Re: Disabling temporary addresses by default?

[Fernando Gont <[e-mail]>]

On 30/1/20 16:14, Michael Richardson wrote:
> 
> Lorenzo Colitti <lorenzo=[e-mail]> wrote:
>      >> Option #2: Reduce the N ratio, but allow ongoing sessions to use invalid
>      >> addresses. This could be worse than legacy RFC4941 (if long lived
>      >> sessions last more than a week), since now temporary addresses would not
>      >> have a limited lifetime (the lifetime would only be limited wrt the
>      >> ability to employ them for new transport protocol instances)
>      >>
> 
>      > ISTM that "using invalid addresses" is poorly defined and likely difficult
>      > to implement. It also isn't not much (if at all) different from the current
>      > status quo with lots of deprecated addresses. The only difference is that
>      > it forbids allow applications from purposely binding to an expired address.
>      > But applications don't do that - why would they? The current privacy
>      > address is better than that address in every way.
> 
> So, I have an FTP control channel open for two weeks.
> I want to bind a new socket for data tranfers... and now I can't bind the IP
> address that I want. 

In that case, set the appropriate socket option, and override the 
default such that all your connections use a stable address.




-- 
Fernando Gont
SI6 Networks
e-mail: [e-mail]
PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492