That didn't help. However, adding a blacklist=floppy
option to kernel boot did the trick. I think floppy is
included in the initrd image and is loaded by the kernel
in the early phases of boot; I also suspect this happens
because I let the Debian installer load floppy when I
set up my system.
I am having problems mounting one of my USB disks by
its UUID.
(snip)
Any hints, pointers, or advice would be greatly appreciated.
Do you really mean USB disk (HD via USB) or USB sticks? I'm just curious; the
USB flash memory may not have a UUID.
A possible workaround is to use 'udevinfo'
Now I can't boot into either the newly installed K6 kernel
or the older 386 version and get the following messages
at boot time from either
pivot_root: no such file or directory
/sbin/init: 432 cannot open dev/console no such file
kernel panic: attempted to kill init !
The trick is to
somebody out there who can give me a helping hand to repair
somehow the hd or point me to a method/way or program to
rescue at least my holiday-photographs from france i put on
the hd two hours before the crash...?
The most important thing is to avoid using the disk as much as possible (do
**snip**
I hoped to reduce the probability of it [tape player]
failing at the same time as the internal HD by
starting it less often
For HDs the R/W head always comes into contact with the disc once the rotation drops
enough. Crack open a dead HD and you'll see nice concentric rings which
What do you recommend as a good, practical WYSIWYG editor
for creating simple web pages, with mostly text, some
graphics and pdf, for debian Lenny please? Thanks.
This may be a bit [OT] since I don't have any WYSIWYG web page editor to
recommend, but ...
Actually WYSIWYG is VERY BAD for web
Userspace services daemons etc, like any program will
perform better on an
i686 processor if they are i686 optimised :-)
It would certainly be interesting to see comparisons of this. The primary
advantage is that more registers are available so there are less games played
with swapping
The message:
PCI: unable to reserve mem region #1:.. for device :00:09.0
is the most likely place where the system failed. The driver could not initialize
properly, so the RAID cannot be used and your system goes to waiting for root
filesystem.
If the message actually tells you what
* a) What are the pros and cons in switching to 64 bit mode?
* - Is it faster?
I don't know - never ran my Opteron in 32-bit mode. I was wishing 64-bit
floating point operations would be faster but they're not because AMD didn't
really improve the math unit. In principle it should be faster if
It sounds like your network script is running while the wireless driver hasn't
finished setting up. I have an EPX form computer out in the field and it's got
an rt2570 in it. Because of the peculiar way that machine operates, I use
pre-up/post-down to load and unload the driver; the driver
(quote)
Bluetooth uses the same frequency range as 802.11b and g ...
(end quote)
That's right, and you're also right about the power meant to be limited to less than a normal 802.11
transceiver. I've mostly seen Bluetooth run as an extra protocol on top of 802.11b/g and thus no extra gizmos
I remember a waiting for root filesystem when I was using a software RAID and my
drivers didn't load. I had to edit /etc/mkinitramfs/modules and make sure my md driver
was added to the initrd image.
It is possible that you need some driver to load to access your RAID? If you
wait, say, 5
(quote)
One of the most common operations performed in a Web page is to detect the
browser type and version. Browser detection is performed to ensure that the
content presented to the browser is compatible and renders correctly
(end quote)
The quote is 100% bullshit of course. The whole
(quote)
I can see the device using udevinfo and I can mount it as
root manually, but why won't volume-manager work anymore?
(end quote)
Did you look at the actual device nodes in /dev to see who owns each device?
If user:group=root:root then you need to read through the /etc/udev{,/rules.d}
(quote)
I still have the problem. ie. The windoze partition is
mounted automatically fine, but I can only cd to it if I am root.
(end quote)
Do:
man mount
Look at the 'Mount options for ntfs'. All your mysteries are explained there -
for example:
... By default, the files are owned by root
(quote)
To get sizes, du is the obvious choice, but you could do the ridiculous:
$ find . -type f -exec cat {} \; | wc -c
(end quote)
I do the slightly less ridiculous:
for X in $(find . -name '*'); do du -b $X mylogfile; done
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of
I would suggest reading the USB docs in these links:
http://www.kernel.org/pub/linux/utils/kernel/hotplug/udev.html
It is convenient to use persistence to ensure that all removable USB devices
retain their name regardless of how many times you plug them in and remove
them. This also allows
Do you use paravirtualization? If not, recompile a kernel with no
virtualization support. The problem here is that the NVidia driver is trying
to use a module symbol which has been declared for use with GPL drivers only.
If I remember correctly that means EXPORT_SYMBOL_GPL(somevariable) is
(quote)
I then upgraded back to the bios version that came with the laptop, but the
unexpected irq trap persists.
(end quote)
Did you enter the BIOS setup and ensure that everything was set back the same as before? A BIOS update may 'update' the actual settings to some default state.
--
To
(quote)
I tries downloading a newer kernel image on another Debian box
...
How do I download the real amd64 image?
(end quote)
The package managers will naturally download a file for their own architecture
because they download a Packages list for their architecture. Fetch a package
with
(quote)
But I can't see this page from linux, only from windows, do you know any way to
see these pages from linux?
(end quote)
More goddamned Microsoft crap is what the problem is. In Iceape 1.1.1, go to Tools-User
Agent Switcher-Internet Explorer 6 (XP) and you will be able to see the
Aside from the numerous pure Debian vendors, if you don't mind 'Ubuntu' you
might want to check if Canonical still offer to send you a free CD. Otherwise,
'wget' has been ported to WinDos and is yet another tool which can resume a
download. If your disk space is really too small then the 'net
(quote)
Could I use Linux to copy the Windows program exe files etc to the Windows fake
partition created by qemu and then go into Windows under qemu and then get it
to install the software?
(end quote)
For many pieces of software that scheme would work, but for copy protected
discs it will
You would have to explain your setup more or we might give you all the wrong
advice. For example:
1. Are you getting internet access through 'cable' (which requires a cable
modem, probably built into the cable TV receiver) OR
2. Is your internet via ADSL, in which case you have an ADSL modem
The beeping may be a coincidental problem. The power supply is the usual
culprit; remove your new power-hungry HDs and see if the system boots normally.
You did make a mistake with copying your boot disk though; using 'dd' like that
makes a copy that looks exactly like your old HD, which
You mean like 'chvt' ?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
There are rugged laptops aimed primarily at the military but also commonly used
in industry. Occasionally you can pick them up at ex-govt auctions for
next-to-nothing and in great condition.
Failing that, how fast a beast do you want? You can get something like a
LBC-GX500 from WinSystems
Oops ... my fault here folks ... I didn't realize that I was linked to a
different mail list from another forum - I thought it was all the same. Blah.
I hate cross-site stuff. Maybe I should activate my debian.org stuff again so
I can see what's going on.
--
To UNSUBSCRIBE, email to
Perhaps you're thinking of the limit on open files? That would be correct -
since even a 'socket' connection is a file, the default limit can be pretty
severe for a server. In a 'bash' environment you would change the limit with
the 'ulimit' command (see bash manual for details). I know I
huh?
I really don't know what the hell you're talking about. I've been leaving the
subject blank lately because people bitch about it for no apparent reason. But
if you really don't want the subject changing, why don't you ask Intnsred if he
can set things so that the subject isn't
Once you add each cd to your source list using apt-cdrom add you do not have
to think about which CD has what - the package manager will ask you for the correct CD.
You will also be able to use 'apt-cache search' to find keywords in packages on all CDs
after they're added.
--
To
Well, your MUA is your responsibility - don't complain when it doesn't behave
the way you want. The 'subject' is mutable so you should expect it to change in
this forum.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Use 'apt-get dist-upgrade'. Even this will not necessarily be successful on
the first pass and you may have to run it up to 3 times.
'ntp' is a special case; ntp has been extensively revised fairly recently and
you'll have to spend (a lot of) time reading through the documentation to see
Do the changes you make via OOo get saved to the remote file or is your problem
only with the disconnect?
I would suggest setting up the ssh server on your etch machine and using
'putty' from Windows. Make sure you can log on (generally as a normal user,
then use 'su' to become root), and
Well, at least in this case the computer is correct. The east coast of
Australia is UTC+10:00 for eastern standard time and UTC+11:00 for DST. The
computer is correct - when it it 16:50, 6 Mar 2007 UTC, Australian EDST is
3:50 7 Mar 2007.
Hmm... then again... is that really saying EST -
1. Are you really using 'lpr' or are you using the 'cupsys-bsd' pretender and
accidentally filtering things through cups?
2. Send directly to printer to see what happens:
cat mydocument /dev/lp (or whatever the parallel device is)
3. Check BIOS and ensure that you have a suitable parallel
Etch is almost 2 years old!
SATA support began in the 2.5 kernels and some modules were backported to 2.4. There have been many huge changes in the 2.6 kernel to date, so Etch should be the one to use. Kernels 2.6.16 to 2.6.18 seem fine, 2.6.19 is buggy, never tried .20 - .21 is in the works.
If you have 2.6 and the Alsa drivers, it all works well - just make sure you
edit settings with the 'alsamixer'.
You'll probably find that everything is working but your headphone channel is
muted. With OSS I have no idea what setting to change (or if this is even
possible). It is also
You probably did not need ndiswrapper in either case.
The 'rt2500' driver has been made open source over a year ago:
apt-cache search rt2500
A small group is currently rewriting the code for a mode generic and friendlier
driver but I don't believe the work is done yet - but the drivers and
If there are already tools like 'apt-get' installed, then you can invoke an
install yourself and maybe get better information on what's failing:
apt-get install console-tools console-common
From your log there is no evidence that initramfs-tools failed; that seems to
be OK.
Just a bit of
Your USB gizmo should be recognized no matter what. A PS2 to USB converter
needs extra circuitry (but this is a tiny chip which they can cast the
connector over). Going the other way is easy because many USB keyboards have
circuitry that will recognize they are not on a USB line and they
http://www.linuxprinting.org/show_printer.cgi?recnum=HP-Color_LaserJet_2600n
Personally I would not buy a (contemporary) color printer under about US$2000 -
in my opinion they are never worth the money. Until recently I would never
have bought a monotone laser printer for under $600 - but
How can you operate with no modem? Isn't your Speedtouch device a router/modem?
Anyway, when you connect to the ATM network the device doing the negotiation
will receive a static address - if that is your router then your router will
have a static IP on the WAN side and a local subnet IP on
If you follow 'kerneltrap.org', 2.6.19 and 2.6.20 are bug-ridden. 2.6.19
shipped with bugs and 2.6.20 hasn't worked them all out yet. But if you enjoy
playing with the latest 'git' data, this is what happens. I usually wait until
a released version is a few weeks old before I decide to play
Just make sure you have the (empty) directories /dev and /dev/pts. The system DOES need a few devices before udev does its magic - for those create the device nodes the usual way. (have a look at the makedevices script for clues) I remember reading instructions on this months ago but can't
Do you have different users for your normal and chroot environments? The next
thing to check is the permissions on the /etc directory - does it have the 'x'
bit set so you can see what's in that directory? Remember that chroot does not
switch the entire system - libraries loaded continue to
If you have a built-in ethernet, make sure it is activated in the BIOS. Older
versions of the kernel will not activate a device disabled by BIOS (but newer
versions will give you an option). Otherwise, check the 'dmesg' logs as
already suggested to see what the device drivers print out.
--
auto eth0
iface eth0 inet manual
Or else try to configure as 'static' with address '0.0.0.0'.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Even Bill Clinton would never be able to convince anyone that MP3 was free, so
I'm not surprised the software author charges for MP3 support - he may have
spent some money to acquire the documentation and also a licensing arrangement.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a
The log suggests a number of things:
1. Is your compiler the same used to build the kernel? (If not, it's easier to
build a kernel than to set up an older compiler.)
2. Do you have other modules currently loaded such as the nvidiafb or riva
drivers?
And then of course as already posted, you
Make sure the RAID device is set to be 'persistent'. Also check backwards
through scripts to see where md2 might fail.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
'Multicore' processors are pretty new - older machines have always had an
enormous number of processors. 80 cores on a single chip would take up most of
the silicon wafer. (and will probably have a few defective cores and rather
interesting cooling problems)
Problems encountered by multiple
There are a number of TeX variants and no doubt one would have been made for
guitar tabs.
Long long ago in a galaxy far far away I used 'musixtex' - but that it used for
piano and orchestral arrangements - I can't recall if it did guitar tabs as
well.
It takes a few days of playing with
Even if you're only using 'etch', try an 'apt-get dist-upgrade'. Did you
notice if you had a long list of 'packages held back' ?
One reason for DCOP failing is permissions on the /tmp directory - just check
that it is 'drwxrwxrwt'. Incorrect permissions will cause many necessary bits
to
For a genuine (full) backup, use 'tar' - that's what it was made for (and it
acts properly on symlinks and preserves attributes etc). Just be aware that
tar originally worked with tapes and just wrote all data into a single stream
with a few markers. That behavior is preserved, so you need
Do NOT change the /lib/modules/(kernel-version) because that is begging for
trouble. You need to maintain the naming conventions and should only break them
if you really know what you're doing. (And most people who do know what they're
doing don't have reason to break it.) The module update
Yes, it can be done, but not with the current default system. You now have to
be an expert to get Debian to run on that class machine. For Debian I would
recommend a minimum 32MB with no X, and minimum 64MB if you run X. However,
there are alternatives (see www.distrowatch.org for too many
You must be accustomed to a strange TeX distribution - all varieties I've
encountered produce dvi files (as Don Knuth originally coded it). Also
remember that there are a number of commercial TeX distributions and they have
quite a few modifications and additional style files and may not
Java runs on 64-bit machines. I think Debian has a package for it now, but I
still get mine directly from the Blackdown website.
'chroot' cannot be avoided when you want to mix 32 and 64 bit libraries.
Furthermore the kernel needs to be compiled with 32/64 support or else the
32-bit system
Wow - did this theme have an animated 3D poochie or clippy?
I wonder if there is a 'SETI' theme out there...
It is also not unusual to find badly designed monitoring gizmos. For example,
a panel meter that polls in a very tight loop - since humans take a long time
to react (and probably
60 matches
Mail list logo
