I missed this post. I have the same situation too, and it isn't so nice to
me that Ctrl-Alt-Del can do it. My situation is so bad that I have to use
Alt-SysRq-B to reboot....... :(
My hangup is much more random.... sometimes freeze at,say 2 minutes, after
reboot(just after started X); and sometimes used for a whole day without
problem.
I am using kernel-2.4.2-20mdk too, and the whole / is reiserfs, except
/boot. No wonder my freeze is so bad ;)
Abel Cheung
On 30 Mar 2001, Steve Fox wrote:
> On 31 Mar 2001 01:31:35 +1000, Con Kolivas wrote:
> > There have been some posts here about machines hanging with
> > reiserFS. As I mentioned earlier I've been getting these
> [snip]
>
> I posted the below a couple days ago with no response. I believe it may
> be tied to your situation.
>
> ----------------------------
> The last couple of days or so I have been using my Thinkpad T20 just
> fine for about an hour. Then suddenly for no apparent reason, some of my
> applications freeze up (seem to be apps which use networking like
> Evolution or X-Chat). From that point on I can't even close them because
> they are all in uninterruptable status ('D' on ps). At that point I can
> not even shut down GNOME except by killing the X server. Then when I
> press Ctrl Alt Del to reboot it looks at me funny and laughs.
>
> Using:
> kernel-2.4.2-20mdk
> kernel-pcmcia-cs-2.4.2-20mdk
>
> With an Intel EtherJet 82557 mini-PCI ethernet card (eepro100 module).
>
> Anyone else hitting this?
> ------------------------------
>
> I have / on ext2 but /home is on ReiserFS. My kernel is on -22mdk now
> and it still occurs. I'm thinking that it is indeed tied to reiserfs
> because I can boot into GDM just fine (ie, I'm not accessing /home just
> yet). Tonight when I logged in GNOME started to load then froze once I
> started up some applications. I killed the X server, logged into
> console, and did a 'ps' to see almost every process started since my
> login was in uninterruptable sleep (aka 'D') status. This is definitely
> a kernel bug. I hope Mandrake stops the press in time to fix this.
> Someone should report it to the kernel mailing list I think.
>
>
