Mailinglist Archive: opensuse (856 mails)

< Previous Next >
Re: [opensuse] interesting reading about systemd
  • From: Andrei Borzenkov <arvidjaar@xxxxxxxxx>
  • Date: Mon, 3 Oct 2016 17:38:37 +0300
  • Message-id: <CAA91j0WBZEcX0ab_-aftEKTFN5wq-m7Uin7gY+83BdAVzfdJ=g@mail.gmail.com>
On Mon, Oct 3, 2016 at 4:26 PM, Anton Aylward <opensuse@xxxxxxxxxxxxxxxx> wrote:
On 10/03/2016 08:58 AM, Per Jessen wrote:
I'm sure that is/was a genuine problem, but allow me to count it as an
exception to prove the rule. I think it is very far from your example
to your general suggestion that

"... that if there is a problem I can not edit an script and solve it.
I have to wait for the devs to solve and distribute it".


Indeed.
As I say in another posting, if you are relying on /etc/fstab and the
generator
that parses it, you end up with the systemd semantics that maintain the
system,
automatically restarting when something 'dies' so as to maintain a constant
picture.


The problem with unexpected (un-)mounts has absolutely nothing to do
with generator or fstab.

If you want different, take the entry out of fstab and write your own mount
file
that does what you want.


Which will have exactly the same problem. Do you ever test what you suggest?

This is not a problem with systemd.
It is doing what it supposed to.

Without leaving user choice of changing it and without even telling
user it will do it. So yes, it is not a bug - it is design decision.
Both design decision and how it is implemented is questionable and
contributes to attitude towards systemd.
--
To unsubscribe, e-mail: opensuse+unsubscribe@xxxxxxxxxxxx
To contact the owner, e-mail: opensuse+owner@xxxxxxxxxxxx

< Previous Next >
This Thread
Follow Ups