Question about MRU and Padding in PPP - PPP

This is a discussion on Question about MRU and Padding in PPP - PPP ; Hello there. I am beginning to learn about PPP. But I got stuck on MRU from start. The following is the paragraph that make me crazy. Please anybody explain me the meaning of the following paragraph. ================================================== ======================== Page 73. ...

+ Reply to Thread
Results 1 to 2 of 2

Thread: Question about MRU and Padding in PPP

  1. Question about MRU and Padding in PPP

    Hello there.

    I am beginning to learn about PPP.
    But I got stuck on MRU from start.
    The following is the paragraph that make me crazy.
    Please anybody explain me the meaning of the following paragraph.


    ================================================== ========================
    Page 73.

    If the peer indicates a larger-thanexpected MRU, there is no harm in
    sending
    Configure-Ack for this value but then sending only the largest
    messages possible,
    even though these messages may be less than the requested MRU.
    In other words, reasonable implementations should reply with
    Configure-Ack
    if the offered value is any value greater than some small lower bound
    (say, 64 octets).
    ================================================== ========================

    "but then sending only the largest messages possible,
    even though these messages may be less than the requested MRU. "

    what I want to know is that when MRU is large, the ppp frame sender
    should pad the PPP frame
    to the size of received MRU.

    I hope your good and easy answer.

    thank you for reading my writing.

  2. Re: Question about MRU and Padding in PPP

    hyskim74@empal.com (Shawn Kim) writes:
    > "but then sending only the largest messages possible,
    > even though these messages may be less than the requested MRU. "
    >
    > what I want to know is that when MRU is large, the ppp frame sender
    > should pad the PPP frame
    > to the size of received MRU.


    No, no padding is ever necessary.

    For example, if the peer offers you an MRU of 1600 in its LCP
    Configure-Request, but your implementation (due to its design) can't
    possibly send a frame larger than 1500 octets, there's no harm in
    saying LCP Configure-Ack to that larger value, even though you'll
    never send anything that large.

    --
    James Carlson, IP Systems Group
    Sun Microsystems / 1 Network Drive 71.234W Vox +1 781 442 2084
    MS UBUR02-212 / Burlington MA 01803-2757 42.497N Fax +1 781 442 1677

+ Reply to Thread