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

Re: from / to /usr/: a summary



md@Linux.IT (Marco d'Itri) writes:

> On Dec 22, Philip Hands <phil@hands.com> wrote:
>> Could we not have a package that checks if a system is going to be
>> unbootable under the circumstances in question (i.e. it has /usr on
>> nfs4, or whatever) and refuse to install on such a system, lets call
>> that package 'early-boot-usr'.
>> 
>> Then for the people that are having to put in extra effort into
>> packaging things that want to assume that /usr is there from early boot,
>> they just need to depend on early-boot-usr.
> Yes, we will need something like this. But sooner or later udev will
> depend on it, so I fear that it will not solve your problem.
>
> -- 
> ciao,
> Marco

This works verry badly with a package. Think of what happens on a system
where /usr is not early mountable. You run dist-upgrade, the new udev
comes in and depends on early-boot-usr, everything gets downloaded and
unpacked and then when configuring early-boot-use fails and a million
packages are left unconfigured. The system is unbootable and lots of fun
ensures.

Having early-boot-usr fail in preinst isn't much better. The problem is
you can't make packages refuse to install on such a system, you can only
make them fail. Not an option imho.


What early-boot-usr should do instead is ensure one of two things:

1) initramfs is used so /usr will be mounted there
2) everything needed to mount /usr is copied to /

The same hook that is used to update initramfs could be used to re-copy
stuff to / on updates too. It might not be pretty but it wouldn't leave
people in the ditch.

MfG
        Goswin


Reply to: