Summary of 2nd question on UUCP on SunOS4.1.1

From: Alek O. Komarnitsky (alek@spatial.com)
Date: Sun Aug 04 1991 - 23:17:28 CDT


> My original question:
> When using our TrailBlazer+ to dial our UUCP site at 19.2K, it sometimes
> hangs after transmitting "some" files. When the uucico is repeated, it
> hangs on that same file repeatadly, with similar behavior as described
> by Patrik (i.e. pkcget: alarm 1 pkcget: alarm 2 etc. until finally
> it disconnects). What is *strange* (at least to my feeble mind), is that
> when I step the Trailblazer down to 2400 baud, it is able to successfully
> transmit the file. Although I can't cause this problem to occur, it's
> behavior is very repeatable when it does.

I believe (not able to test yet) the best answer came from zjat02@trc.amoco.com:
You may need to upgrade the proms on your Telebit. I had an old trailblazer
and the prom upgrade fixed me up. [sorry I lost your full name zjat02]

Matt Cohen at sysnmc@chron.com wrote:
It looks like you're getting errors in transmission. It seems that
UUCP only tolerates a certain number of errors (alarms) and then drops the
connection. Not surprisingly, the error rate is much lower at 2400 baud than
at 19200. Try turning on MNP and V.42 error correction - add the string
"s95=2s97=1s106=1" to your "at" dial string. It worked for me!
> Unfortunately, the last two switch registers are *not* on my TB+,
> so I'll either need the firmware upgrade mentioned above :-), or
> need to buy a TB2500.

John Polstra at polstra!jdp@uunet.uu.net wrote:
Here is something for you to check: Check the flow control
configuration of the Trailblazer, and make sure it's NOT set up for
X-ON/X-OFF flow control. One of my UUCP neighbors was having the same
problems you describe, and that turned out to be the cause. Apparently,
at some point in the transmission an X-OFF would fly by (by coincidence,
as part of the UUCP protocol) and that would shut down transmission.
Once stopped, it of course never started again.
> This is done with register S58 and S68 register. I had allready changed
> these to use "hardware" handshaking, but John's advice would be handy
> to keep around for future reference.

Finally, Partick Faltstrom summarized the response to the original question:
To force zero parity, add the expect-send couplet "" P_ZERO to
your line in /etc/uucp/Dialers. See the manual for examples.
> Which we also allready do, but we found it out the hard way!

mark@deltam.com wrote about phone line quality, but there's not much one
can do about this one. I also received responses from sdb%hotmomma@uunet.uu.net
and another person who's E-mail address I've dropped (even though he was
nice enough to send several inquires - sorry 'bout that).

The Sun-Managers comes through again - thanx,

Alek Komarnitsky 303-449-0649
Software Tools Manager, Spatial Technology, Inc. 2425 55th Street, Bldg A
alek@spatial.com Boulder, CO 80301-5704



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:20 CDT