| 1 |
On Mon, Aug 19, 2013 at 1:04 PM, Alan McKinnon <alan.mckinnon@×××××.com> wrote: |
| 2 |
> On 19/08/2013 11:31, pk wrote: |
| 3 |
>> On 2013-08-19 00:49, Dale wrote: |
| 4 |
>> |
| 5 |
>>> Picking random message sort of. Isn't eudev still going to support a |
| 6 |
>>> separate /usr? That is my understanding. If eudev is not then I may |
| 7 |
>>> have to reconsider some things myself here. |
| 8 |
>> |
| 9 |
>> Yes, that is my understanding as well. But the "decision" to not support |
| 10 |
>> a separate /usr lies higher up in the system hierarchy (as I understand |
| 11 |
>> it). Gentoo as a system will not support a separate /usr if we are to |
| 12 |
>> believe the conversation (I haven't seen any official notice of this |
| 13 |
>> though). That is the sad part. The problem I have, as an engineer, is |
| 14 |
>> that "everybody" says that a separate /usr is broken, that sysvinit is |
| 15 |
>> broken without explaining why. In order to fix a problem you need to |
| 16 |
>> know what is broken... The people who claims the brokenness are, imo, |
| 17 |
>> hand waving and they've managed to convince higher uppers in the Gentoo |
| 18 |
>> infrastructure (as it seems). I guess if you repeat something often |
| 19 |
>> enough it becomes a "truth" or said person(s) just agrees to stop the |
| 20 |
>> nagging. |
| 21 |
> |
| 22 |
> |
| 23 |
> It's not that separate /usr is broken - it's not. |
| 24 |
> |
| 25 |
> The issue is a separate /usr without an initramfs. And the issue ONLY |
| 26 |
> occurs at early-boot time. |
| 27 |
> |
| 28 |
> The problem is that with modern hardware much code that was |
| 29 |
> traditionally stored in /usr may be needed early in the boot sequence, |
| 30 |
> before /usr is mounted. The obvious case is firmware and drivers, and |
| 31 |
> the usual example cited is bluetooth keyboards. If you need keyboard |
| 32 |
> input at this time, you need to have the bluetooth daemon running, which |
| 33 |
> is on /usr, which is not mounted. |
| 34 |
> |
| 35 |
> The solution is to use an initramfs, and on a technical level it's not |
| 36 |
> any different to needing a way to get the ext4 module off disk so you |
| 37 |
> can mount /. |
| 38 |
> |
| 39 |
> Some may argue that bluetooth keyboards are a rarity and that's tough. |
| 40 |
> Well, there's Macbook hardware, and phones which have soft keyboards. |
| 41 |
> But many scenarios could exist, all due to the fact that hot-pluggable |
| 42 |
> hardware can in theory run any arbitrary code to get itself up and |
| 43 |
> running, and if that code is on a volume that is not mounted... The |
| 44 |
> solution is obvious - all that code should be on / somewhere, or should |
| 45 |
> be mountable using an initramfs. |
| 46 |
|
| 47 |
You fail to understand why separate / is required. |
| 48 |
|
| 49 |
Had the argument was: If you have special needs then have /usr mounted at boot. |
| 50 |
I would have agreed. |
| 51 |
This means that if you are using bluetooth keyboard, well you do have |
| 52 |
an extra requirement. |
| 53 |
|
| 54 |
However, because of your specific configuration drop the ability to |
| 55 |
recover from filesystem corruptions or be able to repair is totally |
| 56 |
different issue. |
| 57 |
|
| 58 |
> Personally, I think that splitting / and /usr is a daft idea: |
| 59 |
> |
| 60 |
> a. I have multi-TB hard disks, completely unlike the 5M monsters that |
| 61 |
> Thomson had to deal with in the 70s |
| 62 |
|
| 63 |
You could have mounted several disk at boot even in the 70s. |
| 64 |
|
| 65 |
> b. I haven't had /usr break on me during boot requiring busybox in |
| 66 |
> maintenance mode for at least 5 years. Every startup failure in that |
| 67 |
> time required a rescue cd anyway, and I always have one of those handy |
| 68 |
|
| 69 |
This is your take... and it is totally wrong. |
| 70 |
|
| 71 |
> c. it IS useful for terminal servers, but those tend to have experienced |
| 72 |
> sysadmins, and they really should be OK with an initramfs (or their |
| 73 |
> vendor should ship one) |
| 74 |
|
| 75 |
Who is that vendor? so you along with systemd, udev, gnome, etc... do |
| 76 |
you suggest the same vendor will also provide initramfs for gentoo... |
| 77 |
maybe this is the next stage of systemd... |
| 78 |
|
| 79 |
> I'm often at the front of the Lennart-bashing parade, and what he says |
| 80 |
> often makes sense but only in his narrow view of the world, but in |
| 81 |
> *this* case, I can't help but admit he does have a point. |
| 82 |
|
| 83 |
Again, there is no reason why not support separate /usr configuration, |
| 84 |
people who have special needs, like running systemd or have special |
| 85 |
complex userland hardware that is a must for single user mode can |
| 86 |
always mount /usr at early stage. |
| 87 |
|
| 88 |
But because of the fact that you are using systemd or have bluetooth |
| 89 |
keyboard force everyone to merge /usr is something that is unclear to |
| 90 |
me. |
| 91 |
|
| 92 |
> |
| 93 |
> -- |
| 94 |
> Alan McKinnon |
| 95 |
> alan.mckinnon@×××××.com |
| 96 |
> |
| 97 |
> |
Archives