Hi
Something like this would be valuable for network interface services.
Would be nice to be able to write a generic "udhcpc_at_" service, and then
instantiate "udhcpc_at_eth0" and "udhcpc_at_eth1".
My 0.02$.
/Esben
On 13 January 2016 at 18:50, Olivier Brunel <jjk_at_jjacky.com> wrote:
> On Fri, 25 Dec 2015 11:43:31 +0100
> Laurent Bercot <ska-supervision_at_skarnet.org> wrote:
>
>> Hello everyone, and happy holidays!
>>
>> In the past few years, there have been some bits and pieces of
>> discussion about "instanced services", i.e. some kind of supervised
>> service that would be able to create different instances of the
>> process at will. But it never got very detailed.
>>
>> I'd like to ask you: if you feel that instanced services are
>> useful, what functionality would you like to see ? Concretely,
>> practically, what do you want to be able to do ?
>>
>> Please stay away from mechanism and implementation details;
>> I'm just trying to get a very high-level, conceptual feel for it
>> at the moment - "what", as opposed to "how".
>>
>> Thanks !
>>
>
> Hey,
>
> So I'm not sure this is what you're thinking of, but I can describe
> what "instanced services" are with regards to anopa; And that's
> basically just a way to use the same servicedir for multiple services,
> or more exactly only create/manage a single servicedir, but be able to
> use it for more than one services (or instances).
>
> Can be helpful when you have multiple instances of "the same"/similar
> services, where the only difference would be an argument to the daemon
> or something to that effect.
>
> This gets into mechanism I'm afraid, but I'm not sure how to best
> explain it otherwise: servicedirs would be created/managed in a
> sourcedir, and enabling (or compiling, might be the equivalent in
> s6-rc) means pretty much copying the servicedir from the sourcedir into
> the scandir (or compiled, or whatever), where it becomes "active"/in
> use.
>
> With instanced services, it means when you enable such a service you
> add/specify an instance name. So e.g. the servicedir is "getty_at_" but
> you enable "getty_at_tty2" -- which just means the servicedir getty@ is
> copied under a different name in the scandir.
>
> The intent being that to enable another getty, you just enable
> getty_at_tty3 and that's it. One actual servicedir is being maintained,
> but used many times (as different instances, which really are
> different services in the end). If you need to fix/change the
> servicedir, it's done once and then applied to all instances.
>
> (Obviously the scripts (e.g. run) needs a way to use the instance name,
> as that usually ends up being an argument to the daemon, or to point
> to/read from a different config directory, etc)
>
> Note that this actually isn't limited to longruns, but also works for
> oneshots. Hence why I'm thinking this might be different to what you're
> thinking of.
>
> Anyhow, HTH
> -j
--
Esben Haabendal
Ulstrupvej 7, 9500 Hobro, Denmark
Received on Mon Jan 18 2016 - 10:05:11 UTC