Gentoo Archives: gentoo-dev

From: Joost Roeleveld <joost@××××××××.org>
To: gentoo-dev@l.g.o
Subject: Re: [gentoo-dev] udev and /usr
Date: Thu, 15 Sep 2011 22:13:59
Message-Id: 6080144.TD9jt0YPcQ@eve
In Reply to: Re: [gentoo-dev] udev and /usr by Zac Medico
1 On Thursday, September 15, 2011 01:34:50 PM Zac Medico wrote:
2 > On 09/15/2011 01:03 PM, Joost Roeleveld wrote:
3 > > But, with udev then failing, will there be the /dev-entries to mount the
4 > > different partitions to fix the environment?
5 >
6 > I the preferred approach is to enable CONFIG_DEVTMPFS=y and
7 > CONFIG_DEVTMPFS_MOUNT=y, so that the kernel populates /dev for you
8 > automatically.
9
10 Will this be sufficient for "localmount" to get the system to work correctly?
11 It is my understanding that some udev-scripts are the actual problem that is
12 being solved with this?
13
14 I wasn't aware of these kernel options also being required.
15
16 > >>> Also, I was actually hoping for a reply to the rest of my email as
17 > >>> well, especially the idea for splitting udev into 2 seperate
18 > >>> processes.>>
19 > >> In essence, what your doing here is playing a game of "let's see how
20 > >> long we can delay the mounting of essential filesystems". If you play
21 > >> this game, then again, you expose yourself to the possibility of
22 > >> unsatisfied dependencies. Therefore, the only foolproof approach is to
23 > >> mount all essential filesystems as soon as possible (via initramfs).
24 > >
25 > > True, but I don't have any scripts configured for udev on my desktop.
26 > > My server has some scripts related to Xen, and those are all under
27 > > /etc/xen/...
28 > >
29 > > In this case, would it still be necessary to use an initramfs?
30 >
31 > Well, as long as your essential filesytems aren't mounted before init is
32 > called, there's always the possibility that some issue of unsatisfied
33 > dependencies will arise in the future. Therefore, the most foolproof and
34 > future-proof approach is to have them all mounted before init is called.
35
36 With systemd, one of these is the dbus-stack. Yes, I'm aware of this.
37 But, if systemd isn't used, init should work. Or have I missed something about
38 init being deprecated for systemd?
39
40 I think systemd is nice for desktops/laptops. But on servers it seems to be
41 overkill to me and as I umount filesystems as part of my backup-scripts,
42 having something force-mount them in the background is going to kill those
43 scripts.
44 But this bit is off-topic.
45
46 > >>> If someone can explain to me why my idea won't work, please let me
47 > >>> know.>>
48 > >> If your goal is to expose yourself to the possibility of unsatisfied
49 > >> dependencies, they your idea will achieve it.
50 > >
51 > > No, my goal is to come up with a different solution to this problem
52 > > which, on my system and possibly also on a lot of other systems,
53 > > doesn't actually exist.
54 > If a problem doesn't exist now, that doesn't mean one won't arise in the
55 > future. As said, the most future-proof approach is to have them all
56 > mounted before init is called.
57
58 Or, if I am not mistaken, before udev is started when not using systemd.
59
60 --
61 Joost

Replies

Subject Author
Re: [gentoo-dev] udev and /usr "Michał Górny" <mgorny@g.o>