LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Slackware (http://www.linuxquestions.org/questions/slackware-14/)
-   -   A little instability in my Slackware, Need to decipher these errors (http://www.linuxquestions.org/questions/slackware-14/a-little-instability-in-my-slackware-need-to-decipher-these-errors-663034/)

BobNutfield 08-15-2008 01:21 PM

A little instability in my Slackware, Need to decipher these errors
 
Hello Everyone,

Just lately I have been getting a little instability in my otherwise PERFECT Slack installation. Occassionally (not every boot up, about 1 in ten), the volume controls stop, sound dissappears, and when I attempt to reboot the system hangs on a black screen for about two minutes then finally reboots, but only after I lightly touch the power button.

I realize that this is not enough information for anyone to diagnose anything, but the final part of dmesg concerns me a little. I don't know enough about the kernel boot messages to diagnose what these errors means, so if anyone has any idea what is going on I would appreciate a comment or two.

Code:

powernow-k8: Found 1 AMD Athlon(tm) 64 X2 Dual-Core Processor TK-57 processors (2 cpu cores) (version 2.20.00)
powernow-k8:    0 : fid 0xb (1900 MHz), vid 0x13
powernow-k8:    1 : fid 0xa (1800 MHz), vid 0x14
powernow-k8:    2 : fid 0x8 (1600 MHz), vid 0x15
powernow-k8:    3 : fid 0x0 (800 MHz), vid 0x1a
Intel ISA PCIC probe: not found.
Databook TCIC-2 PCMCIA probe: not found.
r8169: eth0: link down
r8169: eth0: link down
r8169: eth0: link down
NET: Registered protocol family 10
lo: Disabled Privacy Extensions
wlan0: duplicate address detected!
Clocksource tsc unstable (delta = -83335939 ns)
ACPI: PCI Interrupt 0000:01:05.0[A] -> GSI 18 (level, low) -> IRQ 21
[fglrx] GART Table is not in FRAME_BUFFER range
[fglrx] Reserve Block - 0 offset =  0X7ffc000 length = 0X4000
[fglrx] Reserve Block - 1 offset =  0X0 length = 0X1000000
hda-intel: Invalid position buffer, using LPIB read method instead.
APIC error on CPU1: 00(40)
APIC error on CPU0: 00(40)

I am most interested in the ACPI errors and wonder if it has anything to do with the "MBIOS bub - timer not connected" message that I and many laptop owners have.

Any replies appreciated.

Bob

Bruce Hill 08-15-2008 08:18 PM

I think it's related to your CPU and kernel options -- read this thread.

BobNutfield 08-16-2008 12:41 PM

Thanks, Bruce. That would explain that part of the dmesg error. But the thread suggests that it is normal when you are using frequency scaling and a dual core cpu. But, I believe you might be right about the kernel. I am using the hugh smp kernel and most have suggested I should have chosen the generic kernel. It is very stable most of the time and I can live with the little hiccups. I will be more careful in choosing my kernel on the next Slack update.

Thanks for the reply

Bob

Alien Bob 08-16-2008 12:46 PM

You don't have to re-install Slackware to get a different kernel! The 'generic' kernel is already on your hard drive (if you opted for a full install) - all you need to do is update /etc/lilo.conf and create an initrd (initial ramdisk).

Eric

BobNutfield 08-16-2008 02:24 PM

Thanks, Eric. I do indeed have the generic kernel in my boot directory. I also have an initrd.gz file. I don't use Lilo, I use grub. I think I will try to add that kernel to the grub boot menu and see if I can get it to boot.

Bob

BobNutfield 08-19-2008 02:28 PM

Hmmmmm.....I suppose this is more a grub issue than Slack. But, I have the vmlinuz-smp-generic file in my boot directory, I made and initrd for this kernel and added it to grub this way:

Quote:

title Slackware Generic Kernel
root (hd0,4)
kernel /boot/vmlinuz-smp-generic-2.6.24.5-smp root=/dev/sda5
initrd /boot/initrd.gz
But, I get Error 15: File not found. I have been dealing with grub for a long time, and I usually dispose of this error pretty quickly, as it normally from some silly typo error in menu.lst. But not this time, there are no typos. I boot the huge kernel I use with Slack with just such an entry, except that vmlinuz is a symbolic link to the kernel. The only difference here is that I point grub directly to the kernel instead of a symbolic link.

Any thoughts why this won't boot?

Bob

Bruce Hill 08-19-2008 02:54 PM

Perhaps because you have the wrong name for the kernel? It should be:
Code:

/boot/vmlinuz-generic-smp-2.6.24.5-smp
rather than:
Code:

/boot/vmlinuz-smp-generic-2.6.24.5-smp

BobNutfield 08-19-2008 04:26 PM

Well, I don't drink, so I can't blame being drunk when I edited menu.lst, but obviously I did have a typ0. Thanks, Bruce, for reminding me not to be so sure of anything and check and check again. You are, indeed, correct and changing the name boots (well, attempts to boot) the kernel. There are no modules installed so I got a slew of errors and booted into busybox.
The main error was no /sbin/init. So, have to go back and find the module package for this kernel and get it installed before I try this again.

Thank you for reading my post and correcting my silly error.

Bob

Bruce Hill 08-19-2008 04:35 PM

If you did a full install, the modules are installed. Do you have both of these:
Code:

mingdao@silas:~$ ls /lib/modules/2.6.24.5
2.6.24.5/    2.6.24.5-smp/

and do you have this:
Code:

mingdao@silas:~$ ls /var/log/packages/kernel-modules-
kernel-modules-2.6.24.5-i486-2          kernel-modules-smp-2.6.24.5_smp-i686-2

?

I always copy and paste into /etc/lilo.conf to avoid such me errors. :D

BobNutfield 08-19-2008 05:06 PM

Yes, did a full install, and I do have those packages. I guess it must be the initrd.gz I built that is causing the problem. It seems OK, though. I have the initrd.gz file and the initrd-tree files.

More research needed, I guess.

Thanks again

Bob

Bruce Hill 08-19-2008 05:13 PM

Why don't you try using Alien Bob's mkinitrd_command_generator.sh script? You can run
it with the "i" switch, for interactive, and make sure it meets your needs, i.e.:
Code:

root@silas:/boot# sh mkinitrd_command_generator.sh -i
I've found that script generally pretty good.

BobNutfield 08-20-2008 02:39 AM

Once again, thank you, Bruce. That script did the trick. It booted smoothly after rebuilding the initrd with it. I have been testing it out for a day now and there does seem to be a very slight difference in performance, but hardly noticeable. I still get that pesky "Triggering Udev Events" message which hangs the boot time for a good 30 seconds or more.

I am guessing that this is all hardware related specific to this type of laptop (Toshiba AMDX2) as I have read a number of threads about it and most seem to have laptops of this type.

Thanks again

Bob

Bruce Hill 08-20-2008 04:22 AM

This is the first I've read of you having a "Triggering Udev Events" error message.

You should post it so others might have a look.

Do you still have the other problems you mentioned in the OP:
Quote:

controls stop, sound dissappears, and when I attempt to reboot the system hangs on a black screen for about two minutes then finally reboots, but only after I lightly touch the power button.
I'm not there with you, but more than likely you're touching the power button is
the same as pressing Reset.

BobNutfield 08-20-2008 10:02 AM

Well, the exact message is:

Quote:

Triggering udev events: /sbin/udevtrigger --retry=failed
There is at least one other user who reports the same issue:

http://www.linuxquestions.org/questi...ng+udev+events

I have noted that this thread is about wireless networking, and this message appears just before the loading of network modules. I am using ndiswrapper and this may well be the issue. Wireless works fine for me, so I do not believe that has anything to do with the other issues of first posted about.

Quote:

I'm not there with you, but more than likely you're touching the power button is
the same as pressing Reset.
Yes, I am sure that is correct. Lightly touching the power button is in effect a reset, but not like a hard shutdown because it just forces a continuation of the orderly shutdown after the hang mentioned. I installed Slack in April on this laptop and have been using daily since. The problems experienced in my opening post have only occurred half a dozen times, and have not yet reoccurred in the generic kernel. I may in fact solve it using this kernel.


All times are GMT -5. The time now is 10:14 PM.