🚀 go-pugleaf

We experience file transfer rates of around 2.5Mb/minute and
we are transfering files from the UNiSYS host to UNIX machines.
We have also noted that the DCP is 100% busy during the transfers.

Over the past few days I have been investigating the channel between
the 2200 and the DCP, we have seen the following messages:

NCH            11:21:38.447     015/001                    020000 $(11)  NCH
            CODE=102 DCPA
            THE TOTAL NUMBER TIMES THAT THE HOST WAS ASKED TO RESEND A FRAME SINCE 00:03:06 10/02/92 IS  9
            THE NUMBER OF TIMES THAT A HOST FRAME HAD TO BE RESENT BECAUSE A STOP BIT WAS SET IN THE DATA IS 8

from CMSUTILITY, ltanalysis log file runs.


I decided to look at the channel arrangements we have because I
thought that the frame-sz parameter on the FEP NDS might have
been optimized for character based terminal traffic rather than
for burst mode file transfers.


If anyone could explain the NCH log file entries I'd appreciate
it. What FEP frame-sz values do people use in their CMS configs?


Steve Fitch                            o          ,__o        o   /\/\
Open University                       <\        _-\_<,     []<\  /~~\ \/\
Management Services Division        __/__,     (*)/'(*)      /> /    \ \ \
Milton Keynes, England.       +44 (0908)65 2733  s.c.fitch@open.ac.uk