Re: HP DV9540 laptop AMD64 with nVidia G8400M graphic card will not start in X mode
Am Freitag 24 August 2007 schrieb Niels Larsen: On Friday 24 August 2007 02:35:29 Chris Ahlstrom wrote: At last I installed a driver direct from nvidia's site, following their instructions: http://www.nvidia.com/object/linux_display_amd64_100.14.11.html It worked, and I got kde up and running, but then I tried as normal user, and got segmentation error. Then I tried again as root and got the same error. Now I an installing a 32 bit debian, and leaving this list for now. I cannot get neither a K/ubuntu 64 -or 32 bits installed, it installs and then I get black screen, so it seemt as it only is debian which will install on my new laptop. Can see, that I get the same X problem just now on 32 bits debian, so I will have to install a driver from nvidia, and then see. Thanks for your help so far :-) (I learned a lot) N o reason for this. The normal user must be in group nvidia. The Debian package does this automatically, the NVidia-installer NOT. Regards Hans -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Opinions on ext3 vs XFS vs reiserfs for LAMP server
On Fri, Aug 24, 2007 at 12:48:29AM +0200, Michael wrote: I just tweaked all ext3 partitions to something like /dev/sda10 / ext3noatime,nodiratime,errors=remount-ro 0 1 to see what will happen. (The option data=writeback caused boot trouble) So far, anything unchanged but the firefox history where is no 'Today' anymore. MySQL, Apache, and NFS servers seem to be ok. And actually anything is loading or writing faster. For example application launching and disk caching. Besides /home and /var, I have seperate /var/cache with different inode sizes and dir_index, for lots of small files (webproxy). I guess something like that could be handy for other filesystem based databases too. I wasn't able to undelete a whole directory tree from ext3 with autopsy - i didn't try other tools but i did some research and it looked like it's not so easy, maybe impossible. Lots of tools mention they work on ext3 too but they can't do the same on ext3 as on ext2. undelete only works on ext2 since it doesn't have sorted directories. ext3 resorts the directory when you add or delete files, and hence looses the entry for the deleted file that ext2 just left there but marked deleted. I don't think I have ever needed undelete. That is what backups are for, along with being careful. :) Allowing undelete tends to lead to inefficient filesystem designs. If you want undelete, install some tools to replace rm and the like with programs that move files to the trash instead. Doesn't protect against overwriting, but then again undelete never did either. -- Len Sorensen -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Opinions on ext3 vs XFS vs reiserfs for LAMP server
On 08/23/07 04:07:46PM -0700, [EMAIL PROTECTED] wrote: Quoting Jim Crilly [EMAIL PROTECTED]: On 08/23/07 10:03:24AM -0700, [EMAIL PROTECTED] wrote: The problem of zeroing files of XFS still exists, however its not some mythical type of corruption. You'll only see it on files recently written to within seconds (say approx 60 secs) of a hard power off. If you can't risk it, or think you may have encounter the odd hard reset, ext3 might be a better choice. Actually it's been fixed as of 2.6.22: http://oss.sgi.com/projects/xfs/faq.html#nulls Of course that doesn't help you if you're using sticking with the kernel shipped with etch. I'm not so sure its fixed. I just tested with a sid samba box, running 2.6.22 kernel, and XFS filesystem. Connected to it via a WinXP box and copied a word doc file to it. Soft rebooted the samba box to make sure the file was sync'd to hard drive. Re-connected to samba share and opened the word document, added some text lines to it, saved and quit Word, then yanked the power out. Rebooted and re-connected to the samba share again only to find the file full of squares. Ext3 would have at least retained the original contents of the file. I tested the exact same thing again but waited 60 seconds after saving the file, and then yanked the power out. Upon a boot up, the file was intact and the save worked. So you still have about a 60 second window of newly written files and a power loss for data corruption, unless the program can sync it to disk before that. Well I'm only passing on what the XFS devs have said, all of my boxes are on UPSes so I rarely saw the issue anyway. But are you sure the squares you saw in the word doc were nulls? The FAQ page says that you can use the xfs_bmap command to see if it has any extents allocated and if it does then it would likely be another issue. Jim. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Opinions on ext3 vs XFS vs reiserfs for LAMP server
On Thu, Aug 23, 2007 at 09:47:13PM -0500, Neil Gunton wrote: Eh? I've been using replication for years now. It works out of the box, rock solid, and nothing special on the backend at all. Master produces binary log, which is replayed on the slave(s). Nothing questionable about it, it's used in many large scale installations. Again, more FUD. Unless the two servers run in sync, then I don't consider it replication. Better than what postgresql offers though (which is currently no replication attemp of any kind). You can declare any feature to be essential and then denigrate other products that don't have it as toys. Personally I've never missed subselects, which, to me, means they are not exactly essential. Of course, the riposte to that is to claim that I'm somehow not a real DBA, or not running a real database. Whatever evidence is presented to prove that MySQL is a perfectly cabable database manager, the PostgreSQL zealots then simply say Well, that's not a real database application, or Well, maybe it's ok for running a hobby website... ok, whatever. You can't argue with people who have such a religious mindset that no amount of evidence or argument can change their viewpoint. From my end, I can happily accept that PostgreSQL has some cool features, it's no skin off my nose, though most of these features are such that I would never have actual cause to use them. Personally, I like to keep the database simple, and keep my logic in the application, not in the database. Amazingly, the world does not fall down around my ears as a result. In any case, MySQL does now support things like subqueries, views and stored procedure for those who want them. Transactions have been there for a long time now. Well back when I was using databases quite a bit, I stuck with postgresql. At the time mysql's license was a bit obnoxious, and it had lousy locking (whole table at a time or nothing), so postgres was the better fit. I also like the psql command line tool, for which I have found no equivalant for mysql (mysqlcc isn't bad, but I am not a fan of gui stuff in general). Today I only really have to deal slightly with mysql since that is what bugzilla uses (and I did look into converting it to postgres and found too much mysql specific code in use so I decided it was too much work). Some people like to have business rules and logic embedded in the database; other people think it's a terrible idea and stay well away from it. As with many things, it's a matter of taste. You can declare something to be a toy all you like, but many large installations use MySQL for very large-scale, non-trivial tasks. Purists may like PostgreSQL, but people who are simply interested in getting the job done tend to do just fine with MySQL. Certainly. I have personally always avoided logic in the database. I find it leads to scaling problems, and also makes it difficult to move to another database later. If you stick with just plain SQL and don't use any advanced features like stored procedures, then you can pretty much move between most decent databases. All of this patronizing parroting of the same tired dogma is so counter-productive and bitchy... it turns me right off. -- Len Sorensen -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: AMD64 X2 questions
On Fri, Aug 24, 2007 at 01:04:16AM +0200, Michael wrote: Do you know if the kernel tries to balance load really equally ? For my 2core AMD64 it seems cpu0 has to be about 70% before cpu1 gets involved. Another idea is, i always see cpu0 is the most busy one. But wouldn't switching the 'most busy cpu' from 0 to 1 at certain intervals (say, 3 seconds) reduce heat ? If you have a single dual core cpu, then it shouldn't make any difference. Also if the cpu can't handle running at 100% load forever, then the cooling system is misdesigned. It is just normal to always use the first cpu first. And moving processes around causes cache flushes which slows things down, and probably wastes more power moving data around needlessly. In fact if you have three processes to run and in total they need 90% of one cpu, then it is better to run it all on one cpu and let the other one go to power saving mode rather than running half the load on each. That does save power and reduce heat. -- Len Sorensen -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: AMD64 X2 questions
On 08/24/07 01:04:16AM +0200, Michael wrote: Lennart, probably don't have any good reason to. The scheduler does a great job Do you know if the kernel tries to balance load really equally ? For my 2core AMD64 it seems cpu0 has to be about 70% before cpu1 gets involved. Another idea is, i always see cpu0 is the most busy one. But wouldn't switching the 'most busy cpu' from 0 to 1 at certain intervals (say, 3 seconds) reduce heat ? It balances running processes but that's all it can do. So if you have 1 single-threaded process doing something it'll only run on 1 CPU and the other will sit idle. Trying to balance that one process between both CPUs would be horrendous for performance since it would constantly invalidate the cache on each CPU causing that data to be reloaded from main memory. Jim. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: HP DV9540 laptop AMD64 with nVidia G8400M graphic card will not start in X mode
On Fri, Aug 24, 2007 at 02:53:49AM +0200, Niels Larsen wrote: On Friday 24 August 2007 02:35:29 Chris Ahlstrom wrote: At last I installed a driver direct from nvidia's site, following their instructions: http://www.nvidia.com/object/linux_display_amd64_100.14.11.html You really don't want to do that. It WILL break on some future upgrade. 3rd party installers are always a bad idea. Just get the debian packages from unstable instead and install those. That will work fine and upgrade correctly. -- Len Sorensen -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: HP DV9540 laptop AMD64 with nVidia G8400M graphic card will not start in X mode
* Lennart Sorensen [EMAIL PROTECTED] [2007-08-24 09:33:27 -0400]: On Fri, Aug 24, 2007 at 02:53:49AM +0200, Niels Larsen wrote: On Friday 24 August 2007 02:35:29 Chris Ahlstrom wrote: At last I installed a driver direct from nvidia's site, following their instructions: http://www.nvidia.com/object/linux_display_amd64_100.14.11.html You really don't want to do that. It WILL break on some future upgrade. 3rd party installers are always a bad idea. Just get the debian packages from unstable instead and install those. That will work fine and upgrade correctly. The one or two times I've tried to use Debian procedures to update the kernel, it worked okay. But I've also had trouble with it, especially when adding drivers to the mix. I much prefer doing it all myself and understanding where to put every file. But that's just me. -- Tux rox! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: AMD64 X2 questions
* Lennart Sorensen [EMAIL PROTECTED] [2007-08-24 09:27:34 -0400]: In fact if you have three processes to run and in total they need 90% of one cpu, then it is better to run it all on one cpu and let the other one go to power saving mode rather than running half the load on each. That does save power and reduce heat. I found this recent article (by a Microsoft guy!) in Dr. Dobbs to be interesting: http://www.ddj.com/hpc-high-performance-computing/201202924 August 03, 2007 How Much Scalability Do You Have or Need? Orders of Throughput * How Much Scalability? * O(1): Sequential Code * O(K): Explicitly Threaded Code * Comparing O(1) and O(K) * O(N): Scalable Throughput And the Free Lunch -- Tux rox! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: HP DV9540 laptop AMD64 with nVidia G8400M graphic card will not start in X mode
* Hans-J. Ullrich [EMAIL PROTECTED] [2007-08-24 08:27:20 +0200]: Am Freitag 24 August 2007 schrieb Niels Larsen: On Friday 24 August 2007 02:35:29 Chris Ahlstrom wrote: At last I installed a driver direct from nvidia's site, following their instructions: http://www.nvidia.com/object/linux_display_amd64_100.14.11.html It worked, and I got kde up and running, but then I tried as normal user, and got segmentation error. The normal user must be in group nvidia. The Debian package does this automatically, the NVidia-installer NOT. On my desktop, running two nvidia-based graphics systems (one on the motherboard, the other a cheap EVGA-brand card), things are working very well with a 2-week-older version of the driver [NVIDIA-Linux-x86_64-1.0-9755-pkg2.run], and I am not in any nvidia group. That's probably purely a debian thang. My formula for installing third party drivers is to combine them with a manual kernel upgrade: - Download the driver source or install package. - Download the latest stable kernel source. - Copy my old .config and do a make oldconfig. - Review the options via a make menuconfig. - Do a make, make modules_install. - Do a make initrd with the right output and version options. - Copy the System.map, .config, vmlinuz, and initrd.img files to das /boot with the right names (e.g. initrd.img-2.6.21.3-amd64-2). - Add the new entries to /boot/grub/menu.lst. - Reboot and make sure it works; of course, the old video driver will not work with the new kernel. - Make a new /usr/src/linux link. - Run the install procedure for the new driver. - Modify xorg.conf manually. I feel more comfortable doing it myself. I've done it this way on laptops, too, for both nvidia video (nvideo?) and Intel Centrino wireless (which has the extra burdren of writing an init script, installing a regulatory daemon, installing the Intel firmware, and optional configuration stuff such as wpasupplicant, ifupdown, guessnet, and NetworkManager). As an aside, they (the Winbois) that Linux has no documentation, but all these steps are discoverable through the search engine. But it pays to keep a log of what you did, for the next time grin. -- You driver me crazy! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: HP DV9540 laptop AMD64 with nVidia G8400M graphic card will not start in X mode
On Fri, Aug 24, 2007 at 10:14:13AM -0400, Chris Ahlstrom wrote: The one or two times I've tried to use Debian procedures to update the kernel, it worked okay. But I've also had trouble with it, especially when adding drivers to the mix. I much prefer doing it all myself and understanding where to put every file. But that's just me. I like having a working system and never having to reinstall. The nvidia installer is very redhat centric. It doesn't quite understand where debian wants things. This can cause problems. It will also overwrite files beloging to other packages in some cases, which will then overwrite the nvidia ones the next time they are upgraded, in which case the nvidia driver breaks. The debian package for the nvidia driver does the correct diversions to avoid these problems, and is much easier to upgrade. I used to do my own kernels. I stopped bothering around 2.6.0 since debian's kernels work perfectly and they keep track of the security updates, so I have to do less work. module-assistant and the module source packages just work and it only takes me a few seconds of work when a new kernel arives from debian. It is always worth learning to use good tools like the ones debian provides. Learning to do things consistently and efficiently is worth at least as much as doing everything from scratch (in a usually inefficient way). Learning how the tools work lets you learn from people with lots of experience to see how things really can be done. -- Len Sorensen -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: HP DV9540 laptop AMD64 with nVidia G8400M graphic card will not start in X mode
* Lennart Sorensen [EMAIL PROTECTED] [2007-08-24 10:20:13 -0400]: On Fri, Aug 24, 2007 at 10:14:13AM -0400, Chris Ahlstrom wrote: I much prefer doing it all myself and understanding where to put every file. I used to do my own kernels. I stopped bothering around 2.6.0 since debian's kernels work perfectly and they keep track of the security updates, so I have to do less work. module-assistant and the module source packages just work and it only takes me a few seconds of work when a new kernel arives from debian. Maybe I will try it that way next time. It is always worth learning to use good tools like the ones debian provides. Learning to do things consistently and efficiently is worth at least as much as doing everything from scratch (in a usually inefficient way). Learning how the tools work lets you learn from people with lots of experience to see how things really can be done. Good point. You learn something either way. Chris -- Tux rox! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Opinions on ext3 vs XFS vs reiserfs for LAMP server
Quoting Jim Crilly [EMAIL PROTECTED]: I tested the exact same thing again but waited 60 seconds after saving the file, and then yanked the power out. Upon a boot up, the file was intact and the save worked. So you still have about a 60 second window of newly written files and a power loss for data corruption, unless the program can sync it to disk before that. Well I'm only passing on what the XFS devs have said, all of my boxes are on UPSes so I rarely saw the issue anyway. But are you sure the squares you saw in the word doc were nulls? The FAQ page says that you can use the xfs_bmap command to see if it has any extents allocated and if it does then it would likely be another issue. I'm not sure. I'm thinking the nulls thing is indeed fixed, but perhaps its still something different than the very nature of XFS. My understanding is limited of its mechanics, but since XFS will never journal data, only meta-data, then won't there always be the chance that a file can get corrupted on a timely power loss? UPS is a good buddy of mine too. :)
Re: Opinions on ext3 vs XFS vs reiserfs for LAMP server
Hi michael (2007.08.24_17:10:24_+0200) limited of its mechanics, but since XFS will never journal data, only meta-data, then won't there always be the chance that a file can get corrupted on a timely power loss? A file can get corrupted, but hopefully not the file metadata (i.e. the filesystem) UPS is a good buddy of mine too. :) /me has noticed that on reliable power grids (i.e. not here South Africa), UPSs cause as many outages then they save... And I'm not talking about cheap and nasty UPSs. SR -- Stefano Rivera http://rivera.za.net/ H: +27 21 794 7937 C: +27 72 419 8559 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Opinions on ext3 vs XFS vs reiserfs for LAMP server
On Fri, 2007-08-24 at 08:10 -0700, [EMAIL PROTECTED] wrote: Quoting Jim Crilly [EMAIL PROTECTED]: I tested the exact same thing again but waited 60 seconds after saving the file, and then yanked the power out. Upon a boot up, the file was intact and the save worked. So you still have about a 60 second window of newly written files and a power loss for data corruption, unless the program can sync it to disk before that. Well I'm only passing on what the XFS devs have said, all of my boxes are on UPSes so I rarely saw the issue anyway. But are you sure the squares you saw in the word doc were nulls? The FAQ page says that you can use the xfs_bmap command to see if it has any extents allocated and if it does then it would likely be another issue. I'm not sure. I'm thinking the nulls thing is indeed fixed, but perhaps its still something different than the very nature of XFS. My understanding is limited of its mechanics, but since XFS will never journal data, only meta-data, then won't there always be the chance that a file can get corrupted on a timely power loss? UPS is a good buddy of mine too. :) I'd trust XFS with my data, but we have an expensive support contract with SGI, and it's sub-optimal to use their volume management tool (XVM) without XFS -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Opinions on ext3 vs XFS vs reiserfs for LAMP server
On 08/24/07 08:10:24AM -0700, [EMAIL PROTECTED] wrote: Quoting Jim Crilly [EMAIL PROTECTED]: I tested the exact same thing again but waited 60 seconds after saving the file, and then yanked the power out. Upon a boot up, the file was intact and the save worked. So you still have about a 60 second window of newly written files and a power loss for data corruption, unless the program can sync it to disk before that. Well I'm only passing on what the XFS devs have said, all of my boxes are on UPSes so I rarely saw the issue anyway. But are you sure the squares you saw in the word doc were nulls? The FAQ page says that you can use the xfs_bmap command to see if it has any extents allocated and if it does then it would likely be another issue. I'm not sure. I'm thinking the nulls thing is indeed fixed, but perhaps its still something different than the very nature of XFS. My understanding is limited of its mechanics, but since XFS will never journal data, only meta-data, then won't there always be the chance that a file can get corrupted on a timely power loss? AFAIK ext3 is the only one where data journaling is even an option and virtually no one uses it since it's off by default so XFS is no worse in that regard. Also since you were testing via Samba there's a chance that it's doing something funny that 'normal' Linux processes won't do when you save a file. Jim. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: HP DV9540 laptop AMD64 with nVidia G8400M graphic card will not start in X mode
I've installed a amd64 lenny system, and installed just afterwards the nvidia 100 driver directly from the nvidia installer, with the 2.6.21-2-amd64 kernel. Not a single glitch, everything working fine... and lightning fast! Ivan 2007/8/24, Chris Ahlstrom [EMAIL PROTECTED]: * Lennart Sorensen [EMAIL PROTECTED] [2007-08-24 10:20:13 -0400]: On Fri, Aug 24, 2007 at 10:14:13AM -0400, Chris Ahlstrom wrote: I much prefer doing it all myself and understanding where to put every file. I used to do my own kernels. I stopped bothering around 2.6.0 since debian's kernels work perfectly and they keep track of the security updates, so I have to do less work. module-assistant and the module source packages just work and it only takes me a few seconds of work when a new kernel arives from debian. Maybe I will try it that way next time. It is always worth learning to use good tools like the ones debian provides. Learning to do things consistently and efficiently is worth at least as much as doing everything from scratch (in a usually inefficient way). Learning how the tools work lets you learn from people with lots of experience to see how things really can be done. Good point. You learn something either way. Chris -- Tux rox! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- --- Ivan S. P. Marin -- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: HP DV9540 laptop AMD64 with nVidia G8400M graphic card will not start in X mode
On Fri, 2007-08-24 at 17:09 -0300, Ivan Paganini wrote: I've installed a amd64 lenny system, and installed just afterwards the nvidia 100 driver directly from the nvidia installer, with the 2.6.21-2-amd64 kernel. Not a single glitch, everything working fine... and lightning fast! It works fine today. However, in the process of installing, the nvidia-installer overwrites a number of files from assorted Debian packages. Now, assume those packages are updated (e.g. by a security update), reversing the nvidia changes. The nvidia driver will unexpectedly stop working. You try to uninstall, which replaces the overwritten files with the versions that were there oringinally (e.g. 1.0.3 versions of files that were updated to 1.0.4). This will break X completely. Congratulations, you're now one of the hundreds of people who demanded that nvidia-installer was great, right up until it poked them in the ass. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
