[nycbug-talk] ucarp question

[David Lawson]

On Sep 20, 2006, at 10:38 AM, Isaac Levy wrote:

> On Sep 20, 2006, at 10:28 AM, Okan Demirmen wrote:
>> On Wed 2006.09.20 at 10:12 -0400, Isaac Levy wrote:
>>> On Sep 20, 2006, at 9:38 AM, Jeff Quast wrote:
>>>> On 9/20/06, Yarema <yds at coolrat.org> wrote:
>>>>>
>>>>> But I think the main advantage of ucarp over carp(4) is that you
>>>>> can do
>>>>> whatever you need in the up/down scripts.  carp(4) does not have
>>>>> this type
>>>>> of flexibility.
>>>>>
>>>>
>>>> Of course not. Thats what ifstated(8) is for!
>>>
>>> That's a new-ish OpenBSD thing?  Sounds interesting...
>>
>> well, if 2 years and 7 months is "new-ish", then yes ;)
>
> Well, in the scope of the big picture, I'll call that new :)
>
>>
>>> Hrm.... Niiiiice.
>>
>> yes it is.

Huh, that looks really interesting.  I ran into some problems with a  
heavily loaded box running CARP interfaces where some interfaces  
would fail to the secondary and others wouldn't, leading to some  
obvious routing problems.  I couldn't find a pure CARP way of dealing  
with the problem, it looks like this might be something that'd solve  
that issue reasonably elegantly.

--Dave