Darren J Moffat wrote:

> Darren.Reed at Sun.COM wrote:
>
>> Darren J Moffat wrote:
>>
>>> Darren.Reed at Sun.COM wrote:
>>>
>>>> Rather than create N different services for IPFilter, we've
>>>> gone with keeping the existing service name but allowing
>>>> SMF to be used to control what it does at a finer level.
>>>
>>>
>>>
>>> What is the reason for this ?
>>
>>
>>
>> Please review the earlier discussion on this subject.
>
>
> Please point me to the thread it isn't obvious where I should be looking.
>
> I fully understand why you need to be able to control ipf/ipnat etc 
> separately.  I don't understand the rationale for one service versus 
> multiple.


See comments from Michael Shapiro:
http://www.opensolaris.org/jive/thread.jspa?messageID=54238#54238
and Jim Carlson:
http://www.opensolaris.org/jive/thread.jspa?messageID=54183#54183

on why breaking up the SMF service, as originally suggested, was not
seen to be a good step forward.

>>> I don't see anything that description of ipfadm that you can't to 
>>> today with svcadm and svcs, if you used a separate service for each of
>>> the things that make up IPfilter.
>>
>>
>>
>> Sure, you can use svcadm/svcs to achieve those things but how hard
>> are they to do using them?  Not to mention that there are some steps
>> that aren't obvious to new comers (ie. svcadm refresh).  I believe
>> there is worthwhile value added by using this extra layer, as it were.
>>
>> So far as I'm concerned, the use of svcadm/svcs is in this case an
>> implementation detail of the mechanism used to manage the components
>> of IPFilter.
>
>
> The answer I was actually expecting and would have been happy to 
> accept was that IPfilter on all operating systems was going to get an 
> ipfadm command with those options (the implementations would differ).


That is definately possible.

> I'm not sure I like the adding of fooadm when svcadm/svccfg could do 
> it.   It is already confusing enough with inetadm (and some people 
> believe it was a mistake).


In this case the granularity of the service presented by inetadm is at a 
level
that would also be suitable for svcadm.

Darren


Reply via email to