doc. on failure recovery using LVM for RAID1
I've read that LVM can perform RAID1 mirroring. However, I'm having trouble finding documentation of RAID1 failure recovery using LVM (recognizing failures, what to do with LVM after replacing a failed disk, and how LVM rebuilds the mirror). The md/mdadm documentation says how to assimilate a replacement disk/partition into a say, how md rebuilds the mirror, etc. Which LVM documents cover that aspect of LVM? Thanks, Daniel -- Daniel B. d...@smart.net -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: having DHCP use name server from PPP connection?
Raj Kiran Grandhi wrote: Daniel B. wrote: Can dhcpd be configured to pass on (to DHCP clients on a local, private (NATted) network) the DHCP server machine's current domain name server addresses (given to the machine by PPP (etc.))? What happens when your server's current name server changes before the DHCP lease expires? How about answering the question instead of bringing up other concerns? The name server specified in my PPP connections only changes when my ISP decides to change something (on the scale of months). Therefore, it's easy to set the lease length so the private-network DHCP clients have an old name server setting for only a small fraction of the time. To answer your question directly: The same thing that happens any time the current name server changes before the DHCP lease expires. Right? (That is, does it matter whether PPP updates the name server in /etc/resolv.conf automatically or one edits manually? Whether that's manual or automated, if the DHCP server configuration isn't updated, then certainly the DHCP clients will be out of date. If the server configuration _is_ updated, then if clients don't update their leases, then they're still out of date (again, whether the update is manual or automatic). The only way client would get updated is if DHCP has some kind of lease revocation, and, again, is doesn't seem to matter whether the update is manual or automatic.) I see how to hard-code name servers in /etc/dhcpd.conf with the domain-name-servers statement. However, I don't yet see how to have dhcpd report whichever name servers are listed in /etc/resolve.conf (put there by PPP's /etc/dhcpd.conf). (What does dhcpd give to clients if there is no domain-name-servers statement? The documentation I've seen (dhcpd, dhcpd.conf manual pages; Sarge version) doesn't seem to say.) It probably does not give anything. Does anyone know of documentation that specifies what happens? One possible solution to your problem is to run one of the several caching name servers like pdnsd or dnsmasq on your DHCP server and pass on your DHCP servers lan IP as the DNS server IP to the clients. I was hoping to avoid messing with a DNS server, but, yes, that's an option. Thanks, Daniel -- Daniel B. d...@smart.net -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
having DHCP use name server from PPP connection?
Can dhcpd be configured to pass on (to DHCP clients on a local, private (NATted) network) the DHCP server machine's current domain name server addresses (given to the machine by PPP (etc.))? I see how to hard-code name servers in /etc/dhcpd.conf with the domain-name-servers statement. However, I don't yet see how to have dhcpd report whichever name servers are listed in /etc/resolve.conf (put there by PPP's /etc/dhcpd.conf). (What does dhcpd give to clients if there is no domain-name-servers statement? The documentation I've seen (dhcpd, dhcpd.conf manual pages; Sarge version) doesn't seem to say.) Thanks, Daniel -- Daniel B. d...@smart.net -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: how to build from _modified_ source package
Cameron Hutchison wrote: Daniel B. wrote: Are there any instructions for proceeding from having downloaded the source package files and _not_ having unpacked things? (I think my current state is as if I had done apt-get source --download-only xfree86 (I didn't actually do --download-only, but from cleaning up after trying a couple things I think that's the state I'm in.)) Just run apt-get source xfree86 in the directory where you've downloaded the existing files. apt-get will not download them again if it does not need to, I just ran apt-get source xfree86 and it deleted xfree86_4.3.0.dfsg.1.orig.tar.gz and started downloading it again. All 59 megabytes of it. Over a slow connection in my case. Good thing I suspected you didn't know what you're talking about and made a backup. and will unpack the source files. You can then edit the source and build it as described in other messages in this thread. When I started building (before), I got messages about extracting the source. Which build commands re-extract the source files from the source archive before building and which build from the possibly modified source (without re-extracting)? Where's the documentation? Daniel -- Daniel B. [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
how to build from _modified_ source package
How do you rebuild a Debian package from source _with_ local modifications? The instructions I've seen all extract source (original plus patches) and build in one step, not giving a chance to make local edits. Thanks. Daniel -- Daniel B. [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: how to build from _modified_ source package
Kamaraju S Kusumanchi wrote: Daniel B. wrote: How do you rebuild a Debian package from source _with_ local modifications? The instructions I've seen all extract source (original plus patches) and build in one step, not giving a chance to make local edits. I frequently do this with couple of packages. For example, I would like to have readline support, history support in gnuplot. So I recompile it myself in a debianized way. The step-by-step instructions as to how I do it can be found at http://malayamaarutham.blogspot.com/2007/04/build-gnuplot-with-gnu-readline-and.html Thanks. Unfortunately, those instructions address only changing some configuration in the debian/ directory, not editing source. Are there any instructions for proceeding from having downloaded the source package files and _not_ having unpacked things? (I think my current state is as if I had done apt-get source --download-only xfree86 (I didn't actually do --download-only, but from cleaning up after trying a couple things I think that's the state I'm in.)) Daniel -- Daniel B. [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: any pages listing text mode resolutions of video cards?
Douglas A. Tutty wrote: On Tue, Nov 13, 2007 at 12:26:33AM -0500, Daniel B. wrote: Kelly Clowers wrote: On Nov 12, 2007 7:24 AM, Daniel B. [EMAIL PROTECTED] wrote: Kelly Clowers wrote: On Nov 11, 2007 7:46 PM, Daniel B. [EMAIL PROTECTED] wrote: I've been having trouble finding out the text-mode resolutions of video cards. Does anyone know of a good compilation of that information? Here is a list of modes: http://en.wikipedia.org/wiki/VESA_BIOS_Extensions#Linux_video_mode_numbers I think any card that supports VESA probably has to support 1280x1024 and 24 bits (see the table higher up on that page) Actually, I was talking about text-mode resolutions (as I wrote above), not graphics-mode resolutions. Ok, I found it in that Wikipedia article: Modes 264-268 (vesa's numbers, not the kernel's) are text modes. 264 (0108h) is 80x60, 265 (0109h) is 132x25 266 (010Ah) is 132x43, 267 (010Bh) is 132x50 268 (010Ch) is 132x60. Actually, what I meant was the text-mode resolutions supported by specific video cards. That is, information usable to choose which card to get. My old card did 132x60. My new card did 132x44. In case I buy a new card, I want to know which modes a given card supports. I haven't been able to find that information. (I haven't found any compilation for multiple cards; the manufacturer's page for my new card doesn't mention the text modes.) So to be clear, you are specifically __not__ wanting to use a framebuffer which uses graphic mode and is therefore slower. Not necessarily. I'm trying using a framebuffer mode to see how it is. However, I'd like to be able to know what text mode a card supports the next time I buy a video card in case I don't want to use framebuffer mode. Daniel -- Daniel B. [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: any pages listing text mode resolutions of video cards?
Kevin, Kevin Mark wrote: On Sun, Nov 11, 2007 at 10:46:53PM -0500, Daniel B. wrote: I've been having trouble finding out the text-mode resolutions of video cards. Does anyone know of a good compilation of that information? Relatedly, are they any good tutorials on switching from using hardware text-mode for non-X virtual consoles to using ... um ... whatever the name is of the feature of having textual virtual consoles generated using hardware graphics mode? Are you talking about 'frame buffer' programs? fbxine, fbi, ... Yeah, frame buffer is the term I've heard. Thanks. Does using frame buffers let the kernel provide (text-mode) virtual consoles using the video hardware's graphics mode (that is, the kernel tells the hardware what pixels to display, as opposed to telling it only what characters to display)? I take it frame buffer mode lets one use any text resolution that fits within the video card's full graphics resolution? Is there a good starting point in the assorted Debian and Linux documentation for trying frame-buffer mode? And setting vga=771 or similar in your kernel options? Yes. I've been using vga=10 in my kernel options (via LILO) to set the virtual console text mode resolution at boot time. I was using 132x60 on my old video card (a Diamond Viper 330/nVidia RIVA TNT2 AGP card). (Yeah, yeah; I know; it's ancient.) I bought a new, basic video card (an ATI Radeon Rage XL PCI card). (Hey, I did write basic.) (In Windows, the driver for the old card seems to be causing problems (severe file system corruption), and there doesn't seem to be a newer version of the driver, so I bought another, different video card, hoping that the new card's driver wouldn't have the same problem.) Unfortunately, the new card's text modes only go up to 132x44 (losing me 25% of the virtual console height I'm used to). That's why I was asking about where to find the text-mode resolutions of cards (in case I want to buy a different card that handles at least the resolution I had). Of course, if using frame-buffer-based virtual consoles provides the text mode, I don't have to care about the video card's text modes. How slow is frame-buffer mode? I think I tried it a couple years ago and it was noticeably slow enough that I went back to hardware text modes. (Yes, my CPU is old (relatively slow) too right now.) Thanks, Daniel -- Daniel B. [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: any pages listing text mode resolutions of video cards?
Kelly Clowers wrote: On Nov 11, 2007 7:46 PM, Daniel B. [EMAIL PROTECTED] wrote: I've been having trouble finding out the text-mode resolutions of video cards. Does anyone know of a good compilation of that information? Here is a list of modes: http://en.wikipedia.org/wiki/VESA_BIOS_Extensions#Linux_video_mode_numbers I think any card that supports VESA probably has to support 1280x1024 and 24 bits (see the table higher up on that page) Actually, I was talking about text-mode resolutions (as I wrote above), not graphics-mode resolutions. Relatedly, are they any good tutorials on switching from using hardware text-mode for non-X virtual consoles to using ... um ... whatever the name is of the feature of having textual virtual consoles generated using hardware graphics mode? Framebuffer? I am not sure I understand what you are asking, but I think you want your VTs to use higher resolutions and be graphics capable Yes, in my virtual consoles I want to see text in a higher resolution that the video card can provide in hardware text mode. e.g. fbi http://packages.debian.org/etch/fbi Thanks. If that is what you mean, you need to edit your grub menu.lst file and add to the kernel line a reference like vga=791 Is that number (791) a mode specific to frame buffers, or was it just an example? http://wiki.ubuntu.com/FrameBuffer Thanks. Daniel -- Daniel B. [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: any pages listing text mode resolutions of video cards?
Jochen Schulz wrote: Daniel B.: Kevin, And setting vga=771 or similar in your kernel options? Yes. I've been using vga=10 in my kernel options (via LILO) to set the virtual console text mode resolution at boot time. I am not absolutely sure, but I don't think vga=10 gives you a real framebuffer. Right, it does not. That's what I have been using up until now, to select a video card hardware text mode (instead of selecting a kernel framebuffer mode using video card hardware graphics mode (assuming I understand things right)). It just changes the font, which may result in more characters on your display. Actually, it doesn't change the font (or just change the font); it does change frequencies and the total number of pixels available for the hardware to use to display text. ... vga=771 (and similar values) on the other hand really switches display resolution (and color depth) and is most certainly what you want. Yeah, I'm trying that range of numbers and framebuffers now. Hey, a related problem is that in graphics mode, the X server won't use any mode where the dot clock is higher than 125 MHz, even though the video card goes to 230MhZ. (I made a 125MHz mode line and it worked; I changed it to 126 MHz and the server excluded it, and it's not excluded because of horizontal or vertical frequencies outside the monitor's range.) /usr/share/doc/xserver-xfree86/README.ati.gz says ... maximum allowed clock frequency ... 3D Rage adapters. For now, clocks will, by default, be limited to 80MHz, 135MHz, 170MHz, 200MHz or 230MHz, depending on the specific controller. This limit can only be increased (up to a driver-calculated absolute maximum) through the DACSpeed specification in XF86Config. That doesn't quite match up: The driver (or server) seems to have a limit at 125 MHz, not any mentioned in that read-me file. Anyway, I tried the DacSpeed xxx option in XF86Config-4 to set tell the server and driver the limit, but it doesn't seem to have any effect. In fact, when I tried DacSpeed 114, it did _not_ prevent use of a mode with a dot clock of 125 Mhz. It's like it's getting ignored entirely. My X server is Sarge's xserver-xfree86 4.3.0.xxx. The server's driver is the ati driver. The video card is an ATI Radeon Rage XL 8MB PCI card. Thanks, Daniel -- Daniel B. [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: any pages listing text mode resolutions of video cards?
Kelly Clowers wrote: On Nov 12, 2007 7:24 AM, Daniel B. [EMAIL PROTECTED] wrote: Kelly Clowers wrote: On Nov 11, 2007 7:46 PM, Daniel B. [EMAIL PROTECTED] wrote: I've been having trouble finding out the text-mode resolutions of video cards. Does anyone know of a good compilation of that information? Here is a list of modes: http://en.wikipedia.org/wiki/VESA_BIOS_Extensions#Linux_video_mode_numbers I think any card that supports VESA probably has to support 1280x1024 and 24 bits (see the table higher up on that page) Actually, I was talking about text-mode resolutions (as I wrote above), not graphics-mode resolutions. Ok, I found it in that Wikipedia article: Modes 264-268 (vesa's numbers, not the kernel's) are text modes. 264 (0108h) is 80x60, 265 (0109h) is 132x25 266 (010Ah) is 132x43, 267 (010Bh) is 132x50 268 (010Ch) is 132x60. Actually, what I meant was the text-mode resolutions supported by specific video cards. That is, information usable to choose which card to get. My old card did 132x60. My new card did 132x44. In case I buy a new card, I want to know which modes a given card supports. I haven't been able to find that information. (I haven't found any compilation for multiple cards; the manufacturer's page for my new card doesn't mention the text modes.) Daniel -- Daniel B. [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
XFree86 limiting dot-clock to 125MHz when chip goes to 230MHz
My X server seems to be limiting the video modes to a dot-clock to 125MHz even though video adapter's chip goes to 230MHz. The adapter is an ATI Radeon Rage XL PCI card. The chip's dot-clock goes to 230 MHz (per http://ati.amd.com/products/server/ragexl/features.html ). My X server is Sarge's xserver-xfree86 4.3.0.dfsg.1-14sarge5. Is there some configuration to set to tell X that the chip can handle 230MHz? I just upgraded from XFree36 3.3.6 (package xserver-svga) and a Diamond Viper 330 / vNidia RIVA TNT2. That X server allowed at least 202 MHz on that card without any special configuration. Thanks, Daniel -- Daniel B. [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
any pages listing text mode resolutions of video cards?
I've been having trouble finding out the text-mode resolutions of video cards. Does anyone know of a good compilation of that information? Relatedly, are they any good tutorials on switching from using hardware text-mode for non-X virtual consoles to using ... um ... whatever the name is of the feature of having textual virtual consoles generated using hardware graphics mode? Thanks, Daniel -- Daniel B. [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
installed xserver-xfree86, what to do to switch to it?
On Sarge, I just installed package xserver-xfree86 and in the configuration specify to use the xserver-xfree86 server. However, then run startx, I still get the old xserver-xvga server (XF86_SVGA). I even re-did the configuration (using dpkg-recongfigure). What else do I need to do to get startx/initx/etc. to run the xserver-xfree86 server? I notice that the symbolic link /etc/X11/X still points to /usr/bin/X11/XF86_SVGA. Shouldn't the re-configuration have changed that? (Or is that link no longer used? If not, what file stored the choices of which server to run? Thanks, Daniel -- Daniel B. [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: which package contains mkfs.vfat
Mathias Brodala wrote: Hi. Matus UHLAR - fantomas, 02.06.2007 23:26: Serena Cantor, 31.05.2007 23:48: I have trouble using logical vfat partition so I intend to try mkfs.vfat On 01.06.07 00:00, Mathias Brodala wrote: My suggestion: avoid that one and use a Windows system to format with FAT32. This tool thrashed the filesystem of my digital camera twice that much, that not even a Windows system could recover it. Fortunately the camera of a friend was able to rescue the card in my camera. man mkdosfs(8) has some advices concerning digital cameras... Yep, already read that but it doesn’t concern me since I don’t use CF but SD cards. Be careful if you install a package that contains fsck.vfat. It corrupts filesystems, and it ignore the /etc/fstab setting that should tell it to not run. fsck.vfat turns a small VFAT corruption into a big one. (When it finds and tries to fix a small problem, it causes a large one.) Be afraid. Be very afraid. I suffered _multiple_ (repeated) massive corruptions of my Windows (C:) filesystem. For example, the file allocation table's sectors and the root directory's sectors were overwritten. Given that, nothing was recovered by it or Window's CHKDSK. (I ended up with 4 C:/FOUNDnnn/FILE.CHK files (literally--4 directories of files FILE.CHK through FILE.CHK).) What was really strange was that the data that was written over the FAT and the root directory's sectors was data from a file on a _different_ FAT partition. And a later time it corrupted my C: disk, it was the _same_ file (from a different filesystem (partition)) whose contents was written all over the FAT and the root directory (at the exact same sector offset). I wonder whether fsck.vfat tried to check two different filesystems in parallel (I don't recall for sure whether they were on different disks), didn't keep its data straight, and wrote onto the first partition data from the second partition. No, I don't see how that would happen--I don't see why fsck.vfat would read the contents of a file. As I mentioned, another problem is that fsck.vfat ignores the /etc/fstab setting (a zero in the checking-order column) that is supposed to mean to not check the partition. That means that, without further hacking, you can't both have it installed on your system and have it _not_ run automatically (if you have any vfat partitions listed in /etc/fstab). Unless it has been fixed (the Sarge version has not), fsck.vfat SUCKS. (Not if I could only figure out what part of my kernel still sucks and keeps causing more filesystem corruption...) Daniel -- Daniel B. [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Unix-ify File Names
Frank Terbeck wrote:
Mike McClain [EMAIL PROTECTED]:
Frank Terbeck [EMAIL PROTECTED] wrote:
for FILE in `ls *$1` ; do
...
b) it breaks on filenames with spaces (and other special characters).
... Using 'for i in `ls *`'-type loops breaks this and is one of the
main reasons why people think spaces are bad in filenames.
(They are not bad, ...
In what sense are they not bad? Yes, they're certainly legal per the
filesystem and most tools that take filenames. However, they and other
special characters do make it more difficult to handle arbitrary file
names.
For example, if someone wants to use ls's feature of sorting by date
(e.g., ls -t *$1), they cant combine it with the for-loop construct
above (reliably).
Hey, is there any command for taking a filename and escaping/encoding
shell-special characters to make a string that, when parsed by the
shell, specifies that filename? I'm thinking of something that would
work like this:
for i in `encode_for_shell *` ; ...
(mapping each argument to a shell string for the argument's value)
or
for i in `find ... -print0 | xargs -0 encode_for_shell` ; ...
or
cmd=some_command
cmd=${cmd} `encode_for_shell $file_name_with_special_chars`
$cmd
(I'm thinking of something like Java's java.util.regex.Pattern.quote(String)
(see
http://java.sun.com/javase/6/docs/api/java/util/regex/Pattern.html#quote(java.lang.String)
) or Ruby's RegExp::escape(...)
(see http://www.ruby-doc.org/core/classes/Regexp.html#M001216 ), but
escaping/encoding for shell parsing instead of for regular-expression
parsing.)
some people just do not know how to handle them properly.)
You might not be, but it sounds like you're blaming users. Sometimes
it's developers of tools (including designers of formats) that don't
have an escape mechanism to handle spaces or other special characters
(or don't provide support for encoding special characters) who are to
blame.
I am aware that there are HOWTOs and other documents out there
that propagate 'for i `ls *foobar*`' loops. I don't know why their
authors do this. If they didn't know better they shouldn't have
written a shell scripting HOWTO in the first place.
Unfortunately for those they mislead, those authors don't know enough
to know they don't know better. (They must not be the type to dig
into things (e.g., shell syntax) to really understand them, or at
least enough to notice that they don't fully understand them yet.)
Some people use things like this instead:
[snip]
ls * | while read file ; do whatever_command $file ; done
[snap]
This is just a little better than the for loop. It still breaks in
some situations.
I see how it would break with a newline character in a file name.
What other cases break?
There is _no_ reason why 'ls' should ever be used to generate file
lists for loops of any kind.
What about things that ls does that the shell's expansion of wildcards
does not do (e.g., sorting by date or size)?
(Maybe ls should have an equilavent to find's -print0 option.)
Daniel
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: files in /var/tmp
Andrew M.A. Cater wrote: On Sun, Apr 08, 2007 at 03:44:33PM -0700, Kamaraju Kusumanchi wrote: ... My system (Debian Etch) has been recently compromised and I deleted most of the suspicious files. However I am not sure about these. Is it safe to delete them or do you think some process expects them to be there? ... Use Darik's Boot and Nuke to wipe the disk as thoroughly as you can. Then re-install with Etch and clean media. Why would you need to erase the disk like that? (As long as you re-create the systems on the partitions, there's no way the unerased sectors from previous files (from the compromise, before the file-system re-creation) will be strung together to re-create the bad files, right?) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: webcam+amsn
Vamos lá, Os sintomas que você citou são clássicos de uso de modem roteado, acesso condominial HomePNA, Rádio ... etc... onde se usam endereços mascarados... exemplo: 10.0.0.5 Com isso várias portas no Roteador estão travadas, sendo assim, você deve configurar o port-foward do roteador, liberando as portas: TCP: 6989-6900 UDP: 6989-6900 Para liberar as postas no Firewall você deve configurar via iptable através dos comandos no console: iptables -A FORWARD -p udp --dport 6890:6900 -j ACCEPT iptables -A FORWARD -p tcp --dport 6890:6900 -j ACCEPT Com isso provavelmente sua webcam deve funcionar beleza :-) Espero ter ajudado, Daniel B. Reggi -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian User List
s. keeling wrote: Kamaraju S Kusumanchi [EMAIL PROTECTED]: If the list is getting RTFM questions, it also means that the manuals are just not good enough to be understood. So probably trying to improve the Or people aren't finding the documentation. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] a dumb query? pls humor me
Steve Lamb wrote: [EMAIL PROTECTED] wrote: ... Not to mention that such actions are counterproductive. If someone is tortured into confessing to a crime, it is always suspect. Yes, but that isn't exactly what is going on, is it? What's going in is called, if I recall correctly, the short time problem. You know something is going to happen, something horrible, and you know that the person you have has information that will stop it. Once that event passes their information is useless. They won't give it up. What do you do? Do you let hundreds of thousands of people die or do you get the information using all means short of those that incur long-standing harm? That's an unrealistic case that distracts from the rest of the cases. It's rare the you _really_ know. (Remember that 24 is fiction.) So, no, that probably is _not_ the case that is going on. That's a decision that most people will never face in their lifetime and yet they feel they are supremely qualified to judge those who are in the unfortunate position to have to make that call. No. Most of the judging is of those who are using torture in much less definite situations. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Best File System for partitions over 600GB
Mike McCarty wrote: ... If power fails during a write, and the drive scribbles on the disc in a spiral pattern as the head moves toward the parking area, that particular disc is hosed. But the disks almost surely don't scribble on the disk in a spiral pattern. (They'd detect that power is failing (voltage is dropping) and turn off the write current before that happened.) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: a dumb query? pls humor me
Andrew Sackville-West wrote: yeah, but Oregon still doesn't trust me to pump my own gas. Of course, with some of the crap I see around here, that's probably a good thing. Is it that they don't trust you to pump the gas safely, or is it protectionism for gas-station worker as I think it was in some other state? Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT: A Republican!!!!!! (was Re: OT: sponge burning!)
Greg Folkert wrote: ... The Celsius Thermometer wil drop significantly slower the the Fahrenheit one. Only if it has more insulation. Otherwise, the temperature drops at the same speed. Of course, yes, the _numbers_ change at different rates. :-) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT: A Republican!!!!!! (was Re: OT: sponge burning!)
Paul Johnson wrote: Cloudless sky with negligable wind is an absence of weather. Nope. That sounds like clear weather to me. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT: sponge burning!
Greg Folkert wrote: DO NOT USE THE TERM ex-Marines it insults. The correct term is Former-Marines. Sorry. You don't get to re-define English. Ex-something means former something. If someone's a former Marine, he or she is also an ex-Marine. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT: sponge burning!
Chris Bannister wrote: On Sun, Feb 25, 2007 at 09:54:19PM -0500, Roberto C. Sanchez wrote: That is not it. The point is that if a fetus at 22 weeks can survive, who gets to decide when the fetus is actually alive. I say we err on the side of caution and say that it is alive from the moment of conception. You are welcome to your own opinion, however. So masturbation isn't murder? Is that Monty Python I hear singing? :-) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: A Republican!!!!!! (was Re: OT: sponge burning!)
Wulfy wrote: Roberto C. Sanchez wrote: Ooh. Don't even get me started on nuclear power. Cheap, clean, virtually unlimited. We can't use it *because* of the conservationists and environmentalists. Regards, -Roberto Decommissioning nuclear plant... storage of nuclear waste... clean? hardly. Try new-design breeder reactors. Breeder reactors let you extract much more of the available nuclear energy and reduce waste a lot. The new-design breeder cycle also avoids concentrating weapons-grade plutoniun. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Very disturbing feature in icedove
Dave Sherohman wrote:
On Wed, Feb 14, 2007 at 02:30:48PM -0500, Daniel B. wrote:
Dave Sherohman wrote:
On Sat, Feb 10, 2007 at 12:36:55PM -0500, Roberto C. Sanchez wrote:
I was complaining solely about the use of compact to mean delete.
Are you confusing the logical level (what the user almost always deals
with) with the physical level?
I would instead say that the Netscape/IceDove/Mozilla/SeaMonkey developers
are forcing the end user to deal with the details of the physical layer
rather than allowing users to exist, as they normally do, in the logical
realm.
Yes, that seems true.
Of course, sometimes users have to deal with the physical layer. I
guess the question is how much (how much is reasonable).
If you delete files from a disk (past any temporary Trash folder as on
Windows), it's true that you don't have to deal with really deleting
the files in the sense of making the space available for other files,
but you _do_ have to deal with really erasing the data (overwriting the
sectors) if you want to make sure the data is really gone.
The question in the current case is probably whether the user should
control when Seamonkey takes the time to compact folder since the
physical-layer aspect of the time it takes to re-write the files
(presumably) can't be hidden from the user (within the constraint of
using a standard mail-file format and reducing risk of corruption
from crashes or power failures).
At the logical level, the messages are already deleted (from the folder).
There is no way to get them back (from th[e] folder from which they're
deleted) going through the tool (Seamonkey).
If the tool does not provide a means to undelete messages, then I also
find the decision to not make permanent deletion (either when the user
changes folders or exits the program; it doesn't need to be immediate
for reasons which have been repeatedly discussed in this thread already)
a default action to be questionable at best. If you can't undelete it,
then why keep it around?
It might be part of the physical layer that the user does have to deal
with: reliability.
It's not so much that deleted messages _are_ explicitly _kept_ around;
it's that they are _not_ _deleted_ right away. That, of course, is
because trying to delete a message right away by re-arranging the file
(given the current file format, or course) would increase the risk of a
corrupted file in case of crash or power failure during the re-arranging
(and because immediately copying all non-deleted messages to a new file
when a message is deleted would be too slow).
But yes, since (non-hacker) users can't recover messages deleted from
a folder, it would be good if users didn't have to deal with it.
However, given the time it takes to compact a large folder, I don't
know if automatically compacting on, say, exit would be good.
Maybe Seamonkey/etc. should default to compacting on exit or some other
reasonable time, but display a why is Seamonkey compacting message or
button that explains what it is doing and points the user toward the
appropriate settings.
...
But that's the same as deleting a file: Deleting a file tells the file
system to forget about remembering the data, but it doesn't usually
overwrite the data, so it or pieces of it are still on the disk unless
you perform some other operation to actually remove (overwrite) it.
Not really a very good analogy, as the file system will reuse the disk
space without requiring any additional action ('compact', 'purge',
'expunge', whatever) after the file has been deleted.
How is it not a perfect analogy? I didn't talk about the physical aspect
of re-using the space, which the file system _does_ handle for the user;
I talked about the physical aspect of making sure your data was erased,
which users have to be aware of (whether or not they should have to be).
Daniel
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Very disturbing feature in icedove - monolithic files not always bad
Steve Lamb wrote: ... And before we get into this again I only have to ask one question. If a single file is such a bad thing why is it MySQL (and other) databases don't store records per file but, instead, per table? You'd think the corruption problem would be just as bad for them. And yet companies around the world routinely store immense amount of data in monolithic files without much concern. Far more than the piddly amount of mail any individual on here would worry about. Whether a single file is bad (or requires copying to a new file to reliably make changes) all depends on the file format. Making a logical change to the data involves making a set of one or more physical changes to the file. If making the first changes in the set and forgetting to make the rest of the changes (e.g., in case of a power failure) leaves the file corrupted and unrecoverable, then it's not a good format for making incremental changes without copying to a new file for reliability. The mbox format that Seamonkey uses is like that. However, if making the first part of the changes without making the second part leaves the file in a state such that things are recoverable (e.g., either the partial change is backed out or the partial change is completed), then you can make incremental changes without needing to copy to a new file for reliability. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Very disturbing feature in icedove
Steve Lamb wrote: Dave Sherohman wrote: OK, one more time: Delete by default does not have to mean delete *immediately* by default. Look at the underlined text above. I already explicitly stated that I didn't mean immediate deletion and that delete- on-folder-change or delete-on-exit are probably better, Uh, not in my mind. Maybe it stems from my years in the ISP business but generally users only want things deleted when they say they want it deleted. That doesn't mean When they press the delete key or even After pressing the delete key and changing folders or after pressing the delete key and exiting the program. After they say, in its most conservative setting, is after they have configured the client to delete the way they want. And before you or anyone else jumps up with more of your preferences and ignorance about the reality of computers let me remind you of one simple fact. Windows and OSX, by default, require the user to empty the trash. ... ... until the end user twiddles the knob they want to keep as much as possible because the expected behavior is that the user has to tell the computer to delete it, really, and this time I mean it! This is no different. Actually, it is. We're not talking about Seamonkey's Trash folder to which tentatively deleted messages are moved and from which users can recover or really delete those tentatively deleted messages, working like Windows' and OS X's trash folders that you you mention. We're talking about physically deleting deleted copies of messages. (When you logically tentatively delete a message from the Inbox folder and Seamonkey logically moves it to the Trash folder, there's still a physical copy of the data in the file that implements the Inbox folder. That physical copy is never available to the user through the tool.) Surely you're not ignorant of that reality. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Very disturbing feature in icedove
Freddy Freeloader wrote: ... ... Any message that has been deleted in Icedove/Thunderbird/SeaMonkey is recoverable, at least up until the time the folder is compacted or the Trash folder is emptied, from the Trash folder. After that happens then, no, the message is not recoverable. What is so odd about this? I have yet to see a graphical email client that doesn't act pretty much the same way. You might need to distinguish between messages that are recoverable by non-hacker users (via the tool) vs. messages that are recoverable if you know how to clear the appropriate X-Mozilla-Status bit. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attracting newbies
Joe Hart wrote: ... Sorry to butt in here, but I think a point needs to be made. A large number of modern websites do not allow the viewer to choose how to view the page. If the browser window is too large, empty space will appear on both sides. If the browser window is too small, the view will be cut. Yes, that was part of my point--many webpages are written poorly, violating the design and intent of HTML and the web. If the page is well written, scrollbars may appear. What you do mean? (Scrollbars appear normally automatically, so it doesn't a page doesn't have to be written well to make them appear. It just has to to _not_ be written so poorly that it suppresses scrollbars (which you could say is written well enough for them to appear).) Also, the appearance of a horizontal scrollbar frequently is an indication of a page's being poorly written.) Quite frequently webmasters choose fonts that are so small that they are almost non-legible, and one must increase the text size. Yes, that's another widespread problem. Hopefully this is a fad. Unfortunately, it's not a fad which we could hope they'd get tired of and drop. Much of it is ignorance (even of the fact that things look different on other people's screens because their settings (font size, screen size, browser, OS even) are different). Of course, part of it is Microsoft's bad default configuration of IE (to display HTML-default-sized text in a large physical font, inducing many web page designers to start off by trying to reduce the font size, messing things up for everyone with a properly adjusted browser (an adjusted IE or a browser with better defaults)). The whole idea of HTML was to allow the browser to adapt to the user. Definitely (or, more precisely, it's to allow the browser to adapt the content to the user). Someone decided to throw that ideology out the window. Or was too irresponsible to even notice that that was the ideology. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attracting newbies
marc wrote: Daniel B. said... [EMAIL PROTECTED] wrote: On Thu, Feb 08, 2007 at 09:27:40AM -, marc wrote: ... And the user can also provide their own CSS too, should they wish. Right. But the reader shouldn't have to re-write a page's style sheet just to be able to read it conveniently. And no-one suggested that they should. Saying that the user can provide their own CSS suggests that that's an acceptable solution, so, yes, your words DID suggest that. Daniel- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attracting newbies
[EMAIL PROTECTED] wrote: On Thu, Feb 08, 2007 at 09:27:40AM -, marc wrote: Daniel B. said... ... Please note another problem with PDF: The page size and layout are fixed. Not really a problem, more of a feature of the format; the idea being that a PDF renders the same regardless of the display platform (at least, in theory). In many situations, this is a very good thing. It's a good thing only when the exact formatting really matters. However, frequently it's a bad thing. If you have a regular paragraph of words, it doesn't matter exactly where the line breaks are. Therefore, delivering it in PDF with lines broken at some particular length requires viewing it in a window wide enough to see whole lines. Delivering it in HTML allows the browser to break the lines to fit within the the user's chosen browser pane width. That's a heck of a lot more flexible. HTML adapts to the user's browser pane width (well, if the author doesn't break HTML's ability to do that). Again, to be pedantic, it's CSS that controls the layout, hence the author simply provides multiple CSS, which is what it's designed to do. What do you mean by the author simply provides multiple CSS? If _you_ want to look at something in a full-screen browser window and _I_ want to use a half-screen-width browser window (e.g., to see two web pages side by side), how is an author going to provide multiple CSS stylesheets to cover both of us? What about every size in between? Just to be historical, HTML text adapted to the user's browser long before CSS had even been invented. Of course! (Why do you point that out?) The user can choose how much screen width to use for a browser, the browser can wrap regular text and tables to fit, and the user doesn't have to scroll horizontally to read the bulk of the page. And the user can also provide their own CSS too, should they wish. Right. But the reader shouldn't have to re-write a page's style sheet just to be able to read it conveniently. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Very disturbing feature in icedove
Dave Sherohman wrote: On Sat, Feb 10, 2007 at 12:36:55PM -0500, Roberto C. Sanchez wrote: ... I was complaining solely about the use of compact to mean delete. Are you confusing the logical level (what the user almost always deals with) with the physical level? At the logical level, the messages are already deleted (from the folder). There is no way to get them back (from thr folder from which they're deleted) going through the tool (Seamonkey). (It's not like Emacs' RMail where you can undelete messages that you have marked for deletion, because that tool uses user-visible marking for deletion to give the user a chance to recover from errors instead of Seamonkey's method of using a Trash folder.) However, yes, copies of the data still exist in the files. And, yes, although the user doesn't _usually_ deal with that, the user sometimes does, e.g., when you want to make sure the data has been actually been deleted. But that's the same as deleting a file: Deleting a file tells the file system to forget about remembering the data, but it doesn't usually overwrite the data, so it or pieces of it are still on the disk unless you perform some other operation to actually remove (overwrite) it. Just because Microsoft chooses to arbitrarily redefine words does not mean that we should follow them in doing so. True, but Mozilla/Seamonkey is not follwing Microsoft. It follows what Netscape 4.x (and presumbly earlier) versions called it. And yes, something like Purge deleted message or even (if it weren't too long) Compact to erase old copies of deleted messages would be clearer to users than compact folders. Would Purge Deleted Messages work? Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: iceweasel not being recognized by ISP website
Greg Folkert wrote: On Tue, 2007-02-06 at 17:28 -0500, Daniel Barclay wrote: H.S. wrote: The website ... suggests I either download Firefox or IE 6 or 7, all for Windows. They do not support any non-Windows browser at all! Firefox runs on Linux. Or do you mean that website says or implies that it only works with the Windows version of Firefox (and not others, e.g., the Linux version)? http://www.toonamijetstream.com/gen/requirements/firefox/index.html Go on, just try it. That plug-in really works on Linux. Gee, I only asked if he meant that their site (which would include their plug-in) only worked with Windows versions of Firefox, to clear up the seeming contradiction. Is there some reason a simple yes wouldn't have sufficed? Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attracting newbies
Douglas Allan Tutty wrote: On Tue, Feb 06, 2007 at 10:09:30PM -0500, Michael Pobega wrote: ... Pdf can have internal links as well as a table of contents that one can click on. On the other hand, one needs X to read it and a postscript capable printer to print it (yes I know...). Please note another problem with PDF: The page size and layout are fixed. HTML adapts to the user's browser pane width (well, if the author doesn't break HTML's ability to do that). The user can choose how much screen width to use for a browser, the browser can wrap regular text and tables to fit, and the user doesn't have to scroll horizontally to read the bulk of the page. With PDF, the text is wrapped at a fixed width (once you pick a magnification to make the text readable, the text columns and the overall page size are fixed). The user has to scroll horizontally more frequently (possibly even once per line of text depending on font size and page width). PDF is good for it's sleekness really, ... Are you thinking of PDF itself or laid-out documents you've seen in PDF? Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: iceweasel not being recognized by ISP website
Chris Bannister wrote: On Sun, Feb 04, 2007 at 11:42:22AM -0800, Andrew Sackville-West wrote: Raju's point about employment with capitalone is entirely different. CapitalOne is not (at least ostensibly) a web content company. As such they can (IMO) be somehwat forgriven for having non-compliant stuff on the web. That said, their web folks should be slapped around a bit. make sense? http://linuxmafia.com/ What about it? (Is there any particular part of the http://linuxmafia.com/ site you meant to be pointing to?) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Is your KERNEL pattern right?
Mark Williamson wrote:
BUS==usb, SYSFS{product}==Palm Handheld*, KERNEL==ttyUSB[013579],
Shouldn't you be matching [13579]? Matching the 0 as well will match
the first serial pipe to the handheld, not the second. On my Palm zire
I can only hotsync to the second...
...
... (I'm not sure what the
other serial pipe at zire0 is actually useful for, I don't use it).
Does anyone else know that the first serial device/pipe/whatever
is for?
Daniel
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Please stop using Horrendous Coloring (or coloring period)
Francisco Zabala wrote: ... Please, any comments (such as the one above) that you feel beneficial for ALL Debian users who read this list, please feel free (and encouraged) to submit them to the whole list (as opposed to the individual user), as I am certain we can all benefit from it. So _you_ get to send comments such as your previous one to the list and other people, such as me, don't? Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Please stop using Horrendous Coloring (or coloring period)
Francisco Zabala wrote: ... Couldn't imagine how simple coloring would generate so much hatred (with so much passion). To help your weak imagination, consider this possible explanation: Because the message sender effectively reached into Greg's computer and rudely told the his mail reader to ignore his preferred foreground and background colors and size for text, especially because one overriding color was so highly saturated. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Are lilo and grub compatible?
José Alburquerque wrote: ... As far as I know, lilo and grub are mutually exclusive because both are boot-loaders that use a disk's mbr to boot up operating systems. LILO certainly isn't restricted to using the MBR. I've been using it on a floppy for years. (The floppy is a physical boot Linux or boot Windows switch on my dual-boot system.) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: MODERATOR - NEWS GROUP
Douglas Tutty wrote: .. Since the spam doesn't seem to be targeted specifically to *N*X system users, it may be safe to think that their targeted audience mostly is running *doze. Sind *doze people can't handle .ps files easily there's less incentive for the spammers to send .ps files. So the list filter could drop any messages containing parts other than text/html, .ps, and .gz files. What about people who read Linux mailing-list mail on Windows? The other thing I've noticed is that a new spam will have a subject starting with Re: . Could not the list filter verify that the subject Re'd actually exists in the list archive within a set time frame (30 days?). Then anyone who replies to an old message has their message ignored. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: bad anti-aliasing; what thinks my CRT is an LCD display
Florian Kulzer wrote: [ I accidentally sent this message when it was only half finished; here is the full text. Sorry for the noise. ] On Tue, Jan 02, 2007 at 22:53:17 -0500, Daniel B. wrote: Where is the configuration or auto-detection of whether a video display device is a CRT or an LCD? ... It seems that something thinks my display device is an LCD panel, when actually it's a CRT. Check the file .fonts.conf in your home directory. This is the stanza to turn off RGB antialiasing: match target=font edit mode=assign name=rgba constnone/const /edit /match Thanks. I'll try that. Does fontconfig always assume (e.g., upon installation) that it should do RGB antialiasing, or does it try to detect what kind of display I have (e.g., by asking a lower level, say, the X11 server)? Where is the display type supposed to be configured or autodetected? ... you can try different options for fontconfig by running (as root) dpkg-reconfigure fontconfig-config I had tried that, but it didn't seem to be give my any options related to antialiasing or the display type. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
bad anti-aliasing; what thinks my CRT is an LCD display
Where is the configuration or auto-detection of whether a video display device is a CRT or an LCD? I've been getting strange color fringes around text when anti-aliasing is turned on. It seems that something thinks my display device is an LCD panel, when actually it's a CRT. Thanks, Daniel -- Daniel Barclay [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: printing problem - spurious d characters, bad pixels,spurious form feeds - SOLVED
IB. wrote: charles norwood wrote: On Thu, 2006-07-06 at 10:27 -0400, Daniel B. wrote: ... Also, given that the spurious characters appear at different places in the printout when I try again suggests that something random is going on (dropped or extra characters in the output stream), which doesn't sound like having the wrong PPD file. Oh--another thing is that since upgrading to Sarge and 2.6.8 (2.6.8-2-k7-smp), trying to print occasionally hangs the entire system. That suggests a problem with the parallel port driver or something closer to that later. Daniel I don't have an answer, but my experience is random errors are due to hardware failing. No, I'm pretty sure it's not failing hardware: The hardware worked fine with Woody, kernel 2.4, and the pre-CUPs printing system. And it still works fine when I boot into Windows and print. I found the technical problem. As I hypothesized, it was a problem with the kernel in the parallel port driver. CONFIG_PARPORT_PC_FIFO was enabled. The fact that CONFIG_PARPORT_PC_FIFO is experimental leads to another question: Why the hell did a Debian kernel for the _stable_ release (Sarge) use an _experimental_ kernel option?! Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
kernel parallel-post printing problem - spurious characters printed randomly interspersed in printout - since kernel (and Sarge) update
I wrote: Since I upgraded to Debian Sarge and kernel 2.6.8 (2.6.8-2-k7-smp), I've been getting lots of errors in my printouts. The error pattern is that at multiple, seemingly random positions in the middle of the printout, there is a spurious d character... ... Because plain-text documents (seem to) print fi[n]e, it doesn't seem to be a kernel problem with the parallel port. Actually, it DOES seem to be a problem in the kernel (or downstream) after all. If data is written slowly enough to device file /dev/lp0, the printout is fine; otherwise, the printout is corrupted at random locations, seemingly by extra or missing bytes, as I described before [ http://lists.debian.org/debian-user/2006/10/msg00133.html ]. That strongly suggests a problem with handshaking or interrupt handling in the kernel (and not with the parallel port and printer since printing worked fine with Woody and kernel 2.4 and still works fine under Windows). I configured lpd to write to a file instead of /dev/lp0. I printed a Postscript file, and after having it rastered for my printer, lpd wrote the print data to the specified file. Then I copied the file's contents to /dev/lp0 in a way that slowed things down, like this (subject to typos and any rememberos): cat x.dat | ( while dd bs=1 count=1 ; do : ; done ) | cat /dev/lp0 When the block size was 1 or maybe also when it was 2, the printout would be fine. When the block size was 3 or more, I'd almost always get errors in random locations. (Repeated attempts with the same block size would get errors in different positions.) Does anyone know what's going on? Does the kernel have a bug in its parallel-port handshaking or interrupted-handling logic? Do I have something configured wrong? Is my hardware (Asus A7M266-D motherboard (AMD 768 chip)) buggy near the parallel port? Thanks, Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: printing problem - spurious characters printed randomly interspersed in printout - since Sarge/kernel update
Marty wrote: Daniel B. wrote: [With] Debian Sarge and kernel 2.6.8 ... I've been getting lots of errors in my printouts. ... at multiple, seemingly random positions ... there is a spurious d character... ... This applies to files that go through the magicfilter/gs rasterizer, but does not seem to apply to text files (which get written directly to the printer). It seems as if ... something got out of sync between the stream of printer commands from the rasterizer and the printer itself ... Re-printing the same document the same way ... yields errors in different locations. It does not seem to be a cabling or printer unreliability problem: ... Because plain-text documents (seem to) print fi[n]e, it doesn't seem to be a kernel problem with the parallel port. Therefore, it seems to be a printing software problem--except that the randomness ... doesn't make sense. I first noticed this problem when I upgraded to Sarge and kernel 2.6 _and_ switched from lpr to CUPS. ... I removed CUPS and re-created my previous setup that used lpr and magicfilter for printing. Unfortunately the problem still remains ... ... [I suspect] that the d characters ... were supposed to be part of the frame around bitmap data ... as if [extra bytes], or a bad count, is being generated by the Postscript-to-bitmap renderer or an extra byte or bytes are being written out to the printer). Has anyone seen a problem like this? I haven't seen this but I also have a BJ-series printer and mine has DIP switches which I used to experiment with different emulation modes and other settings. I think one emulation mode was Epson compatible, and the other native. What might be causing the random variation in the errors? One guess is that you are using the wrong emulation mode, or the right emulation with the wrong DIP switch setting. I doubt having wrong dip-switch settings is the problem since it worked fine with Woody, kernel 2.4, and lpr/magicfilter/gs, and it still works fine under Windows. (Well, I guess it's possible that the gs filter changed assumptions about the dip-switch settings, but that doesn't seem to fit the randomness.) Also I tried different BJ series .ppd files and I ended up using a .ppd file for a different model, but never understood why it seemed to work better. I'm not using PPD files. I'm using the lpr package and the magicfilter package (and not using any PPD files with the lpr package). How can I intercept the bytes that are being sent by lpd You mean cupsd? Are you trying to run both together? No and no. I mean lpd from package lpr. I'm not running them together: ... I removed CUPS and re-created my previous setup that used lpr and magicfilter for printing. (I switched back because I thought that would get me back to having a working printer. Of course, it ended up telling me that the problem was not in CUPS, and is somewhere else in Sarge (in gs?) or kernel 2.6 (or something else?).) to the printer device (to see the the errors are being injected before that point)? The only way I know to intercept parallel port data ... By device there I meant the character device /dev/lp0 (as opposed to the parallel port hardware). is gdb (possibly with ddd or other front end) or a logic analyzer, but these seem like massive overkill. Definitely. Would pointing lpd to a pipe (or maybe even a file) instead of /dev/lp0 work. Does lpd do anything with ioctls such that it would fail or refuse to work a device file that's not a character device? (In that case, is there any way to log the data that's written to a character device?) The reason I'm thinking about intercepting the data at different points is to see if I can find where the randomness first shows up. For example, if it first shows up in the output of the gs filter, then pretty surely gs is buggy; if it first shows up in what lpd writes to /dev/lp0, then lpd is buggy; etc. How can I check the bytes that are actually being send out from the kernel to the parallel port (to see [i]f the errors are being injected after the byte stream gets to the parallel port device (/dev/lp0)? The simplest way is so swap out the suspect device, but why or how do you think this could happen? Which level of device do you mean here? (I thought you meant the hardware, but that can't be swapped out.) (I meant between the file-system-level device /dev/lp0 and probably the point in the kernel where it's writing out the sequence of bytes to the hardware, or possibly at the level of what the hardware actually got from the kernel.) I was thinking that if the kernel wasn't handling parallel port interrupts correctly, it could be dropping bytes or sending extra bytes. (Of course, that theory (seemingly) is shot down by the fact that plain text (seemingly) prints fine.) Any other ideas? ... Other than that, you might try to see if you can track down what changed to cause the problem. If you have
printing problem - spurious characters printed randomly interspersed in printout - since Sarge/kernel update
Since I upgraded to Debian Sarge and kernel 2.6.8 (2.6.8-2-k7-smp), I've been getting lots of errors in my printouts. The error pattern is that at multiple, seemingly random positions in the middle of the printout, there is a spurious d character, and frequentlyright after the d there sometimes is a column or two of erroneous dots. (Sometimes after the d the printer jumps to the next page; sometimes the printer beeps a lot as if it is rendering ASCII BEL characters.) The printed d seems to be in the printer's native font for plain- ASCII(+) printing. (It's not in the same Postscript font of any text in the middle of which the error occurs.) This applies to files that go through the magicfilter/gs rasterizer, but does not seem to apply to text files (which get written directly to the printer). It seems as if: - the (non-Postscript) printer (a Canon BubbleJet BJ-200ex) had correctly executed a printer command to print out a packet of the bitmap generated by magicfilter, magicfilter's bj200 filter, and/or gs (gs-gpl) - the printer was ready for another character (to print plainly) or bitmap printing command - something got out of sync between the stream of printer commands from the rasterizer and the printer itself - the next byte in the stream was an ASCII d (a 0x64 byte), - the printer, being back in plain-text mode, printed a d - the printer recognized a bitmap-printing command, and got back into bitmap-printing mode, but something wasn't quite synchronized (thus the the next column or so of dots got messed up) - things got synchronized again for a while (since the printout's fine for somtimes several printer lines worth of bitmap dots until the next error). The location of the errors seems to be random. Re-printing the same document the same way (e.g., printing a text document through enscript or printing a web page from a browser) yields errors in different locations. It does not seem to be a cabling or printer unreliability problem: - I haven't even unplugged the printer or printer cable (parallel) since I upgraded from Woody and kernel 2.4 to - Printing still prints fine from Windows. - Plain-text documents print fine (lpr xxx.txt), with no known errors. Because plain-text documents (seem to) print file, it doesn't seem to be a kernel problem with the parallel port. Therefore, it seems to be a printing software problem--except that the randomness (the non-repeating position and the varying quantity of the errors) doesn't make sense. I first noticed this problem when I upgraded to Sarge and kernel 2.6 _and_ switched from lpr to CUPS. Thinking the problem was something in CUPS or foomatic (the PPDs files?), today I removed CUPS and re-created my previous setup that used lpr and magicfilter for printing. Unfortunately the problem still remains (except that plain-text files print fine, since they are printed directly to the printer). The errors always involve a d character. That makes me think that the d characters I see are 0x64 bytes that were supposed to be part of the frame around bitmap data, and that some count of bitmap-data bytes is getting out of sync with the stream of bitmap-data bytes (e.g., as if an extra bytes or bytes, or a bad count, is being generated by the Postscript-to-bitmap renderer or an extra byte or bytes are being written out to the printer). Has anyone seen a problem like this? What might be causing the random variation in the errors? How can I intercept the bytes that are being sent by lpd to the printer device (to see the the errors are being injected before that point)? How can I check the bytes that are actually being send out from the kernel to the parallel port (to see of the errors are being injected after the byte stream gets to the parallel port device (/dev/lp0)? Any other ideas? Thanks. Daniel -- Daniel Barclay [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: dma errors
David Baron wrote: On Thursday 27 July 2006 00:38, Daniel B. wrote: ... I have had similar problems with an Asus A7M266-D motherboard with multiple kernel versions (2.2 through 2.6, I think). From what I've been able to gather, my motherboard's IDE controller (AMD 768?) is buggy, and the Linux kernel doesn't do whatever the Windows kernel does to work around the problem. (I can't run the disks attached to the onboard IDE controller at full sped (i.e., with DMA on) or I get extreme filesystem corruption in Linux. Under Windows, I can run a seemingly full speed without any data- corruption problems.) Check the cables! Why should I waste my time checking the cables again? The motherboard IDE controller and disks work fine with Windows. The problem under Linux has occurred with various disk drives with various cables. And, more importantly, some Linuxer dismissed the A7M266-D motherboard (or the AMD chipset) as broken hardware. Therefore it is not the darn cables! Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: dma errors
Greg Madden wrote:
...
John Schmidt [EMAIL PROTECTED] wrote:
...
I have two IDE drives ... the motherboard is quite old.
...
I get the following error messages from dmesg:
... hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x84 { DriveStatusError BadCRC }
ide: failed opcode was: unknown
hde: dma_timer_expiry: dma status == 0x20
hde: DMA timeout retry
PDC202XX: Primary channel reset.
PDC202XX: Secondary channel reset.
hde: timeout waiting for DMA
hde: dma_timer_expiry: dma status == 0x21
hde: DMA timeout error
hde: dma timeout error: status=0x58 { DriveReady SeekComplete
DataRequest } ide: failed opcode was: unknown
...
What would be the source of these errors and how should I go about
fixing them?
A couple of things, the generic IDE driver doesn't do dma , make sure
you load the module for your chipset. Using 'hdparm' will
show if dma is set for the drives.
Greg:
Clearly John's driver is doing DMA; otherwise all the DMA errors
wouldn't be present.
John:
Which motherboard do you have?
I have had similar problems with an Asus A7M266-D motherboard with multiple
kernel versions (2.2 through 2.6, I think).
From what I've been able to gather, my motherboard's IDE controller (AMD
768?) is buggy, and the Linux kernel doesn't do whatever the Windows
kernel does to work around the problem.
(I can't run the disks attached to the onboard IDE controller at full
sped (i.e., with DMA on) or I get extreme filesystem corruption in Linux.
Under Windows, I can run a seemingly full speed without any data-
corruption problems.)
Daniel
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: printing problem - spurious d characters, bad pixels, spuriousform feeds
charles norwood wrote: On Thu, 2006-07-06 at 10:27 -0400, Daniel B. wrote: ... Also, given that the spurious characters appear at different places in the printout when I try again suggests that something random is going on (dropped or extra characters in the output stream), which doesn't sound like having the wrong PPD file. Oh--another thing is that since upgrading to Sarge and 2.6.8 (2.6.8-2-k7-smp), trying to print occasionally hangs the entire system. That suggests a problem with the parallel port driver or something closer to that later. Daniel I don't have an answer, but my experience is random errors are due to hardware failing. No, I'm pretty sure it's not failing hardware: The hardware worked fine with Woody, kernel 2.4, and the pre-CUPs printing system. And it still works fine when I boot into Windows and print. Besides, random errors are also easily due to bugs in concurrent code (e.g., the kernel and interrupt handlers (device drivers)). My lockups have been due to hw/sw misconfiguration e.g. irqs colliding. What kind of configuration? (BIOS settings? auto-detected Linux configuration? manual Linux configurations?) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: printing problem - spurious d characters, bad pixels, spuriousform feeds
Felipe Sateler wrote: Daniel B. wrote: Can anyone help with this? I got absolutely no replies when I posted it before. Now the problem is much worse, with dozens of errors per page, maybe evening averaging one error per line of text (when printing a plain-text file). I'm taking a _very_ wild guess here: perhaps you are using the wrong PPD file. Have you tried with alternative ones (perhaps downloading a new one from linuxprinting.org)? I'll double check. However, I'm pretty sure I've got the right PPD file. Also, given that the spurious characters appear at different places in the printout when I try again suggests that something random is going on (dropped or extra characters in the output stream), which doesn't sound like having the wrong PPD file. Oh--another thing is that since upgrading to Sarge and 2.6.8 (2.6.8-2-k7-smp), trying to print occasionally hangs the entire system. That suggests a problem with the parallel port driver or something closer to that later. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
printing problem - spurious d characters, bad pixels, spuriousform feeds
Can anyone help with this? I got absolutely no replies when I posted it before. Now the problem is much worse, with dozens of errors per page, maybe evening averaging one error per line of text (when printing a plain-text file). Thanks. --- Since I upgraded to Debian Sarge and switched to CUPS for printing, I've been getting lots of errors in my printouts. The error pattern is that somewhere in the middle of the printout, there is a spurious d character, and right next to the d (I think after it) there is a column or two of erroneous dots. (Sometimes after the d the printer jumps to the next page.) The printed d seems to be in the printer's built-in font for plain-ASCII(+) printing. (It's not in font of any text in the middle of which the error occurs.) It seems as if: - the (non-Postscript) printer (a Canon BubbleJet 200-ex) has going along happily printing out the bitmap generated by gs or whatever my Postscript-to-printer-commands filter is, - something got out of sync (e.g., the printer thought the bitmap data ended), - the next byte was an ASCII d, - the printer, being back in plain-text mode, printed a d - the printer got back into bitmap-printing mode, but something wasn't quite synchronized (thus the the next column or so of dots got messed up) - things got synchronized again for a while (since the printout's fine until the next error.) The location of the errors seems to be random. Re-printing the same document the same way (e.g., printing a text document through enscript or printing a web page from a browser) yields errors in different locations. I'm pretty sure it's not a cabling or printer unreliability problem. I haven't even unplugged the printer or printer cable (parallel) since I upgraded from woody to sarge, and (I think) printing still prints fine from Windows.) I think it's a printing software problem (although I guess it could theoretically be a kernel problem with the parallel port). When I upgraded from Woody to Sarge, I also switched to using CUPS from whatever I had before, which I think involved magicfilter and using gs as the Postscript engine. The errors always involve a d character. That makes me think that the d characters I see are d bytes that were supposed to be partof the frame around bitmap data, and that some count of bitmap-data bytes is getting out of sync with the stream of bitmap-data bytes (e.g., as if an extra bytes or bytes, or a bad count, is being generated by the Postscript-to-bitmap renderer or an extra byte or bytes are being written out to the printer). Has anyone seen a problem like this? Thanks. Daniel -- Daniel Barclay [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
printing problem - spurious d characters, bad pixels, spuriousform feeds
Since I upgraded to Debian Sarge and switched to CUPS for printing, I've been getting lots of errors in my printouts. The error pattern is that somewhere in the middle of the printout, there is a spurious d character, and right next to the d (I think after it) there is a column or two of erroneous dots. (Sometimes after the d the printer jumps to the next page.) The printed d seems to be in the printer's built-in font for plain-ASCII(+) printing. (It's not in font of any text in the middle of which the error occurs.) It seems as if: - the (non-Postscript) printer (a Canon BubbleJet 200-ex) has going along happily printing out the bitmap generated by gs or whatever my Postscript-to-printer-commands filter is, - something got out of sync (e.g., the printer thought the bitmap data ended), - the next byte was an ASCII d, - the printer, being back in plain-text mode, printed a d - the printer got back into bitmap-printing mode, but something wasn't quite synchronized (thus the the next column or so of dots got messed up) - things got synchronized again for a while (since the printout's fine until the next error.) The location of the errors seems to be random. Re-printing the same document the same way (e.g., printing a text document through enscript or printing a web page from a browser) yields errors in different locations. I'm pretty sure it's not a cabling or printer unreliability problem. I haven't even unplugged the printer or printer cable (parallel) since I upgraded from woody to sarge, and (I think) printing still prints fine from Windows.) I think it's a printing software problem (although I guess it could theoretically be a kernel problem with the parallel port). When I upgraded from Woody to Sarge, I also switched to using CUPS from whatever I had before, which I think involved magicfilter and using gs as the Postscript engine. The errors always involve a d character. That makes me think that the d characters I see are d bytes that were supposed to be partof the frame around bitmap data, and that some count of bitmap-data bytes is getting out of sync with the stream of bitmap-data bytes (e.g., as if an extra bytes or bytes, or a bad count, is being generated by the Postscript-to-bitmap renderer or an extra byte or bytes are being written out to the printer). Has anyone seen a problem like this? Thanks. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
printing problem - spurious d characters, bad pixels, spurious form feeds
Since I upgraded to Debian Sarge and switched to CUPS for printing, I've been getting lots of errors in my printouts. The error pattern is that somewhere in the middle of the printout, there is a spurious d character, and right next to the d (I think after it) there is a column or two of erroneous dots. (Sometimes after the d the printer jumps to the next page.) The printed d seems to be in the printer's built-in font for plain-ASCII(+) printing. (It's not in font of any text in the middle of which the error occurs.) It seems as if: - the (non-Postscript) printer (a Canon BubbleJet 200-ex) has going along happily printing out the bitmap generated by gs or whatever my Postscript-to-printer-commands filter is, - something got out of sync (e.g., the printer thought the bitmap data ended), - the next byte was an ASCII d, - the printer, being back in plain-text mode, printed a d - the printer got back into bitmap-printing mode, but something wasn't quite synchronized (thus the the next column or so of dots got messed up) - things got synchronized again for a while (since the printout's fine until the next error.) The location of the errors seems to be random. Re-printing the same document the same way (e.g., printing a text document through enscript or printing a web page from a browser) yields errors in different locations. I'm pretty sure it's not a cabling or printer unreliability problem. I haven't even unplugged the printer or printer cable (parallel) since I upgraded from woody to sarge, and (I think) printing still prints fine from Windows.) I think it's a printing software problem (although I guess it could theoretically be a kernel problem with the parallel port). When I upgraded from Woody to Sarge, I also switched to using CUPS from whatever I had before, which I think involved magicfilter and using gs as the Postscript engine. The errors always involve a d character. That makes me think that the d characters I see are d bytes that were supposed to be partof the frame around bitmap data, and that some count of bitmap-data bytes is getting out of sync with the stream of bitmap-data bytes (e.g., as if an extra bytes or bytes, or a bad count, is being generated by the Postscript-to-bitmap renderer or an extra byte or bytes are being written out to the printer). Has anyone seen a problem like this? Thanks. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: RAID Sizes (was Re: Why do people in the UK put a u in the word color?)
[EMAIL PROTECTED] wrote: On Mon, May 08, 2006 at 01:50:27PM -0400, Daniel B. wrote: Mike McCarty wrote: Well, I used to work as a watchmaker, and I can't think of any context where KB stands together as written with K meaning karat. That's not surprising--in SI, the prefix is the scale factor, and the remainder is the unit. I don't think there are any unit symbols that have multiple uppercase letters. Is Hz for Hertz not standard? Yes, it is standard. Why do you ask? It's certainly not a unit symbol with multiple uppercase letters. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] blink tag effect with animation
H.S. wrote: Arafangion wrote: I HATE BLINKING TEXT. I am sure that this sentiment is shared by many other people. Agreed. However, in case, the user wants a pair of words flashing for a few days to attract attention to a special item. Hmm. Words flashing for a few days. That sounds REALLY annoying. :-) At least do your flashing with animated images--browsers usually let one stop the animations. (One reason blinking text is hated is probably that browsers don't stop the blinking when they stop image animation.) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: RAID Sizes (was Re: Why do people in the UK put a u in the word color?)
Willie Wonka wrote: Serial ATA (SATA) data transfer rate specification = 1500 *mbps* or *mb/sec* (megabits per second). No. Megabits be per second is Mbps (lowercase m means milli). Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: RAID Sizes (was Re: Why do people in the UK put a u in the word color?)
Willie Wonka wrote: ... 1 bit * 8 = 1 byte ^^ I forgot to capitalize my 'B' in Byte above The word byte doesn't need to be capitalized. (Were you thinking of the capitalized letter B by itself when it stands for the word byte?) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: RAID Sizes (was Re: Why do people in the UK put a u in the word color?)
Willie Wonka wrote: ... IOW - Is this how one would correctly display these rates ? 1500mbps = 1.5gbps = 187.5mBps = 1.875gBps ? I think you mean 1500Mbps = 1.5Gbps = 187.5MBps = 1.875GBps As you can see the capitalized 'B' appears a tad ...'out of place'(?), but it's likely /very/ necessary, in order to maintain clarity. If you're at all familiar with SI it shouldn't look out of place. Consider the scaled units mA, GW, kV, mPa, etc. While I may over-annunciate and over-emphasize when referring to Bytes, instead of bits (via my use of GB/MB/KB vs. gb/mb/kb), Why are you changing the capitalization of the prefix letters there? Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: RAID Sizes (was Re: Why do people in the UK put a u in the word color?)
Mike McCarty wrote: Well, I used to work as a watchmaker, and I can't think of any context where KB stands together as written with K meaning karat. That's not surprising--in SI, the prefix is the scale factor, and the remainder is the unit. I don't think there are any unit symbols that have multiple uppercase letters. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
CUPS/foomatic/gs printing problem - plain text - first line too high,gets cut off
I'm having trouble with printing after switching to using CUPS, foomatic, and gs (gs-esp) in Sarge: When I try to print plain text, the system cuts off the top two-thirds of the first line, and the first several characters on the left. From the position of the partially printed characters, it seems that the text is being aligned with the top and left edges of the paper instead of being aligned with the top and left edges of the printer's printable area. Which configuration file (or files) controls the positioning of plain text? Can I set a system-wide default offset for plain text? Note that the CUPs printer test page seems to work fine. Specifically, the border around the page does not get cut off, and it seems to go right up to the edge of the printable area. Therefore, the system knows exactly how wide the unprintable margins are, so at least part of my system is configured right. Shouldn't the formatting of plain text (into PostScript and from there into pixels/dots) use (whether directly or indirectly) the information about the width of the unprintable margins to position the text within the printable area (at least on the top and left)? My printer is a Canon BubbleJet BJ-200ex; I'm using the Canon-BJ-200-bj200 PPD file (selected via CUPS' web interface). Thanks. Daniel P.S. Which configuration file controls the number of lines of plain text formatted per page? -- Daniel Barclay [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: aptitude --target target sarge xxx doesn't limit to sarge
Paul E Condon wrote: Think OP was trying to use sarge, not stable. Right. I think apt does not recognize sarge or woody or etch etc. Does anyone know why? (Is it a bug in APT tools? Something wrong in my local mirror? Something else?) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: aptitude --target target sarge xxx doesn't limit to sarge
Paul E Condon wrote: On Mon, Mar 06, 2006 at 12:47:45PM -0500, Daniel B. wrote: Paul E Condon wrote: ... I think apt does not recognize sarge or woody or etch etc. ... why? ... ... I think it is a design flaw, more than an implementation bug. Why does it seem to be a design flaw? (What makes it seem to be bigger, or harder to fix, than just an implementation bug?) Speaking from the outside of Debian, it seems that the 'design' of apt grew over time. Whether 'stable' is an alias for 'sarge' or 'sarge' is an alias for 'stable' is a big deal in a 'well designed' system, but this issue seems not to have been resolved as apt was implemented. Now to 'fix' it would involve major rewrite ... What about APT suggests that the rewrite would be major? Thanks, Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: What do you do with a .jar file?
Jon Dowland wrote: On Fri, Mar 03, 2006 at 08:22:44AM -0600, Matt Zagrabelny wrote: there should be hundreds of articles on the internet for how to run an application from a jar file. google: execute jar file Yep: there's hundreds of articles on the _world wide web_ for most topics discussed on d-u. Yeah, but what about Usenet newsgroups? They're not part of the web (archives and gateways aside). :-) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: aptitude --target target sarge xxx doesn't limit to sarge
Andrei Popescu wrote: On Wed, 01 Mar 2006 16:43:54 -0500 Daniel B. [EMAIL PROTECTED] wrote: When I run aptitude --target sarge install ..., aptitude doesn't listen to the instruction to install a version from Sarge; it installs a version from Testing. (I have both Sarge and Testing mirrors in my APT sources list file.) What needs to be done to get --target ... (or APT::Default-Release) to work? ... Hhhm... from man aptitude it seems the long option is --target-release (or -t). Aptitude accepts (sufficiently unambiguous) abbreviations of long option names. The form --target appears to be sufficient. (Note in my follow-up message that --target worked when I used the release name stable. Also, aptitude rejects unrecognized options--try aptitude -targetJunkToMakeUnrecognized.) And you didn't specify what is your default release, if you did set one in /etc/apt/apt.conf If you didn't set it maybe you should... Why would I need to specify a default release when I'm specify a target release on the aptitude command line? (Note that it was because setting the default release to sarge in /etc/apt/apt.conf.d/xxx doesn't work that I was trying aptitude -t.) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
CUPS/foomatic/gs printing problem - plain text - first line too high, gets cut off
I'm having trouble with printing after switching to using CUPS, foomatic, and gs (gs-esp) in Sarge: When I try to print plain text, the system cuts off the top two-thirds of the first line, and the first several characters on the left. From the position of the partially printed characters, it seems that the text is being aligned with the top and left edges of the paper instead of being aligned with the top and left edges of the printer's printable area. Which configuration file (or files) controls the positioning of plain text? Can I set a system-wide default offset for plain text? Note that the CUPs printer test page seems to work fine. Specifically, the border around the page does not get cut off, and it seems to go right up to the edge of the printable area. Therefore, the system knows exactly how wide the unprintable margins are, so at least part of my system is configured right. Shouldn't the formatting of plain text (into PostScript and from there into pixels/dots) use (whether directly or indirectly) the information about the width of the unprintable margins to position the text within the printable area (at least on the top and left)? My printer is a Canon BubbleJet BJ-200ex; I'm using the Canon-BJ-200-bj200 PPD file (selected via CUPS' web interface). Thanks. Daniel P.S. Which configuration file controls the number of lines of plain text formatted per page? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
aptitude --target target sarge xxx doesn't limit to sarge
When I run aptitude --target sarge install ..., aptitude doesn't listen to the instruction to install a version from Sarge; it installs a version from Testing. (I have both Sarge and Testing mirrors in my APT sources list file.) What needs to be done to get --target ... (or APT::Default-Release) to work? This is with Sarge's version of aptitude and friends. Thanks, Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: aptitude --target target sarge xxx doesn't limit to sarge
I wrote: When I run aptitude --target sarge install ..., aptitude doesn't listen to the instruction to install a version from Sarge; it installs a version from Testing. (I have both Sarge and Testing mirrors in my APT sources list file.) What needs to be done to get --target ... (or APT::Default-Release) to work? Hmm, I found that --target stable does work. So why doesn't --target sarge work? What does it take to make it work? Thanks, Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How to reclaim space on hard drive not partitioned fully initially?
Michelle Konzack wrote: Am 2006-02-23 08:49:04, schrieb Mitchell Laks: Long ago, in a galaxy far away, I partitioned my 120GB hard drive. fdisk /dev/hda Why not use cfdisk? Does cfdisk give as much control over partitioning (e.g., which partition numbers to use and where to place partitions regardless of the order in which you create the partitions)? If don't know if was cfdisk, but once I tried the default partitioning program presented by the Debian installer (maybe around the time of Slink), and it really sucked relative to fdisk because I couldn't control where partitions would go (and maybe couldn't even see where they were put). Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
timing(?) problem syncing USB Palm OS device connected via udev (pilot-manager and pilot-link)
What does it take to get the timing right in trying to connect to a Palm OS PDA when using udev? I'm having a problem in which connecting seems to work only if I get the timing just right. Sometimes connecting works when I press the PDA cable's sync button before clicking on PilotManager's sync button (or running a pilot-link command such as memos or addresses), but sometimes it doesn't work. (The PDA eventually times out and PilotManager or the pilot-link command). It seemed to depend on the timing (working if starting the Linux- side software in 6 seconds or less), but then I got inconsistent symptoms (it didn't work at all). At one point, it also seemed to depend on whether I started with the PDA turned on or started with it off (pressing the sync button to both turn it on and initiate a connection), but, of course, later I got inconsistent symptoms. (It didn't work in either case.) (I suspect that at least some of the inconsistent symptoms were from accidentally having hung PilotManager processes still hanging around with /dev/pilot open.) (It never seems to work if I nc button (or run a pilot-link command) _before_ pressing the PDA cable's sync button.) (And, of course, it doesn't work if I click on PilotManager's sync button (etc.) before turning the PDA on, since udev hasn't created the /dev/pilot device file yet.) So do PilotManager and pilot-link programs work with udev? Does udev work with program likes PilotManager and pilot-link? (Does udev not properly accommodate for needs of such programs? Do those programs make bad (or just old) assumptions about devices? Is something else the problem?) My PDA is a PalmOne Tungsten T5. I'm running a Sarge system with udev (Sarge's 0.056-3) and kernel 2.6.8 (Sarge's 2.6.8-16sarge1). Thanks, Daniel -- Daniel B. [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: timing(?) problem syncing USB Palm OS device connected via udev (pilot-manager and pilot-link)
I wrote: What does it take to get the timing right in trying to connect to a Palm OS PDA when using udev? I'm having a problem in which connecting seems to work only if I get the timing just right. Sometimes connecting works when I press the PDA cable's sync button before clicking on PilotManager's sync button (or running a pilot-link command such as memos or addresses), but sometimes it doesn't work. (The PDA eventually times out and PilotManager or the pilot-link command). It seemed to depend on the timing (working if starting the Linux- side software in 6 seconds or less), but then I got inconsistent symptoms (it didn't work at all). At one point, it also seemed to depend on whether I started with the PDA turned on or started with it off (pressing the sync button to both turn it on and initiate a connection), but, of course, later I got inconsistent symptoms. (It didn't work in either case.) (I suspect that at least some of the inconsistent symptoms were from accidentally having hung PilotManager processes still hanging around with /dev/pilot open.) (It never seems to work if I nc button ^^ Oops. ... if I click on PilotManager's sync button ... (or run a pilot-link command) _before_ pressing the PDA cable's sync button.) (And, of course, it doesn't work if I click on PilotManager's sync button (etc.) before turning the PDA on, since udev hasn't created the /dev/pilot device file yet.) So do PilotManager and pilot-link programs work with udev? Does udev work with program likes PilotManager and pilot-link? (Does udev not properly accommodate for needs of such programs? Do those programs make bad (or just old) assumptions about devices? Is something else the problem?) My PDA is a PalmOne Tungsten T5. I'm running a Sarge system with udev (Sarge's 0.056-3) and kernel 2.6.8 (Sarge's 2.6.8-16sarge1). Thanks, Daniel -- Daniel B. [EMAIL PROTECTED] -- Daniel Barclay [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: timing(?) problem syncing USB Palm OS device connected via udev (pilot-manager and pilot-link)
Jean-Marie Thomas wrote: On Sunday 26 February 2006 18:27, Daniel B. wrote: What does it take to get the timing right in trying to connect to a Palm OS PDA when using udev? ... My PDA is a PalmOne Tungsten T5. I'm running a Sarge system with udev (Sarge's 0.056-3) and kernel 2.6.8 (Sarge's 2.6.8-16sarge1). Recent udev requires kernel version at least 2.6.14. What about the version of udev that shipped with Sarge? Does it not work with the kernel shipped with Sarge (or the security update version thereof)? In what sense does recent udev require kernel 2.6.14 or newer? (What are the failure symptoms?) Are you sure udev is running ? Yes, udev is clearly running. (No, I don't know for sure that it's doing exactly what it is supposed to, but it seems to be workding fine.) Daniel -- Daniel Barclay [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [**solved by a reboot**] moving (and losing?) partitions with cfdisk
Levi Waldron wrote: 3. after changing your partition table, you really do have to reboot - at least this is my best guess as to what the problem was. Sometimes you can avoid the need to reboot: If you can unmount every other partition that is on the disk whose partition table you are modifying, then when you write the partition table with fdisk or whatever, the kernel can re-load the partition table and you won't get the message about needing to reboot for the changed partitioning to take effect. Of course, that doesn't work for the disk containing your root filesystem partition (because you won't be able to unmount that partition because it's in use), but it can save rebooting if you're repartitioning a different disk. But here's a question: Why can't the kernel handle some types of changes to a partition table (e.g., a new partition) while some partitions are mounted? It seems that the kernel could compare the current partition table with the previous state of the partition table (using its in-memory data structues derived from the partition table) and determine whether any in-use partitions have changed. If there have only been deletions of partitions that are not mounted and/or additions of partitions, then couldn't the kernel assimilate the changes (deleting data structures for deleted partitions and creating data structures for new partitions)? So why doesn't the kernel do that? Is it just that no one felt it was a useful enough feature to implement? Is it that the kernel's data structures (or code using them) aren't set up for deleting and adding partitions incrementally? Or is there some bigger limitation? Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
where is APT::Default-Release documented?
I'm having trouble getting APT::Default-Release to work. Where is the documentation for it (in Sarge)? The apt.conf(5) manual page sees to be where it should be documented (e.g., in its section The APT Group). However, that manual page doesn't mention it, anywhere. The aptitude(8) manual page mentions APT::Default-Release, but only to define something else in terms of it. /usr/share/aptitude/README has only the same mention as aptitude(8). /usr/share/doc/apt/examples/configure-index.gz contains an example, but only a blank one, and no documentation. So where is APT::Default-Release documented? (What exactly does it do? How can I see the difference it makes without proceeding all the way with installing something? For example, change should I see in aptitude as I'm browsing packages or if I partially select packages for installing (but before I confirm and proceed with installing them)?) Thanks. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: minimizing bandwidth taken be spam delivery attempts even when identifiable by To: address
Paul Johnson wrote: It sounds like you're already at a spot where you can't reasonably reduce bandwidth used by email any further. Yeah, that seems to be true, unfortunately. For example, does your network have a caching HTTP proxy? If not, you're literally flushing bandwidth down the toilet. That statement's a little strong (untrue). It's only true if or when something is making HTTP requests that could be serviced from the cache. (By the way: Ouch! Don't abuse the word literally like that.) Most organizations have far, far more to gain from caching HTTP than from cutting corners on email. Ah--now I see the assumption/perception that led to your statement above. No, I'm not (representing) an organization--my questions weren't about a machine for an organization; they were just for a single-user machine. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: minimizing bandwidth taken be spam delivery attempts even when identifiable by To: address
Glenn English wrote: On Wed, 2006-02-15 at 16:08 -0500, Daniel B. wrote: Clive Menzies wrote: On (15/02/06 12:58), Daniel B. wrote: What anti-spam method minimizes the network bandwith used by spam delivery attempts? sa-exim with spamassassin rejects mail at SMTP time which may solve your problem I was _already_ talking about rejecting mail at SMTP time. There's a rule in my mail server's firewall's input chain that looks for new incoming connections to port 25 and sends them to the spammer and asia chains where the connections are rejected on IP. bogofilter identifies spam and directs it to the spam box. Then a Perl script goes through that mail every couple hours and puts the IPs in the spammer chain for a couple weeks. The script also scans the currently active mail log for rejects due to the Spamhaus RBL. ... Yeah, that's the kind of thing I was looking for. Thanks. If there's a less bandwidth way than at the firewall, I'd like to know about it. My idea (which I don't is practical) was to query RBLs in real time after receiving a TCP connection (SYN?) packet and before sending an acceptance (or rejection) packet. Relative to your suggestion, that would save the bandwidth of receiving the first spam message from the spam host, although it would add the bandwidth cost of querying the RBLs and would depend on the spammer's having already been identified. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
minimizing bandwidth taken be spam delivery attempts even when identifiable by To: address
What anti-spam method minimizes the network bandwith used by spam delivery attempts? How hard is it to refuse incoming TCP connections to the SMTP port based on DNSBL, using exim4? Would refusing connections reduce the overall traffic (maybe even causing spammer machines to think I no longer run an SMTP server?)? Or do spammer machines usually check DNS MX records and would they deliver mail to the backup mail server, which would then try to deliver it, using up the same bandwidth? Is rejecting the message at the RCPT command the best action, or is something else better? NOTE: My main problem here is not spam to valid e-mail addresses-- it is spam to invalid (non-existent) addresses. Although it s easy enough to recognize the spam messagse and refuse them, just the network traffic from the many spam delivery attempts seems to use up a lot of my (modem-based) network bandwith. Also, I've tried configuring DNSBL querying in exim4, but the extra DNS packets seem to use lots more bandwidth. (No, I haven't set up a local DNS server/cache yet.) ADDITIONAL NOTE: This is it not for my main e-mail (which goes to my ISP's SMTP server and stays there until I POP it down). This is for a server I run on my own machine so I can define my own e-mail aliases (which all forward back up to my ISP's SMTP server and my ISP-based e-mail address) so I can avoid giving out my real e-mail address to everyone and can detect who sells/rents/gives my e-mail address (one of the aliases) to others. Thanks. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: minimizing bandwidth taken be spam delivery attempts even when identifiable by To: address
Clive Menzies wrote: On (15/02/06 12:58), Daniel B. wrote: What anti-spam method minimizes the network bandwith used by spam delivery attempts? sa-exim with spamassassin rejects mail at SMTP time which may solve your problem I was _already_ talking about rejecting mail at SMTP time. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: minimizing bandwidth taken be spam delivery attempts even when identifiable by To: address
Paul Johnson wrote: On Wednesday 15 February 2006 09:58, Daniel B. wrote: ... How hard is it to refuse incoming TCP connections to the SMTP port based on DNSBL, using exim4? That is easy, and I run my own DNSBL instead of trying to figure out exim4's ACLs in great depth. Would refusing connections reduce the overall traffic (maybe even causing spammer machines to think I no longer run an SMTP server?)? Yes, it will reduce traffic by rejecting the message before DATA. Uh oh; it sounds like you weren't paying attention to my question. I wasn't talking about rejecting _messages_; I wa talking about rejecting TCP _connections_, somehow injecting blacklist checking between receiving a TCP SYN(?) packet and replying (either accepting the connection or refusing it (generating connection refused))). No, the spammer won't think you no longer run an SMTP server because that assumes spammers care about delivery notifications (a fact not in evidence). Yep, you weren't paying attention: How would delivery notifications be relevant? If I don't even accept the TCP connection, there's no SMTP conversation, so there's no delivery notifiation to talk about, right?. I was talking about what the spammer would do if it got connection refused when trying connect to my SMTP port: Or do spammer machines usually check DNS MX records and would they deliver mail to the backup mail server, which would then try to deliver it, using up the same bandwidth? Well, whatever method you use to filter email should be used on all your inbound MXs, or you're just defeating your own efforts. Okay, thanks for confirming my impression there. Is rejecting the message at the RCPT command the best action, or is something else better? Any time during the SMTP connection is a good time. Not if I'm trying to minimize bandwidth. (Rejecting right after recognizing a bad address is clearly better than receiving a whole message.) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: switching to kernel 2.6 - diald error: No pty found ...
I wrote: John Hasler wrote: Daniel writes: Is Sarge's diald supposed to work with kernel 2.6.8 and udev? Do you really need diald at all? What does it do for you that dial on demand can't? I don't know; maybe nothing. However, I'd like to avoid changing too many things at once. (At the moment I'm trying to try a 2.6 kernel to see if it finally avoids severe filesystem corruption when I try to enable DMA on my AMD IDE controller.) So, does anyone know if there a solution to the No pty in range pty[p-s][0-9a0-f] problem? Okay, never mind the pty question; I did drop diald and use pppd's demand-dialing option. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: switching to kernel 2.6 - manual modprobe lp to dev /dev/lop
Mark Fletcher wrote: Daniel B. wrote: In trying to switch to kernel 2.6(.8) and udev on Sarge, I found that device node /dev/lp0 (for a parallel-port printer I have) doesn't get created unless I manually run modprobe lp. Is the printer (and/or parallel) port supposed to be recognized automatically and is /dev/lp0 supposed to be created automatically? Or is it expected that module lp needs to be loaded explicitly (in /etc/mod...whatever). ... The only explanation I can immediately think of for this is that in your previous kernel, parallel port support was compiled into the kernel whereas in your 2.6.8 build you have built it as a module. No, I don't think it's that. (All my kernels (my previous patched 2.4.18 configured like Debian's 2.4.18, Sarge's 2.4.27 kernel, and Sarge's 2.6.8 kernel) had/have parallel port and lp support built as modules. I think things changed only when I went to 2.6.8 and started using udev) (By the way, note that I didn't say that the parallel-port modules weren't loaded: modules parport and parport-pc are loaded; it's module lp that isn't loaded automatically.) Actually, I think part of the difference might be that before udev, the device node (/dev/lp0) already existed, so when lpd tried to access /dev/lp0, that would trigger the kernel to load module lp. Now, since udev doesn't create the device node /dev/lp0 until the module is loaded, lpd startup can't trigger loading the lp module (and lpd fails because there is not /dev/lp0). My real question is whether udev (or module loading) is supposed to be loading module lp automatically (and therefore something's wrong with it on my machine). *In particular*, I wonder if it's simply that the kernel can't know what's attached and listening to the parallel port (presumably because there's no standard probing protocol for parallel-port devices), so it can't know to load the lp module (which would trigger udev to create /dev/lp0), and therefore manually forcing the lp driver into the kernel (whether by loading the module or reconfiguring and recompiling the kernel) is the expected way to get things to work. I am assuming you built your kernel from source and didn't take a pre-compiled binary one here -- if you didn't of course you won't KNOW what's modular and what's not -- just one of the many excellent reasons for building your own. Actually, it is pretty easy to know what's a module--Debian kernel image packages include the kernel configuration file used to compile it (and there's all the module files in /lib/modules/version/...). In any case, adding lp to the list of modules to load on startup in the file /etc/modules will fix it -- Right. I'm just wondering if that's the right way or if there's a better way to fix it ... but if you know you will always want parallel printer support on every boot you probably should have compiled lp into the kernel -- hardly a disaster but it's worth considering carefully what goes in the kernel and what gets compiled as a module so you avoid confusions like this one. ... well, within the realm of leaving it as a module. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: udev is ruining my life
John W. M. Stevens wrote: Udev was a response to devfs. Sadly, BOTH systems were poorly thought out. ... Udev was the user space devfs, but unfortunately, it was also designed to cover all of dev, instead of just the sub-set of hot attach/detach devices that make sense for a dynamic device file system. Obviously, better interaction with existing kernel infrastructure is necessary before udev can go live. What wasn't thought out well with udev? (I'm asking whether you mean there's a problem in its core design or whether you just mean that the implications weren't all thought out and handled fully before users were exposed to it.) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: help with quoting/expansion in bash
Mike Bird wrote: On Sun, 2006-02-12 at 11:23, David Berg wrote: I'm trying to write a for loop that descends into a list of directories and runs a command. I can't seem to get the quoting right though. Most of the directories have spaces and they are making things difficult for me. Here is what I have: for DIR in dir\ 1 dir\ 2 dir\ 3; do cd $DIR pwd cd .. done The spaces in $DIR aren't escaped. Any help, including a pointer to the relevant part of TFM, is much appreciated. cd $DIR Regarding the FM pointer request: See man bash and the section titled QUOTING. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Are pre-2.6.15 kernels incompatible with udev?
Paul Dwerryhouse wrote: On Sun, Feb 12, 2006 at 10:02:47PM +, David Jarvie wrote: I have updated my etch installation with the latest updates, which include kernel 2.6.15 and udev. I find now that when I boot up an older (customised) 2.6.12 kernel, X fails to start up because it can't find the device /dev/psaux which is used for the mouse. Can somebody tell me how I can run my older kernel with etch and udev? At a guess, udev won't start at all when it sees the older kernel, and That sound unlikely; udev works on 2.6.8, and 2.6.12 is newer than 2.6.8. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: instructions for switching to kernel 2.6 on Debian Sarge?
Digby Tarvin wrote: ... Perhaps I was just lucky and by installing 2.6 from the start the installation process did everything for me... Oh--do you mean to installed Debian from scratch (as opposed to upgrading a previous installation)? (Mine was an upgrade from Woody to Sarge, and then trying kernel 2.6.8.) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: instructions for switching to kernel 2.6 on Debian Sarge?
Andrei Popescu wrote: On Thu, 09 Feb 2006 16:59:23 -0500 Daniel B. [EMAIL PROTECTED] wrote: ... the documentation that I found did warn me that my mouse device (/dev/psaux) would change (and therefore gpm and X might break), but the device the documentation said the mouse would change to (/dev/input/mouse...) doesn't exist, and the documentation didn't say anything about what one needs to do to get it to exist ... udev will take care of that. Just install the kernel ... Actually, the first time I tried it, udev hadn't taken care of it. (I had installed udev, and when I looked for /dev/input/mouse..., it didn't exist.) However, when I tried booting into 2.6 again, /dev/input/mouse... did appear. Hmm... I'm pretty sure I had installed udev _before_ I booted 2.6.8 for the first time. However, it is possible that I booted 2.6.8 and _then_ installed udev, which means that first time I looked for /dev/input/mouse... I hadn't yet booted 2.6.8 normally to give udev a chance to do its boot-time actions. (and don't uninstall the 2.4). If something doesn't work, you can still boot with 2.4 and research. Definitely. (I still have bootable clone of my Woody system from before I upgraded to Sarge.) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
switching to kernel 2.6 - diald error: No pty found ...
In trying to switch to kernel 2.6(.8) and udev on Sarge, I'm encountering a problem with diald. It says: No pty in range pty[p-s][0-9a0-f] and then fails (saying Diald is dying with code 1). Any ideas (what to do to make such device nodes available, or maybe how to reconfigure diald to use whatever those device nodes turned into)? Is Sarge's diald supposed to work with kernel 2.6.8 and udev? Thanks, Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
switching to kernel 2.6 - manual modprobe lp to dev /dev/lop
In trying to switch to kernel 2.6(.8) and udev on Sarge, I found that device node /dev/lp0 (for a parallel-port printer I have) doesn't get created unless I manually run modprobe lp. Is the printer (and/or parallel) port supposed to be recognized automatically and is /dev/lp0 supposed to be created automatically? Or is it expected that module lp needs to be loaded explicitly (in /etc/mod...whatever). Thanks. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
switching to kernel 2.6 - splay Failed to open sound device /dev/dsp
In trying to switch to kernel 2.6(.8) and udev on Sarge, I found that splay/xsplay fails, saying Failed to open sound device. Using strace, I see that opening /dev/dsp is failing: open(/dev/dsp ... ) = -1 EBUSY (Device or resource busy). I'm not knowingly running any sound daemons, so why would the device be busy? Does this sound like a change because of udev or just because of 2.6? (Is /dev/dsp a device from the old sound drivers? If so, why is it device busy instead of something like no such device?) Thanks. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: switching to kernel 2.6 - diald error: No pty found ...
John Hasler wrote: Daniel writes: Is Sarge's diald supposed to work with kernel 2.6.8 and udev? Do you really need diald at all? What does it do for you that dial on demand can't? I don't know; maybe nothing. However, I'd like to avoid changing too many things at once. (At the moment I'm trying to try a 2.6 kernel to see if it finally avoids severe filesystem corruption when I try to enable DMA on my AMD IDE controller.) So, does anyone know if there a solution to the No pty in range pty[p-s][0-9a0-f] problem? Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: instructions for switching to kernel 2.6 on Debian Sarge?
Digby Tarvin wrote: I didn't think there was much more to it than just doing a apt-get intall kernel-image-2.6.8 Actually, there is (or sure seems to be). As I mentioned, the documentation that I found did warn me that my mouse device (/dev/psaux) would change (and therefore gpm and X might break), but the device the documentation said the mouse would change to (/dev/input/mouse...) doesn't exist, and the documentation didn't say anything about what one needs to do to get it to exist (install packages? run a new equivalent of MAKEDEV? mount /sys or similar?). Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
instructions for switching to kernel 2.6 on Debian Sarge?
Where are instructions on how to upgrade to kernel 2.6 (ideally, specifically in Debian's Sarge release)? The instructions I've found so far are quite sketchy. (For example, although I have found instructions that tell me that my mouse is no longer at /dev/psaux but is now at /dev/input/mouse (or .../mouse0 or .../mice), they don't say what all I need to do to get /dev/input or /dev/input/mouse to appear in the first place.) Are there any Debian-specific instructions out there (e.g., specifying which packages (beside udev and its prerequisites) I need to install and what other changes I might need to make (e.g., mounting new special file systems (/sys? /dev?), dropping or changing old special filesystems or real directoryes (/dev?), etc))? If not, are they any non-Debian-specific instructions that explain what all is needed? (This is for a fairly basic desktop system--non-USB keyboard and mouse, USB only for camera and other lower-priority devices.) Thanks, Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: why do PPP connections die with heavy disk usage?
Miquel van Smoorenburg wrote: On Thu, 2005-07-21 at 14:04 -0400, Daniel B. wrote: ... ... the reason I'm using PIO mode in the first place is because I get massive file system corruption when I use DMA mode with IDE controllers on my motherboard (Asus A7M266-D; AMD 762(?) chipset). ... I don't know, I try to avoid broken hardware :) Is the AMD 762(?) chipset broken? Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Missing public keys in aptitude
Andreas Janssen wrote: ...If you use Sarge (without backported apt from somewhere else!), this problem shouldn'd occur, because not only apt-key, but the whole GPG stuff is not implemented in your apt. debmirror seems to be a different issue, however. It looks like debmirror from Sarge /does/ check GPG signatures, and you probably can solve the problem by importing the matching keys using gpg. Oh, okay. Can you point me to how to import keys into gpg (e.g., which commands man page/info page/etc says how to import and maybe how debmirror might be querying gpg for the key)? Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: debmirror can't find public key to validate Release files
Hans Ekbrand wrote: On Wed, Feb 01, 2006 at 04:50:47PM -0500, Daniel B. wrote: Andrew M.A. Cater wrote: On Wed, Feb 01, 2006 at 11:10:56AM -0500, Daniel B. wrote: ... I can't tell if I deleted a key I had before (in purging and/or re-installing some things I shouldn't have) or if upgrading debmirror got me a version that now checks against a key I never had in the first place. What can I do to get debmirror working again? ... apt-get install debian-archive-keyring ?? Nope, that doesnt work. (There's no such package in Sarge or the Sarge security updates (unless my mirror is broken).) What else? get debmirror from sarge manually (not apt-get). install with dpkg. Huh? I already have debmirror from sarge (debmirror version 200502027). That's the version of debmirror that is complaining about the lack of a public key. Did you mean to download something else from Sarge? Or did you mean to download debian-archive-keyring from something other than Sarge? Or something else? (Is debian-keyring (in sarge) the old name of the debian-archive-keyring that was referred to about, or is it something else?) Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Missing public keys in aptitude
Florian Kulzer wrote: David Kirchner wrote: On 2/2/06, Andreas Janssen [EMAIL PROTECTED] wrote: man apt-secure, man apt-key Neither are found on my Sarge install, and I don't see them in aptitude. Install the 2006 archive signing key. This has been explained plenty of times on the list, search the archive if you need a longer explanation. I've been watching the list, reading most of these threads. I've seen plenty of people having trouble with it but few solutions. Can you send a link to the archived post that includes a solution? The original message by Joey Hess: http://lists.debian.org/debian-user/2006/01/msg00291.html That is not a link that includes a solution--that message says to use the command apt-key, which, as David pointed out, does not exist in Sarge. Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
debmirror can't find public key to validate Release files
Where does Debian (or debmirror?) store the public key that debmirror uses to validate Release files? In upgrading to Sarge, I purged (or upgraded) a few too many things, and now when I try to run debmirror to keep my local mirror updated, it says: ... [0%] Keeping: dists/sarge/Release.gpg gpg: Signature made Sat Dec 17 05:46:27 2005 EST using DSA key ID 4F368D5D gpg: Can't check signature: public key not found Release signature does not verify. ... I can't tell if I deleted a key I had before (in purging and/or re-installing some things I shouldn't have) or if upgrading debmirror got me a version that now checks against a key I never had in the first place. What can I do to get debmirror working again? Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: debmirror can't find public key to validate Release files
Andrew M.A. Cater wrote: On Wed, Feb 01, 2006 at 11:10:56AM -0500, Daniel B. wrote: ...when I try to run debmirror to keep my local mirror updated, it says: ... [0%] Keeping: dists/sarge/Release.gpg gpg: Signature made Sat Dec 17 05:46:27 2005 EST using DSA key ID 4F368D5D gpg: Can't check signature: public key not found Release signature does not verify. ... I can't tell if I deleted a key I had before (in purging and/or re-installing some things I shouldn't have) or if upgrading debmirror got me a version that now checks against a key I never had in the first place. What can I do to get debmirror working again? apt-get install debian-archive-keyring ?? Nope, that doesnt work. (There's no such package in Sarge or the Sarge security updates (unless my mirror is broken).) What else? Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
