[SM-Discuss] HAL strangeness

Ethan Grammatikidis eekee57 at fastmail.fm
Sun Apr 19 22:32:18 EDT 2009


On Sun, 19 Apr 2009 11:29 +0200, "Treeve Jelbert" <treeve at scarlet.be>
wrote:
> My linux system has been out of commission for a couple of days. 
> Everything 
> would freeze when entering run level 5.
> 
> Eventually I found an error message from xorg, refering to HAL.  I then 
> discovered that hald had crashed.
> 
> I tried recompiling hal, dbus, etc.  I tried different kernels. I tried
> the git 
> version of hal. I enabled debugging in hal which generated masses of
> data, 
> mostly unintelligible.  I did see some messages about the cd drives near
> the 
> end of the debug messages.
> 
> Finally I removed one of the cd drives.  Everything now works again.  
> I suspect some sort of race condition with dual cd drives (1 sata, 1
> pata) and 
> a dual core processor.

I get a flood of errors from my pata dvd-rom in dmesg. This on a quad
core, only the 1 dvd-rom on pata and just 1 hard disk on sata. In the
past the drive used to stop working after some hours of uptime, but this
no longer occurs with kernel 2.6.26.2.

The flood of messages I get are all repeats of this:

hda: status error: status=0x58 { DriveReady SeekComplete DataRequest }
ide: failed opcode was: unknown
hda: drive not ready for command

Anecdotally, this is a similar message to that which I saw every boot
some 10 years ago. The drive in question then was a whopping 200MB and
worked flawlessly regardless of the error.

-- 
wow.. wondering how to make it more general and less hard-coded is making my head hurt. that's a good sign that it's my code




More information about the SM-Discuss mailing list