On Fri, 2008-11-14 at 17:23 -0500, Seth Vidal wrote:
On Fri, 14 Nov 2008, Dan Williams wrote:
> On Fri, 2008-11-14 at 17:10 -0500, Seth Vidal wrote:
>>
>> On Fri, 14 Nov 2008, Jeremy Katz wrote:
>>
>>>
>>> Also, there's this weird "running daemons are bad" mentality
which I'm
>>> not really sure what the right way to approach is. But it's one that
>>> I'm not sure how true it is in our current world of increasingly moving
>>> functionality out of the kernel and into userspace in which case you
>>> have to have something running in addition to the kernel.
>>>
>>> Maybe to try some examples -- irqbalance is a daemon and not in the
>>> kernel anymore[1], does that make it "not useful"? Or for another
side,
>>> various kernel threads are really just daemons... maybe we shouldn't
run
>>> them either?
>>>
>>
>> if you replace useful with necessary in his argument I think it ends up
>> making more sense.
>>
>> If the daemon is not NECESSARY for the task it is fulfilling then why have
>> it running?
>
> So I guess it boils down to how you define "task" then...
Actually it boils down to what the task is.
If the task is:
[snip]
Then everything else is unnecessary to have running.
If the task is to deal with complicated dhcp and wireless configurations,
multiple and complicated vpn configurations and notify the
console/desktop user about all of these things (which a lot of laptops
fall into that category) then NM is extremely useful and I'd argue
necessary for a convenient computing experience.
But there's a cost to maintaining two systems that are entirely
divergent and with little sharing/modularity. I don't see the value of
"don't run a daemon" as outweighing the cost of "keep another system
maintained".
Jeremy