1 |
24.02.2014 16:39, Mark David Dumlao пишет: |
2 |
> On Mon, Feb 24, 2014 at 3:11 PM, Yuri K. Shatroff <yks-uno@××××××.ru> wrote: |
3 |
>> 24.02.2014 02:32, Alan McKinnon wrote: |
4 |
>>> [1] For lack of a better term, let's just call systemd here a "system |
5 |
>>> controller". What is this ONE thing a system controller should do and do |
6 |
>>> it well? |
7 |
>> |
8 |
>> |
9 |
>> An init daemon generally does one thing well. |
10 |
> |
11 |
> it's obvious you haven't thought this through. |
12 |
> |
13 |
> consider, for a moment, that the "one thing well" that an init daemon |
14 |
> is supposed to do is |
15 |
> "run programs that do arbitrary things to get the system to an arbitrary state". |
16 |
> |
17 |
> do you not see a problem? |
18 |
|
19 |
No. As you say, ``an init daemon is supposed to do is "run programs``, |
20 |
until here you're right, but then you start talking about things the |
21 |
init doesn't do but the programs do. In your wording, an init daemon is |
22 |
also a DBMS, an MTA, a network startup daemon, a firewall, a getty and |
23 |
whatever program runs on the system. |
24 |
There was a post in this thread with a link to an opinion what an `ideal |
25 |
init` would do: just fork and exec anything in /etc/init.d or somewhere |
26 |
else. |
27 |
In the real world, it's of course not so simple. But it doesn't mean you |
28 |
may imply init's responsibility for `arbitrary` tasks. If I write an ASM |
29 |
program with an illegal instruction, is it the init's problem? If my |
30 |
mail/web server is DDOSed, is it the init's problem? If my HDD dies, |
31 |
also the init's problem? |
32 |
|
33 |
-- |
34 |
Regards, |
35 |
Yuri K. Shatroff |