another process

[Graham Williams]

Received Sat 01 Jun 2002 10:16am +1000 from Joey Hess:
> Graham Williams wrote:
> > I've got myself stuck somehow with upgrading some packages (am current
> > with "unstable"). After a recent apt-get dist-upgrade I now have:
> > 
> > debconf: DbDriver "config": /var/cache/debconf/config.dat is locked by
> > another process
> 
> Well it seems that you have another debconf instance running somewhere.
> Perhaps you're reconfiguring a package on another virtual console? Only
> a running process can keep this lock held, so find it and kill it and
> you should be ok.

Thanks for the advice. I've found a "dpkg-reconfigur" running. Killing
that it fixed it.

Regards,
Graham


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

Re: debian gnome splash screen

[Graham Williams]

Received Fri 08 Jul 2005  7:14pm +1000 from Rakotomandimby Mihamina:
> Hi,
> 
> I would like to customize the debian gnome splash screen in order to add
> the name of our organisation.
> 
> I did not see how to... would you know?
> 
> Thank you.

I've got some details at 

http://www.togaware.com/linux/survivor/Splash_Screen.shtml

Hope it helps.

Regards,
Graham


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