Re: SATA drives not recognized on new install
On Sat, 28 Apr 2007 15:30:55 -0700, Kenward Vaughan wrote > Is this likely a problem of not having the right support built into the > installation disk, and if so how can I get around that? > I'm guessing your using the latest etch installer, which comes with a 2.6.18 kernel. You might want to try a debian installer with a more recent kernel. Perhaps something from here: http://kmuto.jp/debian/d-i/ Perhaps there is a custom etch installer out somewhere with 2.6.20 ? You could always install the sarge one from above and upgrade to etch afterwards. hth, Mike -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
SATA drives not recognized on new install
Hi, I just built an amd box with a 5600+ processor on an MSI K9A Platinum MB. The latest (as of yesterday) netinst version of debian-amd64 was used to boot up the system. I saw none of the BIOS features which were mentioned on the installation page as things which had to be disabled, ran the memory up to 800 Mhz, etc., etc. The boot process times out when trying to communicate with the SATA drives (I have 2 WD drives, both recognized by Windows when I installed that later). The message given is along the lines of "ata.1: SATA drive failed to IDENTIFY (I/O error errmask=0x104)" Other lines earlier in the boot included SATA max udma/133 cmd 0xC200ED00 crt 0x0 bmda 0x0 irq 1277 and SATA link SStatus 123 SControl 300 (I have no clue whether these are helpful, though.) This happens for all 4 slots (with only 2 being populated, of course). I tried to get an idea of what people did with similar problems (amdforums), and tried a variety of boot options (noapic nolapic; noapic acpi=noirq|off ; etc.), none of which helped. One time I was able to shft-pgup and saw a line about a BIOS bug, along with the letters MCFG, but I couldn't catch it again at other times. Looking at dmesg later on following a normal boot showed no such line in the part of dmesg which had been saved. It appears that most if not all other HW is recognized, including the LAN. The K9A has the ATI RD580 north bridge chip and SB600 south bridge, and is a Crossfire board (which I don't care about). The BIOS version is 1.3 (Dec. '06). Is this likely a problem of not having the right support built into the installation disk, and if so how can I get around that? Can I build an amd64 kernel on my 32 bit kernel machine and somehow build a bootable disk off of that? (a flash drive perhaps?) Many thanks for any help! Kenward -- With or without (religion) you would have good people doing good things and evil people doing evil things. But for good people to do evil things, that takes religion. --Physicist and Nobel Laureate Steven Weinberg -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: My new Etch 2.6.18-4 install segfaults and freezes!
"Pascal Giard" <[EMAIL PROTECTED]>: > 1) resetting the bios? Try that first anyway in such cases, if you or someone ever played with it before. (Enter the BIOS pressing F2 or some similar key, as mentioned on the first or second 'boot' screen, and look for something like 'Reset to defaults'. Usually this is completely safe. Leave BIOS with 'Save new settings and exit'.) Memory DIMMs: Sometimes people do not insert them correctly, thy have to snap in (have a close look) and you often need some surprising force to be successful. If a mem DIMM bar is defect then you should see BIOS error at bootscreen, or incorrect amount of mem reported. Note that BIOS usually reports detected size in Kilobyte (long number ! with a "K" at the end). > 4) running memtest86? I believe in most cases this is not necessary. It's time consuming, too. > On 4/3/07, marie s <[EMAIL PROTECTED]> wrote: > > I don't even know where to begin to try to fix this. I If you really want to get help then please what are you doing ? Which kind of machine, and whatever you think is non-standard: Like, what kind of mainboard, and, if IDE or SATA (or PATA) controllers (if you know), which Debian version, any unusual partition setup or filesystem formats. You can try to look up system with a knoppix CD, it would be interesting if this one runs with less problems. Lookup /var/log/syslog and kern.log For example, enter a shell or terminal, and su root cd /var/log grep -n 'disk\| ide\|mem\| ram' * > /tmp/grepfile Attach /tmp/grepfile to mail. If you don't know the right tools and how to debug, just ask on any debian list most debian people know > > But since the "corruption" was reported in different > > files every time I ran the test What if harddisk or mainboard is defect. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: googleearth
Am Samstag, 28. April 2007 schrieb Michael: > Thanks Hans. > > > Add the path to the libs in ld.so.conf. > > Looks like didn't have to tweak anything manually (after using the --force > option), maybe cause this is googleearth-package 0.6 then ? > > > Remeber to install nvidia-glx-32 > > I did. (Do i really need to? I not even roughly understand what each of the > various nvida packages is for) The normal one are for real 64-bit applications, the nvidia-glx-legacy has something to do withe the licenxse, the nvidia-glx-32 is for running 32-bit-applications, which are runnning under 64-bit (either with libs from chroot or ia32-libs). > > At some point xorg couldn't find a proper nvidia driver, no matter i > recompiled the module from the source for kernel 2.6.18 with m-a. I could > solve it by updating anything to kernel 2.6.20 and to nvidia- > 0.9755, from unstable. I recompiled the new sources, dunno why but now it's > working. > Yes, but I could never compile the nvidia-kernel-module on 32-bit system with kernel 2.6.20 (there is a bugreport by me). In 64-bit, everything was o.k. > > NOT install it in the chroot. > > ok. > > The nvidia-glx in chroot seems to conflict with the nvidia-glx-32. This is, because the kernel or the application wants to run the chroot libs and/or the nvidia-glx-32 together (Don´t blame me, this is, what I guess) Anyway: deinstallling those in the chroot is working. > m° Regards Hans P.S.: I am using googleearth 0.6 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: cpu overheating
Or fan isn't correctly wired -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: googleearth
Thanks Hans. > Add the path to the libs in ld.so.conf. Looks like didn't have to tweak anything manually (after using the --force option), maybe cause this is googleearth-package 0.6 then ? > Remeber to install nvidia-glx-32 I did. (Do i really need to? I not even roughly understand what each of the various nvida packages is for) At some point xorg couldn't find a proper nvidia driver, no matter i recompiled the module from the source for kernel 2.6.18 with m-a. I could solve it by updating anything to kernel 2.6.20 and to nvidia- 0.9755, from unstable. I recompiled the new sources, dunno why but now it's working. > NOT install it in the chroot. ok. m°
Re: run away processes.
Hi Chris, > Ok I've got a program called kerry that runs a daemon: beagled Do you really need it ? I guess you can install the components of kde-extras manually, without beagle... > Then there is: kio_pop3 running from Kontact/Kmail. I never heard such problems with kmail before. Are you on a dialup or is your network (wlan?) suffering from huge latency ? Still you could try, for example, a smart mailer like Sylpheed-claws, which has great advanced featues for custom filtering and actions, or (more common but not smart) evolution or thunderbird. > Then there is artsd, but that doesn't happen all that often, but will > definitely eat more than it's share, but not by much. Weird. Are you sure it's not kernelspace or the sound backend (like alsa) causing problems ? Yet, you can tune down the level of desktop sounds in kcontrol. > Openoffice.org once and a while will entirely stop the desktop Openoffice always had some minor problems, in my experience, expecially wiuth window managers. But it's not eassy to keep such a complex thing ready for different OS and archs...But there is no real alternative. Be patient. > need to ssh in to restart KDM Not even ctrl alt F1 possible ? Weird. Make sure there are no kernel problems behind (look into kern.log and dmesg). m°
Re: Re:
Want get banged by your dick with tall woman? Open www.2211122. And add COM after dot at the end. cazzeggiare sottoterra. Ho voglia di una sigaretta. Brucio la prima bionda della giornata e rifaccio il del suo passaggio. Bastava togliere la sicura e attendere sei secondi.inizio ad assaggiarla; voglio sentire il gusto di ogni suo centimetro, fissarmi nella mente il suo che i poliziotti non sono riusciti a scoprire in ferro. Dietro di lui, i passi si fanno pi� vicini e il silenzio � irreale. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
