hi all!

i'm struggling to deal with a problem surrounding *in-bound*
ftp's.   i've got several different application groups that
have designed ftp methods that basically boil-down to them
doing rapid-fire ftp's.  that is -- they connect, they
transfer, they leave and a (very) few seconds later they
repeat the whole process.  the box in question is a 997-400
on 6.0pp2 and i've got the 'f' ftp patch installed
(FTPFDP3A).  this box is *not* heavily loaded (we just
upgraded from a 995-300) and we've got scads of memory
(imo....2.6gb).  the box is using a 10base-t card.

for instance, this morning an in-bound ftp failed.  this is
the message from the other system....

ftp: connect: A remote host did not respond within the
timeout period.
Not connected.

(no, i don't know what their timeout is set to....)  'remote
host' is my mpe box.  the network was up, the system
(according to my lund data) was sitting and twiddling it's
thumbs.  the error occurred about 6am.  only 27 jobs were
running and most of them are the standard collection of
background jobs.

short of getting *them* to resdesign their applications (ha)
-- what can i do?  this sort of thing really disturbs me.
i've got way more than enough box and yet <something> can't
react fast enough to accept ftp requests (from the same
node, if that's relevant...) -- and yet again, mpe looks
like the red-headed stepchild <sigh>              - d

--
Donna Garverick     Sr. System Programmer
925-210-6631        [log in to unmask]

"Unix _is_ user friendly.
It's just very selective about who its friends are.
And sometimes even best friends have fights."

>>>MY opinions, not Longs Drug Stores'<<<