Re: Driver in Lenny for ASUS 802.11n Network Adapter?
On Wednesday 01 July 2009, Andrei Popescu wrote: > Though upgrading the kernel seems like a very big deal, sometimes > it's actually safer than other packages. 2.6.29 doesn't require newer > packages, so you don't actually need to backport it. And if you get > in trouble you can always boot 2.6.26 instead. Yeah, that is true, but isn't there a greater risk that things are not that well tested, so that I may experience more obscure problems? I used to run more experimental and patchy stuff back in the day, but nowadays I have so little time for tweaking, that's why... Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer kje...@kjernsmo.net Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Driver in Lenny for ASUS 802.11n Network Adapter?
All, My new motherboard has a wifi card which is identified as a USB device, and lsusb says this about it: Bus 004 Device 002: ID 0b05:1742 ASUSTek Computer, Inc. 802.11n Network Adapter I have found that it can be used in Lenny with these instructions: http://wiki.debian.org/rt2870sta That involves using a backported Linux 2.6.29 kernel, and I would like to keep as much as possible on Lenny on this box, and the kernel is something I would certainly like to avoid updating. However, I haven't got any 802.11n devices, it would be perfectly OK to have just 802.11b/g. So, my primary question is if there is a driver in Lenny that would support this card with 802.11b/g? Alternatively, is it possible to compile just a kernel module rather than backporting the whole kernel? And if so, how? Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer kje...@kjernsmo.net Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Dependencies with dh-make-perl for CPAN packages
Hi all! I have found it very useful to use dh-make-perl to install CPAN packages on Debian and derivatives, but there is one thing I can't get to work well: dependencies. Makefile.PL does contain (or should anyway) the distribution's dependencies, and I see that there is some code in the dh-make-perl script that seems to be there to take advantage of it, but it doesn't add anything to the package, it seems. Nor have I found any documentation for it. Are there any "user-level" documentation for dh-make-perl, or would one need to read up on how maintainers work? Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
[OT] MPEG2 on Windows (was Re: How can I play avi, wmv, mov videos on a standalone DVD player?)
On Thursday 29 December 2005 17:45, Ron Johnson wrote: > kino (which is a non-linear video editor) *might* be able to do > it, as onw of it's side functions. Tangential on this topic, I encoded a movie from and AVI created by my digital camera on my Sarge box to MPEG2 to using Kino. It plays fine on all my systems, but I've sent it to friends, and they can't play it on their windows boxes... I wonder what MS has done to mess up that, it should be a no-brainer? It has to be their fault... :-) I don't have access to any windows-boxes now, has anybody else had this problem? Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Getting a list of installed packages
On Tuesday 27 December 2005 21:53, J Merritt wrote: > Is there a way to get a list of installed packages in Debian, > preferably from the command line and preferably in a text file? Yup, you could do dpkg --get-selections | grep install > file.txt or dpkg -l | grep ^ii > file.txt depending on your needs. See also the dpkg manpage. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Debugging severe performance problems
On Tuesday 27 December 2005 11:53, Gonzalo HIGUERA DÍAZ wrote: > Given the three scenarios you give, specially the first and maybe > also the third, disk access is a likely candidate. Check with hdparm > whether the HDD-chipset is using the best transfer mode available > (some DMA mode, given the hardware). Uh-ouch. Yeah, I really thought that I used DMA, but when you said it, I checked, and indeed it wasn't turned on in the kernel config. I needed a recompile anyway, and I was sure to enable it. My first impression is that this really did the trick! Thanks a lot! Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Debugging severe performance problems
Hi all! One of my systems has a really severe performance problem. It has an AMD Athlon XP 1700+ CPU. It has 768 MB RAM, and only very rarely swaps anything. It has a 3Dfx Voodoo Banshee (rev 03) PCI card with 16 MB on-card RAM. The performance problems has been there since the summer. For some time, it ran woody with KDE 3.1 on top, then I had it tracking sarge in April. It performed well at that point. I've tried many different things, but it all boils down to that I think the hardware is OK, but that I really have no clue where to start. It is my father's system, so I'm not bothered with it too often, but the performance is insanely bad. When it does updatedb, the system pretty much freezes for the time it takes to scan the disk. Firefox takes more than a minute to start, and a regular apt-get upgrade with trivial security fixes makes the load go to above 5. My own system is similar to this system, but has a somewhat slower CPU, and performs well. It is definitely not because it doesn't have the hardware. Not to get hung up in this, but I think I first saw the performance problems as xfree86 4.3.0.dfsg.1-13 went into sarge. I have therefore been working under the hypothesis that it is connected to some X problems, and it seems like it just takes a long time to draw anything. It is hard to tell, of course, if that's a symptom or a cause. So, I tried replacing the 3Dfx PCI card with my own well-performing Matrox G450 AGP card, but that had no appreciable effect. Reducing the resolution to 800x600 did however help a bit. Now, I think X might not have anything to do with it. I have also tried a full memtest86 scan, but no problems were spotted by it. I've been looking through all the logs I can think of, but seen nothing that meets the eye. One thing that has been haunting me is this error message: ldconfig: Cannot mmap file /lib/libslang.so.1-UFT8. I suppose that came in with a typo (UFT8 vs. UTF8), since I forgot to change my sources.list to stable shortly after woody was released back in the day. However, I've lived with this since then, and it had never any performance input before. It's quite funny that it remains though, I've upgraded the whole system since then... So, well, I have a system here that borders on unusable and no idea how to fix it. Where would you start if you had a *really* slow system? Best, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Subversion repository permissions
On mandag 07 november 2005, 06:20, Joey Hess wrote: > The simplest solution is to switch it to use the fsfs backend, which > uses some simple tricks (and a much more sensible design) to utterly > avoid the whole class of permissions issues that affect bdb. OK, cool! After reading up, it does indeed seem like migrating to fsfs is the most sensible thing to do. Thanks! Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: corrupted sources.list
On søndag 06 november 2005, 23:53, Sam Rosenfeld wrote: > W: You may want to run apt-get update to correct these problems Have you tried this? (it does the trick for me when I get this error) Best, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: Subversion repository permissions
On søndag 06 november 2005, 10:54, Simo Kauppi wrote: > Check your /etc/logrotate.d/apache2 (or /etc/logrotate.conf). I've investigated further, and it definitely has nothing to do with logrotate or postrotate. This is clear because the problem occurred in a period when logrotate was not run. The problem is not _that_ strongly connected to Apache. The real problem is the user that runs svn at the time when the Berkeley DB happens to rotate its logs, and that has nothing to do with logrotate, I'm pretty sure. In fact, I'm not sure it is "logs" in the common sense at all, it may be svn giving it that name. If that user happens to be www-data, I need to tell www-data that it must create the file with 664, that's the crucial thing, I think. But I can't see how. Other than edit apache2ctl, but since there is nothing about svn in there, I'm not sure it would even help... Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: Subversion repository permissions
On Sunday 06 November 2005 10:54, Simo Kauppi wrote: > Check your /etc/logrotate.d/apache2 (or /etc/logrotate.conf). > You should have something like: [snip] > create 640 root adm < change the mode 664 Ah, right. It isn't actually the apache logs that get the wrong permissions, it is the Berkeley database that Subversion use. That's why I haven't been thinking about this before. That file is owned by www-data:svn, it doesn't have anything to do with root:adm. Nevertheless, I guess you could be right, it may actually be the postrotate that causes it, as Apache2 is restarted. I'll try an see how it goes! :-) Cheers, Kjetil -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Finding old versions
On mandag 31 oktober 2005, 10:57, Rogério Brito wrote: > > Is there any way I can list this kind of old packages? > > One way is to look at the packages that are listed by aptitude on the > "Obsolete and Locally Installed Packages" or to use > apt-show-versions. Ah, great, apt-show-versions gave me what I needed! Thanks a lot! :-) Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Subversion repository permissions
Hi all! There is just one thing that's painful about Subversion, but that's rather painful too: permissions... I have often thought I have got it right, but there is one thing that remains: There seems to be some rotating of the log.* files, and if that rotation is prompted by me using ssh, everything is fine. But if it is prompted by the web server, it seems like the new file is not created with 664 like it must be, but with 644, which will cause problems later, requiring manual intervention and database recovery. The Book, http://svnbook.red-bean.com/en/1.0/ch06s05.html suggests modifying the apachectl script, but I'm guessing that's not the Debian Way. I need to tell Apache2 to use umask 002 when rotating these logs somehow, but I'm out of ideas... Any? Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Examining Swap
On torsdag 03 november 2005, 22:35, Thomas wrote: > Is there a way to move the contents of the swapfile back into the > (free) mem manually? I'm not positive about this, but I think you can trust the OS to do it properly. There are parameters that you can tune, but generally, if something is swapped out, it is because there had been sudden spikes in the memory use, and the stuff in the swap just hasn't been needed for ages. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Does increasing RAM 512 MB -> 1 GB lead to better performance?
On onsdag 02 november 2005, 02:55, [EMAIL PROTECTED] wrote: > How do I know if I'm regularly filling up my RAM? If you're using KDE, ksysguard is nice. You can add a ksysguard applet to the panel, and you can fire up the ksysguard application. From there, you can drag an drop a sensor showing swap use at any time. Best, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Does increasing RAM 512 MB -> 1 GB lead to better performance?
On tirsdag 01 november 2005, 14:40, Tshepang Lekhonkhobe wrote: > > How often do you deal with large data sets? > > What is a 'large data set'? I often see this term but not very sure > what it means... In my master's in astrophysics, I created, and subsequently analyzed files that were around 100 MB. Nowadays, I'm also playing with data from the Shuttle Radar Topography Mission, http://www2.jpl.nasa.gov/srtm/ and it was the reason why I updated to 1.5 GB RAM. As for the original question: I would keep an eye on the swap, by for example running top now and then or install a panel of ksysguard in the kicker panel if you use KDE. If it is much used, get more RAM. If you have the money, it is as simple as that, I think... Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: What's your favourite FLOSS?
On lørdag 29 oktober 2005, 17:00, Tshepang Lekhonkhobe wrote: > mathematics [ octave ] I'm, contrary to everybody else, going to comment just on this. My favorite here is without a doubt R. r-recommended in Debian and http://www.r-project.org/ on the web. It has some minimal, but very cool, OO concepts (as opposed to e.g. the proprietary IDL, which is supposed to be OO, but is (was) entirely misunderstood), and has a very straightforward mapping between math and code. E.g. for loops are unneeded in R, and that's actually a good thing. Highly recommended! Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: Getting Debian updates onto a slow bandwidth system
On lørdag 29 oktober 2005, 04:52, David E. Fox wrote: > On Fri, 28 Oct 2005 22:18:12 +0200 > > Kjetil Kjernsmo <[EMAIL PROTECTED]> wrote: > > A somewhat related question: Can jigdo, when building CDs, take the > > updates and replace the packages that has been updated, in the > > install image? > > It's supposed to be able to do that, if you put the CD in the drive > mountpoint and type in its path, it should scan the contents of the > CD when building the image. Of course, you need to write the new ISO > (because it contains new files) onto another media. Hmmm, I couldn't get this work. I'm home now, but I tried it at work, due to more decent bandwidth and a really fast local mirror. Building the .iso itself went fine, updating it with security updates didn't. I mounted the .iso loop on /mnt, as per the howto, and then I tried variations of http://security.debian.org/ as the APT mirror. jigdo-lite then goes on to scan, and it does in fact scan for all the files, but gets a 404 for each one. Problem is, the updates are in a subdir "updates" under "pool", and I can't find a way to get that in. Had it been updates/pool, it would have been fine, but it is pool/updates :-| So, no luck there. Also, I think it seems to look for exactly the same filename, so the security updates, with sarge1, etc appended, wouldn't be included. Any further suggestions? Best, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Finding old versions
Hi all! I'm going to make a clean install of Sarge on my new hard drive, and the idea was to use dpkg --get-selections to get what I had on the old. However, I bet I have lots of woody-era, even potato perhaps, software on that old disk, and I only know how I can get rid of libraries that nothing depends on (deborphan/orphaner). Presumably, I don't need gcc-2.95 or gcc-3.0 anymore, for example. Is there any way I can list this kind of old packages? Cheers Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Stock vs. Debian kernel sources
On søndag 30 oktober 2005, 06:52, John O'Hagan wrote: > What do I need to be aware of if I use stock sources? I think I remember somewhere around when 2.6 released that Linus said something about him not seeing it as his role to provide end-user kernels. That's for distributors to do. I haven't been following the developments since but since then, and since not needing very specialized patches anymore, I've built my own kernels, but built them on kernel-source packages rather than kernel.org releases. If you like being on the edge, you may be bitten. But OTOH, by being on the edge, you get the experience that's needed to provide stable kernels for the rest of us, so, by all means do it. I think it boils down to that, the issue is simply that unexpected things may happen, but if you are equipped to deal with it, you're making everyone a favor. Cheers Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: Getting Debian updates onto a slow bandwidth system
On fredag 28 oktober 2005, 19:27, [EMAIL PROTECTED] wrote: > > Is there a possibility to copy all sarge updates > > to, say, a usb-drive, takes this home and update the debian > > system from this drive. > > I have limited bandwith (debian) and I want to use another > > host (windows) for this. > > The Debian package 'apt-zip' does exactly that. A somewhat related question: Can jigdo, when building CDs, take the updates and replace the packages that has been updated, in the install image? I'll be doing a clean install next week, where I'll install partly from CDs, and don't want to waste bandwidth or space on the CDs for packages that will be updated immediately anyway, and there are lots of them this time. Best, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [OT] SATA vs. SCSI
On fredag 28 oktober 2005, 11:10, Miquel van Smoorenburg wrote: > If I've learned one thing about disks in the last few years, it's > that you should never ever buy the largest disks available. Right. I do the same thing, but for a different reason: Usually, the GB/price-ratio has a max, not for the latest and biggest, but for the somewhat smaller. I have a 250 GB drive waiting for me at the post office now, that has the best ratio now, a month ago, it was a 200 GB drive. As for the SATA vs. SCSI, I guess there is a reason why SCSI disks usually has a 5 year warranty, whereas *ATA just one. But they are also 5 times as expensive per GB as SATA. I have personally no experience with SCSI, but at my old institute, they were used exclusively, and we had 3 TB of SCSI disks when I finished my studies. Since I knew sysadmin well, SCSI disks blow up too, so, what I would do in this situation is to go for 3ware RAID controllers and SATA with lots of redundancy. You could have 4 redundant disks and still be within the cost of SCSI, unless, of course, the disks live up to their one-year warranty period, which hasn't happened to me yet... :-) Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Timeline for 3.1r1 release?
Hi all! I'm wondering if anybody has heard something about the timeline for the next stable release, 3.1r1? There is http://people.debian.org/~joey/3.1r1/ but the "proposed timeline" link there is 404. You see, I ordered a new 250 GB IDE disk and a new IDE cable (cross fingers) today, and I'm going to do a clean install, and my bandwidth at home isn't _that_ good, so I'd like to burn two CDs at work, and use that as a start. But with the huge amounts of security fixes lately, particularly X, it would be nice to bring along those updates on CD... And I can't get them with jigdo, can I? Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: spurious mailman messages
On torsdag 20 oktober 2005, 19:14, Joe Mc Cool wrote: > # > This message was created automatically by mail delivery software. > A message that you sent has not yet been delivered to one or more of > its recipients after more than 1152 hours on the queue on benburb. > > The message identifier is: 1EBDwj-0007sZ-2Y > The subject of the message is: subscribe > The date of the message is:Fri, 02 Sep 2005 16:59:17 +0100 > > The address to which the message has not yet been delivered is: Hmmm, ok. > pipe to |/var/lib/mailman/mail/mailman subscribe sheep > generated by [EMAIL PROTECTED] > Delay reason: pipe_transport unset in system_aliases router Hmmm, OK, I haven't used pipe transports in ages, but I suppose, you've gotten rid of all that, so the only things that remain are things on your queue? So, you are on Exim4, right? Then I have a really brutal way of deleting everything...: cd /var/spool/exim4/msglog/ exim4 -Mrm `ls` Since /var/spool/exim4/msglog/ happens to contain files named after the message ID for each email, this will remove them all... You have to make sure that there aren't anything there that you don't want to delete of course... And don't hold me responsible for any lost email... :-) Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: OT: Damaged harddisk and/or disk controller - ps
On torsdag 20 oktober 2005, 18:38, Marty wrote: > That looks familiar. OK! > In my case it also warned of imminent drive failure, It said nothing about that... > smartctl -H indicated that the disk was failing, OK, I get: SMART overall-health self-assessment test result: PASSED (PASSED on hdb too) > and the > BIOS SMART function warned about the disk. Swapping out the power > supply seems to have cleared up all problems and the disk now passes > all tests. Cool! :-) > Of course there's a chance your drive motor could really > be failing. Yeah... > Surge performance has nothing to do with steady state power or heat > dissapation. It's the ability to handle transient loads, and depends > mostly on the quality of the design and components in the power > supply. It may also be related to component aging or other > degradation of the power supply. Yup, sure. > > Is there any way I can measure the actual consumption? > > You probably need a storage scope and current probe. Hmmm, OK. I don't have that here. I suppose I could find a lab at my old uni and scare someone into helping though, but I'm not sure it is worth the effort. > > I don't have a more powerful > > PSU available, and the cost of that means it gets even harder to > > decide what is the right thing to do... Since I need a larger disk > > anyway... > > If it's the power supply, then you may need to retire it anyway. Yup. But in that case, I'd like one of these...: http://www.ups2.com/ and I have been holding off on that, under the assumption that my current PSU is actually OK... Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: OT: Damaged harddisk and/or disk controller
On torsdag 20 oktober 2005, 05:40, Mike McCarty wrote: > My recommendation: > > Get all the data you possibly can off that disc ASAP, and buy a > replacement disc for $50. Then, and only then, start thinking about > whether that drive can possibly be used for data storage. Maybe > you could put /tmp on it or something. Hmmm. OK. But I can get a 200 GB disk for NOK 675, or USD 100, and that will be all the disk this motherboard will ever see, I suspect... So, I don't know. My data isn't worth a lot, really, because I have redundancy and backup. My time _is_ worth a lot, but I enjoy the challenge of figuring out what to do about it too... So, that's why I'd like to grok first and act later... :-) Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: OT: Damaged harddisk and/or disk controller - ps
Thanks for the words of wisdom, Seeker! On torsdag 20 oktober 2005, 11:32, Seeker5528 wrote: > Using the Maxtor without it being paired to another drive is an easy > test to do before getting into the things others have suggested that > require additional parts or test equiptment. H, well, it isn't that easy... Since it happens on average just once per month, and it has been several months since the last time I had issues, I could easily run it for half a year, and wrongly conclude that this was the cause... Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: OT: Damaged harddisk and/or disk controller - ps
On torsdag 20 oktober 2005, 01:27, Marty wrote: > It could be a spin-up problem due to a worn out motor. Right. It is actually something like that that's my primary suspect. > This would > probably be reported by smartctl from the package smartmontools. Ah, thanks for the pointer! Got the daemon installed now, and did a run, I'm seeing these errors: Error 149 occurred at disk power-on lifetime: 1024 hours (42 days + 16 hours) When the command that caused the error occurred, the device was in an unknown state. Error 148 occurred at disk power-on lifetime: 1024 hours (42 days + 16 hours) When the command that caused the error occurred, the device was in an unknown state. Error 147 occurred at disk power-on lifetime: 1017 hours (42 days + 9 hours) When the command that caused the error occurred, the device was in an unknown state. Error 146 occurred at disk power-on lifetime: 1017 hours (42 days + 9 hours) When the command that caused the error occurred, the device was in an unknown state. Error 145 occurred at disk power-on lifetime: 1017 hours (42 days + 9 hours) When the command that caused the error occurred, the device was in an unknown state. Does that mean anything to you? > It's also possible that your power supply can't handle the powerup > surge, and this can also mimic the motor spin-up problem. I've > recently had this problem. Try swapping the power supply, preferable > with a more powerful one. OK, it could be, but I put my hand in there last night after Alvin's post, and again this morning after power-up, and the PSU is cool, so it is nothing to indicate it is over-heated, at least. Furthermore, it is a 340W PSU, which should, according to spec, give more than enough power for my system, unless it is something wrong with it, of course... Is there any way I can measure the actual consumption? It would be interesting from several perspectives. I don't have a more powerful PSU available, and the cost of that means it gets even harder to decide what is the right thing to do... Since I need a larger disk anyway... Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Linus' backup strategy...
On torsdag 20 oktober 2005, 00:59, Hendrik Boom wrote: > On Thu, Oct 20, 2005 at 12:44:06AM +0200, Kjetil Kjernsmo wrote: > > Linus' backup strategy, > > Is that the one wher you posting it all on usenet and find it later > in the archives? Almost: "Backups are for wimps. Real men upload their data to an FTP site and have everyone else mirror it." -Linus Torvalds Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: OT: Damaged harddisk and/or disk controller - ps
On torsdag 20 oktober 2005, 00:53, Hendrik Boom wrote: > Normally good advice, but his hard disk seems to be working fine when > it does power up. Yup! And I have done nasty things to the system, just to see. For example doing aide --update, updatedb, an intensive write process and a CPU-intensive computation process simultaneously. It should put maximum stress on the system, both power-wise, disk use and CPU... Never seen any errors, and certainly no other problems... It is just that it isn't detected at startup... "Detecting primary master None" In fact, I have been planning to put this disk in my server system to make it software RAID with the disk it has allready, since that system is rebooted very seldomly... :-) Thanks for all the kind advices, keep them coming, it is time for a nap around here... :-) Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: OT: Damaged harddisk and/or disk controller - ps
On torsdag 20 oktober 2005, 00:34, Alvin Oga wrote: > hi ya kjetil > > On Wed, 19 Oct 2005, Kjetil Kjernsmo wrote: > > Hehe, no, it was the latter, a it was a well-recognized store, but > > they were caught shipping disks that had been returned as new. > > sounds like dell .. they paid millions for that boo-boo Hehe, no, it wasn't them. Dell does annoying things like turning cable connectors around, so they don't get my money. Local store-thingie, but the same mistake... > other possibilities > > you cannot ( should not, as in expect systema and corrupt data > problems if you do ) > - do not mix ata-33 ( cdrom ) with hard disks Checked. > - do not mix ata-66 with ata-100 > - do not mix ata-100 with ata-133 Hmmm, OK. Is it that fragile...? The slave is a 4.5 GB Western Digital something someone threw after me... I dump things that are semi-important onto that for redundancy. > i'd try a different ( new ) power supply Hmmm, ok. May I ask why? > as henrique(?) said ... make backups .. Sure. All my important work is checked into a remote SVN repository. I have a few big things on the disk, but it is public domain data, using Linus' backup strategy, and allthough it would be painful to download all that again, it can be done. E-mail, --get-selections and a handful of letters is all I need backup for, that goes on the slave disk regularly, which is occasionally burnt to CD. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: OT: Damaged harddisk and/or disk controller
On onsdag 19 oktober 2005, 23:35, Alvin Oga wrote: > hi ya kjetil Hey, and thanks for the quick response! :-) > i'd replace the ide cables ... with new 80 conductor cable Yup, that was the first thing I did... :-) It is the cheapest thing to do, so I did that some time ago. No effect... > > What I see is that if the primary master disk isn't detected, the > > primary slave isn't either. > > that'd depend on: > - the jumper settings on your disk > - the manufacturer of the disk > - the disk controller > > - it is normally, supposed to find the slave disk, even if the master > is dead or non-existent Yup! But I mean, if the controller itself was damaged, you'd expect it to find neither... > > > However, if I pull the IDE cable from the > > primary master, the primary slave is detected. > > that probably means your jumpers on the disk is not right > - set the jumpers on the disk to be both cable select > or explicity set master on the disk at the end of the cable Hmmm, I think I have that right... *looks inside the running box*. Yup, jumpers are correct on both disks Also, the problem happens just randomly, on bootup (this is a noisy box sitting under my bed, I turn it off at night), and if it had anything to do with jumpers, I would be expecting consistent problems, not something that would appear after the box had been stable for a couple of years and with no changes... It feels more likely it is due to wear and tear, but what is getting worn...? > using master on the middle of the cable can create whackyness Yup, it is on the end. Slave is on the middle. > > The primary disk, a Maxtor 40 GB disk, has been a bit unreliable > > from the start, > > maybe you have a bad disk, but is it unlikely... > and also depends on where yo bought the disk from ... > mom-n-pop-me-too stores vs an iso9001 certified outfits where they > supposedly don't throw things around or do your 6' drop tests before > shipping Hehe, no, it was the latter, a it was a well-recognized store, but they were caught shipping disks that had been returned as new. People got disks with data on. Whooops... So, well, I can't be really sure, but at the time I bought this, 40 GB disks were the latest and greatest, so it isn't very likely they could have had time to ship it too many times... However, there are things... For one thing, I initially mounted the disk with the wrong screws, so it probably had a lot more vibrations its two first weeks than what was good for it. That's when I found the right screws. Also, Norway's electrical grid is really b0rked, so the chances this box has seen non-sinusoidal AC is pretty high... :-( > > 90% of the time .. its just bad cables and jumpers or bad mb > (controllers) Well, I have excluded cables and jumpers as the cause, so then, it is the controllers... But if it was the controller, why would it detect the slave when the master was dead...? Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
OT: Damaged harddisk and/or disk controller
Hi all, My question is, I suppose OT, since things go wrong long before bootup, but with the many fine folks here, I hope you will permit me, and perhaps even provide me with an answer! :-) For a long time, I have had problems before boot: My primary master disk is sometimes not detected, and thus, the machine doesn't boot. It happens about once a month, and it is completely magic to me: I usually pull the power cable, fiddle a bit with the IDE cable to the primary disk, reconnects and that does the trick. If these actions have an effect, I really do not know, and sometimes, like this morning, a lot more fiddling was required... I don't even know if it is the disk that's the problem or the disk controller, and that's my primary concern that I hope you can help with. What I see is that if the primary master disk isn't detected, the primary slave isn't either. However, if I pull the IDE cable from the primary master, the primary slave is detected. This too, could be random, as I have not seen it often enough to tell, but it seems so, indicating that the controller is fine. I have also bought a new IDE cable, so it is not the cable. Today, I pulled the power from the disk, and when I inserted it again, it gave a sound like "I'm spinning, I'm spinning!", and that was what did the trick, it seemed. Could it be that occasionally the disk simply doesn't start spinning at all after a shutdown? In the list of devices on bootup, the IDE controller always shows with a single entry, so it is not that the IDE controller is totally absent. lspci also shows it as: :00:04.1 IDE interface: VIA Technologies, Inc. VT82C586A/B/VT82C686/A/B/VT823x/A/C PIPC Bus Master IDE (rev 06) The primary disk, a Maxtor 40 GB disk, has been a bit unreliable from the start, so I suspect it is to blame. As some may remember, I have been thinking about buying a RAID controller for this box, and this is part of the equation, allthough I have pretty much suspended the idea for now. If the IDE controller is flaky, then I would want to buy a new SATA controller and a SATA disk, but if it just the disk, then I'd go for just a new IDE disk, but it would of course be sad to have gotten a IDE disk if the controller was bad... So, it is a financial question too... :-) Any ideas on how to attack this issue to bring more certainty as to the cause? Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: NFS shares over the internet
On tirsdag 18 oktober 2005, 21:05, David Dawson wrote: > the remote machine can ls the share, send and receive small files > (say 100 bytes) but when an attempt to send or receive a larger file > is made, the konqueror process reports 'stalled' and goes into > uninterruptable sleep. Uh, ouch. Well, I think conventional wisdom teaches that you wouldn't want to expose NFS over the open network, allthough I personally don't see many bad things when you've firewalled it off like that. However, you might want to have a look at SFS: http://www.fs.net/sfswww/ There are debian packages. It takes quite a lot of configuration, but works great for me. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: How to install R 2.2.0 Debian 'unstable' package in otherwise 'sarge' system
On onsdag 12 oktober 2005, 23:04, Dan Davison wrote: > What is the best way to do this? If I don't want to upgrade > to unstable, must I compile R myself, or is there some way to install > this as a debian package? Then, I'd recommend using the R's own backports, this line would probably do the trick: deb http://cran.us.r-project.org/bin/linux/debian stable/ and allthough it isn't there yet, it'll probably be there soon. However, I played with apt-build the other day, and produced my own build. It was quite fun, so reading up on apt-build can be a nice exercise. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: What to do with all the doc files in /usr/share/doc?
On onsdag 12 oktober 2005, 22:47, Jim Lynch wrote:
> For years now, I've been seeing the vast quantities of files in that
> directory and other than going in and zmoreing them, I haven't
> figured out how to read them.
I've been thinking about unzipping them all, but I figured it wasn't
worth the disk space.
But if you think that's OK, you could probably do it with
find /usr/share/doc -name '*.gz' -exec gunzip {} \;
or something like that...
Cheers,
Kjetil
--
Kjetil Kjernsmo
Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer
[EMAIL PROTECTED]
Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: gpg: export just my keys, not whole keyring?
On onsdag 12 oktober 2005, 13:36, Matt Price wrote: > interesting, the depths of my ignorance continue to berevealed to > me... I'm not sure what a keyid is! Ah, OK! Hmmm, I'm going searching... It seems you have an (old) key http://pgp.mit.edu:11371/pks/lookup?search=matt.price%40utoronto.ca&op=index and for that key, the KeyID is 80161724. It is just something that (with reasonable probability, possibly) uniquely identifies that key. Since there are subkeys and stuff (these things are beyond me), that have their own ids, this can get fairly complex. Also, you'll see longer keyids now, but to make a long story short, it is a hex number consisting of the last 8 numbers of you fingerprint. So, if you find your fingerprint, you've got it. DEADBEEF is the metasyntatical variable (i.e. example) keyID, since it is all hex and a funny word of the correct length. > But I did try this: > gpg --export "Matt Price <[EMAIL PROTECTED]>" > ring.gpg -- and > I ended up with something much smaller. So maybe that's good enough. Cool! :-) > But one thing is, I'm still not sure that the key I export is the one > I'm actually using to sign my debian packages & files. I know I've > made a number of keys inthe past (part of my wanderings in the dark) Right! :-) > and now I don't know which one is automatically used when I invoke > gpg. jeez, how do I figure that out? E, /me looks around the room for backup... :-) Try gpg --list-keys "Matt Price <[EMAIL PROTECTED]>" They will have dates and stuff. And you'll see the associated keyid. Also, you can have a look at mine gpg --recv-keys 6a6a0bbc should import it, and gpg --list-keys 6a6a0bbc should give you an idea of how it looks, 6a6a0bbc being the keyid. > > > oh, and finally -- if I figure out which key I want to export, do I > add it to the ubuntu-keyring just with > cat mykey.gpg >> ubuntu-archive-keyring ? > or will that screw with the binary file format somehow? I don't know, but I would be careful... I think I would do something like (untested) gpg --no-default-keyring --keyring ubuntu-archive-keyring --import mykey.gpg But then, I'm not quite sure you would want to import it in the ubuntu keyring... Perhaps you'd rather import the ubuntu-keyring into your pubring.gpg, if that's your default. I must admit that I'm a little rusty here, though. Also, I found, way back, that importing the whole debian keyring into my pubring was a bad idea, this was back in the day when GPG would totally lock KMails interface, and with the large Debian keyring, it would do it for half an hour or something. It won't be that bad now, but I also learnt that unmerging the most interesting keys was a non-trivial task, so nowadays, when I come across a key, I do gpg --no-default-keyring --keyring seen.pgp --recv-key DEADBEEF gpg --recv-key DEADBEEF to import it into both seen.gpg and the default keyring. > sorry for the ignorance. No problem. We've all been there. Still is, quite often, in fact! :-) Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: debian vs ubuntu and knoppix
On onsdag 12 oktober 2005, 11:53, Antony Gelberg wrote: > Of course, I'm also tempted to install Sarge with KDE, OpenOffice, > Mozilla etc. How un-user friendly can that be. ;) Not unfriendly at all! :-) If you are prepared to manage the box... It is installing and tweaking config options to be _Exactly Right_ that's hard in Debian, using it, when well tweaked is no harder than any other distro or Windows for that matter for someone reasonably intelligent and willing to learn. And given that you're their offspring... :-) My parents run Debian. The workstation has only Debian, the laptop has Debian and XP, but XP is hardly ever in use. I've set them up, and it works fairly well. There are issues that I haven't had time to corrrect, so they call now and then. I've promised I'll fix it for real soon... :-) And that works fine until your father decides he is a geek too, after all, and goes out to buy a new toy without checking the hardware specs and driver availability that we do... :-) Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: gpg: export just my keys, not whole keyring?
On onsdag 12 oktober 2005, 06:30, Matt Price wrote: > I get a huge file (don't have room for it!) > even : > gpg --export [EMAIL PROTECTED] > ring.gpg I think you need to use the KeyIDs, not the address, so, something like gpg --export DEADBEEF > ring.gpg would do it, I think... But yeah, gpg isn't that easy to use as of now... Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: automating lynx
On lørdag 08 oktober 2005, 00:24, Mark Lijftogt wrote: > > I am trying to get create a cron job that will go fetch a doc file > > from a particular web site. > > I would suggest to use wget instead of lynx in this case. Yep, I would suggest that too, but it sounded like the OP had to get links and then retrieve a specific page from that, and that adds to the complexity. If it gets really complex, I might suggest writing an Perl LWP script. The LWP packages are rather imposing, but once you look at it, the examples are excellent, so even with software that powerful, you can get some speed quickly. Hmmm, right now search.cpan.org had problems, but that's the place to go to read the documentation in a nicely formatted variant. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: EXIM Anyone can help me ?
On tirsdag 04 oktober 2005, 22:08, Sylvain MARTINS wrote: > I want to send mails from my network using my debian box and it's ok > if the recipient is local but not if recipient is non local :( > > i just want to authorize any hosts on my network to send mails all > over the world as they can do from webmail on debian box... Right. Hmmm, well, for one thing, you appear to be running exim version 3. I find it strange that it is included in Debian at all, because the developers has discouraged people from using it for at least a year. However, you might get away with installing exim4-daemon-light Furthermore, the thing you probably want is SMTP AUTH. That's supported by Exim. Exim 3 had it, but I never got it to work, but with Exim 4, it just worked, rather magically... Problem is, I can't explain to you how I did it, because it just suddenly worked, I don't quite know what I did... But it involves setting a file with passwords and stuff. Once you have Exim 4 installed, you could look in /etc/exim4/conf.d/auth/30_exim4-config_examples I think that's it. But takes some work, good luck! :-) Best, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: apache2 mod_perl sarge not working
On mandag 03 oktober 2005, 03:55, marek wrote: > Is there any documentation on how mod_perl2 differs in Sarge from the > official release? There is the renaming documentation: http://perl.apache.org/docs/2.0/rename.html which details the renaming that was performed on the Release Candidate that was sent out after the version that is in sarge, so you can interprete that document as "what didn't happen". Then, there is the changelog: http://search.cpan.org/~gozer/mod_perl/Changes The version in Sarge is 1.999_21, so there is everything above that. > In addition does anyone know when the "official" release of mod_perl2 > will be included in Sarge? Most probably, the answer to that is "never". However, all hope is not lost, because Norbert Tretkowski maintains a backport repository, and you can get the packages from there: mod_perl 2.0.1 itself: http://www.backports.org/pending/libapache2-mod-perl2/ and quite likely you'd want this too: http://www.backports.org/pending/libapreq2/ They are (AFAIK) not apt-getable yet. I have not tested these packages, but I believe they have not changed since he made the packages we run at work in production, which has proved themselves under very high loads. So, these are recommendable. > And could I help in getting the official release into Sarge? It is a highly complex question. The policy for updates to Sarge is here: http://people.debian.org/~joey/3.1r1/ Now, I'd argue that using these packages _would_ result in an overly broken system. However, if there are people out there, and there may well be, who have developed their apps knowing full well that the final release never made it to sarge, updating these packages will break their system. And even though I would never run the packages in Sarge in production, I can understand people making an argument that it should not happen in a point release. That's the stability you get, you'll never break a system that works by an upgrade. So, well, I think the simple solution here is to use the backports... Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: apache2 mod_perl sarge not working
On lørdag 01 oktober 2005, 03:47, marek wrote: > Thanks for pointing out my mistake in the directive. > I have corrected that and added the Options +ExecCGI. Now > the script is functioning. There's one thing you might want to note with mod_perl2 in Sarge, though: Debian has an alpha of mod_perl2 that had a different API than the final release. Thus, anything you develop using mod_perl2 in Sarge will not work for any others, and it will stop working once you upgrade to the official release, and anything any others develop for mod_perl2 will not work under Sarge. Therefore, I would strongly advice against using mod_perl2 in Sarge. I have discussed it with the upstream developers, and that's advice too. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: bayesian filter training question
On torsdag 29 september 2005, 21:51, Roberto C. Sanchez wrote: > So, I finally decided to get with the 20th century and install > spamassassin (acutally spampd hooked through postfix) to do site-wide > spam filtering for my server. Yiiihaaa! > My question is this. As I am training > it with sa-learn, is it (good|bad|indifferent) to train it on spam > that has already been flagged as spam. That is, will this reinforce > spamassassin's notion of spam or ruin it? No, that's fine. In fact, SA has this autowhitelist concept that does exactly that (it's not really a whitelist, though, more an "evening out weird things that may happen", I'm not using it). You should have a good look at bayes_ignore_header, so that it won't train on things that are obviously in spam. SA is pretty good it this itself, but if you see spam that has been filtered elsewhere a lot, be sure to use it. I'm guessing that you, like me, are doing this for your family. In that case, I have found that it is quite sufficient to train a single database with the spam and ham of the entire family. If you have more diverse users, you would probably need to have a per-user configuration. For example, a friend of mine has an uncle who is a psychiatrist working with people with gambling obsessions, and SA was pretty catastrophic for him until he got a per-user config. Finally, I found that SA, in it's default 3.0-form was much too conservative about the assigned scores, so I have a bunch of rules that I have adjusted the score of. You'll get some experience about that in time, I guess. Also note that SA 3.1 has been released upstream. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: Checking `bindshell'... INFECTED (PORTS: 3049)
On onsdag 28 september 2005, 07:45, [EMAIL PROTECTED] wrote: > ... > Checking `bindshell'... INFECTED (PORTS: 3049) > ... > > What I need to do ? Links are welcome. chkrootkit is a good thing to run, but often it can be scary. You will sometimes get false positives with it, and bindshell is the thing that most commonly jumps up. So, I'd just google first and see what shows up. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: Backing up installed packages.
On torsdag 22 september 2005, 20:57, R. Clayton wrote: > I would be interested in hearing opinions and suggestions about a > general approach to backing-up and reconstituting package archives, > as well as opinions and suggestions about the particular approach > I've outlined above. I haven't actually used it for anything, but I regularly do dpkg --get-selections > get-selections-2005-09-22.txt You should then be able to restore the system with dpkg --set-selections < get-selections-2005-09-22.txt if needed. I think, since I never had to restore. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: What can I do with six new publicly available computers?
On onsdag 21 september 2005, 15:59, Mitja Podreka wrote: > At the moment we are only offering web access. We would like to > create a place where people will come and do some creative stuff with > computers. What exactly depends on people itself. Sounds great! For one thing, you allready have trivial access to programming tools, compilers and stuff, for those who wants to get engaged in that. For people who want to get engaged in writing, there is of course Wikipedia. I think one of the more interesting things you could do are in the field of geographic information systems. You have, for example the OpenStreetMap project: http://www.openstreetmap.org/ Setting up a center to coordinate map production in your area would not only benefit your area, but would give you valuable experience the rest of the world would be interested in. There are many good mountaineers in Slovenia, and I know someone who is gathering up data for a complete global digital elevation model, and I'm sure Slovenian mountaineers could help in validating these data. As for clustering, a real cluster is probably not something you would pursue right now, but I would still recommend looking at things like Condor: http://www.cs.wisc.edu/condor/ I have just barely used it myself, many years ago, and I can't tell the state of it now, but it is under active development. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Experiences with LSI / Adaptec cards
On tirsdag 20 september 2005, 08:34, David Harrison wrote: > I've been googling around to find out which HW RAID cards work best > with Debian and from what I can see the LSI cards seem to do best. I recently went through the same thing, but probably with a different objective than you, I just wanted a cheap thing on my desktop. Allthough it doesn't seem to mention the cards you listed, I found http://linuxmafia.com/faq/Hardware/sata.html to be very interesting. As for HW RAID, I found that 3ware seems to be the thing. Actually, I recommend looking in the documentation of the kernel source. I found that an interesting resource too... :-) Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Mirrors for security.debian.org?
On tirsdag 20 september 2005, 06:15, Joe Smith wrote: > Q: Why are there no official mirrors for security.debian.org? > A: The purpose of security.debian.org is to make security updates > available as quickly and easily as possible. This policy is quite understandable, but as an aside: Anybody know if the developers have started preparing a point release? With both the number of packages in proposed-updates and the size and popularity of some of them, notably X and Apache, it would have been nice to have a point release, both for the mirror problem and for installation from CDs onto low-bandwidth systems. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Install USB Mouse, Lose Keyboard
On mandag 19 september 2005, 23:12, Mike McCarty wrote: > However, Debian no longer recognizes her keyboard. It sees the > mouse fine, but not the keyboard. Windows still sees and uses > both the mouse and the keyboard with no problem. Weirdness. With my job computer, a Dell box, the order of which I inserted the mouse and keyboard and the order of the ports that I inserted them in actually mattered... I thought part of the point with USB was that you shouldn't worry about stuff like that, but... You might want to try different combinations of those. Also, you could try cat-ing the devices and see if anything happens when you press keys or move the mouse. This may amount to do as root: cat /dev/input/* Best, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: digital camera software
On mandag 19 september 2005, 10:12, Bayrouni wrote: > I need nice software for my digcameraital . > But I have'nt any idea which software is fine for accessng, reading, > ... all enabled info and data on the camera. If you use KDE, I can recommend digikam. Even though there are many features on my wishlist for it, I still find it usable and rather cool. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Apache2: httpd.conf or apache2.conf?
On mandag 19 september 2005, 18:13, Jared Hall wrote: > No, httpd.conf is not obsolete. Anything you stick into it will > still operate as it did before (default configuration). Yep, you're right, and indeed it is Included. I also found that I enabled RewriteEngine in there... :-) It might look as if you should not need to modify apache2.conf, but put configuration that's valid for all your virtual hosts in there, and leave apache2.conf pretty much as it is. Best, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: problem with automatic upgrade (changed conffile)
Now that the original poster has found is answer, I guess an aside is ok, On mandag 19 september 2005, 17:56, Angelo Bertolli wrote: > But if the is compromised, won't they also have > the key? No, they would have to get to joey's (or another member of the security team's) secret key, and compromise his passphrase. I doubt he has his secret key on the Debian machines, and even if he had, it should be hard to get the passphrase. There'll be more hops for them to go through to score... Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Apache2: httpd.conf or apache2.conf?
On mandag 19 september 2005, 06:53, Steve Dondley wrote: > The README file that came with Debian doesn't cover the difference > between the httpd.conf file and the apache2.conf file. Is the > httpd.conf file basically obsolete? I haven't had time to look into it at depth, but I think the answer is yes. I've got everything in apache2.conf file, friends said it should be there, and independently at work, they've deleted httpd.conf entirely. So, that's the safest bet, I think. (CC-ing since you set the Reply-to) Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: problem with automatic upgrade (changed conffile)
On mandag 19 september 2005, 00:18, Hans Ekbrand wrote: > > You must > > expect issues like these, it is a feature... :-) > > Not getting security updates automatically installed a feature? Not > in my world! Well, imagine the security.debian.org box getting compromised, and the attacker pumping out a trojanned "security" upgrade. You install it automatically before the Debian folks take the box out. The attacker has your IP too... That's a serious single point of failure for the entire community, you know... I prefer to read and understand the DSA, and check that the DSA is signed with a key I trust (I'm just a hop from joey) before I do a manual apt-get upgrade on affected machines. But YMMV, that's just me. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: problem with automatic upgrade (changed conffile)
On søndag 18 september 2005, 23:34, Hans Ekbrand wrote: > I have changed /etc/kde3/kdm/kdmrc, but so what? Why does dpkg fail > here? There are certain limitations to doing things automatically, and in this case, it detects certain changes that it needs your input on. It is a config files that's changed. I believe it is possible to tweak something to make it do more things automatically, though. You must expect issues like these, it is a feature... :-) Note that there is a package cron-apt, that you may want to install instead. It can be configured to do automatic upgrades. However, I know that not everbody thinks it is a good idea to do it, and IIRC the documentation of that package has a discussion about it, you should probably look into it. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: UPS recommendations?
On lørdag 17 september 2005, 17:01, Colin wrote: > Then get an APC one. I have two of them for my systems (LS 500, CS > 500) and they've been working great. The best part is the linux > software you need is already part of Debian. The package is called > "apcupsd". That's pretty neat! Since it is brought up, does anybody have experience with Inform UPS Guard: http://www.informups.com/pages/prod/guard.html They are really cheap here (cheapest I can see), and I've heard they are quite straightforward to get working under Linux. Also, does anybody have experience with the internal UPSes and PSU here: http://ups2.com/ If I was building a new system, I'd really want one of those I wonder why it isn't fairly "standard". Probably because for the dominant OS power outages aren't the only reason for downtime... :-) Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: Overwhelmed newbie
On lørdag 17 september 2005, 13:23, Antony Gelberg wrote: > > Oh, there is plenty of reason. Stable has its issues, as any other > > complex software system. > > Ok, what reasons are there to get burnt by using stable? The point > of stable is that it is very unlikely to burn a clueful user. For the newbie, the 30-something known (by me) crashbugs in KDE is likely to bite him (say, for example the bug that crashes KMail if you delete all mails in a folder fast, #285794). Yes, I'm using KDE despite of this. The person looking to Debian as a server platform may be bitten by the fact that Debian contains a mod_perl 2 release that's incompatible with everything else out there. Code developed for mod_perl 2 on Debian won't work in the rest of the world and vice-versa. Just two examples that even a clueful user would have to deal with. No, I'm not complaining, for a system of Debian's size, it is only to be expected. But I'm thinking about Debian stable more in terms of "won't suddenly jump up an bite you unexpectedly because something changed" rather than "everything will always Just Work and work reliably". > So what? The OP didn't say he wasn't willing to learn. We could > give him the option. Of course, of course. And I very much respect your dedication towards this, and indeed the OP indicated that he was willing to learn. However, I suspect that what it would take for Debian to be a recruitment platform to free software, it would mean that if a newbie silently gives up, someone will have to send him an email and ask how it goes, and drag him back in. I suspect few even have the time to do that. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: Overwhelmed newbie
On fredag 16 september 2005, 21:09, Antony Gelberg wrote: > Please reply onlist. Errr, well, I don't consider this an important topic for the list, and it is also rather rude to respond to a off-list message on-list. Please refrain from that in the future. > It's because if people use stable, there is no reason to get burnt. Oh, there is plenty of reason. Stable has its issues, as any other complex software system. > There is no reason to be overwhelmed by anything if the user is > willing to learn. I understand that Debian is not geared at newbies, > but that doesn't mean that newbies can't use it. Well, speaking as someone who has his parents on Debian, I can attest to that newbies can use a Debian system fine. I also speak as someone who did his first Linux install ever with Debian. Been there, done that, didn't get a lousy teeshirt. However, in my parents' case, it is because I have made most of the complex, technical choices for them, and in my case, it was because I was dedicated and had a good friend and experienced sysadmin by my side throughout the process. Frankly, I find that to be efficient on Debian, one should read many hundred pages of documentation. I'm perfectly fine with that, because I realized from the start that it would pay in increased productivity once I'm up to speed. However, one must acknowledge that this is an imposing task for most users. > Most importantly, this is _debian_-user. If you want to advocate > other distributions willy-nilly, it's not the place. That kind of isolationism is something I think you will find very little support for in the free software community. I think most people will agree that we want to direct people to what is best for them. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: Overwhelmed newbie
On fredag 16 september 2005, 14:20, Fritz Brown wrote: > Help! I have recently begun an attempt to install Debian on a Sony > laptop (Mobile AMD K6-2 550MHz, 64MB RAM), and am thoroughly > overwhelmed with choices about which I know nothing! Yeah, I can really see that. It is huge, and not easy to gain an overview of. Others mention Sarge, the 3.1 version, and yeah, I wouldn't install Woody, for a desktop, it was allready outdated when it was released. But in fact, I would rather suggest you install Ubuntu, which is not Debian, but based on Debian. Have a look at http://www.ubuntulinux.org/ They'll send you a free CD too! Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: Sarge problem with Matrox video driver ?
On torsdag 15 september 2005, 16:40, michael bailey wrote: > Running Debian Sarge on an Intel P3 machine, twice in > the past two weeks the screen has frozen completely > and the Caps Lock and Scroll Lock green lights have > started flashing rapidly. The mouse and keyboard then > both don't work and you can't move anything on the > screen. Oooo, I'm seeing exactly the same problems, and had set aside the evening to make a writeup about the problem. My card is a dual-head G450, and I use both heads. I have exactly the same symptoms as you describe, except my LEDs don't blink and it doesn't happen 100% of the time, only 90%. Also, I had been running sarge for about 1.5 year before the release, and the problem was unknown to me before the last weeks before the release. In fact, I suspect it sneaked in with the last X upgrade in Sarge. Like you, I can hit Ctrl+Alt+F1 and make the system freeze. Moreover, running halt as root can freeze the system too, not only logging out through the window manager. I have been exploring many different possible causes, but has not found a single thing that can reliably let me shut down the computer. I have been thinking that it may be because I run my dual screens in non-Xinerama mode (because I like two independent virtual desktops better than one big virtual desktop). Also, there was this bug in artsd that could freeze systems on logout, but I shouldn't be affected, and I even applied some workarounds for it. Still, I have been trying to find out if it could have been something with having sound when logging out. Since it has been a daily problem now since (before) Sarge was released, I have tried so many things, and I'm pretty much out of ideas. Also, I have tried to look in logs, but I really can't see anything suspect in any of the system logs. So, I have just praised journaling filesystems, and hoped for the best... It is interesting to note I'm not alone, though, and I hope by joining forces and with the help of the fine folks here, we can figure it out. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC pgpDs9Jirrbqd.pgp Description: PGP signature
Re: to not apt-get upgrade certain packages
On torsdag 15 september 2005, 03:34, Roberto C. Sanchez wrote: > For dpkg and others you will need to read the man > pages, or maybe someone else can explain how. Yup, I think echo "libwine hold" | dpkg --set-selections echo "wine hold" | dpkg --set-selections echo "wine-utils hold" | dpkg --set-selections will do the trick. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Hardware RAID advices needed
On lørdag 10 september 2005, 23:02, Hans du Plooy wrote: > Kjetil Kjernsmo wrote: > >The first is a dirt cheap Sunsway Sil 3112-based card, i.e. > > fakeraid. > > How "old" is your PC? Pretty exactly 4 years old. It has the last generation of Athlon T-Bird CPUs. > On my 2ghz AthlonXP I have onboard VIA > (fakeraid) plus a PCI Sil 3112a card. Raid-5 over four SATA discs > give me in excess of 50MB/s sustained write, depending on what I'm > copying, with a system load of about 2.00. That's pretty neat. > >The second option is a 3ware 8006-2PL, > > You can't go wrong. Cool! :-) > >The third option is to buy a SATA non-RAID controller and run SW > > RAID entirely based on that. > > I'm not sure if you even get that. I've never seen a SATA controller > that didn't have Raid. I think they exist. This, for example: http://www.adaptec.com/worldwide/product/proddetail.html?prodkey=ASH-1205SA > But then, if you think about it, it's all fake > raid, so all of them are non-raid. Yup. I found that even those cheapest SATA controllers that were advertized as non-RAID had a SiL3112. Grok it those who can. > http://cgi.ebay.com/3ware-8006-2LP-64-bit-PCI-SATA_W0QQitemZ523855901 >7QQcategoryZ39968QQssPageNameZWDVWQQrdZ1QQcmdZViewItem > > If your data is worth $45 to you, then this will see you through. OK, cool! Thanks a lot for all the advices, now I know a lot more about my options. Still have to count my pennies, though, and figure out what I really want. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Re: ServerAlias considered harmful
On mandag 12 september 2005, 17:47, Angelo Bertolli wrote: > Ok, but what's the Apache configuration answer to this? I write out > my virtual hosts without using mod_rewrite, and I really don't want > to write TWO virtual host blocks just for the sake of redirecting > domain.com to www.domain.com. I'm afraid that's been my solution :-( It is somewhere on my TODO list to make a writeup on this, but I'm afraid it has been pushed back... Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
ServerAlias considered harmful
On mandag 12 september 2005, 02:07, David Clymer wrote: > but wouldnt it > be much easier to add a ServerAlias to the vitualhost config? Without trying to respond to the original question, I would like to point out that ServerAliases are vil and architecturally broken. The redirect should have been a 301, otherwise it results in URI aliasing, which is bad: http://www.w3.org/TR/2004/WD-webarch-20040816/#uri-aliases I asked Dan Connolly (one of the main web architects) about ServerAlias once, and he said he could see valid uses for it, but the general use as it is now is just architecturally broken. To see why, my previous project at work involved finding and counting which URIs of what sites is broken in Opera, and that was complicated by widespread URI aliasing, it was simply impossible to count that reliably. Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Hardware RAID advices needed
On lørdag 10 september 2005, 12:52, Hugo Vanwoerkom wrote: > off your back? Eh, right! > > You did not get any recommendations. Your reason for purchase is > appealing. But the price differences are huge. Yep, I didn't get any very specific recommendations, but a lot of good information, many thanks to all respondents, so... > What are you going to do? I haven't actually decided. However, I'm still evaluating three options. The first is a dirt cheap Sunsway Sil 3112-based card, i.e. fakeraid. Sunsway produces mostly low-end stuff, but I'm very satisfied with the (two) USB2+FireWire card I have from them, so, well, I still consider that possibility. The second option is a 3ware 8006-2PL, which seems excellently supported and has good reviews from everywhere. Right now, this is actually what I find most appealing. However, the only local store that I've heard favourable things about doesn't have it in stock right now, will get in three weeks, which gives me more time to think. The third option is to buy a SATA non-RAID controller and run SW RAID entirely based on that. There are some high-endish cards that support expansion and level migrating, notably a HighPoint card. However, these are not well supported it turns out, and so it isn't very appealing right now. Also, my CPU and mobo is now 4 years old, and it may soon need an upgrade, future extensibility might not be such a big concern after all. The 3ware solution is rather expensive, but not scarily so, that's the main downside to it, and of course, the concern "what if the card breaks", a situation which isn't quite clear. Further advices are still appreciated! :-) Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC
Hardware RAID advices needed
Hi all! I'm finally back after getting a nasty project on my back, and I have a question... I'm thinking about buying a hardware RAID controller for my desktop. The main reason I don't want to go with software RAID is that my mobo IDE controllers are allready full with other stuff, and one of them may have a problem. So, I would need some kind of controller anyway. I'm not doing it for performance, just for reliability. Yeah, I have additional backup, but I like the idea that the data will be accessible even if a disk crashes. So, that's main priority. Now, what I was thinking: If I buy a four channel RAID card now, and two S-ATA disks (which is what I want), set them up in RAID 1, can I add a third disk later, and convert this system seamlessly to RAID 5 without a reinstall? Every Linux geek I know says "no, I don't think so", but I think my old sysadmin did that routinely on Tru64 years ago... If a disk blew up, he'd just throw it away and replace it, no worries. I'll go with a HW RAID even if I can't do exactly this, but then I think I'll go with a 2-channel controller. Can I have your recommendations, please, for a controller for a simple desktop system, with the main object of having redundancy in case of a disk crash? I think I'll go with the cheapest there is for the task. I've seen a 2-channel RAID 1 Adaptec AAR-1210SA, that's a really cheap one, I can go a bit higher than that. The cheapest 4-channel RAID 4 I've seen is HighPoint RocketRAID 1640 S-ATA. Same range is a number of Promise FastTrak cards. Then there's the Adaptec 2410SA, which I have seen recommended, but I think that if I go that high, it should be able to do the RAID 1 to 5 seemlessly thing, or have a good reputation for taking good care of my data. I've heard that some HW controllers can make it really painful to access the data, even if just one disk crashed. Is this a problem with any of the mentioned? Finally, would you buy disks from the same or different manufacturers? I would guess that similar disks are easier on the controller, but OTOH if there is a flaw in one, it is likely to be in the other too Your recommendations are most welcome! :-) Cheers, Kjetil -- Kjetil Kjernsmo Programmer / Astrophysicist / Ski-orienteer / Orienteer / Mountaineer [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/ OpenPGP KeyID: 6A6A0BBC pgphPeAfzjLgk.pgp Description: PGP signature
Upgrading a libdb2 db to PostgreSQL
Hi all! I've killed my largish and very effective bayesian spamassassin db... :-( The problem is described in Debian Bug #188997 (and duplicates), so it is a well-known problem, but I haven't been successful in working around it, so I seek your help in doing that. Briefly, the problem (as I understand it), is that perl 5.6.1 was linked against libdb2, whereas perl 5.8.4 is linked to libdb4.0, and there is no simple upgrade path between them. SpamAssassin itself would have an nice upgrade path if I just had perl 5.6.1 and could read the db, but I can't... I tried to do it with the db4.0_upgrade program as suggested, it didn't work, but really what I'd like to do is get the db from libdb2 to PostgreSQL, so I might not need to go through libdb4.0 anyway, so I haven't spent much time on that. I suppose there must exist a way to get the data out of the libdb2 database (it is still on my box) other than going through Perl, but I don't know enough about how stuff like that work... Can anyone suggest a way to do this? Cheers, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: chkrootkit output
On fredag 20. august 2004, 09:32, Adam Funk wrote: > /usr/lib/blender/.bfont.ttf > > Why are these files flagged as suspicious, They all begin with a period, so they don't show with ls (you need ls -a) to see them. dot-something are frequently used to store config information in user's home directories, but have little use beyond that, AFAIK. But since they are slightly hidden, I would guess it is the only reason why they are flagged. > and should they concern me? Can't give an authoritative answer to that, but usually, no. However, you might just have a look and them and see if they look suspicious... Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Configuring CUPS to allow all users to (re)start printer
On fredag 20. august 2004, 16:02, Tim Kelley wrote: > Yes; configure the "admin" location in cupsd.conf to allow valid > users, then they will be able to use their user accounts to do that > ... > > > AuthType Basic (or Digest) > AuthClass User Uhm, OK, thanks. So there isn't any way to narrow it down to the specific action of just starting the printer (which is really what I was looking for)? > Though this opens up a whole new can of worms ... Yup. It doesn't really have security issues, since all valid users are trusted, but they can manage to mess things up... Oh well... > However, the real problem is why do they have to restart the printer? It is sort of weird: If for some reason the printer is turned off or is otherwise not responsive, and a printer job is sent (or in progress), the printer will be stopped, and clearing out the problem will not help. Even a reboot won't do it. This is a HP OfficeJet T65, so it is not among the Best Supported Printers[tm], I suppose. Cheers, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sending RBL mail to email account instead of deny
On lørdag 21. august 2004, 07:29, Kent Andersen wrote: > is there any way to tell exim to forward RBL blacklist mail to an > email account for further review? Dunno, but in general, I advocate the use of e.g. Spamassassin to evaluate more aspects of an email than a single blacklist. You may set several different thresholds, and deny if it is above a certain score. I feel that single, absolute rules are a Bad Thing[tm]. You would need Exim4 to do this, though, there is documentation at: http://duncanthrax.net/exiscan-acl/ Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC
Re: query about logrotate
On fredag 20. august 2004, 10:29, vivek misra wrote: > acctually I am making a log rotate for log files and ionly want that > all data except last six days are removed how could I done this, > Is log rotate file is placed in crontab If I understand your question correctly, you want to keep logs only for six days? I would suggest achieving this with a combination of daily and rotate directives. From the man-page: daily Log files are rotated every day. rotate count Log files are rotated count times before being removed or mailed to the address specified in a mail directive. If count is 0, old versions are removed rather then rotated. So, I think if you set daily and rotate 6, it would do what you expect. All this happens in /etc/logrotate.conf and /etc/logrotate.d/ Please have a good look at the files in there, they are rather self-explanatory once you know where to look. You may have to edit many files in /etc/logrotate.d/ Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Configuring CUPS to allow all users to (re)start printer
Hi all! My parents have the problem that the printer sometimes stops itself for various reasons. It is easy enough to restart it with the KDE print system, just go "Start Printer" (or something like that). The problem is, it asks for password, and unfortunately, it will only accept the root password. I'm not going to give them root access, (the terrible mess they would do with a root password...), but they shouldn't have to call me every time the printer stops... Instead, I'd like to allow all local users to restart the printer if it stops, but I haven't find anything concrete in the docs to do that, and while I found it easy enough to set the printer up with CUPS, it isn't actually very transparent when you need to do something a bit more out of the ordinary... :-) Anybody know how to do this? Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: xfree86 4.4.0
On fredag 2. juli 2004, 15:24, jack kinnon wrote: > Is Xfree86 4.4.0 available for Debian stable? Apart from that 4.4.0 has license problems and will probably not go into Debian in any form, the answer to such question is usually a search on apt-get.org. If you can't find it there, the answer is most likely "no" :-) Cheers, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC
Re: install: kdelibs3 - error in unpacking
On torsdag 1. juli 2004, 12:45, Alexandr Rosen wrote: > I am new to linux and debian and kde, and I have a problem > with installing the system - the message > > kdelibs3 - error in unpacking > > occured during install, and "dpkg --configure kdm" says that > errors were encountered while processing. Is there any easy > solution? Hi! This isn't much of an answer, I guess, but perhaps better than nothing: Debian is sort of three different distributions, stable, which is really stable, testing and unstable. Stable is for those who do not want to spend a lot of time figuring out things that are wrong. It just works, with no glitches. Testing and to an even greater extent Unstable is bleeding edge. They are probably among the most up-to-date things you can find, but as you've noticed, you'd bump into issues now and then that are harder to analyze. For that reason, they are mainly for testing and development, for those who can address that kind of issues easily on their own. There's a page about KDE in Debian at http://wiki.debian.net/?DebianKDE but I can't see anything of help to you there. If you can't perhaps Debian is not for you. It can be pretty tough to work with, and there are many things that need to be learnt. If you want to learn it to the bottom, there will always be people willing to help. I started right into Debian as a newbie, but I had 7 years of experience as a UNIX user and a good friend in meatspace to hold my hand... :-) I suppose that since you're new, you would probably want to find a flavour of Linux that's easier to get started with. People here tend to recommend Xandros http://www.xandros.com/ or Libranet http://www.libranet.com/ as they are both based on Debian. Cheers, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Getting no sound from ALSA under 2.6.7
Hi all! I finally decided it was time to upgrade to the 2.6 kernel. So I have compiled myself a kernel, based on Linus' 2.6.7 with the Ruby patches. I need those for my local multiuser setup. It all seems to work very nicely, except that I lost my sound. The problem is, I'm not really getting any error messages I'm able to see. Everything seems quite nice, it is just that I'm not hearing anything... Has anybody got ALSA working with the alsa-base and alsa-utils currently in Sarge? There are some examples lacking (bug reports have been filed on those, I'm sure they are working on it), so I'm feeling I'm just blindly trying different things but having no clue what they are supposed to do. I guess a good start is to cite the output of my lsmod: Module Size Used by appletalk 35700 - ipx28688 - p8022 2244 - psnap 3912 - llc 7704 - nfs 105556 - lockd 63816 - sunrpc152100 - ppp_deflate 6148 - bsd_comp5924 - ppp_async 12068 - ipv6 247040 - ppp_generic29080 - slhc7748 - ipt_MASQUERADE 3716 - iptable_nat25040 - ipt_LOG 6372 - iptable_filter 2724 - parport_pc 34432 - parport41032 - snd_cmipci 32996 - snd_pcm_oss53832 - snd_mixer_oss 19332 - snd_pcm97028 - snd_page_alloc 11468 - snd_opl3_lib 10980 - snd_timer 25000 - snd_hwdep 8928 - snd_mpu401_uart 7812 - snd_rawmidi24928 - snd_seq_device 8012 - snd53412 - apm17904 - 8139too24356 - mii 4868 - We see snd_cmipci, and cmipci is good, that's the driver for my sound card. First I tried compiling most stuff into the kernel, but the one I'm running now, the whole sound system are in modules. How does this look? I've been experimenting with files here and there to get it to understand that I'm using ALSA 1.0, for example: [EMAIL PROTECTED]:/etc/alsa> find . . ./modutils ./modutils/0.9 ./modutils/0.9/modules-0.9.conf ./modutils/0.9/file4k6mBy ./modutils/0.9/fileygRrGd ./modutils/0.9/1.0 ./modutils/0.5 ./modutils/1.0 ./modutils/1.0/1.0 ./0.9 And symlinks here and there: [EMAIL PROTECTED]:/etc/modprobe.d> ls -l [snip] lrwxr-xr-x1 root root 22 Jun 30 15:12 alsa -> /etc/alsa/modutils/1.0 -rw-r--r--1 root root 86 May 2 12:50 alsa-base ...and [EMAIL PROTECTED]:/etc/modutils> ls -l [snip] lrwxr-xr-x1 root root 22 Jun 30 15:09 alsa -> /etc/alsa/modutils/1.0 -rw-r--r--1 root root 26 May 2 12:50 alsa-base -rw-r--r--1 root root 121 Apr 6 2002 alsa-path Then, I've ran alsaconf, and it just comes out telling me everything is OK. Also, I tried to insert some magic in one of these 1.0 files and run dpkg-reconfigure. It too told me everything was OK. update-modules and restarting ALSA I've done it all... But not in a very structured manner, since I don't have much idea what all these things are supposed to do... :-) Then, there's the mixer. I've yanked everything I could up to max. Interestingly, as I push the volume of the "PC speakers" up to 100, I (increasingly) hear a lot of noise on the speakers. Furthermore, it is clearly generated by the PC, as the noise follows the noises from the box, e.g. hard drive noises... So, it seems there is something coming out from the sound card, but it isn't the music I was expecting to hear... :-) For some time, I thought it was mostly that I didn't get modules loaded or something. I supposed that that's what the missing examples are about. But from the lsmod, to me it looks like they are... I've got quite a lot of trouble with arts lately. I'm using KDE, and usually play music with noatun, and arts seems to have some problems at startup, killing itself and stuff. However, I've been testing with playing "known good files" with alsaplayer, so this shouldn't have any impact, I thought. I've tried ALSA before, with varying degrees of success, I've got it working for some periods, and some periods not, and I can't really point out why. Recently, I've been sticking to OSS. Except for just checking that I could compile and boot 2.6 in the late -testing period, it is the first time I've tried 2.6. Finally, I saw some RC bugs have recently been closed. Anybody know if those would have had any effect, and if we'll see new uploads soon? Any hints, please? Best, Kjetil -- Kjeti
Re: Where do I start?
On Tuesday 24 February 2004 14:11, [EMAIL PROTECTED] wrote: > I would like to build my own Linux flavour. Where would be good > places to start? I think http://www.linuxfromscratch.org/ is the place to start if you're serious about that. Cheers, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: better mail organisation
On Friday 20 February 2004 13:06, martin f krafft wrote: > One of last year's Sysadmin issues (or was it USENIX's ;login:?) had > an article on mail organisation. The short story was that the guy > had his mail system configured in such a way to automatically > maintain a folder hierarchy of correspondents, where each such > folder receives all mail exchanged (sent and received) with that > correspondent. I remember seeing something like that too, and when I got KDE 3.2 installed a few days ago on my laptop, I noticed KMail had "Search Folders". I just briefly looked at it thinking "oh, so they implemented this", had a little bit of look around, before I went on to more urgent matters. I didn't explore it in any depth, and so I don't know if it does what you intend, but as I said, it appeared to me to be at least intended as an implementation of these ideas. So, you might want to check it out. Cheers, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Unintended loss of spam...
Hi! Being used to the steady stream of e-mail from debian-user and other lists, the natural reaction when I came back from lunch to "No new messages" was "WTF?" So I began looking into logs, and while it seems that there were no messages between about 14:01 and 14:51 (that's OK), there is one message that I haven't seen: 2004-02-13 14:34:23 1ArdSU-00056O-Nw <= [EMAIL PROTECTED] H=pat.uio.no [129.240.130.16] U=7411 P=esmtp S=5086 [EMAIL PROTECTED] 2004-02-13 14:34:23 1ArdSU-00056O-Nw => kjetil <[EMAIL PROTECTED]> R=local_user T=mail_spool 2004-02-13 14:34:23 1ArdSU-00056O-Nw Completed The message ID smells of spam, and while I have exiscan-acl to reject spam above 12, it will state it plainly if it does. In the syslog, it seems that it was indeed marked as spam: Feb 13 14:34:18 pooh spamd[13139]: connection from localhost [127.0.0.1] at port 40657 Feb 13 14:34:18 pooh spamd[19617]: info: setuid to nobody succeeded Feb 13 14:34:18 pooh spamd[19617]: checking message <[EMAIL PROTECTED]> for nobody:65534. Feb 13 14:34:23 pooh spamd[19617]: identified spam (6.9/5.0) for nobody:65534 in 4.7 seconds, 3380 bytes. So, it seems I lost a spam message. So what, you may say...? :-) Well, I don't like the idea of loosing messages without intending to... I have seen it err at the opposite end as well, failing to scan spam, it happens with about 1 in 5000 messages, I would estimate. Easier to live with, because at least I know when it happens. I'm using the backports.org package: 2.63-0.backports.org.1 Well, it isn't a lot to go on here, I guess, but I thought I'd share it just to hear your opinion. Cheers, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: apt-get sources.list
On Friday 13 February 2004 11:53, Hans du Plooy wrote: > What do I specify in sources.list to get access to *all* the packages > in http://www.gonkgonk.com/mirrors/backports.org/debian/dists/woody/ > ? > > I currently have: > deb http://www.gonkgonk.com/mirrors/backports.org/debian stable > spamassassin postfix I think you want deb http://www.gonkgonk.com/mirrors/backports.org/debian stable all Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sarge dist-upgrade wants to remove KDE
On Thursday 12 February 2004 18:09, Adam Aube wrote: > On Thursday 12 February 2004 11:20 am, Kjetil Kjernsmo wrote: > > I'm tracking Sarge now, and for the past couple of days, when I > > tried dist-upgrade, it has wanted to remove pretty much every KDE > > package on my system... > > KDE is currently broken in Sarge. The 3.1.5 packages coming from Sid > should fix this soon, OK, yeah, I knew that it was broken in Sarge, but I was surprise that this would happen just now, as the first 3.1.5 packages was going into Sarge. I was tracking Sid up to new years eve, and since then I have been tracking Sarge, and haven't seen any particular problems since then. > Add this to your sources.list: > > deb http://snapshot.debian.net/archive pool lm-sensors > > The problem is a dependency of ksysguardd was removed from Sarge, > which this line will provide. Thanks, but really, that is a problem I corrected long ago, I think I fixed that problem when I first upgraded to Sid (after having a mostly-woody system since I got into Debian as a newbie). But I am OK, really, I just need to avoid a dist-upgrade, since I have KDE working well. So I was just curious if it was some reason I was seeing this now. Cheers, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Sarge dist-upgrade wants to remove KDE
Hi! I'm tracking Sarge now, and for the past couple of days, when I tried dist-upgrade, it has wanted to remove pretty much every KDE package on my system... Like: [EMAIL PROTECTED]:/home/kjetil> apt-get -s dist-upgrade Reading Package Lists... Done Building Dependency Tree... Done Calculating Upgrade... Done The following packages will be REMOVED: ark artsbuilder brahms cervisia flashkard kaboodle kaddressbook kalarm [etc] Some of these packages are very recently installed, some are older (probably from before I upgraded to sid around new-year). For example (I did an -s to see): Remv flashkard (4:3.1.5-2 Debian:testing) Remv kaboodle (4:3.1.2-1.4 ) Anybody else seeing this? I suppose there is some conflict here, how should I go about finding out how to resolve that? Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: whew
On Wednesday 11 February 2004 18:07, David wrote: > On Wed, 11 Feb 2004 03:51:37 -0500 > > "~Honorable Dr Lou Who" <[EMAIL PROTECTED]> wrote: > > What a mess! > > [snip miles of babble] > > Uhhh... this guy's _gotta_ be trolling, hasn't he? Yup. > No one could be this clueless.. could they Well, 1500 partner websites... Where have I seen that before...? > However, this could be a prime example of the class of people who > definitely _should_ stick with windows.. And finance MS' robbery of our rights, including our right to receive and impart information and ideas through any media and to participate in the cultural life of the community, to enjoy the arts and to share in scientific advancement and its benefits? As much as one hate the trolls and the abusive newbies, the alternative, that they continue to pay MS money and allow them to retain the power to strip us of our rights, is so much worse that I feel it is not really an option. Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: How to Print to Printer from g++ program?
On Wednesday 11 February 2004 06:27, Thomas H. George wrote: > > If this question is regarded as wildly off target for this list, It probably is, and so, I'll not respond directly to your question, but rather ask a new one... :-) > As a physicist I wrote large programs in Pascal before retiring in > 1994 but nothing since until last week. I thought of a problem I > wanted to solve, read up on C++ and got my program working with only > a few struggles with syntax Great! > but how to write directly to the printer has eluded me. I see. Well, I know very little about C++, but I also wrote a program some time ago in Pascal that produced pretty graphics on HP LJ III-series of printers. Nowadays, I think you would use a library that provides you with an abstraction layer to the printer. I have no experience there, but I think Qt has a print library, and it is written in C++. However, are you sure that you are using the right tool for the job? The nice thing about the Free Software community is that most problems are solved allready, the code is available, and it has undergone extensive though informal peer review. If it is a science problem, I would strongly recommend the R system. I used it for my thesis in astrophysics, and it is really very good. It has a very 1-to-1 correspondance between math and code, and makes any code you write easy to understand and it is very easy to get a good overview. It is really a statistics system, but as far as I'm concerned, it is good for any kind of science. It has a very strong community consisting of some of the main authorities in numerical statistics developing it. See http://www.r-project.org/ It is also packaged in Debian, just go apt-get install r-recommended. It has a large and improving graphics library that will let you make good graphs of your data. It also has bindings to C (and through that, C++) and FORTRAN, so you can reuse existing code within the R framework. If it is not a science project, well, I find that programming at the relatively low level of C++ is usually quite painful and seldomly necessary. I tend to return to Perl programming whenever I need something done. There is a bunch of Postscript modules on CPAN: http://search.cpan.org/search?query=postscript They will probably assist you greatly, should you choose a Perl path. Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Downgrading to recent mailman security release
Hi all! I'm in a bit of trouble concerning installing the recent security upgrade of Mailman. A few days after the release of Woody, I had still "testing" in my sources.list, so, I got a few upgrades that I didn't intend to have (yep, I was very new at the time). Probably, among those, Mailman 2.0.12-1. The recent security release of Mailman is 2.0.11-1woody7, so if I want that, I need to downgrade. I presume that the version I have, allthough newer, is vulnerable to the problems mentioned in the advisory. So, I downloaded the deb of the security release and went dpkg -i on it, but it refuses: Downgrade detected, from version 0x2000cf0 to version 0x2000bf0 This is probably not safe. Exiting. Well, OK, I guess it might be unsafe, and I would really hate it if it breaks Mailman... I really don't have the time to go through a reconfiguration now, and besides, I intend to migrate to Sympa once I get the time. So, advices on what to do with this is very welcome! Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: logrotate question
On Tuesday 10 February 2004 13:45, Dr.-Ing. C. Hurschler wrote: > Hi, > > I'm running Debian Testing, and have noticed that my syslog, > kern.log, and debug.log files are too large (over 100Mb). Yeah, I've also run into trouble like this. I tried to avoid some problems by adding # Rotate logs larger than 2M size 2M to my logrotate.conf > I looked > in logrotate.conf and logrotate.d, but didn't find anything that > would rotate these files. Aren't these files rotated in a "standard" > installation? I also noticed that the logrotate entry in cron.daily > only points at logrotate.conf, but what about the logrotate.d > entries, how do they get loaded? Somebody else answered that question, but I have noted that even though my configuration tells logs to be rotated at certain intervals and/or sizes, apparently they aren't rotated on my workstation and laptop. Up to now, I haven't understood why, but your post made me do some googling. My home workstation is off at night, so cronjobs are not run. At least that's a feeling I've had for some time, but not known how to deal with. Sounds familiar? So, five minutes ago, I discovered anacron. I installed it, and there is something going on my system now... Perhaps it is actually performing the housekeeping long neglected by usually having the system off at night...? In that case, apt-get install anacron may be what you're looking for? Also, BTW, a few days ago, I suggested that it would be interesting to have a way for logrotate to know how much disk space it can be allowed to use in total, to avoid it filling up an entire partition. More people interested in this? Cheers, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Two Apache instances; IPTables to redirect ports
On Monday 09 February 2004 18:04, Adam Aube wrote: > On Monday 09 February 2004 07:43 am, Kjetil Kjernsmo wrote: > > So, how do I set up my box to run two different Apache > > instances...? > > Install your demo system from Apache source in /usr/local/apache, and > make sure it uses a different port, log & pid files, etc. Oh, OK... Isn't it a viable alternative to run apache with -f config, with another config file...? I thought that would be the most elegant path to take, but I have no clue on how to implement it practically, so perhaps you're right. > > The other question is related: I have to run the demo server at a > > different port, say 8081. But, I figured it should be easy with > > IPTables to redirect this so that incoming traffic to my demo > > server test.skepsis.no at port 80 is redirected to 8081, without > > interfering with the other traffic to the other virtual hosts on > > this machine. > > Impossible. IPTables works at too low of a level to see the hostname > in the request sent to the server. OK. >What you need is a reverse proxy. > I know Squid can be used for this, but I've never done it, so I can't > give you any advice on how to go about it. OK, I have some hints for using yet another Apache instance to do that, I'll probably look into that. Thanks! Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Rejecting viruses the Right Way[tm]
On Monday 09 February 2004 17:52, Steve Lamb wrote: > Derrick 'dman' Hudson wrote: > > If a message is either rejected (during the SMTP dialog) or bounced > > (after accepting and queueing the message) then the same innocent > > third party receives some junk mail.[1] The difference is only in > > which server is sending the bounce message. > > The presumption being, of course, that the other side is a real > MTA and not the virus/worm itself. Rejecting is acceptable as the > onus is on the other side on what to do. You're not generating the > bounce. If it is a virus/worm then it isn't likely to generate a > bounce. If it is an MTA then they had best get their act together > and not propigate viruses. Right. This is, I think, the idea behind rejecting the message (both viruses and spam), is that it is not you who are generating the bounce, so unless the virus is programmed to deal with the rejection, nobody will get the bounce. Same with spam, it is the actual spammer who will get the rejection. He may generate a bounce to whoever he has forged into the MAIL FROM:, but he has no reason to do so, has he? Also, he has to clean his lists now and then, otherwise, he would spam mostly dead addresses. So, one would think that he would use the rejection to clean the lists, but I can't see that happening. So, basically, if gluck had rejected the message on RCPT TO: rather than send it over to master, a bounce message would presumably not have been generated by (in this case) 213.222.144.148. But it raises the question, what if viruses are doing something to deal with a rejection, and does it with malice, could they use it to do bad things? Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: logs rapidly fill /var partition
On Monday 09 February 2004 16:12, Haines Brown wrote: > > > The problem is that each of these logs ends up 120 Mb in size and > > > the remaining space in my 500 Mb /var partition quickly > > > disappears. As an aside, wouldn't it be cool if you could tell logrotate how much diskspace all the logs combined could be allowed to use, so that it could rotate, compress and delete logs before they fill up the entire partition? It would perhaps not help in your case, but I have experienced a couple of times that logs have grown large in a few days because I haven't been paying attention to what happened, and completely filled the partition (on critical systems, I have a seperate /var/log/ partition, so that it won't crash things if full, but I haven't filled that). Is this possible? If not, what do think, should I file a wishlist bug? Cheers, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: What's up with http://www.fs.tum.de/~bunk/debian
On Monday 09 February 2004 15:24, Christian Schnobrich wrote: > My first guess would be that "bunk" is no longer there, and his > account has been closed. Could also be that he exceeded some quota or > whatever. Doesn't sound very likely that it would happen overnight just like that, and without any information about where he'd go. It seems like he posted to debian-devel from that address two days ago and to LKML 25 hours ago. (In case you're not aware, Adrian Bunk maintains one of the largest repository of backports to Woody, a lot of people use them and they are really good). I would be surprised if there isn't something wrong that's beyond his control. Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Two Apache instances; IPTables to redirect ports
Hi all! I mainly wondering if anybody has experience with running two different Apache instances on a single machine under Debian? The reason I'm asking, is that I have to run a production server and a demo server on the same machine, and while their configs could be very similar, the demo server should run my test modules, perl modules, that is. So, each instance would have to have at least different @INCs, and I've been told that's impossible with a single instance. So, how do I set up my box to run two different Apache instances...? And, while I'm at it: What's the best way to set @INC? The other question is related: I have to run the demo server at a different port, say 8081. But, I figured it should be easy with IPTables to redirect this so that incoming traffic to my demo server test.skepsis.no at port 80 is redirected to 8081, without interfering with the other traffic to the other virtual hosts on this machine. Is this as easy as it sounds...? :-) Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: video card compatibility
On Sunday 08 February 2004 22:14, Ashley Mervyn Graham wrote: > so, i'm asking for suggestions of video cards under debian/linux, i > would like it to work out of the box, i would like it to be > relatively cheap ($75-$150), it doesnt have to be new, or handle > games, as i use consoles for that. but i want video playback, also, a > nice resolution wouldn't be that bad. I'm happy with my Matrox Milliennium G450 DH. I'm not sure about it's video playback capacity, but check it out, at least one of its immediate successors should be good at it. It has explicit Linux support, and Matrox has a mgapdesk tool, which is GPLed and in Debian. It's not a serious gamer's card, since the 3D support is not very good, but it is crystal clear in 2D, and that's important when you stare at the screen for most of the day... Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Rejecting viruses the Right Way[tm]
Hi all! The recent MS viruses has been bothersome for quite a few of us, I presume, because of the noise it creates. I have configured my Exim4 install to reject MS executables at SMTP-time, so I don't see a lot of the actual virus. But I suppose I get a few bounces, like everyone, because MyDoom tries to send itself to "common name"@some.domain, with a forged return path it found somewhere. This is rather annoying, so it is important to configure the servers to avoid it, I figure. If you agree, please read on! :-) If I've understood the configuration I have tried to make correctly, if you reject the virus in the SMTP-dialog, either due to a unknown username (in the RCPT TO) or because it has a MS executable (in DATA), that bounce should not go to the address in the return-path or MAIL FROM: Which is good, because it is trivially forged, and so, a bounce that goes to the addresses there will often end up at an innocent third-party. If, OTOH, you first accept the message, _then_ bounce, the bounce will go to that innocent third party. So, one shouldn't do that. If the message is accepted, it is too late to bounce. I've seen quite a few of these bounces, and since I'm not very experienced myself, whenever I've seen bounces from Exim4 installs, I've dropped the postmaster of that domain a note, telling them what happen, and that this must be an error, and I'd like to discuss it, in case my install behaves similarly. I haven't had a response from anyone, though. But now, I got a bounce from Debian's servers, and I thought "et tu, Brute": [EMAIL PROTECTED] unknown local-part "linda" in domain "debian.org" -- This is a copy of the message, including all the headers. -- Return-path: <[EMAIL PROTECTED]> Received: from gluck.debian.org [192.25.206.10] by master.debian.org with esmtp (Exim 3.35 1 (Debian)) id 1ApPpz-0005Vz-00; Sat, 07 Feb 2004 04:37:23 -0600 Received: from catv-d5de9094.bp04catv.broadband.hu (learn-orienteering.org) [213.222.144.148] by gluck.debian.org with esmtp (Exim 3.35 1 (Debian)) id 1ApPps-00050t-00; Sat, 07 Feb 2004 03:37:18 -0700 So, it is pretty clear that it didn't come from me, eh... :-) But I got the bounce. Why is this happening? Well, for one thing, it seems like gluck.debian.org has accepted the message and sent it on to master.debian.org. Is that the reason? Since gluck accepted the message, there's nothing master can do about rejecting it and the bounce wrongly ends up in my mailbox. gluck is the 2nd MX, as far as I can see, but would the same thing have happened if master had been handling the message? My main worry is of course that my own setup does the same thing, so that my bounces end up in random people's mail boxes. I haven't got a 2nd MX (but I hope to get one soon), and the rejection at RCTP TO I haven't tweaked, it just rejects unrouteable addresses. My rule to reject MS executables looks like this: deny message = $found_extension files not accepted (may contain MS virus) demime = com:exe:vbs:bat:pif:scr and can be found in my ACL config. I believe this should only reject at SMTP time. So, under what circumstances could this reject a message after it has been accepted and result in a bounce to an innocent third-party? If I get around to get a 2nd MX, would I have to set up this 2nd MX to reject viruses and spam at SMTP-time, or is this something that could be left to the primary MX? From the above story, I think it looks like the 2nd MX would have to handle the rejection, that it can't be handed over to the primary MX. Your thoughts on this subject? Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Pinning to testing
On Thursday 01 January 2004 12:01, Steinar Bang wrote: > Mine look like this. It seems to be working. I've no idea why > (pinning is a mystery to me): Hm, it seems like it will remain a mystery to me as well... It seems to work now, after I installed the python packages apt insisted had to come from unstable, it doesn't pull anything else from unstable, but goes with testing instead, as desired. I have no idea why, but I guess it could have been something with those python packages... Thanks anyway! Vennlig Tiddeli-bom, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [OT[UnOfficial Unsubscribe FAQ
On Thursday 01 January 2004 16:49, Joseph A. Nagy, Jr. wrote: > 2. But what if I send it to just debian-@lists.debian.org? > > I'll shoot you. I have a confession to make: I once managed to send an unsubscribe to debian-security. Background: I had just gotten into a situation where I had to run out of the door, and was to be offline, or have very little bandwidth available to me for some time, and I had thus five minutes to unsubscribe from a bunch of lists. Most of them went fine, but with debian-security, I managed to send the unsubscribe request directly to the list. I apologized a second later. It can happen to most people when you're stressed. Most non-braindead list systems have realized that you need to protect your list-members from both negligence and mishaps, by intercepting messages that looks like unsubscribe requests. You need to do that because it is a fact of life that if you don't, there is a lot of noise caused by the unsubscribe-requests themselves, but also by people posting content-less FAQs. Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Pinning to testing
On Thursday 01 January 2004 12:01, Steinar Bang wrote: > Mine look like this. It seems to be working. I've no idea why > (pinning is a mystery to me): > > Package: * > Pin: release a=testing > Pin-Priority: 990 Oh, ok, that would make some sense, since 990 is above the default pin, it could have something to do with that, but it didn't actually work for me... Any others...? Cheers, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: The Darkness
On Wednesday 31 December 2003 11:58, Antony Gelberg wrote: > The back label mentions "digital copy protection", which I find > rather worrying... Then, I would say, return the (C)D to the artist with a long letter explaining why this approach is fundamentally flawed, and can only work by controlling the senses of both the artists and the listeners. Controlling *their* senses. That's what I'll do if one of my own favorite artists would end up doing something like that. Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: about latex and tables
On Saturday 20 December 2003 08:43, lakshmi wrote: > I have a table that is too big (too long) to fit on one page. Can > LaTeX break this table by itself ? How ? Yep, check out supertabular, I've had good experiences with that. See also longtable, it's said to be easier to use. You'll find it all on CTAN: http://www.ctan.org/ Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Pinning to testing
Hi all! I was wondering if anybody could explain pinning to me... I know several sources, yes, but just one more time... Please...? OK, here's the problem: I've been tracking unstable on my workstation for a few months now, and I'm pretty happy with it, it's up-to-date now and I have the stuff I need. It's time to stabilize a bit (after kdm has been uploaded). So, from now on, I'd like to be tracking sarge, but certain packages should still be installed from unstable if I ask for it explicitly. Take for example python2.3. I've had that on hold for some time now, so what I have installed is 2.3.2.91-1. Unstable has 2.3.3-1, whereas sarge has 2.3.2-2. The idea is that if I go apt-get upgrade, I would get sarge's version, but if sarge's version is older than the one I currently have, it would just leave it and wait for a higher version to become available. I don't want to downgrade anything. Pinning is for this purpose, so I tried to create a preferences file. Initially, it started like this Package: * Pin: release a=unstable Pin-Priority: 302 Package: * Pin: release a=testing Pin-Priority: 550 After unholding python2.3, and go apt-get upgrade, it started to pull 2.3.3-1 from unstable. Hm. Back to reading. I've read http://www.argon.org/~roderick/apt-pinning.html and the APT HOWTO: http://www.debian.org/doc/manuals/apt-howto/ch-apt-get.en.html#s-pin The latter says: | Priorities 0 to 100 denote packages that are not installed and that | have no available versions. Hm, this doesn't really make much sense to me in the context of apt-pinning: Why would one set the pin to <100? Just to say that it isn't installed...? And that it won't be installed even if I ask for it explicitly? That got me really confused. But then: | Priority 100 is the priority assigned to an installed package - for | the installed version of a package to be replaced by a different | version, the replacement must have a priority greater than 100. OK, so in spite of my confusion above, it seems like if I set the priority to below 100 for unstable, it will not replace allready installed version, because the priority for unstable will be less than the priority of allready installed packages. Sarge, OTOH will have priority of 550, so it will be installed. That's what I want. So, now my preferences look like: Package: * Pin: release a=unstable Pin-Priority: 99 Package: * Pin: release a=testing Pin-Priority: 550 apt-get upgrade, however, doesn't agree with my reasoning... It starts pulling python2.3 2.3.3-1 from unstable archives... I've also tried reordering the entries, run apt-get update, etc. They don't make any difference... One concern I get out if this as well, it sounds like if I keep a pin <100 for unstable, it is not going to be installed even if I ask for it? That's not what I set out to achieve... So, if anybody would care to explain this just another time, I would be happy! :-) A Very Happy New Year to everyone! Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
depmod trouble when installing new kernel
Hi all! I'm installing a new 2.4.23 kernel on my main server (that I compiled some time ago), using Marcello's tree, but the same config I used for the Debian 2.4.22 kernel previously. The kernel was compiled on my workstation (with make-kpkg) and then uploaded to the server. When I try to install it, I get lots of error messages on the form: depmod: *** Unresolved symbols in /lib/modules/2.4.23.2003-12-01.1-pooh/ kernel/net/wanrouter/wanrouter.o depmod: *** Unresolved symbols in /lib/modules/2.4.23.2003-12-01.1-pooh/ kernel/net/x25/x25.o There was a problem running depmod. This may be benign, [snip] I'm not actually using the things mentioned there, I don't think I have it compiled in at all, but I noticed it complained about other things I am using, such as ext2 (but I think I have compiled into the kernel). Because this is a machine that it is hard to get physical access to, I'd like to know if it is benign before I proceed... How do you address this...? Best, Kjetil -- Kjetil Kjernsmo Astrophysicist/IT Consultant/Skeptic/Ski-orienteer/Orienteer/Mountaineer [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] Homepage: http://www.kjetil.kjernsmo.net/OpenPGP KeyID: 6A6A0BBC -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
