Date   : Thu, 07 Sep 2006 23:11:43 +0100
From   : jgh@... (Jonathan Graham Harston)
Subject: Re: Tube File Copying

>Message-ID: <E57D6D3E50DF42A9A4D27EAB5EA8F2ED.MAI@...>
 
"Ian Wolstenholme" <BBCMailingList@...> wrote:
> However I still get a timeout problem on the Master after a "chunk"
> has been saved across Econet.  My routine uses OSGBPB to
> read part of the file into the buffer and then *SAVEs the buffer.
 
You could try using smaller "chunks". Even though you have a 16M buffer,
doesn't mean it's efficient to try and transfer 16M in one go. Econet
bridges barf on packets larger than 8K, so try using 8K chunks. Read as
much as you can, then send it in 8K chunks.
 
Also, you could try changing the NFS timeouts. Unfortunately, details of
OSWORD &13 calls with XY%?0>8 are hard to find. Here they are:
 
XY%?0 = 12 - Read printer server name
XY%?0 = 13 - Set printer server name from $(XY%+1)
XY%?0 = 14 - ?
XY%?0 = 15 - Read file server retry,
   XY%?1=not listening, XY%?2=no reply, XY%?3=machine peek
XY%?0 = 16 - Set file server retry,
   XY%?1=not listening, XY%?2=no reply, XY%?3=machine peek
XY%?0 = 17 - Translate net number
 
So, use OSWORD A%=&13, XY%?0=16, XY%?2=255 to set timeouts to max.
 
-- 
J.G.Harston - jgh@...                - mdfs.net/User/JGH
In 1939 $50 of groceries would fill three station wagons. Today I
can lift $50 of groceries with one hand. I must have got stonger.