Bug#401916: Bug 401916
On Thu, Mar 15, 2007 at 12:17:33AM +0100, Michael Prokop wrote:
> * David Härdeman <david@hardeman.nu> [20070314 20:58]:
>
> > Feel free to install the new initramfs-tools, then apply the
> > above patch to udev, install, regenerate the initramfs image and
> > try booting with rootdelay=20 and see if it pauses 20 secs at the
> > right stage of the boot.
>
> Ok, just verified it:
>
> Using current version of initramfs-tools (0.85f) with current udev
> and lilo plus udev-support_rootdelay.patch applied works when using
> something like "rootdelay=7" as bootoption on my external USB
> harddisk installation. *Without* the rootdelay option the system
> definitely fails to boot (as documented in #401916 in more detail),
> *with* the rootdelay option it works as expected.
cool.
> JFTR: $KERNELSOURCE/Documentation/kernel-parameters.txt documents
> the option as well:
>
> | rootdelay= [KNL] Delay (in seconds) to pause before
> | attempting to mount the root filesystem
>
> .. so this feature should hit Etch anyway.
yeah i know it's an official bootparam.
> Further suggestions:
>
> It would be fine to provide some output while running the "sleep
> $ROOTDELAY" so users without usplash_write are notified about this
> issue as well.
well that is nice to have,
but not a killer feature we just do what the user asked aka sleep.
> If mounting the rootdevice *fails* there should be some basic
> documentation available on the screen how to fix this issue. The
> rootdelay stuff should be mentioned as well as using features like
> root=UUID=$UUID_OF_DEVICE (which works fine as well).
you are overshooting here,
that is support request on some irc or mailing list.
also d-i will soon install by uuid by default.
we provide already a help message when someone gets dropped to
shell and the message gets improved for 0.87.
also we can't put too much text there otherwise the interesting
error info scrolls off, yes nobody uses serial anymore..
thanks + best regards
maks
Reply to: