On Fri, 1 Jun 2007 17:54:18 -0400
Joey Hess <[EMAIL PROTECTED]> wrote:

> In my tests installing the laptop task in lenny, uswsusp still prompts
> with uswsusp/continue_without_swap.
> 
> I've looked at this some. It seems that the first time the config script
> runs, all is ok. uswsusp/resume_device is not set, and
> /etc/initramfs-tools/conf.d/resume exists and gets parsed and the d-i
> hack used.
> 
> The problem occurs the second time the config script runs. Remember, the
> config script typically runs once during preconfiguration, and a second
> time when the package is actually installed.

I hadn't remembered... Also during my test it didn't show up. I tested
it by going to a shell in the installer and installing the package by
hand. In such a situation it doesn't get pre-configured, I guess.

I don't know a way out at the moment. I seems we're back at trying to
figure out if we're run under d-i or not. I could write to some state
variable the first configuration run, and then clean this var when we
have run postinst. But this is becoming more and more ugly...

I'm tempted to wait until d-i finally dumps the devfs style naming. And
I don't think that a change that tries to fix this will be allowed into
an update for etch, do you?

grts Tim



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Reply via email to