Index: [Article Count Order] [Thread]

Date: Mon, 26 Dec 2005 10:42:30 +0900
From: "haoda" <haoda@64translator.com>
Subject: [dhcptest:00053] Re: Can'tDecodeRelayReplyMsg
To: "'amit agarwal'" <amit_agarr@yahoo.co.in>
Cc: <dhcptest@tahi.org>, <haoda@64translator.com>
Message-Id: <200512260142.jBQ1gdkO023762@bahamas.64translator.com>
In-Reply-To: <20051223070013.33997.qmail@web8514.mail.in.yahoo.com>
X-Mail-Count: 00053

Hi, Mr. Amit,

Sorry, the reply is delayed because of the holiday from 12/23.


>If i send relay reply message to a multicast address, he is 
>comparing it with an unicast address...
>HDR_IPV6_relay_reply_nut_relay.DestinationAddress 
>received:ff02::1:2 = fe80::200:ff:fe00:a4a4
>But if i send the relay reply to the unicast address 
>fe80::200:ff:fe00:a4a4, the packet is not being received. 
>The IP address is not reflected i feel so because Neighbour 
>Solicitation is being sent to this address and it does not 
>recieve any NA back(This happens only when the interface
>does not have such an IP address).

I think you need change the value of field "Type" from "router"
to "host".
The TN can not accept the message from the NUT(Server) 
when it is set to "router" in current edition.
Please refers to the file named "00Readme".  

>And also please tell me whether the server should 
>send the relay reply message to multicast address 
>or the unicast address of the relay agent from 
>where it got the relay forward message 
In this test script, I think the server should send the relay-reply
message to the unicast address.
Please refers to RFC3315, P37-38, line 2068
"The server unicasts the Relay-reply message directly to the 
 relay agent using the address in the source address field from
 the IP datagram in which the Relay-forward message was received."



>This test cases are failing becuase script  
>is sending invalid SID in renew message
>./S_RFC3633_12.2_ClientInitPDRenew.seq 
>./S_RFC3633_12.2_ClientInitPDRebind.seq
>./S_RFC3633_12.2_ClientInitPDLifetime0.seq
>./S_RFC3633_12.2_ClientInitPDNoBinding.seq
>I feel that there is problem in the script please verify it .....
Thanks for your report.
I think them are bugs in current scripts when process SID from
the reply message.
I 'm  fixing them now.


Sincerely yours,
Haoda