| 1 |
On 07/17/2012 07:02 PM, William Hubbs wrote: |
| 2 |
> On Tue, Jul 17, 2012 at 05:20:13PM -0400, Richard Yao wrote: |
| 3 |
>> An often cited benefit of the /usr merge is the ability to put |
| 4 |
>> everything but /etc on NFS and for that reason, we need to force an |
| 5 |
>> initramfs on people happily using /usr without it. |
| 6 |
> |
| 7 |
> This is not quite correct. The initramfs is required because of [1]. |
| 8 |
|
| 9 |
What is [1]? |
| 10 |
|
| 11 |
>> Interestingly, the /usr merge changes made to genkernel permit us to |
| 12 |
>> mount /etc from a genkernel-built initramfs by putting /etc on a |
| 13 |
>> separate mount point in fstab and then doing `echo /etc >> |
| 14 |
>> /etc/initramfs.mounts`. |
| 15 |
> |
| 16 |
> That doesn't negate putting /usr on nfs and making it possible for |
| 17 |
> different hosts to share it. |
| 18 |
|
| 19 |
People can still have different hosts share / with host-specific stuff |
| 20 |
(e.g. /etc) mounted by genkernel. |
| 21 |
|
| 22 |
>> I have also been told that the /usr merge is necessary because upstream |
| 23 |
>> will force it on us. Interestingly, most of @system on Gentoo Linux is |
| 24 |
>> GNU software, which would need to stop supporting things in / in order |
| 25 |
>> for that to happen. As far ass I know, systemd does not work on GNU HURD |
| 26 |
>> and it would be incapable of functioning if the GNU project made this |
| 27 |
>> change. Hell will freeze long before that happens. |
| 28 |
> |
| 29 |
> This is basically not relevant since we do not support HURD. |
| 30 |
|
| 31 |
It is relevant because it guarantees that the GNU stuff in @system will |
| 32 |
continue working. That allows us to narrow our focus to the non-GNU |
| 33 |
things required to use Gentoo Linux. |
| 34 |
|
| 35 |
Looking at @system and what it typically pulls into @world, the only |
| 36 |
thing that might cause a problem is udev, although virtual/dev-manager |
| 37 |
is in @system, rather than udev. If that happens, we have a few ways of |
| 38 |
dealing with that: |
| 39 |
|
| 40 |
1. Patch udev. |
| 41 |
2. Fork udev. |
| 42 |
3. Consider breaking people's systems then. |
| 43 |
|
| 44 |
Until then, doing what RedHat wants is unnecessary. |
| 45 |
|
| 46 |
>> Lastly, don't tell me to read systemd's case for why we should break |
| 47 |
>> people's systems. I have read it and I find it flawed. There is |
| 48 |
>> absolutely no need for us to make this change. |
| 49 |
> |
| 50 |
> Without elaboration on why you find their case flawed, this sounds |
| 51 |
> like the typical, "if it isn't broke, don't fix it" argument. |
| 52 |
> While that has merrit, if there are advantages to doing |
| 53 |
> something, like I think there would be to doing the /usr merge, it may |
| 54 |
> be worth the transition, especially if we can make it as smooth as |
| 55 |
> possible. |
| 56 |
|
| 57 |
The cost to benefit ratio is simply too low for "lets change it because |
| 58 |
it could be better this way" to merit making the change. The things that |
| 59 |
I have heard are going to break existing systems that I have gone |
| 60 |
through some trouble to support. I really don't want to see that. |
Archives