Re: [tcpm] New version of TCP Option for Transparent Middlebox Discovery available

[Joe Touch <touch at ISI.EDU>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1



Andrew Knutsen wrote:
> Joe Touch wrote:
>> Two other observations:
>>
>> - 'transparent' (that don't want to be seen) middleboxes will just
>> ignore this too
> 
>    True. Lets just say they are not of any of the defined "device types"...
> 
>>
>> - this option is (IMO) as likely to result in a dropped or RST
>> connection at firewalls, nats, or enpoints that incorrectly implement
>> options (i.e., that treat unkown options as errors rather than
>> ignoring them) as it may be to actually work
> 
>    But the situation will be better than the present one...  and this
> problem is actually surprisingly rare with the current invalid kinds
> used by all vendors in the WAN optimization market. I suppose this is
> partly because people don't deploy equipment that breaks their network.

I'm also asking whether they deploy equipment that breaks anything for
the end user (and sure, vendors do this all the time - they're called
NAT boxes ;-)

Joe
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (MingW32)

iEYEARECAAYFAkrDuFsACgkQE5f5cImnZrscHgCgkFstfA5gy6HBUbuK1wSp05oR
nYsAnicM5mEt9TZSsY/UyJhC0MwA6THS
=S8C8
-----END PGP SIGNATURE-----