Re: Query regarding number of paths for Multi-homing scenario

[Michael Tuexen <[e-mail]>]

On Aug 5, 2010, at 2:43 PM, Padmalochan Moharana wrote:

> Hi Michael,
> 
> Thanks for the quick answer. I have one more doubt.
> 
> In below scenario, suppose the network cable connected  to NIC having
> primary IP 192.168.1.2  is unplugged. Should endpoint B send data to
> endpoint A using secondary IP i.e. 192.168.1.3?
This is implementation dependent. You can not rely on it.
Most likely your setup is not substantially better than a
single homed one.
Also having multiple IP addresses from the same subnet
on one or more interface of a host results in platform
specific behaviour on the IP layer.

Best regards
Michael 
> 
> Br,
> Padmalochan
> 
> On Thu, Aug 5, 2010 at 5:59 PM, Michael Tuexen
> <[e-mail]> wrote:
>> On Aug 5, 2010, at 2:15 PM, Padmalochan Moharana wrote:
>> 
>>> Hi All,
>>> 
>>> I have some doubt regarding the number path use by an sctp association
>>> during multi-homing scenario.
>>> 
>>> Suppose end point “A” is single homed (having IP 192.168.1.1) and
>>> endpoint “B” is multi-home (having IP 192.168.1.2 and 192.168.1.3). As
>>> per definition of path in the rfc-4960 (section 1.3.)
>>> 
>>> Path: The route taken by the SCTP packets sent by one SCTP endpoint to
>>> a specific destination transport address of its peer SCTP endpoint.
>>> Sending to different destination transport addresses does not
>>> necessarily guarantee getting separate paths.
>>> 
>>> As for endpoint “A” two destination address is available then two path
>>> is available to send packets to endpoint B from A.
>>> 
>>> As for endpoint “B” only one address is available then only one path
>>> is available to send data to endpoint A.
>>> 
>>> Again as per the section
>>> 6.4.1. Failover from an Inactive Destination Address
>>> 
>>> When retransmitting data that timed out, if the endpoint is
>>> multi-homed, it should consider each source-destination address pair
>>> in its retransmission selection policy.  When retransmitting timed-out
>>> data, the endpoint should attempt to pick the most divergent source-
>>> destination pair from the original source-destination pair to which
>>> the packet was transmitted.
>>> 
>>> 
>>> My doubts:
>>> 1-    How many paths are available for endpoint B to reach endpoint A?
>> There is only one destination address available for endpoint B.
>>> 2-    Should endpoint B send HEARTBEAT to endpoint A from 192.168.1.3
>>> (secondary IP of B) in regular interval to test the reachablity of A
>>> using that IP?
>> HEARTBEATs are used to monitor idle destination addresses. So endpoint B
>> supervises only the one IP address of endpoint A.
>>> 3-    Can endpoint B send DATA to A (192.168.1.1) from IP 192.168.1.3
>>> (secondary IP of B)?
>> Sure. Endpoint A will accept all SCTP packets, no matter if they come
>> from 192.168.1.2 or 192.168.1.3. However, the ports must be OK and the
>> v-tag must be OK.
>>> 4-    Should endpoint A discard the DATA that received from
>>> 192.168.1.3(secondary IP of B)?
>> No.
>> 
>> Best regards
>> Michael
>>> 
>>> Regards,
>>> Padmalochan
>>> 
>> 
>> 
>