[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: hibernate / ACPI suspend: help needed



On Tuesday 22 March 2005 17:32, Rurik Wahlin wrote:
> Thanks for the sugestions.
>
> > That's because the swappartition contains the memory image. It's highly
> > recommended that you add mkswap to your init scripts, otherwise your
> > risking severe filesystem corruption of you suspend, then boot a
> > non-suspend aware kernel and then try to resume afterwards. That's
> > because the filesystem is in inconsistent state during hibernate, booting
> > and mounting it will modify it, but a resumed kernel won't notice that it
> > has been booted and modified in the meantime and will thus cause
> > corruption.
>
> Yes I have been very carefull to restore the swap between each new try...
> ;)
>
> > Try to get rid of the initrd, i.e. compile in everything necessary to get
> > the system booting, and it should work.
>
> I'll try that. Do you think the 2.6.8 will be ok? Someone suggested
> earlier that I should use 2.6.11. I had some problems when trying to run
> with 2.6.10 and would prefer the kernel I know work without any problems
> with the rest of the system.

You can at least try 2.6.8, but there have been a couple of improvements over 
this version in more recent kernels: drivers cause less problems, swsusp 
should be a little faster and suspend more reliable. Also I'm not sure if 
2.6.8 already supported suspening x64, but newer kernels are supposed to. So 
especially if you run into problems, try later version (but I think that's 
obvious). There's a lot of work being done WRT powermanagement, so at least 
that should only be getting better.

FWIW, I myself am not using swsusp on my x64 hardware (yet), since it's 
running 24/7 anyway, but am a happy user of suspend2 on my x86 notebook.

Cheers,

sebas
-- 
  http://vizZzion.org   |   GPG Key ID: 9119 0EF9 
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
The world is moving so fast these days that the man who says it can't be done 
is generally interrupted by someone doing it. - E. Hubbard

Attachment: pgpWtUib9i4u4.pgp
Description: PGP signature


Reply to: