On Thu, Mar 23, 2000 at 04:39:47PM -0600, Manoj Srivastava wrote:
>
> Radim> Better is to define new standard as found in some packages:
> Radim> restart=method 2
> Radim> force-restart=method 1
>
> That may violate the principle of least surprise. I would find
> it strange if restart didn't start the service.
I would also be surprised if restart didn't start a service.
Especially if I was calling it because I suspected that something
might have killed it, but didn't feel like weeding through ps to find
out when I could just restart it with no problems.
> I do agree that these two behaviours are valid, and are
> required in some situations. What I do not have a handle on yet is
> which is the more common case, and thus which should be the default
> behavior.
I browsed /etc/init.d briefly. Overwhelmingly, no checks are made
to see if the service is running. When a check is made, it is
generally to see whether or not the service needs to be stopped before
being started, not to see whether a start should be made. Granted, I
only made a brief perusal, but I didn't see any cases where restart
would not have started a package that wasn't running.
> Either we need restart+force-restart, or we need a
> restart+maybe-restart.
I suggest "static-restart", static in the sense of not changing
current behaviour.
--
Zed Pobre <zed@debian.org> a.k.a. Zed Pobre <zed@resonant.org>
PGP key and fingerprint available on finger; encrypted mail welcomed.
Attachment:
pgpbaG4ehJ_Cr.pgp
Description: PGP signature