On 8/15/06, David Lee <[EMAIL PROTECTED]> wrote:
On Tue, 15 Aug 2006, Alan Robertson wrote:
> David Lee wrote:
> > [...]
> > Alan: Could you briefly describe, please, the differences between the two
> > things:
> > (1) differences that are external black-box view (e.g. as documented in
> > those URLs);
> > (2) internal differences (e.g. allowing more interfaces).
> >
> > It still feels that we ought to be able to keep things to a simple, single
> > end-user (black-box) interface that, in internal construction, could take
> > best advantage of the particular OS/environment (including offering as
> > many interfaces as necessary).
>
>
> If you use IPaddr, then you typically get an interface for the address
> that looks something like eth0:0.
>
> If you use IPaddr2, then you get NO named interface - at least by
> default. And, it runs faster as a result.
>
> So, with IPaddr, if you do ifconfig afterwards, the interface for the IP
> address shows up. With IPaddr2, ifconfig will claim that no such
> address is managed.
Many thanks, Alan. Appreciated.
Was this difference (visibility of resulting IP/interfaces) a deliberate
design choice? Or was it a side-effect?
The difference in using ip2 instead of ifconfig that works much
faster. As far as I understands IPAddr2 won't work on other OSes,
Solaris for example.
My instant, gut reaction (admittedly never having used the "...2" variant)
is that this hiding of the IP/interface from the system's "ifconfig"
command feels unclean.
Does this difference ever get used?
What is the "it" that runs faster? (And how much and how importantly?)
IPaddr script? (Relatively unimportant.)
network activity on new IP/interface? (Could be important.)
It seems that these two RAs are still externally very similar (even if
wildly different internally). The principal differences seem to be:
1. Internal: Completely different implementation (but "...2" unlikely to
work on non-Linux).
2. External: "...2" produces no named result (IP/interface hidden from
ifconfig) and the hiding functionality may be unavailable on some OSes.
3. Internal (reflects to external): "...2" can support more interfaces.
> Lars and company know this. They just ignore it - presumably because
> they think it unimportant.
Aside: Looking in CVS, I see that Lars actually wrote "IPaddr2" (at least,
he did the first check-in). So it is likely that if the hiding of the
IP/interfaces is deliberate, he might know about it. It would be
interesting to get his own view here.
On track again:
How important is it (hiding the IP/interface)?
Suppose there were a single user-visible "IPaddr" which permitted a user
option to attempt to hide the resulting 'eth9:9' IP/interface? And which
internally detected the OS/environment and used the most appropriate
implementation to do all its functionality?
Thanks again. Best wishes.
--
: David Lee I.T. Service :
: Senior Systems Programmer Computer Centre :
: Durham University :
: http://www.dur.ac.uk/t.d.lee/ South Road :
: Durham DH1 3LE :
: Phone: +44 191 334 2752 U.K. :
_______________________________________________________
Linux-HA-Dev: [email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/
_______________________________________________________
Linux-HA-Dev: [email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/
