Clarification on continuously incrementing sequence count

Kevin Jones kev at
Thu Sep 25 04:58:49 PDT 2008

* From the T10 Reflector (t10 at, posted by:
* "Kevin Jones" <kev at>
FCP makes no restriction on the use of "Continuously Incrementing
Sequence Count". I cannot figure out from the standards (FCP, FC-FS)
exactly how this would work. Below is a contrived example of a WRITE
command with a sequence level retry:
->FCP_DATAx4 (SEQ_CNT=1,2,3,4)
->FCP_DATAx4 (SEQ_CNT=5,6,7,8)
FCP_DATAx4 (SEQ_CNT=0,1,2,3)
Here are some questions:
1. FC-FS defines continuously incrementing sequence count to
    apply to data_frames. data_frames have R_CTL.ROUTING = 0,3,4.
    This means all FCP frames (CMD, DATA, XRDY, RSP) and SRR
    are affected ?
2. FC-FS is slightly vague on whether there are one or two
    SEQ_CNT's for an exchange. I am making the assumption that
    there is one for each direction.
3. FCP allows SEQ_CNT to be zeroed for retransmission, but that's
    just a "may" so what happens if a sequence is retransmitted in
    a class3 connection and SEQ_CNT is "continued". I assume in
    the above scenario, the first frame of the retransmitted
    exchange would have SEQ_CNT=10 ?
Kevin Jones
Fibre Channel Tape Firmware
Hewlett Packard,
kevin.jones at
Hewlett-Packard Limited registered Office:
Cain Road, Bracknell, Berks RG12 1HN
Registered No: 690597 England
The contents of this message and any attachments to it are confidential
and may be legally privileged. If you have received this message in error,
you should delete it  from your system immediately and advise the sender.
To any recipient of this message within HP, unless otherwise
stated you should consider this message and attachments as
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at

More information about the T10 mailing list