RE: /scripts/local: Waiting for root file system...

["Petr Kohts" <kohts@yandex-team.ru>]

>  
> > The solution I'm proposing is cleaner and requires
> > less hand work for the end user. Here's the `diff -u`
> > (tested and working):
> 
> no it's even more dirtier.
> first of all you don't know which and what scripts are local-top
> second you sleep 3x, too lazy to do the math but something
> like rootdelay * rootdelay due to while loop.

I was not quite clear when defined the udev fix as dirty.

It's dirty not because it doubles the time to wait,
but it's dirty because it doubles the waiting logic.

Here's the excerpt from initramfs-tools man page:

local-top OR nfs-top: After these scripts have been executed,
the root device node is expected to be present (local) or
the network interface is expected to be usable (NFS).

So we know what are those scripts. The last thing
we need to check is whether we can run those
scripts several times without breaking thigs --
but this is guaranteed by the underlying tools
which they are using (mdadm, vgchange, etc.)

I'm sorry I don't have enough time to argue with you,
hope that you will find some time to think about the issue
and take the right decision.

Last thing I'd like to ask you -- invent something
which will work for the default installation,
not forcing user to alter kernel parameters.


Good luck,
Petya.