SORT BY:

LIST ORDER
THREAD
AUTHOR
SUBJECT


SEARCH

IPS HOME


    [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

    RE: FCIP Multiple Connection Management



    
    Charles,
    
    Agree 100% - this was discussed in the following thread,
    although, to my recollection, we did not explicitly 
    state a consensus and the text is still found in 
    Annex E of rev 0.3.
    
    http://www.pdl.cmu.edu/mailinglists/ips/mail/msg04662.html
    
    > -----Original Message-----
    > From: Binford, Charles [mailto:CBinford@pirus.com]
    > Sent: Monday, July 09, 2001 12:25 PM
    > To: ips@ece.cmu.edu
    > Subject: FCIP Multiple Connection Management
    > 
    > 
    > From pages 41-42 of FCIP rev 03 
    > *******************************************
    > 5.5 Multiple Connection Management 
    >   
    >    A pair of FCIP device endpoints MAY establish a certain number of 
    >    TCP connections between them. Since a Virtual ISL 
    > potentially maps a 
    >    fairly large number of FC flows (where a flow is a pair of Fibre 
    >    Channel S_ID, D_ID addresses), it may not be practical to 
    > establish 
    >    a separate TCP connection for each Fibre Channel flow. In order to 
    >    address this, an implementation MAY choose to manage a pool of TCP 
    >    connections for a single Virtual ISL and map Fibre Channel 
    > flows to 
    >    TCP connections of that ISL. However, while assigning 
    > Fibre Channel 
    >    flows to TCP connections, an implementation SHALL follow the 
    >    following rules: 
    >   
    >    1)  Once a Fibre Channel flow is assigned to a TCP 
    > connection within 
    >        the virtual ISL, it SHALL send all Fibre Channel 
    > frames of that 
    >        flow on that connection. 
    > 
    >    2)  When an FCIP endpoint processes any response traffic from a 
    >        particular target, the Endpoint SHALL send the response on the 
    >        same connection on which the request was sent. 
    >   
    >    3)  Any class 2 ACK frames SHALL be sent on the same connection in 
    >        which the original frame was sent. 
    >   
    >    These rules are in place to honor any in-order delivery guarantees 
    >    that may have been made between the two end points of the Fibre 
    >    Channel flow. 
    > *************************************
    > 
    > Given the above rules, consider the following:
    >                            |-----|             |-----|
    >                  --------  |FCIP |--TCP_Con_1--|FCIP |   -------
    > FC_Node_A -------|FC-SW |--|Dev1 |--TCP_Con_2--|Dev2 |---|FC-SW| ----
    > FC_NODE_B
    >                  --------  |-----|             |-----|   -------
    > 
    > FCIP devices 1 and 2 have a single Virtual ISL composed of 
    > two separate TCP
    > connections.  Assume FC_NODE_A and FC_NODE_B both send PLOGI 
    > to each other
    > at approximately the same time.  FCIP_Dev1 happens to choose 
    > TCP_Con_1 for
    > the FC Flow of S_ID_A-to-D_ID_B.  FCIP_Dev2 happens to choose 
    > TCP_Con_2 for
    > the FC Flow of S_ID_B-to-D_ID_A.  So far so good, both FCIP 
    > devices are
    > following the above rules.
    > 
    > Now, FC_Node_A sends the PLOGI ACC.  What is FCIP_Dev1 
    > supposed to do?  Rule
    > 1 says choose TCP_Con_1 (this is the S_ID_A-to-D_ID_B flow 
    > and that flow has
    > already been established to be TCP_Con_1).  Rule 2 says 
    > choose TCP_Con_2
    > (this is a response to a frame that was received on TCP_Con_2).
    > 
    > I believe rules 2 and 3 are not needed.  What the combination 
    > of all three
    > rules really say is there must be a defined negotiation or 
    > hash algorithm
    > that both FCIP_dev 1 and 2 have to follow to keep all of the 
    > traffic between
    > FC_Node_A and B on the same TCP connection.  I believe this is totally
    > unnecessary.
    > 
    > If both sides obey only rule 1 then "any in-order delivery 
    > guarantees that
    > may have been made between the two end points of the Fibre 
    > Channel flow"
    > will still be met.
    > 
    > Charles Binford
    > Pirus Networks
    > 
    


Home

Last updated: Tue Sep 04 01:04:20 2001
6315 messages in chronological order