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

Re: sysvinit is still here, and here to stay for jessie (was Re: systemd is here to stay, get over it now)

Hash: SHA512

On 07/04/2014 04:52 AM, Philip Hands wrote:

> The Wanderer <wanderer@fastmail.fm> writes:
>> ... particularly because I use rather fewer things than many other
>> people, and don't use most fancy GUI elements. (For example, I
>> don't have a graphical "power button" at all; I shut down by
>> exiting my window manager, logging out of the console where I had
>> originally run startx, logging in as root, and running 'shutdown
>> -h'.)
> So, let me get this straight:
> You're saying that if, having decided to postpone rebooting after an
> upgrade where any reasonable person would expect to reboot,

This part is precisely what I'm objecting to. I don't consider being
expected to reboot *in order to maintain existing functionality* after
an upgrade to be reasonable. (Being expected to reboot in order to use
the new functionality, for a sufficiently low-level component of the
system, is of course entirely reasonable.)

At the very least, in the very rare case that rebooting is actually
required, a prominent pre-install "this install will require you to
reboot ASAP" notification would be appropriate.

The situation is different in Windows (where such "please reboot now"
notifications are very common post-install, including with ordinary
programs rather than low-level components), of course, but I've
considered that to be an example of one of the advantages of *nix over

> you hear rumours that features you don't use would have been broken
> if you had them installed, you'll be highly displeased -- is that
> right?


I'm saying that if something I do use is broken during that period
between upgrade and reboot, I'll be highly displeased.

It's possible that nothing I do use will be affected, but it's also
possible that something I use will indeed be affected. It's not remotely
clear yet (at least to me) what features will be broken, or indeed even
which of the many systemd-related packages is expected to potentially
cause such breakage.

I would not be in the least bit surprised if I were not the only one who
would be displeased at a "continuing to use your computer indefinitely
after upgrading and before rebooting is not a scenario which is expected
to work" situation, given the historical tendency for people to chase
their own personal uptime records.

> and this is on a laptop, where you run testing, and which generally
> gets rebooted by power outages, rather than any UI component that may
> or may not be working at the time.

No. This is on both a laptop and a desktop; the desktop generally gets
rebooted by power outages, and the laptop by "the battery died because I
left it suspended for too long without plugging it in".

This also isn't about reboot methods which may or may not get broken;
it's about *other* things which may get broken. Nothing has been said to
narrow down what "other things" those might be, AFAIK, only that
GUI-based reboot methods must *not* get broken.

> and you're inflicting this nonsense on the thousands of readers of
> this list for what reason?  If it's to gain our gratitude and
> respect, I'm afraid it's not working on me.

(Why on Earth would anyone think that someone would raise an objection
in order to get gratitude from those to whom the objection is being

I spoke up because I consider a scenario of "everything works, a routine
upgrade occurs, now something is broken until a reboot is performed" to
be undesirable and unacceptable, for pretty much any reason, and I had
never previously encountered a suggestion that such a scenario might be
considered "normal and expected" in any *nix environment.

If it were guaranteed that the "systemd" upgrade would *not* be a
"routine" upgrade, but a special one done with full knowledge of what is
happening, that might mitigate the problem somewhat - but we've already
seen that systemd can easily be pulled in without the user realizing
that it's going to happen, or indeed noticing that it *has* happened
until after the fact.

- --
   The Wanderer

Secrecy is the beginning of tyranny.

A government exists to serve its citizens, not to control them.
Version: GnuPG v1
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/


Reply to: