Re: [NAILS REQUESTED PREVIEW]: Re: SCTP failover

From: David Lehmann <lehmann@ulticom.com>
Date: Mon Feb 14 2005 - 11:05:32 EST
('binary' encoding is not supported, stored as-is) ('binary' encoding is not supported, stored as-is) David Lehmann wrote:
> Janardhan Iyengar wrote:
>
>>>> Switching over to an alternate on an FR may give you really good
>>>> failover,
>>>> but will likely hurt performance when there's no failure (the common
>>>> case). Optimizing for the common case makes sense.
>>>
>>>
>>> Actually, if a failure occurs, you will not be able to fast rtx because
>>> the ack clock will be lost. So switching to an alternate on an FR won't
>>> even improve failover time.
>>
>>
>>
>> Yeah, good point. In fact, turning that reasoning around: when an FR
>> does occur, since an ack clock exists, the ack clock demonstrates that
>> there is no failure!
>>
>> To clarify, an FR occurs only when there are acks coming back; meaning
>> data is getting through and so are acks. Therefore, failure couldn't have
>> happened if an FR happens!
>
>
> Well, we tested it with a real failure. To be honest, I have not looked
> at that code in a long time. So maybe my follow-up comment was wrong.
> I have to look into it.

OK. As a clarification... FR only causes faster retransmission, but should
not cause the error count to increase nor the RTO to be adjusted...
Correct?

-- 
David Lehmann                          Ulticom, Inc.
AOL/Yahoo IM: davidULCM                1020 Briggs Road
1-856-787-2729                         Mt. Laurel, NJ 08054   USA
Received on Mon Feb 14 11:29:59 2005

This archive was generated by hypermail 2.1.8 : Mon Mar 13 2006 - 15:22:24 EST