[Iccrg] Question on RFC 2988 - TCP Retransmission timer

[Lars Eggert]

On 2007-9-20, at 14:45, ext Mark Allman wrote:
>> On 2007-9-20, at 0:58, ext Sushant Rewaskar wrote:
>>> (ii) Reduce minimum retransmission timeouts to 200ms or lower
>>
>> did any of these traces include GSM or UMTS access links? Because
>> they have transmission delays that can be significantly higher than
>> 200ms.
>
> I am not sure I follow here.  Are you saying that the min RTO that is
> (/was) preventing some sort of bad behavior in networks with GSM or  
> UMTS
> components?

Hm.

When I wrote the above, for some reason I was worried about this  
causing spurious timeouts. But thinking about this again, it probably  
won't in most cases, because it's the minimum that we're talking  
about, i.e., what you round up to when the timeout is calculated to  
be smaller.

(It might still cause spurious timeouts in some rare cases, such as  
when the path delay suddenly jumps from < 200ms to >> 200ms, but  
that's a corner case.)

Lars
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2446 bytes
Desc: not available
Url : http://oakham.cs.ucl.ac.uk/pipermail/iccrg/attachments/20070920/976bdc3b/smime.bin