CARP behaviour

[Yar Tikhiy]

On Sat, Mar 03, 2007 at 11:44:12PM +0100, Andre Oppermann wrote:
> Yar Tikhiy wrote:
> >On Sat, Mar 03, 2007 at 12:03:05AM +0000, Bruce M Simpson wrote:
> >>During testing of M_PROMISC I noticed a couple of issues with our CARP.
> >>
> >>1. carp doesn't seem to maintain input/output statistics on its ifnet.
> >
> >This should be OK.  A carp(4) interface is just a place for CARP
> >settings to live.  No real traffic passes through it.
> >
> >>2. carp doesn't seem to detect that the underlying route to the subnet
> >>  its address is exposed on changed to another interface.
> >
> >I seem to recollect that I ran into this issue, too.  Of course,
> >it's a bug.  CARP apparently caches the old route forever.  It
> >should track such a change if it doesn't yet.  We have a sort of
> >notification for the event, don't we?  The userland has the routing
> >socket for that, but I don't remember about the kernel.
> 
> We shouldn't cache route pointers anywhere anymore.  It has been completely
> removed from the PCBs and things like gif and others.

Sounds like a good way to go, too! :-)  Thanks!

-- 
Yar