[plug] Random troubleshooting (half upgraded knoppix).

[Pabs]

On Mon, 15 Dec 2003, Cameron Patrick wrote:

> On Mon, Dec 15, 2003 at 02:11:46AM +0800, Pabs wrote:
>
> | I get this during bootup:
> | Sat Dec 13 20:27:23 2003: ******************************************************
> | Sat Dec 13 20:27:23 2003: * Warning: The dma on your hard drive is turned off. *
> | Sat Dec 13 20:27:23 2003: * This may really slow down the fsck process.        *
> | Sat Dec 13 20:27:23 2003: ******************************************************
> | What is dma, how does it make the hard drive faster, how do I turn it on?
> | would it be in the bios? where should i look?
>
> DMA stands for "Direct Memory Access" or some such - basically it means
> that the hard drive (and IDE controller) writes (or reads) data directly
> into (from) main memory, rather than the CPU feeding it a few bytes at a
> time.  It can make maybe a 10x speed increase, depending on your drives,
> CPU, etc.  Having DMA enabled is a Good Thing :-)

thks for the info

> It /might/ be disabled in the BIOS, but more likely because the kernel
> doesn't support it on your IDE controller or has been configured
> conservatively to disable it by default (DMA support has been buggy on
> some drives and controllers in the past).  If you could post any
> IDE-related bits of dmesg that could be useful.

hmm, will check bios tomorrow

root at chianamo:~# dmesg | grep -i ide
BIOS-provided physical RAM map:
ide_setup: hda=scsi
ide_setup: hdb=scsi
ide_setup: hdc=scsi
ide_setup: hdd=scsi
ide_setup: hde=scsi
ide_setup: hdf=scsi
ide_setup: hdg=scsi
ide_setup: hdh=scsi
Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
PCI_IDE: unknown IDE controller on PCI bus 00 device 48, VID=10de, DID=01bc
PCI_IDE: chipset revision 195
PCI_IDE: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0xb800-0xb807, BIOS settings: hda:DMA, hdb:pio
    ide1: BM-DMA at 0xb808-0xb80f, BIOS settings: hdc:pio, hdd:DMA
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
ide-cd: passing drive hdd to ide-scsi emulation.
scsi0 : SCSI host adapter emulation for IDE ATAPI devices

> | when starting up all the filesystems in /etc/fstab are not mounted
> | mount -a helps, but it used to do that all by itself. wtf is going on?
>
> Perhaps /etc/init.d/mountall.sh is screwed up, or not getting called?

hmm:

root at chianamo:~# dpkg -S /etc/init.d/mountall.sh
initscripts: /etc/init.d/mountall.sh
root at chianamo:~# apt-get --reinstall install initscripts
Reading Package Lists... Done
Building Dependency Tree... Done
0 upgraded, 0 newly installed, 1 reinstalled, 0 to remove and 96 not
upgraded.
Need to get 0B/15.1kB of archives.
After unpacking 0B of additional disk space will be used.
Do you want to continue? [Y/n] y
(Reading database ... 100991 files and directories currently installed.)
Preparing to replace initscripts 2.85-7 (using
.../initscripts_2.85-7_all.deb) ...
Unpacking replacement initscripts ...
Setting up initscripts (2.85-7) ...

root at chianamo:~# less /etc/init.d/mountall.sh
/etc/init.d/mountall.sh: No such file or directory

> | When I do a halt/shutdown, init sez at the end of the sequence:
> | INIT: no more processes left at this runlevel
> | it used to just poweroff, can anyone shed some light?
>
> That also sounds like an init-scripts problem.  Perhaps the symlinks in
> /etc/rc0.d/, /etc/rcS.d/, /etc/rc1.d/ are dodgy?

eek: looks like the initscripts package is horribly broken, or has been
split into some other packages, hmm, the deb is ok. ok, purging &
reinstalling worked, fixed mountall.sh too. well that was fun, esp typing
in "Yes, do as I say!".

> No idea there, sorry :(  Perhaps something related to the setserial
> package (e.g. /etc/init.d/setserial not being run, or configuring the
> wrong thing)?

hmm, reinstalled it, will test tomrrow after i power on again

> | Damn, I can't wait for sarge, so I can get rid of knoppix.
>
> It was released two weeks ago ... in some parallel universe where aj's
> pronouncements bear some correlation with reality :-(  In the mean time
> there are release critical bugs breeding and multiplying, a
> debian-installer which sort-of half works on i386 and powerpc only,
> antique versions of lots of software (being held up by bugs in
> everything else)... *sigh*

Well the compromise kinda put a damper on things. Reading the bits from
the rm (pointed to off debianplanet) it seems the old apathy bug is
hitting debian a little. /me rounds up some cheerleaders to send to dds.
perhaps that will motiveate them a little.


Thanks for the info and help.
muuuusssst sleeeep

Bye,
Pabs