Re: 6rd vs. interfaces(5)
Ivan Shmakov i...@gray.siamics.net writes: Rick Thomas rbtho...@pobox.com writes: On Jul 21, 2011, at 3:29 AM, Ivan Shmakov wrote: […] And in the case of NAT'ed IPv4, it's still possible to register for a free-of-charge tunnel service at http://sixxs.net/ and use AICCU (# apt-get install aiccu.) This (SIXXS) is what I use at home. It works a treat for me. Easy to set up. Easy to use. Fully connected to the IPv6 internet. One of the ISP's here has finally started to offer IPv6, namely: 6to4 (AIUI, they run their own gateway for that) and 6rd. Unfortunately, 6rd is only available for Linux 2.6.33 and later (as per Wikipedia), which isn't in Squeeze. It is actually pretty easy to compile and use your own kernel. Perhaps surprisingly, the kernel interface is pretty stable and you do not need to upgrade the rest of your world (unlike, say upgrading an important shared library). But anyway, is there a way to add a 6rd tunnel to interfaces(5)? TIA. -- FSF associate member #7257 -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/8739hxm8gk@emmy.axel.nom
Re: IPv6 and DNS
Rick Thomas rbtho...@pobox.com writes: On Jul 19, 2011, at 6:13 PM, John Hasler wrote: Go Linux writes: How is [IPv6] going to work on DIALUP! I wrote: Just fine. What makes you think it wouldn't? Johan Kullstam writes: The fact that it doesn't work anywhere else? :- Works fine here. Here too. What kind of problems are you having, Johan? My ISP does not offer IPv6. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87ei1ksgch@emmy.axel.nom
Re: IPv6 and DNS
John Hasler jhas...@debian.org writes: Go Linux writes: How is [IPv6] going to work on DIALUP! Just fine. What makes you think it wouldn't? The fact that it doesn't work anywhere else? :- -- John Hasler -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87zkkandh0@emmy.axel.nom
Re: Does IPv6 preclude use of a NAT gateway?
Camaleón noela...@gmail.com writes: It's probably not the best thing, but I depend on the NAT gateway for a lot of my security--with IPv6, will I still be able to do that? NAT and security do not match. You better put a good firewall and/or IPS system in between ;-) With IPv6 it is not an issue. It's not like IPv6 is actually routable across the internet. :- Hmm, I need to knock on some wood--wait, I can knock on my head--that is about the same ;-) So far, NAT seemingly has provided pretty good security for me. :-) Yes, most of us -wrongly- believe that our NATed router is like a wall between our computers and the dangerous external web (because indeed it is hidding somehow) but this is not a security measure per se but security by obfuscation: that we can't see it does not mean we can't reach it. There can be still holes in router's firmware or bad configured DSL devices that may expose the user regardless NATed or not ;-) OK, you put a new XP box directly on the internet and I'll put one behind a NAT router box. Do you wish to take bets on who can update before getting owned? NAT is a firewall. Maybe not a great one. But it does function as such. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87y5zzcah9@emmy.axel.nom
Re: Downgrading Udev
Andrej Kacian andrej+deb...@kacian.sk writes: On Fri, 8 Apr 2011 11:44:17 +0100 Anthony Campbell a...@acampbell.org.uk wrote: On 08 Apr 2011, Anthony Campbell wrote: After an upgrade today to base-files 6.3 I rashly assume things would be fixed, but not so. My /run is no longer there but now the 2.6.38 kernels won't boot - they stop at the nouveau driver. Luckily I still had a 2.6.37 kernel which isn't affected. Anthony Apologies for following up to myself, but I thought I should say that when I added /run myself the 2.6.38 kernel booted without problems. So... is this udev167-1 safe to upgrade to on Wheezy (base-files 6.1), or should I wait? You can safely upgrade both base-files and udev. Just make sure you delete the /run directory before any reboot. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/874o65pmrr@emmy.axel.nom
Re: Monitor question
Bob Proulx b...@proulx.com writes: Johan Kullstam wrote: I had an old thinkpad t42 with a 14 1440x1050 and it rocked. It weighed only 4.5 lbs even with cd drive. For me, it was an optimal size and weight. The current offerings are all inferior - they are heavier, have less vertical screen dimension and worse resolution. And that is exactly why I am still using my IBM ThinkPad T42 with exactly that configuration. It does everything I need a laptop to do. It has the best keyboard of any laptop I have ever used. But mostly because all of the newer machines are less suitable. It is hard to upgrade to something that isn't as good. I would still be using mine but lightning wiped it out. I got a strike near the house and it came through the cablemodem and ethernet. I have a t500 now. It has a much faster CPU, a decent screen, but it is bigger and heavier. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87aajnxkub@emmy.axel.nom
Re: Monitor question
Gilbert Sullivan whirly...@comcast.net writes: On 12/28/2010 09:40 AM, Klistvud wrote: It's also horrible for web browsing, and for many other tasks. It actually only has two uses I can think of: widescreen movies and side-by-side document viewing. Given that movies are best viewed on large TV sets anyway, the usefulness of widescreen computer monitors is further reduced to just side-by-side document viewing. Arguably, even for that task, dual-head setups are better. On the other hand, there are those of us who must use portable systems for side-by-side document reading and/or tiled terminal window use while traveling and are, thus, limited to a single screen. Widescreen works better for us. Good for you. My gripe is that one can no longer choose. It's shortscreen or nothing. I had an old thinkpad t42 with a 14 1440x1050 and it rocked. It weighed only 4.5 lbs even with cd drive. For me, it was an optimal size and weight. The current offerings are all inferior - they are heavier, have less vertical screen dimension and worse resolution. My portable systems have 1920x1200 LCDs. I'm so accustomed to them that I don't bother with multi-monitor setups at home or at the office any more. Just one widescreen setup suffices, and I don't have to fiddle around switching between multi-monitor and single monitor setups any more. My totally unbiased and scientific $.02. ;-) -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87ei8zygqo@emmy.axel.nom
Re: linux-image for i586
Arnt Karlsen a...@c2i.net writes: On Fri, 12 Nov 2010 16:57:09 +0700, Sthu wrote in message 4cdd0f80.ce7c0e0a.2046.a...@mx.google.com: Thank You for Your time and answer, Arnt: What are the hardware specs of the machine you have? Exactly which CPU, how much RAM, what hard disk (size and rpm), etc. Please provide the following output: cat /proc/cpuinfo lspci free -m dmidecode (If your machine is very old, dmidecode may not work). ..'top|head -n 20 ' to see what's eating your iron. Could You please tell me what is for? What do we want to conclude from the given info - that will be a lot - speaking about the last utility? On Wed, 10 Nov 2010 18:41:26 -0500, Johan wrote in message 87aalgiw61@emmy.axel.nom: The i486 kernel will run on an i586 box. Even if you compile the kernel for i586 how much speed are you really going to get? A few percent? And how much time do you spend in kernel as opposed to userland? Maybe 3-20 percent depending on what you are doing right? So, in rough numbers, that's a 10% off on 10% of what you do for a net of roughly 1%. ..it depends, 'top |head ' tells us another wee story: ;o) Correct me if I am mistaken. Looking at your numbers, it seems that most of the time is spent in userland and not the kernel. I take it that the us field represents time spent in userland and sy is kernel, but you probably can also add in wa wait and hi, si to handle interrupts. st is, according to man top, stolen by other like hypervisor. In the first case, you are 96% in user and only 4% in the rest. In the second, top reports 60.7% user and 38.6% idle, and the rest sums to under 1%. So, making the kernel faster would really only help you by upto 4%. Ideally, the kernel should be zero (although that is clearly unrealistic). Some tasks like running a serial port or a heavily used network filter can cause more kernel activity. Usually, I find that the kernel is mostly out of the way. You could gain significantly by speeding up your useland process as that is where, quite rightly, most of you CPU is going. top - 21:36:01 up 11 days, 20:14, 1 user, load average: 1.08, 1.02, 1.01 Tasks: 57 total, 1 running, 56 sleeping, 0 stopped, 0 zombie Cpu(s): 95.9%us, 0.8%sy, 0.0%ni, 1.5%id, 0.2%wa, 1.3%hi, 0.3%si, 0.0%st Mem: 155764k total, 149516k used, 6248k free, 8400k buffers Swap: 0k total, 0k used, 0k free, 111296k cached PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND 12009 root 20 0 51868 15m 1612 S 89.1 10.0 952:08.72 motion 12691 arnt 20 0 2320 1004 780 R 5.0 0.6 0:00.08 top 1 root 20 0 2024 488 396 S 0.0 0.3 1:09.26 init a...@bacdor:~$ ..the cool breezy competition: top - 21:49:26 up 11 days, 20:13, 1 user, load average: 0.36, 0.38, 0.36 Tasks: 60 total, 1 running, 59 sleeping, 0 stopped, 0 zombie Cpu(s): 60.7%us, 0.2%sy, 0.0%ni, 38.6%id, 0.1%wa, 0.4%hi, 0.1%si, 0.0%st Mem: 92336k total,76504k used,15832k free, 4772k buffers Swap: 265032k total, 2696k used, 262336k free,47080k cached PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND 22145 root 20 0 49820 13m 1692 S 42.1 14.7 697:25.47 motion 22966 root 20 0 2324 1036 808 R 1.9 1.1 0:00.04 top 1 root 20 0 2028 236 208 S 0.0 0.3 0:35.17 init cm46:~# -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87r5eoajq9@emmy.axel.nom
Re: linux-image for i586
Sthu Deus sthu.d...@gmail.com writes: Good day. Just wanted to ask, Why there is no linux-image package for i586, especially w/ MMX support - but only for i486, then for i686 and up? I don't think all that many people have i586 boxen. They were current in the timeframe 1995 through 1997. That's over a dozen years ago. The i486 kernel will run on an i586 box. Even if you compile the kernel for i586 how much speed are you really going to get? A few percent? And how much time do you spend in kernel as opposed to userland? Maybe 3-20 percent depending on what you are doing right? So, in rough numbers, that's a 10% off on 10% of what you do for a net of roughly 1%. If you actually care about speed, you might upgrade to a new cpu. People should be giving away pentium-iii and early athlon boxes for free. (Depending on where you live c.) Thank You for Your time. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87aalgiw61@emmy.axel.nom
Re: Give root password for maintence during every boot of system
Anthony Campbell a...@acampbell.org.uk writes: On 23 May 2010, Merciadri Luca wrote: Hi Mitchell, You might add == |su:S:wait:/sbin/sulogin| == in /etc/inittab. But why was the OP getting the message in the first place? A couple of weeks ago I was also getting it and then it went away as mysteriously as it came. I don't know why you suddenly started getting it, but it is what you have when booting to single-user-mode. The grub menu calls it recovery mode and you can always get to it with telinit 1. In my grub menu list of kernels, I see the recovery mode coming right after the normal mode for each kernel. I could see that if the line with normal mode was disabled, missing or broken, grub might just skip down to recovery mode which gives you the behavior. -- Anthony Campbell - a...@acampbell.org.uk Microsoft-free zone - Using Debian GNU/Linux http://www.acampbell.org.uk - sample my ebooks at http://www.smashwords.com/profile/view/acampbell -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87632eu5c5@emmy.axel.nom
Re: OT - Backup of HA server on external drive
Ivan Marin ispma...@gmail.com writes: Hi all, I have to make a backup plan for a server that is physically very far away from me right now. If for some reason this server goes south, I have to have a plan and it has to be done quickly. The problem is that the personnel that is on site doesn't know squat about Linux (or computers, for that matter), so it must be something dead simple. I was thinking of getting a spare hard drive, connect it to the working server, do a dd of the entire disk to the new disk, and disconnect the disk (the people there can swap hard disks). Something like dd if=/dev/sda of=/dev/sdb bs=1024 dd is a nice tool, but it is a bit low level for this type of task. Linux isn't windows where the latter is picky about being moved around. Why not format and mount the new disk and use cp -ax to shift the data? That way the filesystem can lay everything out nicely and your new disk could be a different size. assuming that sda is the working disk and sdb is the new, unformatted and unpartitioned disk. So if hte machine breaks, I can get the new disk and put in a new machine and everything should work. This server is doing firewall and openvpn, etc, no X, no fancy stuff. Is this going to work? What do you guys think? Cheers! Ivan -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87d3xp81uf@emmy.axel.nom
Re: Intel Core i5 integrated graphics
Pasi Oja-Nisula p...@iki.fi writes: I just retired my 9 year old machine to home use and got a Gigabyte H55-UD3H motherboard with Intel Core i5 processor. Well, the first impressions are that maybe I should have done a bit more research since there are Realtek net and audio chips that might have some issues. Pretty much all motherboards come with realtek ethernet. It's well night impossible to avoid. I have a gigabyte board too. The sound is fine even though its a realtek. Anyway, I have a 3com card for net That sounds like a good plan. and don't necessarily need audio at all. The biggest concern now is with graphics in Xorg. I'm using Debian Lenny and obviously the integrated graphics are not supported there yet. But this is strictly a business machine (browser and a bunch of xterms mainly), so I think I would manage a while with even vesa driver. And I even got the vesa driver to work with 1280x1024 which fits my monitor, so this would be a fine solution for short term. Unfortunately the screen looks like this: 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 So the right side of the picture is shown at the left. I just don't know how I could adjust this. I'm using a vga cable btw, maybe dvi cable would help? Anyone have any experience with the graphics in new Intel processors? Should I just leave this alone and get a low end ATI card (suggestions for something that would work out of the box)? This would require a bit of research since the low end has moved up and the xorg drivers typically lag behind. My x1950 works fine, but it's now too ancient to find and you can probably do much better now. What is the sweet spot for cheap ATI card these days? -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/8739zlj1gi@emmy.axel.nom
Re: Intel Core i5 integrated graphics
Stan Hoeppner s...@hardwarefreak.com writes: Johan Kullstam put forth on 3/27/2010 9:14 AM: Pretty much all motherboards come with realtek ethernet. It's well night impossible to avoid. That statement is absolutely false. There are plenty of mobos on the market with Broadcom, Intel, Marvell, nVidia, Via, and other ethernet on board. If you'd have said Pretty much all _cheap_ motherboards come with Realtek ethernet I'd be more inclined to agree with you. However, I've seen a $55 ($40 after rebate) ECS board on Newegg with Broadcom ethernet on board, so not even all cheap boards use Realtek ethernet chips, although _most_ do. http://www.newegg.com/Product/Product.aspx?Item=N82E16813135246 Hitting $75 USD and climbing up from there you'll see many more boards with non Realtek ethernet. Thanks. I just wish that newegg would put ethernet with realtek as a powersearch field so I could more easily reject the crabcard. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87y6hdhb05@emmy.axel.nom
Re: Correct binary for Intel Core i5
Stan Hoeppner s...@hardwarefreak.com writes: Peter Tenenbaum put forth on 2/21/2010 10:01 PM: Stan -- It sounds like, if the Realtek drivers are not present on the Debian distribution, I have at least two options: going to the Realtek site and downloading their linux 64 bit drivers, or compiling my own kernel from source on kernel.org. Does that sound about right? As far as video cards are concerned, I have a (probably) ignorant question: how do I put the integrated northbridge video support to use on these motherboards, since they do not appear to have any video output spigots on them? Given the length of one of my recent posts, you may have missed this at the bottom. To save yourself a lot of potential headache getting Debian installed and running, and keeping it running, you should really get this Intel motherboard, and not the Gibabyte or any others, specifically due to the onboard Intel Pro 1000 GigE chip. It'll give superior performance compared to either the Realtek or Marvell, and you'll never have a compatibility issue with it WRT Linux/Debian drivers. The Realtek 8111/8168/8169 has had such recent problems. http://www.newegg.com/Product/Product.aspx?Item=N82E16813121394 I've never heard of a Linux problem relating to an Intel ethernet chip. They're just rock solid. I use nothing else in my Debian servers due to problems with DEC 211x0 chips and some 3COM chips in the past. 3COM used to be the gold standard, but as I said, I had Linux problems with some 3C59x and 3C90x cards and switched to Intel. I've never looked back. Lightning zapped my 3c905b. The cumputer had a surge protector, but the cable side of the cable modem didn't. It wiped out my whole wired network. DEC tulips were great but the drivers have suffered bit rot since their heyday in the 90s. After having trouble with the Realtek motherboard built-in ethernet, I just sprung for a PCI Intel ethernet. They are only like $30. I hate Realtek ethernet, but it is very hard to find a motherboard with anything else on it. Thanks for the information. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/873a0snkv7@emmy.axel.nom
Re: Correct binary for Intel Core i5
Mark Allums m...@allums.com writes: On 2/20/2010 5:18 PM, Stan Hoeppner wrote: Peter Tenenbaum put forth on 2/20/2010 4:51 PM: Hello there -- I am planning to assemble a new computer for my home, and to run debian linux on it. I'm planning to use an Intel Core i5-660 CPU in this computer. From the documentation, it looks like the correct binary to use is the AMD64. Is this right? Does anyone know of any problems running debian AMD64 on Intel core-i5 processors? This is the proper binary for the core i5, but that's the least of your worries. Your real concern should focus on whether Debian/Linux has all the drivers for the various hardware devices on the motherboard and in the expansion slots, including, but not limited to: 1. GPU 2. Network interface 3. SATA controller 4. Sound chip If the MB uses the Intel ICH10 Southbridge, the SATA, at least, will work. If the Northbridge is Intel, as it virtually *must* be with the Core i5, the GPU will be okay. If the GPU is external, then buying an AMD/ATI or NVIDIA card will be safe. Most boards have Marvell NIC chips, that's covered. Really? All I seem to be able to find is Realtek. It is easy to get an actual card. And sound will almost certainly be Intel Azalia HD; that's covered. I think you'll be fine. AMD64 is the designation for the 64-bit chip architecture, not the actual chip manufacturer. AMD devised it, and Intel did the sensible thing and copied it when they expanded their x86 architecture to 64-bit; it was a very good design. The 32 bit i386 is also viable. There are a few things which are not yet available for amd64. These are all non-free, but some people want acrobat reader. And hulu won't work with the 64 bit flash. Stay away from the IA64 port. That's for the Itanium. I did that. It took me like an hour to figure out why my machine wouldn't boot off the disk. D'OH! Mark Allums -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87ocjgst1h@emmy.axel.nom
Re: Downgrade libxi6 -- HowTo? (Bug 515734)
Klistvud quotati...@aliceadsl.fr writes: Greetings, fellow Debianites! I've recently installed Debian 5.0.3 on my other machine and am experiencing bug http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=515734 (keymaps messed up in X). Now, the suggestion is to downgrade the libxi6 library. I have several noob questions about that: You will want to know the bad version for step 2. Use aptitude show package to print information. Copy the version string, call it version. 1. How do you go about downgrading a package in Debian -- especially if it's a fresh system that's never been upgraded in the first place? With ftp over to your debian repository and download a past version. I just did this on my sid boxen with xserver-xorg-core and xserver-common. I just fetched the older testing versions. There is also debian-snapshot which will give older versions. Use dpkg -i packagedeb to install it. The dpkg command will warn you about the downgrade. There's probably some fancy way to do the above with aptitude. 2. Won't the Debian maintainers fix the issue in a future official update anyway? Wouldn't downgrading the package on my own accord potentially interfere with such official update? This is the trick: forbid the offending version. Do aptitude forbid-version package=version This says you don't want *that* version, but that future updates are acceptable. 3. How come there still hasn't been such an update? The bug is marked as grave and it probably wouldn't be hard for the maintainer(s) to force the reccommended downgrade of the libxi6 library in a regular update? Sometimes bug fixes take longer than you'd like. It's a vollonteer effort. The maintainer has to be available, find a solution and test it, wait for build and then debian seems to release new filesets twice daily. That said, when a clearly broken version gets released, detected and acknowledged, there could perhaps be some automated mechanism to revert to previous. Forgive my noobness, but I still haven't quite grasped the Debian procedures and mechanisms. But I'm doing a lot of reading, so I hope I'll catch up in time. I already know how to open a terminal and type ls. TIA -- Certifiable Loonix User 481801 -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: SOLVED Re: Sid: Grub2 faling to boot: Unknown device UUID
Curt Howland howl...@priss.com writes: Just to continue to with the saga, today I decided to just screw it and reintall, to try to install Grub2 first since upgrading from Grub to Grub2 just would not boot. Yeah, I just ran afoul of the grub-pc no floppy bug. After a fight with the rescue disk, I managed to install the newer version to the MBR. I mounted /dev/sda1 as my root, did mount -a and used /usr/sbin/grub-install '(hd0)'. Now it pauses and takes an age to show the menu. Then it takes another age to start booting the kernel. I installed Grub2, with only /dev/hda plugged in so that /dev/hdb would not be formatted since the Debian installer refuses to NOT repartition disks. Fine. With just /dev/hda plugged in, Grub2 booted just fine. However, as soon as I plugged /dev/hdb in and tried to boot, Grub choked with error 2. Whatever that is. I used the rescue function, mounted /dev/hdb1, put it in the fstab, had everything working and executed update-grub. But it still would not boot, error 2. I tried twice more, Grub and Grub2, no luck. As soon as I plugged the second IDE drive in, Grub would refuse to boot. LILO works. So my server is back to using LILO, because GRUB just won't work. I'm sure it's just my server, being old and crotchity, but I'm lost as to why Grub worked last week, before Grub2 installed the UUIDs and messed things up, but Grub doesn't work this week. I hate it when that happens. Welcome back, LILO. Long time no see. Not only does lilo work, but it has documentation and howtos. Searching the web for help with grub is an exercise in frustration. What was grub supposed to bring us anyway? -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: OT: launching jobs in a combined serial parallel way
Kamaraju S Kusumanchi raju.mailingli...@gmail.com writes: I have three programs - say proga, progb, progc. proga, progb are completely independent. They take couple of hours to finish. The time to complete proga, progb are not same. progc should to be launched only after both proga, progb are finished. progc takes another couple of hours to finish. What is good way to automate this problem (that is no manual interaction)? I prefer to use nohup since sometimes I have to log out of the machine before the whole process finishes. Currently I have a shell script that works as below. 1) launch proga, progb in the background using nohup. 2) Ask proga, progb to write a file when they finish. 3) Every five minutes check if these files are present. If they are present, launch progc. Use make. I like to use make for all sorts of things, not just compiles. And, since you are already using the existance of a file to check for done, you are practically there. Make with the -j parameter will spawn parallel jobs for you. Write a makefile similar this: proga.out: proga progb.out: progb probc.out: proga.out progb.out progc Then you can do $ nohup make -j 2 Maybe a redirect of the text output to a file. Now you can walk away. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: To synchronize system time witn NTP-server with no winter time shift whole year - how to?
Chris Jones cjns1...@gmail.com writes: On Mon, Mar 30, 2009 at 01:09:10PM EDT, Paul E Condon wrote: But exact spot? That would imply different clock settings in different rooms of one's home. Not for me. Fancy that.. under our latitudes, when your house is a few hundred yards wide.. never mind.. I've overstayed my welcome on this here list. What does Santa Claus do? His house has all the lattitudes. What do the people at the Amundsen-Scott do? For extra credit, consider that daylight and night are each six months long. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: realtek RTL8168C/8111C performance issues
Mikael Rudberg mik...@iced.se writes: Hi I've been using Debian for approx 5 years. I'm in the process of building me a new NAS machine. To that end i have bought a small NAS chassis and a DG945GCLF2 wich has a atom 330 and realtek RTL8168C/8111 NIC. I have everything hooked up to a Dink DGS-1008D gigbit switch [cut litany of realtek woes. also complaint that external card won't fit in the case.] My Gigabyte board (ga-p35-ds3r) has a realtek on it. It never worked right. It would randomly not come back up after a reboot. I dug up an old 3com 3c905b which works just fine for my home network. Unfortunately, nearly all motherboards seem to have this crap realtek chip. I guess this doesn't actually help you. I just wanted to let you know that you are not alone in suffering realtek. Does anyone have these things working reliably? -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: Ext3 for flash drive
Masatran / Deepak, R. masat...@freeshell.org writes: Recently, I re-partitioned my flash drive. I made one FAT32 partition, and one Ext3 partition. The problem is that when I transfer files from my laptop to my work computer, the UIDs on the Ext3 partition are used for the permissions, so I am not able to access the data. How can I fix this? Both computers run Debian Lenny. The laptop runs Sawfish while the work computer runs Gnome. I manually mount the flash drive in Sawfish, and I have a FSTAB entry to allow this without Sudo. Gnome does an automatic mount. I don't have superuser privilege on the work computer. I am willing to use non-Ext3 filesystems, I just want RWX-RWX-RWX-style file permissions. I use tar. Leave the format in FAT. Make tar file (with compression usually). Copy tar file to USB drive. The tar file will contain permissions. You might even be able to use tar straight to the flash drive like tar cvf /dev/sda files... I used floppies like that. But that gives you only one archive per USB. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: PCI ethernet card from ebay doesn't work at all. Possibly fake, but which?
A. F. Cano [EMAIL PROTECTED] writes: Hi, I needed a PCI ethernet card for an old P3 system and so I did some research to make sure it was supported and got the first one that showed up on an ebay search. At least it was cheap. The sellers all claim that they were RealTek (RTL8139D) and they are detected as such. This is what lshw says: Argh. RealTek. That might be your problem. My mobo has a realtek on it and it never works right. I went and found a 3com 905c. I just checked google shopping and you can still find them for about $20. It's a shame that all the mobos bundle such cheap crap. Unfortunately, the bundling has killed the market for add-on NICs. So when the mobo ethernet is too sorry work, you are left with nothing. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Width of line
Andrius [EMAIL PROTECTED] writes: Ron Johnson wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 03/08/08 16:46, Andrius wrote: Andrei Popescu wrote: On Sat, Mar 08, 2008 at 10:31:08PM +, Andrius wrote: Lads, would you be able to explain one simply thing. About width of line. How many characters should be in the row of typical UNIX (or txt) text file? Were made a signature in ASCII in vim, and when it was attached to Thunderbird, some rows was broken and ASCII art together. Were suspect that it has happened because of too many characters in a row, and when some characters were removed, everything starts to look OK. How many chars should be in text file please? Regards, Andrius Usually less then 80, but =72 would be a safe bet. BTW, signatures should generally not exceed 5 rows (netiquette). Regards, Andrei 78 was too big. After 72 looks OK. Is 5 rows really netiquette? Yes. It's not as important now, though, as when modems were *slow*, monitors small, and disk space expensive. Just don't go crazy. - -- Ron Johnson, Jr. Jefferson LA USA The kniiife..., the kniiife... The life of the wife is ended by the kni-ife. Stewie Griffin Eliza Pinchley -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH0x/cS9HxQb37XmcRAntdAKDoOG7W0YeXPiwNAkPhxtZam5sTiACg2dI/ iWFkK3Acvv17MDuAyTgUyD4= =bXtH -END PGP SIGNATURE- Look at that. It is a signature for one politics forum. Looks cool, is it? Regards, Andrius -- __ __ ___ |/ /_ _ __ /|_/ /_ __ `/_ __ \ __ \ _ / / / / /_/ /_ / / / /_/ / /_/ /_/ \__,_/ /_/ /_/\/ ____ ___ /___ ___ / __ __ \ __ `/_ /__ ___/ __ `/_ ___/ _ /_/ / /_/ /_ / _(__ )/ /_/ /_(__ ) /_.___/\__,_/ /_/ // \__,_/ // _ __ ___/ ___ __ \_ ___ /__ _ _ / / /__ / __ /_/ / __ `/_ //_/_ ___/ __ `/ / /_/ /__ /_ _ // /_/ /_ , _(__ )/ /_/ / \__,_/ _/ /_/ \__,_/ /_/|_| // \__,_/_ /_/ /\ _| _ _ _ | _ _ __ _ | _ . _ _ _ | . _ _|_ _ . /~~\(_|(_|| | |||_|_\ \/| (_| | |(/_|(_||| | |(/_ |_|(/_ ||_|\/(_|| / .-. .-. .--' / \ '--. '--. \ ___ / .--' \ \ .- -. / / \ \ / MADE IN USA \ / / \ / for Lithuania \ / \| .--. .--. |/ | )/ : | | : \( | |/ \__/ \__/ \| / /^\ \ \__'='__/ |\PRESIDENT/| |\'Adamkus'/| \ `` / `-._.-' / / \ \ / / \ \ / / \ \ ,-' ( ) `-, `-'._) (_.'-` -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] Does anyone else remember alt.fan.warlord? -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Changing Ethernet Drivers [Solved]
Travis Crook [EMAIL PROTECTED] writes: On Tue, 22 Jan 2008 14:53:47 -0700 Travis Crook [EMAIL PROTECTED] wrote: On Tue, 22 Jan 2008 22:38:11 +0100 Jörg-Volker Peetz [EMAIL PROTECTED] wrote: Travis Crook wrote: [...] First one in the list, after running modprobe forcedeth. The question is: how do I tell the nic (which is a RealTek 8111 Gigabit on-board adapter) to use the forcedeth driver instead of the r8169, which is automatically loaded on boot? Thanks! Try to put the module name r8169 into the file /etc/modprobe.d/blacklist before rebooting. I tried the line blacklist r8169 and rebooted - the module still loaded. Well, once again, the solution was to boot into Windows, tell it to behave itself, then boot into linux and the whole world is happy. Here is the reference: As of 27 May 2007, in kernel 2.6.21.3, you may experience the issues with the r8169 driver if you dual boot Windows on some systems. Windows by defaults disables the NIC at Windows shutdown time in order to disable Wake-On-Lan, and this NIC will remain disabled until the next time Windows turns it on. The r8169 driver in the kernel does not know how to turn the NIC on from this disabled state; therefore, the device will not respond, even if the driver loads and reports that the device is up. To work around this problem, simply enable the feature Wake-on-lan after shutdown. You can set this options through Windows' device manager. I had no end of trouble with that realtek ethernet. Sometimes it would just disappear from the list (lspci). Every so often, even windows wouldn't see it. I scavanged an old 3com 905b pci card from a retired machine in my basement. Then I disabled the realtek from the boot bios. Problem solved. Btw are there are any motherboards *not* using this POS realtek? Every new board I have bothered to check had it. I guess some might have an intel ethernet chipset. Unfortunately, newegg didn't give a way to search motherboard on the basis of on-board ethernet make. and it's found here: http://gentoo-wiki.com/HARDWARE_RTL8168 Thanks for all of the help and tips! It's always a learning experience... -- Travis Crook Visions Beyond www.VisionsBeyond.com 208-478-7836 -- Johan KULLSTAM
Re: debian unstable, stable enough?
Jordi Carrillo [EMAIL PROTECTED] writes: I'm using Debian testing and I was thinking about switching to unstable. Is Debian unstable, stable enough for a Desktop system? Are there broken dependencies in unstable? I run unstable/sid rather than testing. You have to be aware that in unstable you sometimes get minor glitches (not often, really only every few months for me) that need to fixed up. For example, just this past couple of days, the sysv-init package broke by not installing /etc/rc?.d stuff properly. A new init package came out yesterday or so with a warning and some instructions. I had to figure the packages that have init.d stuff that was recently installed. Then I just had aptitude reinstall them. This is the kind of thing that happens in unstable. You kind of need to be aware and keep on top of things and be able fix the stuff that goes pear shaped. I haven't had a major hosing of the system but that's always possible. Testing has an aging process which prevents stuff like the above init problem from entering. However, the aging process is automatic and has no override (unless you drag down stuff from unstable explicitly yourself - which is a very valid mode of operation). This means that if a bug does enter testing, it can take a while for the fix to arrive. If dependencies churn, it can sometimes take a really long time for that fix to arrive. And did I mention that the aging/dependencies is automatic? No help will come to debian testing for broken packages except through the aging or your explicit installation of unstable packages or compile your own. Furthermore, security bugs are not fixed in testing until the package ages properly. Truely, testing is the distribution in which you are most on your own. I figure I am better off running unstable than testing. My personal opinion is that stable - if you run a server or system upon which you depends (e.g., a computer used for your job that you really don't want to have issues with -- even if that's a desktop) or don't want to be on top of upgrade details. unstable - if you want new stuff and can handle the occasional breakage testing - not for user, only for people assembling the next stable Others will of course disagree. Thanks -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] Win2k blocks ports to my ISP's pop3 smtp
T [EMAIL PROTECTED] writes: hi Please don't flame for this waaa~y OT question -- I'm still trying to solve my smarthost problem and was forced to use M$ outlook, I suspect is it my ISP's problem but have to prove it, anyway, long story short: What do you want to use for mail on the win2k system? Thunderbird? I recommend downloading wireshark (was called ethereal) for win2k. That helps me sort stuff out. - My Linux is able to telnet to my ISP's pop3 smtp respective ports Can your win2k telnet into them? Fire up cmd.exe and run telnet. Have wireshark capture so you can see what you get. - My Win2k (fired under WMWare) is not able to telnet to my ISP's pop3 smtp respective ports, though it can ping them without problem. And web browsing is fine. I think you mean you can ping the host by sending an ICMP message. I don't know how you'd ping the TCP ports. - This seems not be the WMWare's problem, because my new grml0.8 launched within WMWare can telnet to my ISP's pop3 smtp respective ports without problem. - I've done some homework about Win2k ports blocking, and have verified that my TCP/IP filtering IPSec is not used. - The Win2k is out-of-box vanilla SP4, with bare minimum configuration, no firewall or anti-virus SW. It has been so long that I'm not using Windoze, I don't have a clue where to look for potential problems now, and where to go for help. Yeah networking on windows is a can of worms. Much easier for me to configure linux but that may be because I am more used to it. If you could give me any hints, it'd be very much appreciated. Wireshark is your friend. thanks a lot PS. plz don't flame this poor guy in this terrible situation. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CUPS dependencies
Roger Leigh [EMAIL PROTECTED] writes: Johan Kullstam [EMAIL PROTECTED] writes: [CUPS] I am still a bit disturbed by the web interface and punching in my root password at my browser (even if it was just local). Just add yourself to the lpadmin group and use your own username and password. Thanks for the hint! I am already running a firewall. It is easy to allow/restrict access with iptables (once you have the firewall infrastructure in place). You can also restrict admin access to particular hosts, subnets, DNS domains/subdomains, etc, and allow passwordless admin access for trusted hosts/networks. You can do the same thing for normal access, too. FWIW what I really hate is when portmap, fam c get dragged in by some random dependency. That's not (AFAICT) a direct or indirect CUPS dependency. No, it's not. But I still hate it when all that wants to get dragged in. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CUPS dependencies
Jan Willem Stumpel [EMAIL PROTECTED] writes: I have been using lpr (later lprng) without any problems, since 1995, using many different kinds of printers (whether connected to my own Linux box, or shared through a 'home network' with Windows PC's). I hear you. I was thinking the same thing. I have been running lprng since frog knows. However, more and more Debian packages seem intent to force me to use CUPS. I don't need CUPS, therefore I don't want CUPS; and when I see the many messages on this list by people having problems with CUPS, I always count myself lucky for simply using lprng. But many Debian packages install CUPS components, because of dependencies. I finally bit the bullet and did CUPS last week. Actually, it was fairly painless. I found config to be easier that with lprng. And really, CUPS doesn't seem stupendously heavyweight. Firefox will now select amongst my printers (which it didn't do with lprng). I am still a bit disturbed by the web interface and punching in my root password at my browser (even if it was just local). I am already running a firewall. It is easy to allow/restrict access with iptables (once you have the firewall infrastructure in place). FWIW what I really hate is when portmap, fam c get dragged in by some random dependency. Is there a way to specify don't ever try to install this package to apt? I am not sure this help you much, but sometimes it's nice to know others share your pain. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Can I stay in testing without going etch?
Paul E Condon [EMAIL PROTECTED] writes: On Sat, May 20, 2006 at 11:10:13AM -0500, Owen Heisler wrote: On Thu, 2006-05-18 at 23:37 -0400, Roberto C. Sanchez wrote: Adam Hardy wrote: You mean 'testing' and 'etch' are interchangeable as far as the sources.list entry goes? lrwxrwxrwx1 200 200 4 Jun 06 2005 testing - etch The stable/testing/unstable names are just symlinks to the actual codenames. So for now, testing and etch are the same. Once etch is released, the symlinks will be upadted and testing will no longer be etch. -Roberto I have been following this thread as I am working on switching to Debian, and have just these questions: If I install Debian stable and have stable in the sources.list file, will updates keep happening, even across releases? I think it would be great it I never had to reinstall, yet could still have a completely up-to-date system. Also, is the same true for unstable and testing? Thanks for clarifying. Look at http://www.debian.org/releases/ Concerning unstable: Sid is unstable is sid is called unstable is 'Still In Development'. Don't use it until you have some experience with stable and testing. In my opinion, you have two choices, whatever the stable is and unstable/sid. If you go with stable, just use a named stable in your sources.list - like sarge since it's easier to control the dist-upgrade when an new release comes. For unstable or sid it doesn't really matter if you call it unstable or sid in the sources.list - it ends up being equivalent. Sid can have some brokeness from time to time but bad packages also get fixed quickly. Only run it if you can handle minor incoveneniences and such. Testing should not be used. It is the last to get security updates - and due to the forced aging, you can't get them in a timely manner. Usually, it has few broken things than sid. However, broken packages that get into testing can stay broken for nearly unbounded lengths of time. There is no manual intervention from debian to force important fixes into testing if they haven't aged properly. Testing, in my opinion, is only for those on the devel team doing quality assurance for the next stable. You can run testing and be prepared to pull in various things from unstable to fix the things that go wrong. I figure that it's just easier to run sid and be done with it. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: test
steef [EMAIL PROTECTED] writes: Roberto C. Sanchez wrote: steef wrote: test after provider-disturbance s. Please don't send test messages. They are rather pointless. If you are subscribed to the list and don't get at least several dozen messages a day, there is a problem. If you post to the list and you don't see your post in a few hours (also don't forget to check http://lists.debian.org/debian-user to see if it was received) then there is a problem. Please don't clutter everyone's mailboxes with such nonsense. -Roberto i disagree in this case completely, roberto. had an urgent problem with two providers here and *had* to test in this way because i work for a living and had to use *zonnet* on another way. A second time in, let's say, ten years. It only takes a couple of minutes to find some random posting on this list and reply to it. Your reply doesn't even have to make much sense, just ask some inane question. That way no one knows you are testing and comes down on you. ;-) so: don't piss me off again or i 'll come down on you! ha, ha. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: time incremented one hour with every boot.
Matus UHLAR - fantomas [EMAIL PROTECTED] writes: On 20.04.06 11:02, Joaquin wrote: I'm experiencing a little problem with the system clock since I entered into daylight saving time: Every time I start debian it adds one hour, so it has became some kind of time machine... If any of you is entering into my system and making fun of me, please say it, I offer a beer or a coffee if you stop doing so... just a joke... ok that wasn't funny... :) When I boot with a Ms Windows it doesn't increments the date, so I think it isn't a hw issue. first remove /etc/adjtime, maybe it instruct your system to change the time after reboot. second, it's bad that windows can't use UTC in hardware clock, you will probably have a little problem each time your computer boots. Not true. Just it up for Greenwich Mean Time (UK) and no daylight savings time. Windows will always show Zulu time, but at least it's somewhat sane and stops playing with your clock. Set up linux to use UTC. Linux is smart enough to display times in your local time zone. Since windows can't do time sanely, I figure this is the least losing mode. third, if you are permanently online, you can install ntpdate and probably configure it to set the clock after boot. in such case you should also install ntp-server to keep the clock in sync. -- Matus UHLAR - fantomas, [EMAIL PROTECTED] ; http://www.fantomas.sk/ Warning: I wish NOT to receive e-mail advertising to this address. Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu. Saving Private Ryan... Private Ryan exists. Overwrite? (Y/N) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: about scp
Nevruz Mesut Sahin [EMAIL PROTECTED] writes: hello friends I am to trying to server ssh -p 19666 www.myservername.com and I am connecting sucsacfully to server. but if I try to copy some file from server or server to my computer scp does not work may be I am using wrong command. could you please help me about command . I once ran afoul of a problem with the same symptom. If your bash_profile (or other init script) prints something in non-interactive shells^*, then scp and ssh with command will fail, but you can still log in with straight ssh. Can you issue a command using ssh? $ ssh example.com ls You should see the output of ls. [*] bash will go out of its way to clear PS1 in non-interactive shells. In my .bashrc, I have ## bash_profile or bashrc put init stuff here, no printouts or ssh/scp will get upset [ -z $PS1 ] return 0 PS1='[EMAIL PROTECTED]:\w\$ ' put your aliases and printy junk after the test -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: firefox - how to change menu font
LeVA [EMAIL PROTECTED] writes:
Hi!
Is it possible to change firefox's (1.5) font, which is used to display it's
menus and dialog windows?
Thanks!
Yes. Make a .gtkrc-2.0 (or is it .gtkrc-2.2?, i made a hard-link
between the two.) Change the gtk-font-name variable to name and size
of something you perfer. Here is mine:
##-
include /usr/share/themes/Crux/gtk-2.0/gtkrc
gtk-font-name = Georgia 11
gtk-key-theme-name = Emacs
style default
{
base[SELECTED] = { 0.529, 0.808, 0.922 }
fg[SELECTED]= { 0.00, 0.00, 0.30 }
bg[SELECTED]= { 0.529, 0.808, 0.922 }
text[SELECTED] = { 0.00, 0.00, 0.00 }
## sky blue (see /etc/X11/rgb.txt for colors)
##base[SELECTED] = { 0.40, 0.60, 0.80 }
##bg[SELECTED]= { 0.40, 0.60, 0.80 }
}
##-
Daniel
--
LeVA
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
--
Johan KULLSTAM
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: about remove directory
Daniel Webb [EMAIL PROTECTED] writes: On Thu, Dec 22, 2005 at 03:30:08AM -0700, Daniel Webb wrote: rm triva: And another trivia: are there any other GNU tools where option position causes different behavior and both ways are valid? find Find has a lot of weird options which prune and such and the order does matter. tar For example: tar cvfb /dev/tape 200 tar cvbf 200 /dev/tape do the same thing. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Many packages missing from testing
Scott [EMAIL PROTECTED] writes: Mitch Wiedemann wrote: Joona Kiiski wrote: Hi! Now for about two weeks there have been many packages out of testing. I'm must wondering what's the point? Those missing packages prevent me from upgrading because there are many among those which I desperatily need and I don't want to start hacking apt. Wouldn't it be better to have an unstable version of packages in testing than no version at all! Okay, you are pros, I'm just a newbie and there must be a good reason for this, this situation is just irritating. Maybe you could consider having four versions of debian in transition phase, like: stable, testing, testing-new, unstable. And when 99.5% packages would have entered testing-new it could replace testing. Just an idea, maybe it would just make things too complicated for developers and maintainers. Check out this thread. Some perspectives on the missing packages in testing... http://lists.debian.org/debian-user/2005/11/msg00683.html Ah-ha! I'm glad to see it's a temporary problem. I just installed Debian for the first time this past week and had planned to stick with etch, but wound up with Sid becaue of all the missing packages. All this is IMHO. Warning rant ahead: 1) testing not for users. It is for debian maintainers putting the next stable release together. There is a mechanical aging process which lets packages come over from sid. A package could get updated, wait, and just when it's about to land in testing, it gets some new minor update. The package may work great the whole time, but it doesn't gets into testing for a long time. When sid is revving heavily, you might never get an update since some dependency somewhere is getting an update. I think that security is now doing updates for testing. They used to not do it and under that case testing was positively foolhardy. While it is usually solid, a breakage in testing can sometimes take ages to get fixed. This is the nature of testing. There is no manual override of the aging process. 2) Use sarge or sid instead. Sarge is stable and everything works and is included. Right now, it's not even superannuated. Sid gets quick updates. It might be broke once in a while, but it isn't broke for long. (Of course it might really blow up and clobber your system if, e.g., libc.so get hosed.) 3) Do not use stable in your apt sources since that could surprise you when we get a new stable. Stable releases are rare enough that manually changing /etc/apt/sources.list is not a problem. Hence, the fixed name is better. 3) If you want to use testing, put etch into your apt sources. Of course, I could be extra perverse and argue that if you are a user who would be surprised you have no business running testing anyhow. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Many packages missing from testing
loos [EMAIL PROTECTED] writes: [snip rant against testing] I just totally agree with you. A little difference, I switch my production machines (stable) to testing somewhere during the frozen time (of course using testing real name. I prefer having a manual control on the oldstable-newstable update. I am around since ham and this worked without problems for me. I agree. The fixed names are much better. There was a thread here a while back (6 months, a year?) about making the default be a fixed name like woody, sarge or etch, rather than stable. I think that would be a much better default. My desktops use unstable. As are mine. Sid is pretty solid for me so far. I can recover from most of the mishaps. But at work I do worry since I could lose hard if things go really badly. I guess that's why they make stable. But that's so boring ;-). The problem is always the same: Newbies don't understand the sense of the word unstable as used by Debian. In fact they lack understanding what a distribution is, and therefore what a stable (or unstable) distribution is. Exactly. I was using testing for a while and got tired of losing when a package broke and wouldn't get fixed for ages. Of course, a savvy user could default to testing and drag in unstable (with whatever pre-reqs) whenever a breakage occured. Perhaps this method could be made more known. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Newsreader: Best of the bunch?
Steve Lamb [EMAIL PROTECTED] writes: Chris Bannister wrote: That sounds more like a religious statement. Gnus is a message reader, i.e. news, mail, local files, ... And to run it I need to insta what? No test editors are involved. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: should etch be Debian 4.0 ?
Mark Fletcher [EMAIL PROTECTED] writes: On Sunday 10 July 2005 21:55, Joris Huizer wrote: Johan Kullstam wrote: Let me see if I understand you correctly. Your reason for having the ambiguity of wether to call it 3.2 or 4.0 is just to keep people from assigning etch a number? I think this is quite logical, as there is some structure in those numbers - 4.0 means a big leap, 3.2 means smaller change; nobody can tell right now how big the step is from sarge to etch, as it's development has just started ofcourse, it's just up to the debian development team to decide wether the changes are big enough to call it 4.0 (anyone know why sarge became 3.1?) just some thoughts Joris I'd add that it's not deliberate ambiguity as a means to any particular end, so much as it not being an appropriate stage of the development of etch for the decision to be made if a major or minor version upgrade is appropriate. This does matter; this list wouldn't take long to hear from a whole tribe of people with nothing better to do than complain about unimportant things if they decided it was to be 3.2 now and then it turned out that the changes were massive and the upgrade path difficult... likewise if they decided 4.0 now and then it turned out the changes were small and relatively minor . Are people really going to look at the version number and say, I've got sarge now and since new number is 3.2 i'll upgrade but if it were 4.0 i'd sit still? Have people done this in the past? Releases come every 3-4 years so why not let the release notes explain the changes. A version number might make sense for automated things where cron downloads and installs a minor increment but not major one. This is so seldom that manual intervention isn't too much to ask for. Since the difference is subtle, why have the distinction? Why not use next release is 4.0 and the one after that 5.0 and so on *no matter how small the update*? -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: should etch be Debian 4.0 ?
Hendrik Boom [EMAIL PROTECTED] writes: On Sun, Jul 10, 2005 at 11:23:10AM -0400, Johan Kullstam wrote: Mark Fletcher [EMAIL PROTECTED] writes: On Sunday 10 July 2005 21:55, Joris Huizer wrote: Johan Kullstam wrote: Let me see if I understand you correctly. Your reason for having the ambiguity of wether to call it 3.2 or 4.0 is just to keep people from assigning etch a number? I think this is quite logical, as there is some structure in those numbers - 4.0 means a big leap, 3.2 means smaller change; nobody can tell right now how big the step is from sarge to etch, as it's development has just started ofcourse, it's just up to the debian development team to decide wether the changes are big enough to call it 4.0 (anyone know why sarge became 3.1?) just some thoughts Joris I'd add that it's not deliberate ambiguity as a means to any particular end, so much as it not being an appropriate stage of the development of etch for the decision to be made if a major or minor version upgrade is appropriate. This does matter; this list wouldn't take long to hear from a whole tribe of people with nothing better to do than complain about unimportant things if they decided it was to be 3.2 now and then it turned out that the changes were massive and the upgrade path difficult... likewise if they decided 4.0 now and then it turned out the changes were small and relatively minor . Are people really going to look at the version number and say, I've got sarge now and since new number is 3.2 i'll upgrade but if it were 4.0 i'd sit still? Have people done this in the past? Releases come every 3-4 years so why not let the release notes explain the changes. A version number might make sense for automated things where cron downloads and installs a minor increment but not major one. This is so seldom that manual intervention isn't too much to ask for. Since the difference is subtle, why have the distinction? Why not use next release is 4.0 and the one after that 5.0 and so on *no matter how small the update*? Well, no matter whether the etch release next year or later is going to be a big or a little upgrade, etch isn't stable yet, and so if it's going to be 4.0 or 3.2 or 3.1415 or whatever, the numbers (if any) for what we have now are going to have to be less than that. So for the time being, 4.0 is probably inappropriate as a version number. I don't see why 4 is inappropriate. 4 has been the successor to 3 for a few thousand years. I don't think that will change in the next decade. 3.2 might be OK. 3.1.9 is probably too close to numbers for sarge with security upgrades. But still, I ask, why the minor and major numbers? Does this help anyone in the least? Others ask why a number at all. That's a good question. But does it even need a version number right now? Or will it need a version number ever? Seriously, what is the point? Etch.0 etch.1 c might be ok. Call it etch.0 while in testing. Then etch.1 at release. Roll-up upgrades are etch.2 etch.3 c. If it were a package, aptitude and its friends would have to know it. But it isn't. Or is that one of the big changes we're going to see? That the entire distribution becomes a package? -- hendrik -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: should etch be Debian 4.0 ?
Santiago Vila [EMAIL PROTECTED] writes:
On Sat, 9 Jul 2005, Nigel Jones wrote:
On 08/07/05, Steve Langasek [EMAIL PROTECTED] wrote:
On Fri, Jul 08, 2005 at 11:57:25AM +1000, Drew Parsons wrote:
I'm already seeing documentation referring to Debian 3.2 (etch).
Where is this? It's certainly wrong for documentation to make assumptions
about the release version number at this point, and is the kind of thing
that makes it harder to change later.
And after all, isn't the point of codenames to avoid third-parties
incorrectly attaching a version number to a not-yet-released version?
http://ru.wikibooks.org/wiki/LOR-FAQ-Debian seems to be saying Etch is 3.2
Also http://www.computerbase.de/lexikon/Debian seems to be saying the same.
(Got these from a google search of etch 3.2 debian (page 8 onwards)).
Those references should be changed, then. It's *not* ok to refer to etch
as Debian 3.2, as the version number for etch has not been decided
yet.
Why the mystery?
What message is being transmitted by calling it 3.2 versus 4.0?
If there is no message, why the distinction?
So what we have now is current version of debian is N.K with next
version of debian being N.{K+1} or {N+1}.K according to some
inscrutible random variable dependent upon the phase of the moon and
other chaotic factors.
The only effect as far as I can see is to cause confusion about the
version number of the next release.
I suspect some sort of Schödinger's cat experiment where the next
version number is in some sort of half-incremented
half-not-incremented superposition state.
Does this state of affairs actually help anyone? ANYONE?
--
Johan KULLSTAM
Re: should etch be Debian 4.0 ?
Mark Fletcher [EMAIL PROTECTED] writes:
On Saturday 09 July 2005 23:56, Johan Kullstam wrote:
Santiago Vila [EMAIL PROTECTED] writes:
On Sat, 9 Jul 2005, Nigel Jones wrote:
On 08/07/05, Steve Langasek [EMAIL PROTECTED]
wrote:
On Fri, Jul 08, 2005 at 11:57:25AM +1000, Drew
Parsons wrote:
I'm already seeing documentation referring to
Debian 3.2 (etch).
Where is this? It's certainly wrong for
documentation to make assumptions about the
release version number at this point, and is
the kind of thing that makes it harder to
change later.
And after all, isn't the point of codenames to
avoid third-parties incorrectly attaching a
version number to a not-yet-released version?
http://ru.wikibooks.org/wiki/LOR-FAQ-Debian seems
to be saying Etch is 3.2 Also
http://www.computerbase.de/lexikon/Debian seems
to be saying the same. (Got these from a google
search of etch 3.2 debian (page 8 onwards)).
Those references should be changed, then. It's
*not* ok to refer to etch as Debian 3.2, as the
version number for etch has not been decided yet.
Why the mystery?
What message is being transmitted by calling it 3.2
versus 4.0?
If there is no message, why the distinction?
So what we have now is current version of debian is
N.K with next version of debian being N.{K+1} or
{N+1}.K according to some inscrutible random variable
dependent upon the phase of the moon and other
chaotic factors.
The only effect as far as I can see is to cause
confusion about the version number of the next
release.
I suspect some sort of Schödinger's cat experiment
where the next version number is in some sort of
half-incremented half-not-incremented superposition
state.
Does this state of affairs actually help anyone?
ANYONE?
Erm, OK. Coming back to earth for a second, I think the
reason why some people object to a version number being
attached to etch is because of the stage of its life
that it's at, it could be argued not to be an official
release yet. (Pardon me, has it even made it to
testing yet?).
Refraining from giving early-stage upcoming versions of
software an official version number until it gets to a
certain stage of maturity is pretty common practice in
largescale software development. And it doesn't get a
lot more large scale than a worldwide open source
project.
Let me see if I understand you correctly. Your reason for having the
ambiguity of wether to call it 3.2 or 4.0 is just to keep people from
assigning etch a number?
--
Johan KULLSTAM
Re: XFree86 packages mess, Xkb broken, how to clean up
Andrey Andreev [EMAIL PROTECTED] writes: Hi, Some time ago I did something dreadfully stupid on my Debian - I upgraded a repository, which was trully not meant for installing on a normal Debian: #maemo deb http://repository.maemo.org/ maemo ossw deb-src http://repository.maemo.org/ maemo ossw In case you have been wondering it's a repo with a lot of things to be run in a sandbox to provide environment and tools for developing apps for the new Nokia 770 Linux tablet. It is meant to be used inside the sandbox. Well, I put this in my desktop's /etc/apt/sources.list and dist-upgraded (quite a few packages got upgraded) and installed some packages on top of it. Next, I completely forgot what I have installed. Next, I figured what I have done, and removed the repo from my sources.list, but as everything seemed to be working OK, I decided to clean it up some other day. A month later, after some upgrades of X my xkb stopped working. I noticed that /usr/X11R6/lib/X11/xkb does not exist, so I made a symlinked it to /etc/X11/xkb/. Now, to my surprise X would seem to start OK, but my keyboard did not work properly (I could do Ctrl-Alt-Bksp, or toggle NumLock, or even switch xkb groups with RWin, but I could not type any letters or numbers, nor could I do Ctrl-Alt-F1, etc.). Removing the link made the keyboard work properly, but I was obvoiously unable to switch keyboard layouts. In this situation XFree86.0.log was reporting that: Couldn't load XKB keymap, falling back to pre-XKB keymap I spent many hours trying to find packages that came from the maemo repository (is there any proper way to do it? If yes, please tell me), and downgrading every package I found to the one in Debian unstable (which is what I run). It did not fix the xkb issue. I diffed my /etc/X11/xkb/ to one on a working Sid and found no differences. Now, as dpkg -S /usr/X11R6/lib/X11/xkb tells me that the file I was missing belongs to the libx11-6 package, and when I hit the problem I was running a version of the package that came from maemo, I am fairly confident that upgrading from that repo is the reason why xkb stopped working. I have reinstalled many X-related packages with no success. I would not mind reinstalling the whole lot and more again, and even losing some configuration data, as my alternative at the moment seems to be a from-the-scratch install, and this sounds wrong (although it is so easy, I would prefer not to do it). So, all ideas are very welcome. Regards, Andro I also lost my xbd files. The problem is that the xfree86 package (xlibs to be exact) contains the files, but, for some unknown reason, fails to actually install them (I am not sure if the very latest rev fixed this). It does make a directory tree /etc/X11/xkb... but many of the files in it are missing. Download the .deb (apt-get will fetch it for you and then find it in /var/cache/apt/archives). Then unpack it manually using dpkg. Copy the relevant file tree to /etc/X11/xkb. Hope this helps. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Is 64MB enough?
John Moore [EMAIL PROTECTED] writes: Thanks but the old clunker's motherboard is not expandable to 256M :-( Star/Open-Office is not going to be pleasant. TeX, on the other hand, will run like a treat. Apache will probably be fine on its own. I am not sure how heavy tomcat/java is. C, perl, python are viable. From: Paolo Alexis Falcone [EMAIL PROTECTED] Reply-To: Paolo Alexis Falcone [EMAIL PROTECTED] To: Debian Users' Mailing List debian-user@lists.debian.org Subject: Re: Is 64MB enough? Date: Mon, 9 May 2005 21:30:14 +0800 On 5/9/05, John Moore [EMAIL PROTECTED] wrote: Thanks. The primary purpose is to study and understand Linux and the programming environment. I'd also like to install Apache and Tomcat and do some Java programming (again, just experimental stuff - nothing commercial grade). I'll definitely need to run a browser and an email client. The ability to run (Star/Open)Office would be nice. in that case you'd need more memory. 256MB is cheap these days. As you know, It's not as cheap if you need 72-pin EDO SIMMs memory. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: X.org ----- Xfree
Hanspeter Kunz [EMAIL PROTECTED] writes: Hi Johan, I am using the radeon driver supplied with xfree86 4.3 in debian/sid. I can set the display to 1400x1050 but there is a band about an inch or so (roughly 100-200 pixels in width) along the rightmost edge where the picture is distorted by odd shadings. On my Sony Z1, radeon card, xfree is running on 1400x1050 without problems. From XFree86.0.log, it seems I am using ATI Technologies, Inc. RV350 [Mobility Radeon 9600 M10] The laptop specs call it an ATI Radeon 9600. Is this the same video as you have? Going to google, I see referrences to M6. Would you care to share some more details about how you got it to work? -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: X.org ----- Xfree
Micha Feigin [EMAIL PROTECTED] writes: At 12 Dec 2004 16:08:07 -0500, Johan Kullstam wrote: Thomas Adam [EMAIL PROTECTED] writes: --- Johan Kullstam [EMAIL PROTECTED] wrote: for it. Look up debian and ibm thinkpad t42 2378fvu sometime. I find it more interesting that because of the Xorg/XFree86 issue, more and more people are saying that they *do* need Xorg, just because it is available. I am having trouble understanding the relevance of your comment to my post. You deleted most of it such that it is impossible to figure out the context so I will repeat once more, slowly, MY HARDWARE ISN'T SUPPORTED BY XFREE86 BUT IT IS BY XORG. THIS IS WHY I NEED XORG. No need to shout :-) Well, *you* are able to read and take input, but there are a few who have preconceived notions and are unwilling to countenance that they might be mistaken. Hence, I am not sure about lack of need to shout for some people. Anyway, are you sure that your hardware isn't supported with the latest dri code (which can be compiled for xfree and there is a debian package for it). DRI as in the kernel support for various chipset accelerations? I am not trying to get faster 3D support or support any fancy rendering. I am simply trying to have working 2D. I am using the radeon driver supplied with xfree86 4.3 in debian/sid. I can set the display to 1400x1050 but there is a band about an inch or so (roughly 100-200 pixels in width) along the rightmost edge where the picture is distorted by odd shadings. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: X.org ----- Xfree
Thomas Adam [EMAIL PROTECTED] writes: --- Johan Kullstam [EMAIL PROTECTED] wrote: for it. Look up debian and ibm thinkpad t42 2378fvu sometime. I find it more interesting that because of the Xorg/XFree86 issue, more and more people are saying that they *do* need Xorg, just because it is available. I am having trouble understanding the relevance of your comment to my post. You deleted most of it such that it is impossible to figure out the context so I will repeat once more, slowly, MY HARDWARE ISN'T SUPPORTED BY XFREE86 BUT IT IS BY XORG. THIS IS WHY I NEED XORG. I bet you, you'd have made do quite happily with XFree86. :P Sorry, you lose that bet. What do I win? -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Radeon 9200 issues
Dominique Dumont [EMAIL PROTECTED] writes: Antony Gelberg [EMAIL PROTECTED] writes: Nope, file is attached. I have also tried inserting the amd64_agp module instead of via_agp, and disabling all the X extensions - no joy. Again - this config works great with my Radeon 9000 card. I can see that when the screen flicks every second or so, I get a glimpse of the desktop, then a monitor OSD message saying something like cannot set this video mode, please set 1600 x 1200 @ 60hz. Video 1600x1200 is not logged by X as possible for your monitor. I am going to guess that 1600x1200 is natural size in pixel of an LCD panel on a laptop. If so, it is not only *a* possible resolution, but the *only* resolution possible (besides 800x600 by grouping pixels 2x2). Other resultions seem possible but the pixels are only kind of sort displayed that way. This resultion has been fairly common for a couple of years now. By the way, I have a laptop with a radeon 9600 and 1400x1050 display. My desktop video card and monitor are also capable of high resolution. The only problem is a distinct lack of fixed width fonts which are readable when laptop monitor is 14 (and desktop monitor is a 19 CRT). Are there any good fonts which show up at least 3mm high at 120 dpi? You should try a more conservative resolution. Once you have a display, then slowly increase the resolution (and/or frame rate). When I get home I'll send you my Xfree config file (radeon 9200 with 1280x1000 video). Cheers -- Dominique Dumont Delivering successful solutions requires giving people what they need, not what they want. Kurt Bittner -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: X.org ----- Xfree
Marc Wilson [EMAIL PROTECTED] writes: On Wed, Dec 08, 2004 at 07:51:47PM +0100, Roelof Wobben wrote: Is X.org better than Xfree ?? No. It's different. Someday it may be better, for some values of better, but not yet. Does someone has X.org getting on work with debian Sarge ?? Xorg will wait until Sarge+1 (namely, Etch). You don't need it anyway. Sure I do. I have a laptop with radeon 9600 mobility in it. XFree86 4.3 cannot display 1400x1040 properly - I have a bunch of wierd boxen on the right most edge and lose about 1/10 of my screen to them. It also has difficulty displaying to an external attached monitor. I read on the internet about having to download and compile the x.org for it. Look up debian and ibm thinkpad t42 2378fvu sometime. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian and spam
Joao Clemente [EMAIL PROTECTED] writes: William Ballard wrote: Including nospam in your email name helps a lot. And how do you subscribe to the list with nospam in the e-mail address?!? You'll never receive an e-mail to that address! You could make a username with nospam in it. Then you can receive email to it. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Linux Installation with SCSI Drive
Tim Kelley [EMAIL PROTECTED] writes: On Monday 06 September 2004 00:00, Ronald wrote: I often try out (installing) Linux distributions, so this is an I/O intensive operation (writing to the drive). Would using SCSI drive significantly improve the installing time? (halve it?) Yes, from CD/DVD of course. Are there any DVD SCSI devices? AFAIK there are IDE to SCSI adapters which attach your DVD to a SCSI bus. Unless, you are carrying some older SCSI devices (I have a pair of Plextors which still work just fine), I would just attach the CD/DVD to the IDE system. FYI currently I use Seagate 7200.7 80 GB and to get full 3 GB install takes around ~10 to 20 minutes. Hmm, a full (everything that doesn't conflict i guess) install of debian is more like 12GB! This must be woody ... Motherboard is Abit KG7-Raid which doesn't have inbuilt SCSI adapter, so I have to buy a separate adapter. CPU is Athlon XP 2400+ with 512 MB RAM. And the drive I want to get is probably Seagate Cheetah 15K.3: 18 GB, 15,000 prm Ultra320 SCSI http://www.seagate.com/cda/products/discsales/marketing/detail/0,1081,619,0 0.html You do realize that the usual PCI slot is limited to 120MB/sec (and real throughput is somewhat less). Unless you have 64 bit or 66MHz PCI, there is *no* point in U320 *at* *all*. Stick with a U160 controller. Also, your typical SCSI disks are not that much faster than typical IDE disk (of same model year). The follow is just typical and does not necessarily follow from the interface SCSI/IDE but are features of the disks you will find -- caveat is here to make explaination easier. The SCSI disk spins faster but has a lower bit density. The SCSI disk will have a fast seeking arm and use a smaller disk to reduce arm travel distance. This means that access times for the SCSI disk are up to twice that of the IDE (less time to wait for sector to spin around to you, less time for head to move into position). On the other hand IDE offers greater total capacity. The higher density of IDE designs is offset by the slower rotation speed which means that sustained transfer rates are close. SCSI costs up to 2-3 times as much as IDE per disk. Also expensive is all the SCSI infrastructure, controller card, cables, termination, SCA cage, etc. SCSI is totally *not* worth it for a single disk system. What SCSI does allow you to do is stuff many more hard drives into your system than IDE will. More spindles = less having to seek. I put system on one disk, my home on another, swap on a third. Now, I can launch and app (disk1), have it read data from my home (disk2) and push some dormant application onto swap (disk3) with a minimum of seeking. RAID5 is also done with SCSI, but this is going to be expense extreme. Really it probably isn't worth it. My system is all SCSI (hardware raid and SCSI dvd and cdwriter) and although it is very responsive and never jerky (I mean never), it really doesn't justify the cost on a home pc. I assembled a Frankenstein system though, of somewhat older equipment that was cheaper (most from ebay). I like URL:http://www.centrix-intl.com for some cheap surplus server gear. Be aware that some of their offering is a bit ancient these days. If you mobo has onboard ide raid I would just go with that, unless you got money to burn. Agreed. If you do I would get an adaptec or bus logic controller though, the linux drivers are very good. I think you mean LSIlogic (formerly Symbios and before that NCR) rather than Buslogic (now Mylex I believe). The LSI controller is likely to be a fair bit cheap than the adaptec although both are fine controllers with good linux driver support. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: New HD advice
Mike [EMAIL PROTECTED] writes: John Summerfield wrote: Mike wrote: David Baron wrote: Thanks for all the good advice (obviously I am considering moving stuff to a new drive). A few more questions: 1. (Might be elementary, not matter, but ) what is best, place files on partition and mount to the target directory or directory(ies!!) on the partition and mount to the parent directory? 2. I have smart monitor running. Got no bulletins from it--might need to configure it differently. The old drive has less smart-capability than the new--temperature being very important! 3. /home, /var ... others ( /local, /usr/src)? SIze recommendations? I recall seeing very detailed recommendations somewhere a few months back. 4. Swap file vs swap partition--I did not know there was such an alternative in linux. IF a file is better, how do I change over? (Swap is almost never used, it seems.) A word about swap space. Although you can create swap space as a file, instead of the kernel handling a raw partition to swap, all swapping must go through the file system, which slows swapping considerably. Not so in 2.6. What do you mean by that? I will give you that the VMM is much better in 2.6. But what's Not so in 2.6? This doesn't address this particular issue, but it does explain why it might be so: http://www.osdl.org/docs/linux_journaling_filesystems_and_workloads.pdf I found it looking for a comparison of filesystems; that's about page 28. Ideally, you would always want swap as a partition. I do not believe that is so on single-drive systems. Consider the amount of time seeking between data and swap paritions, I can't see how any optimisations can overcome the laws of physics. What laws of physics pertain to the issue of swap being a file versus swap being a partition? It really doesn't matter which type of file system you use. One may be faster that the other. One may be better for recovering from a system crash. One may be better for logical volume management. But none are faster than the kernel using a partition for swap in raw mode. That's why AIX, Solaris, HPUX and Linux, by default, use swap partitions, not files. With that said, if you have 2 disks, it's important to have them both installed as IDE primaries, never having two disks on the same IDE channel as primary and secondary. The idea is that your system can read/write to the two primaries at the same time, whereas with a primary and secondary setup, the read/write must alternate between the primary and secondary. Now, if you have your two disks installed and IDE primaries, and you create a swap partition on both disks, and both disks are set to the same priority, say 1, then the net effect is that when swapping occurs, data is written to both swap areas interleaved. This is the fastest way of handling swap space. NO NO NO. best place for swap, least-used drive. Best parition - busiest. Give more of an explanation than this. I suppose that setting a single disk to be used solely for a swap partition would be good although a bit impractical. If I thought it'd help, I'd add an ancient seagate 4.5GB 10krpm SCSI drive for $9 plus whatever centrix-intl is charging for shipping. I could use most of the drive for swap and reserve 1-2G for assembly of cd-rom images. Swap is really slow compared to memory. It is nice to have for dormant program storage, in which case the slowness isn't much of an issue. If you are actively bringing stuff in and out of swap, it is time for more memory. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Compiling kernel without -O2 flag
Sorav Bansal [EMAIL PROTECTED] writes: Hi all, There is a curious thing that I noticed in the kernel. The kernel does not compile if you remove the -O2 flag in the Makefile. It gives an error in an ASM directive mentioning that a register is being spilled. The x86 has a very few registers. With in-line assembly you can easily tie some registers up. I guess gcc got into a corner it couldn't get out of. I would guess it's a bug in gcc, but at least it had the presence of mind to fail loudly. Is this normal behavior? I suppose it is. Have you tried -O1? I know that the kernel depends upon certain functions being inlined for proper behavior. The kernel has some hard time constraints since it needs to deal with the hardware. You fiddle with the compile options at your own risk. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Where should I start ?
Paul E Condon [EMAIL PROTECTED] writes: On Thu, Jul 08, 2004 at 12:00:10AM +1000, [EMAIL PROTECTED] wrote: Hello, I have a simple question (well maybe two). Which version of Debian should I download, sarge or woody ? (I am limited to 12Gb download a month) If I do download the CDROM images is it possible to convert them to DVD images (or vice a versa DVD to CD)? Thank you Michael Woody CDs amount to about 4.5GB. Sarge doesn't have a CD set yet (I think), but does have a new net install CD image (ie one CD only). 12 GB per month seems to me to be not an important issue. Also, for Woody, you do _not_ need all those CDs to install. Most people need only the first CD. I needed the first and second on one machine because it happened to be pure SCSI, and 1st Woody CD only boots using IDE. I really like the minimal CD netinst images. Depending upon what you want to install, a working system needs somewhat less than 14 GB of downloading. With the netinstall you download only what you need. With the multi CD complete set, there will be a lot of stuff you'll never need. Still, the netinst method can easily be a few GB (especially if you have a false start or two) and you don't have many to burn. You might consider buying a CD set. Upgrading from Woody to Sarge is easy, but takes an hour or so. If you are new to Debian, and hope to commit to it, either version is OK as a start. With Sarge you get a head start on using the version that will soon be stable. With Woody, you start learning what Debian has been in the recent past, and about upgrading and dist-upgrading to get to the new stable. I say choose between stable or unstable/sid. I think testing is a bad choice for a running system. Debian developers use testing to create the new stable. For testing, all packages must have a try-out period in unstable and are not accepted into testing until dependencies and such are also met. This means that glaring obvious bugs are rare, however, there are no other mechanisms for security or bug fixes to enter testing. Once a problem enters testing it can stay there a long time -- sometimes many months. A functionality bug for six months is annoying. A security bug for six months could be more than annoying. I have seen suggestions about running testing and then manually picking from unstable to fight the bugs that do get through. This might be a very good strategy but requires more work on the part of the administrator. Unlike other distributions, with Debian, you really do upgrade from one version to the next. You never need to re-install, except if you discover that you did something really really incompatible with your vision of what you want. It is rather difficult to go back to and older version of debian. If you upgrade to sarge or unstable/sid and want to go back to stable, then a re-install is needed. You might also consider using Knoppix or some other 'live CD' version of Debian. This gets you up and running quickly, but the transition to a truly policy compliant Debian installation is not so straight forward as with Woody or Sarge. HTH -- Paul E Condon [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: automatically restarting dying daemons?
Will Trillich [EMAIL PROTECTED] writes: problem: xinetd, after working just fine and dandy for weeks at a time, gets dozens of unexpected signal (source unknown) and gives up the ghost. questions: 1) what's the best way (e.g. debian way) to monitor active daemons and restart them when necessary? maybe some utility already exists for this? or /proc/something? or `ps ax`? 2) how can i track down the source of the signals specific to this case and make it stop? xinetd chugs along nicely for the most part, and then -- poof! -- it dies a sudden death: Give it to init. This kind of stuff is what init is for. man init. man 5 inittab. Use the respawn action. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How hard would this be?(Learning LaTex)
Jon Dowland [EMAIL PROTECTED] writes: On Sat, 26 Jun 2004 23:21:41 -0500, cecil [EMAIL PROTECTED] wrote: Someone told me today at lunch that what with my wierd obsession, as he called it, to perhaps go without a gui(X), I should try that latex thingie. My buddy is a real wordmaster. LOL. I did some reading up on it; it's interesting. I never knew that you could do all that with no window system. Does anyone here use it on a regular basis, and if so, how hard is it to use, setup, print, etc? I used it to write up my 3rd year project, and I think it probably ended up being as useful as writing it in microsoft word (not taking into account of course the unavailability of word for linux). Much of the power that latex has exists now in word: e.g. automatic contents building. How about automatic equation numbering like this? Sample x = 3 (4) and y = 14 (5) I want little numbers at the end of the line, with auto-increment please. I do NOT want some horrible Equation 4 caption under the equation. I do a lot of maths and find word to be rather inadequate and very awkward. I used bibtex for references which was a blessing and a curse. The appearance was significantly more attractive than anything I could concoct in word; and the typesettings is second to none. However some of the limitations/annoyances really built up and got on my nerves. I adopted a vim script called `vim-latexsuite' mid way through and found it very helpful; especially `auto-folding' which collapses each section up until you ask it to be expanded. Find and print out a good reference sheet and try out vim-latexsuite, an emacs equivalent if thats your poison or an environment such as kile or lyx. -- Jon Dowland -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian has turned unusable.
Freivald, Joseph A, GVSOL [EMAIL PROTECTED] writes: Hello, I'm new to the Debian community, but I have used RedHat for about 8 years, and Gentoo for almost two. I must say, Debian is quite good compared to these other distro's. Perhaps RH is more stable than Sid/Sarge, but there is NO way to install a base system from a RH CD. The smallest install I was ever able to get was over 450MB and included LOTS of extras that I really didn't want. I see lots of people advocating Sid(unstable) as a desktop, but shouldn't people who are not developers/maintainers gravitate to Sarge? In a word - no. IMHO testing is *only* for developers who run a 2nd box to check integration and freezing into stable. Testing usually is OK and typically suffers from fewer bugs than unstable/sid. However, bugs that do make it into testing are often not fixed and can linger for a long time due to the semi-autonomous migration policies from unstable to testing. For example: last year sometime, gv got updated and needed a new library version. The new library had some bug or other but the gv didn't properly depend upon it. Thus gv entered testing but was broken. No new library was coming because it got kept back. Despite numerous bug reports, *testing* never got fixed until much later. It got sorted in unstable within days. There is no mechanism for actually fixing bugs in testing -- bugs are fixed in sid and trickle down, *eventually*. Testing is last in getting security updates for this reason as well. Isn't testing/debugging Sarge supposed to be a priority? Perhaps, but in reality it does not appear be such. I tried using testing but was unhappy with it and finally went to sarge. I feel stable is for servers and people with 5 year old video cards. Unstable is for everyone else (most people). Testing is for people working on assembling the next stable on their spare boxen. Also, since packages automatically drop into sarge from Sid after 10 days (unless there is an unresolved issue), you are likely to get all the great new apps that you want, but without someone dropping in a new, buggy version by mistake. Likely, but in case of bug, there is no direct mechanism whereby testing gets fixed. If, e.g., libc happens to rev during the 10 days that a fix needs to wait, you can wait a really long time. Also, this would make more bug reports get filed against Sarge, which would help to progress it to the next stable. Bug reports against sarge do not result in action fixing sarge directly, but are filtered through sid and automated process. Stable and sid are fixed directly. This makes testing the least maintained of the three flavors. I realise that I have written these in a somewhat argumentative form, And I have, perhaps cynically but honestly, responded argumentatively. Hope the comments have helped. but read them as questions. As I said, I'm new here ( 3 months ), but I have read up as much as I can find on the releases and the procedures for advancement. I have used Sarge for about 6 installs now (including upgrade from Woody and the new installer), and I'm very pleased with it's performance and package features. I used Woody for my file server (which now has a local Debian Mirror!), mostly because I don't care about the desktop on it, and I like to have the security patches, but I have Sarge running on two laptops, three desktops and a DB server. Also, I'm running Kernel 2.6.3 with the proprietary Nvidia driver and VMware Workstation on my work laptop. I note this because these things were exceptionally problematic on other distros, but were cheezy-eazy on Debian. --JATF -Original Message- From: news [mailto:[EMAIL PROTECTED] Behalf Of Monique Y. Mudama Sent: Monday, April 12, 2004 7:59 PM To: [EMAIL PROTECTED] Subject: Re: Debian has turned unusable. On 2004-04-12, Adam Aube penned: Monique Y. Mudama wrote: Well, more unstable than the stable distribution takes a lot longer to type and wouldn't fit on a CD volume label =P What about current, then? This would encourage people to use the unstable distribution, which by definition isn't considered ready for prime time. The truth is that there are tradeoffs; a one-word name just isn't going to capture those tradeoffs. If anything, the right term for unstable might be head or tip -- or would that be experimental? But what do I know? I'm just a random user. It does seem to me that we've had the name game a few times, and every time a dev has strongly indicated that we should leave well enough alone. -- monique -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Stable vs. Testing Vs. Unstable
s. keeling [EMAIL PROTECTED] writes: Incoming from Loren M. Lang: I'm curious about how many people are actually using Debian Unstable or Testing to Stable for normal desktop use or even a production server. I've never felt any need for anything but stable. Have you ever owned a video card that's under four years old. I need the hardware support of newer releases. Nearly 2 years ago I had to load sid in order to support my radeon 8500. That card was long in the tooth then and hardly newer today. I notice that stable still doesn't support it. I've never been one to think newer is necessarily better. I update for security or stability. I've no need for the bleeding edge. Every upgrade comes with its own brand new set of bugs, so why not stick with the ones you've got if you can? If you can. Some people like to buy reasonably modern hardware for all sorts of reasons. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Still more X trouble from the install!
Jim Higson [EMAIL PROTECTED] writes: I'm still having these problems, on my main machine, which has resisted XFree from the start. As sugested, I've updated the kernel to the prebuilt image 2.4.18-k7 from the woody isos and confirmed the update with uname. It still won't recognise my usb mouse. The mouse in question is a logitech MX700. I find it hard to belive that the iso kernel builds do not support somthing as comon as a usb mouse, maybe there is another problem here? Google groups says nothing much about this mouse and debian [http://tinyurl.com/2tvzo] and in windows it works as a standard usb mouse. I have this mouse. Just plug it into the PS/2 mouse port. Are there computers which have USB but not PS/2? What is the big fuss about USB mice? And do there exist _normal sized_ keyboards which use USB? All I've seen keyboard-wise is those tiny cramped jobs and the mega internet wireless console jobs. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Still more X trouble from the install!
Jim Higson [EMAIL PROTECTED] writes: On 13 Jan 2004 08:20:24 -0500, Johan Kullstam [EMAIL PROTECTED] wrote: Jim Higson [EMAIL PROTECTED] writes: I'm still having these problems, on my main machine, which has resisted XFree from the start. As sugested, I've updated the kernel to the prebuilt image 2.4.18-k7 from the woody isos and confirmed the update with uname. It still won't recognise my usb mouse. The mouse in question is a logitech MX700. I find it hard to belive that the iso kernel builds do not support somthing as comon as a usb mouse, maybe there is another problem here? Google groups says nothing much about this mouse and debian [http://tinyurl.com/2tvzo] and in windows it works as a standard usb mouse. I have this mouse. Just plug it into the PS/2 mouse port. Are there computers which have USB but not PS/2? What is the big fuss about USB mice? A few motherboards don't have PS/2. USB mice also have a much better refresh rate, IIRC serial mice have a refresh rate of only 35Hz or so, which is pretty unpleasent when you're used to 125Hz like the MX700 can do. Worse still, for gamers the frame rate is effectively limited by the mouse rate, there's no point in updating the screen 120 times per second when input is only recieved 35 times per second. Ah OK, thanks. I thought I'd ask why. I suppose it comes down to not wanting to use an OS if it doesn't let me snipe in quake! It might seem petty, but I bought the MX700 specifically for games. I've got the mouse sorted now, but KDE is crashing the X server, I posted about this elsewhere in the thread. And do there exist _normal sized_ keyboards which use USB? All I've seen keyboard-wise is those tiny cramped jobs and the mega internet wireless console jobs. Sure. Most new keyboards I see use usb. I still use an aincient IBM model m with a din plug, but there are plenty of normal usb keyboards. If I bought a new keyboard now I'd expect it to be usb. I haven't seen them. I didn't see any at the recent computershow I visited. Moreover, I cannot seem to find them on the net. For example, go to newegg.com and lookup USB keyboards. I see about a half dozen but they all have non-standard formats like extra rows of buttons on the top and handrest that protrudes from the bottom. I have a keyboard holder on my desk and it will only accept normal, boring rectangular format keyboards. I also do not want a truncated or compacted keyboard. I can't find USB in the kind I'm looking for. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: daemontools-installer
ahuisman [EMAIL PROTECTED] writes: Hi, I have installed daemon-tools a couple of times on the same machine without any problems.After a dselect update i upgraded to gcc (GCC) 3.3.2 (Debian)and installed daemomtools-installer, djbdns-installer again.This time when i do a build-daemontools i get the following output : [snip] env - /bin/sh rts.tests 21 | cat -v rts At this point i only see a blinking cursor. After 10 minutes i interrupt: Notice rts growing without bound. It's not catastrophic since it grows slowly and your disk has a few MB to spare. And if you look inside rts you'll see a bunch of stuff about GLIBC_2.0 errno not being found. Afaict DJB hasn't kept up with newer libc versions. Fortunately, there is a patch. Head over to david sill's site, life with qmail. URL:http://www.lifewithqmail.org/lwq.html You will want to get netqmail-1.04. This contains patches for qmail, daemontools and ucspi-tcp. Apply the daemontools errno patch, compile, enjoy. Hope this helps. make[1]: *** Deleting file `rts' make[1]: *** [rts] Interrupt make: *** [build-stamp] Interrupt I didn`t experienced this before. Is the problem gcc-3.2.2 ? I think it's libc but it may be gcc libraries. I haven't chased much further than the above. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: A newbie's confusion about GPL
Arnt Karlsen [EMAIL PROTECTED] writes: On 21 Oct 2003 08:00:40 -0400, Johan Kullstam [EMAIL PROTECTED] wrote in message [EMAIL PROTECTED]: I am using gnus/emacs. It treats mail like news. This way, I get mail lists each in their own group and the discussions are threaded. ..agreed, and emacs and gnus _is_ on my todo list. I like to use 2 different apps for these 2 tasks though, as I like to _see_ what I do in case I forget. ;-) I found gnus/emacs baffling at first. I was using elm and tin for a long time. I gave gnus a try but was overwhelmed. I switched to mutt and slrn which I used for a couple of years. Using procmail to split your mail and mutt to read it is a good solution. I think it may best for someone who uses a fast starting non-emacs editor, e.g., vi. Fwiw, slrn has borrowed heavily from gnus so moving from slrn to gnus is not as bad as going from tin to gnus. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: A newbie's confusion about GPL
Tom [EMAIL PROTECTED] writes: ..as in; Where _is_ Osama and Saddam?. And playing the west bank settler games on the Iraqis, is _not_ gonna help. I'm beginning to see this as an issue over which rational and reasonable people can disagree. For me, this all ties back into my Illusion of Technique philosophy. The positive argument the anti-Bush people make is: Iraq is a waste of time; we must eliminate al Quaeda itself. Saddam Hussein is/was a bad man and perhaps should have been eliminated. However, the basis for that is not the fight against terrorism. I answer: show me your SPECIFIC plan, don't just wave your hands and say: We'll do it. What country is Osama bin Laden from? What country provides Al Qaida with most of its money? What country provided 15 of the 19 9-11 hijackers? I know what country is responsible. Our tanks should have rolled *south* out of Kuwait. When you press the anti-War people on just *how* they are going to prevent cataclysmic terrorism, they generally bash Bush and say we'll use police techniques to prosecute them. Sounds nice, but in actuality, tough choices must be made. Maybe in 50 years the Muslims will be turning out killer cars like Germany or killer stereos like Japan. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: A newbie's confusion about GPL
Arnt Karlsen [EMAIL PROTECTED] writes: On Mon, 20 Oct 2003 10:48:07 +0100, Colin Watson [EMAIL PROTECTED] wrote in message [EMAIL PROTECTED]: Could this thread please be taken to some other list or private mail? I think it's clear that it's got hopelessly off-topic. ..any way these off-topic rants can be flagged as such other than OT in the subject? My problem is the time I spend reading etc, so _I_ should really read DU off a newsserver than get 11000 mail since joining DU. ..dial-uppers may wanna enjoy being able to filter on, say OTNO, for OT, news only, and save time DL'ing their mail. ..on a newsreader, you just pick the threads you _wanna_ read, so I would tend to be far more tolerant towards flamewars on my newsfeed than on my mailfeed. I am using gnus/emacs. It treats mail like news. This way, I get mail lists each in their own group and the discussions are threaded. I have auto-expire set to delete old posts and so on. I don't know how anyone can receive very active lists like debian or linux-kernel with a similar type reader. It doesn't help the download time. However, text is pretty fast and even a long thread is unlikely to be as heavy as, e.g., one SWEN. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT - Programming Languages w/o English Syntax
Tom [EMAIL PROTECTED] writes: On Fri, Oct 17, 2003 at 02:01:24PM -0500, Ron Johnson wrote: On Fri, 2003-10-17 at 12:29, Monique Y. Herman wrote: On Fri, 17 Oct 2003 at 11:15 GMT, Tom penned: [OT, sorry -- but question is obscure, will be hard to google] Are any non-english-speaking readers aware of High-level programming languages using non-English syntax? Like, could I find a French C compiler that uses pour instead of for and si instead of if? [snip] You're right; the anglo-centric nature of most programming languages is distressing. It would be fun to code in a language based on a totally Distressing What an over-reaction. Guess what? When French/German/Chinese/Spanish/Portuguese/Japanese Computer Scientists decide to write a programming language in their own native language, there will be programming languages in those languages. But then, why did Niklaus Wirth use English key words, even though he is Swiss/German? I know, there was a swedish version of basic around in the 70s-80s timeframe. Other places would have had similar things. I expect the russians had some russian/cyrillic based thing. Okay, I started this OT thread, I'll try to end it. *I was interested in languages with alternate semantics, not just alternate syntax. *20 years ago I read an article comparing programming languages with the nationality of the author. (Pascal-Wirth-German: highly structured syntax. C-Americans: fast and loose, 20 ways to say the same thing.) You can say in Latin in 7 words what takes 11 words to say in English. Chinese can be incredibly terse (e.g., ancient chinese business documents) or incredbily expressive (e.e., chinese poetry), so that might make an interesting programming language. *Most westerns think in Subject-Predicate terms; i.e., the subject is primary, the predicate modifies it, so we all have Object.Method languages. Cherokee indians use Predicate-Subject; predicate is the central term, subject modifies it, like functional languages. But all of this is terribly OT for the thread, so let it die. Thanks to all who answered. Have you tried Common-Lisp? I found it to be very different from the other languages I know, C, matlab, pascal, basic, fortran c. Even though I don't use it much for my work, learning Lisp was mind expanding. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: xf86 3.0, i845, no go
Joe Paxton [EMAIL PROTECTED] writes: I am using the vesa driver of XFree86 3.0 ^^^ I see this as a simple typo and you are using 4.3.0. from experimental with i845 chipset; when doing 'startx', one of two things occurs: either the server aborts, giving a no screens found error, or else multiple lines and dots of various colors appear on the screen and the machine locks. You might try the i810 driver instead of vesa for X. I have this chipset on a machine at work. I used the i810 driver in the XF86Config-4 instead of vesa. This gave me a few modes such as 1024x768 and 800x600. I tried adding a modeline to give 1152x864 but wasn't able to get that resoltion working (X kept ignoring my modeline, but I do not know why). I have used debconf to try several different combinations of color depths and resolutions. None have thus far been successful, but, as I said, the results that they yield are consistent. The multiple lines of various colors (and machine locking) happens at color depths =8bpp, and the no screens found error occurs at colors depths 8bpp -- resolution doesn't seem to matter in either case. No other fatal error messages (besides the no screens found) occur. I have attached the error logs for each situation, and I have also attached the XFree86Config-4 for each situation (all named so as to hopefully avoid ambiguity). I am using kernel-image-2.4.21-4-686 on a 2.4GHz P4 with 256MB of RAM. I have the kernel AGP module loaded, and there doesn't appear to be a DRM module for the i845 chipset with the kernerl that I am using. I would greatly appreciate any help that could be provided. I am using 2.6.0-test4 and there are agpgart, intel-agp and If you chose the intel-agp, there is some DRI i830 module you can build as well. -- Joe Paxton -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How stable is SiD ?
Paul Johnson [EMAIL PROTECTED] writes: On Fri, Sep 05, 2003 at 03:27:47PM +0200, Joris Lambrecht wrote: Can anyone advise on starting to use SiD as resource for my Debian Workstation ? Doesn't it have to many issues left open, broken dependencies etc. If you have to ask, sid is not stable enough for you. Perhaps his video card isn't supported with the woody shipped xfree86? That's why I went straight for sid last August. (ATI Radeon 8500 needs 4.2.1.) At work I just got some crap corporate box with i845g graphics. I need xfree86 4.3. Please advise. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How stable is SiD ?
Karsten M. Self [EMAIL PROTECTED] writes: on Sat, Sep 06, 2003 at 01:23:43PM -0400, Johan Kullstam ([EMAIL PROTECTED]) wrote: Paul Johnson [EMAIL PROTECTED] writes: On Fri, Sep 05, 2003 at 03:27:47PM +0200, Joris Lambrecht wrote: Can anyone advise on starting to use SiD as resource for my Debian Workstation ? Doesn't it have to many issues left open, broken dependencies etc. If you have to ask, sid is not stable enough for you. Perhaps his video card isn't supported with the woody shipped xfree86? That's why I went straight for sid last August. (ATI Radeon 8500 needs 4.2.1.) At work I just got some crap corporate box with i845g graphics. I need xfree86 4.3. Please advise. 1. Report this problem to your video card vendor. I don't care if you're CTO of IBM or a one-week-temp and Spam-R-Us or Joe's Internet Taco Cafe. Vendors need to know that their customers are using their products with GNU/Linux, and demand XFree86 support. Do this regardless of any subsequent steps you take to resolve your problem. XFree86 supports it since February this year. Unfortunately, debian doesn't have 4.3.0 in its lineup. Woody uses 4.1 which came out in June 2001. XFree86 seems to put out reasonably timely releases. Where do I point these card manufacturers to a time machine vendor so that I can get proper debian support? 2. Consider the option of using a supported card. It's not going to cost much -- last time I tried this trick, it consisted of walking into CompUSA and asking for the lowest-end video card they had (suitable for most office work; who are you kidding about your Gnumeric FPS score), cost was on the order of $25. Submit the receipt for reimbursement, and pass along _this_ request to your internal purchase manager or whitebox vendor. Cannibalizing an existing dead box might get you the card for free. This is what I did. I ordered a cheap-ass card. This is my own money for a machine doled out to me at work. No, I do not have control over what they give me. I am very on the edge even daring to run linux. 3. Go ahead and use 'testing'. I am using sid on my home machine to support the radeon 8500. You're insulated from most of the borkenness of unstable. On the other hand using testing at work and sid at home i've noticed that things break more often in sid, but when broken in testing, they can stay busted in testing for a long time. I am not sure which is better. While the theory gainsays it, I am leaning towards sid being, in practice, more stable. Using 'pinning' and apt preferences, you can include unstable and testing sources, pull from testing by preference, but install selected unstable packages on an as-needed basis. This is what I do. It works most of the time. Note that using pinning to bridge the testing/unstable gap works pretty well (they're relatively close), but bridging stable to either testing or unstable is a real mess. In the latter case, you're trying to bridge 1-2 years of software development, often with large changes in basic foundations. Based on two minutes' hunting through Google and Google Groups, you're going to need XF86 v 4.3, which may not yet be in testing, And neither is it in sid/unstable. but is available via unofficial debs from http://www.apt-get.org http://tinyurl.com/mhvh The general answer is this: GNU/Linux remains disadvantaged by hardware support policies, If by this you mean that GNU/Linux stable cannot support hardware younger than about 3 years old due to a very slow release cycle policy, then yes, I would agree. many of which have been historically shown to be influenced improperly by Microsoft. You can blame MS for a lot, but this one is squarely on debian's head. Yes, I could perhaps lend a hand instead of bitching. However, my complaint is about the attitude behind If you have to ask, sid is not stable enough for you. comment more than it is about the debian X strike force. Free software is a powerful tool for circumventing this situation, but it requires a positive response on your part: telling the vendor that you want support, researching whether and how support is available, and taking necessary steps to configure your system. If you're not willing to do this, then you're better off sticking to known supported hardware, or accepting the rule of an illegal software monopoly over your computing platform. Resistance isn't painless. It has its rewards. I settled on compiling the xfree86 4.3.0 source myself and installing over top of what I had. This seemed to be the easiest way to keep my testing library set. It works for now, but I can't get the video modes I want. (I have also ordered a suitibly obsolete and cheap AGP video card which is supported so this will be taken care of shortly.) Look, when I buy
Re: [OT] SCO is going all out now
Rich Johnson [EMAIL PROTECTED] writes: On Tuesday, July 22, 2003, at 12:32 AM, Brian McGroarty wrote: SCO has made no claims against the 2.2 kernels. If worst comes to worst and SCO finally show some incriminating code in 2.4, stepping back to 2.2 until the relevant bits are purged from 2.4 is all anyone should need to do to cover their assets in countries where this becomes an issue. InformationWeek ( in http://www.informationweek.com/story/ showArticle.jhtml?articleID=12801004) reports: SCO Group claims that Unix has been used to accelerate the development of Linux in two key ways--line-by-line copying of Unix System V source code into the Linux kernel and copying derivative Unix code that enables multiprocessing capabilities. 2.0.0 had dual processor capability. when i found an overflow quad box, i put 2.1.124 on it. later on, 2.0.X kernels would do 4-way as well, but i never bothered going back. so if 2.2 kernels are clear then SMP predates any SCO claim. I have no multiple processor machines. Why on earth would I be pay a penny to license multiprocessing capabilities which I can't use? I guess I'll be going back to 2.2 until this nonsense blows oversigh. and 2.2 works fine on multiprocessor machines. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Nic Issues..
Tom Kloppel [EMAIL PROTECTED] writes: First of all, im using a DECchip 21041 card, it is supposed to use the 'Tulip' module. I installed debian with the normal woddy kernel(2.2.20-idepci) and the network card worked just fine. Last night I compiled/installed/upgraded to the 2.4.18 kernel, rebooted, and the network card wasnt loading, and the little pc light on my cable modem was OFF... So I tried to compile again and i was sure to add support for Decchip 2x04x Tulip module. Now the PC light does turn on, but i still cant get the card to function correctly. What I find wierd is that when i do 'insmod tulip' it says it cant find the module. Should that be happening? You might try the de4x5 driver instead. I have a DEC DE450 with 21041 chip and kernel tulip driver has been broken with this card for a year or so. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: considered harmful (was [off topic] Learning Shell from an old UNIX book)
John Hasler [EMAIL PROTECTED] writes: I know the shortcomings of csh have been discussed elsewhere in this thread. But tcsh is enhanced csh. I have no problem with csh (or tcsh) as a login shell. It is just not suitable for scripting. I do. If csh (and tcsh) suck for scripting, then why on earth use them as your shell? If you make them your shell then you have to spend time learning their broken brain damage. That's a waste. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Iptables: not recommended by Debian?
Jamin W. Collins [EMAIL PROTECTED] writes: On Sat, Feb 22, 2003 at 07:27:59PM -0800, calyth wrote: It may have been the iptables package which provides the init.d script. the comments in the /etc/defaults/iptables says that not even the author likes it. IIRC, that is that the author doesn't like the default script, and doesn't recommend it's use. Unfortunately, the iptables package maintainer doesn't give any other suggesting than the init.d script. Not that the author doesn't like the iptables tool. Rather large difference there. Right. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Iptables: not recommended by Debian?
Colin Watson [EMAIL PROTECTED] writes: On Mon, Feb 24, 2003 at 08:03:55AM -0500, Johan Kullstam wrote: Jamin W. Collins [EMAIL PROTECTED] writes: On Sat, Feb 22, 2003 at 07:27:59PM -0800, calyth wrote: It may have been the iptables package which provides the init.d script. the comments in the /etc/defaults/iptables says that not even the author likes it. IIRC, that is that the author doesn't like the default script, and doesn't recommend it's use. Unfortunately, the iptables package maintainer doesn't give any other suggesting than the init.d script. That seems a little unfair. Have you read /usr/share/doc/iptables/README.Debian.gz? Yes. My bad. That seems to contain some stuff about adding a pre-up and/or post-up for /etc/network/interfaces. I must have missed that the first time I looked at iptables. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Good C/C++ IDE
S Yuval [EMAIL PROTECTED] writes: Hello, Can anyone suggest a good C/C++ IDE? I've tried Anjuta, KDevelop, KStudio and have been disappointed. The only somewhat useful IDE I could find was QIDE, which is a trial product and only supports C. I like emacs. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Swap capslock-control in console?
I have swap-caps-ctrl in X, why shouldn't I want it in console too? How do I set things up so that I swap caps-lock and control in the console? Everytime that console-data updates it tries to clobber my setup (explained below). Is there a debian way to configure this? I hacked myself a new keymap which does this and called it /etc/console/us+.kmap.gz. Then changed /etc/init.d/keymap.sh to point to this by setting CONFFILEROOT=us+ instead of the usual boottime. Every so often upgrading stuff will ask about a modified keymap.sh. Since the package is kind enough not to outright clobber keymap.sh and it graciously makes a new keymap.sh (called something obvious but unmemorable like keymap.sh-dpkg.new or somesuch, but the exact name isn't important here) so I can then manually update things. I had tried replacing boottime with my own keymapfile but console-data will unceremoniously and silently clobber that. Is there a better way to get this? Is this a reasonable thing to want by way of console configuration? -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: A7M266-D CM8738 problems: Pcm control of various mixers don't work
Daniel Barclay [EMAIL PROTECTED] writes: Is anyone familar with this problem?: Since I've upgraded from a system with a SoundBlaster 16 to a A7M266-D motherboard with an on-board C-Media CM8738 chip, the Pcm control of mixer programs (tkmixer, aumix, wmmixer, kmix) no longer work. I have this motherboard and am using the onboard sound. I don't have the mixers you mention installed, but I do have gnome-volume-control (aka gmix) and xmixer. Some simple test by bringing up these two and doing cat /etc/passwd /dev/audio show that adjusting PCM adjusts the volume. (I am using the topmost of the three jacks for sound output if that makes any difference. Not sure if it can be used any other way.) Also, no control besides the master volume control affect the volume of audio from the computer. (The CD slider does work for the volume of audio directly from the CD-ROM drive.) Additionally, the overall output volume seems a little low, as if there is a PCM mixer setting (in the hardware) that is initialized to less than a full-scale value Yes. I get this too. I think it is there to give you a little room to increase volume. I'm currently using a 2.4 kernel with the cmpci module. I am using 2.4.20. Is there any special setup for the CM8738 that I should be aware of? I didn't do anything special. I am using the kernel stock driver. -- Johan KULLSTAM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
