🚀 go-pugleaf

From: Henk Peek <henkp@nikhefk.nikhef.nl>
Date: Sun, 16 Jun 91 18:17:56 MET
Subject: Re: DRSI bug extingushed



>I dont plan any further changes or bug fixes to the DSRI driver - several
>folks have expressed interest in support for multiple boards but thats
>not on the agenda.

>The ball is passed on to ..... (you perhaps ?)

The scc driver from PE1CHL already supports multiple boards, the DRSI board
and is ported to NOS.

>Oops !, sadly I could have told you the fix !  The version of NOS (in
>actuality the 11/30/90 release) is built with IBUFsize set to 512 - 
>considerably more than is needed for an AX25 frame of 256 bytes + header.

The scc driver of PE1CHL links the Ibuffers together when it receives
large measages. The most users use 1 or 2 OptoPcScc boards, 4 or 8
scc channels, with one pool of 32 to 64 Ibuffers of 128 bytes.



The current versions of PE1CHL net are very stable and is used at a lot
of backbone network sites in and outside the Netherlands. An old version
can be found at hp.col.com. Some of the last half year added:
more backbone netrom support, full ftl0 broadcast support (do you want
to use UUSAT14 and packsat BBSes? it can automatically forward 
complete received bulletins to the BBS network or the public ftp.)
and a lot of bug fixes and new commands as: scc calibrate <interface #>
(generate 30 sec. square wave signal to test and adjust modems.)

There are two IBMPC versions: medium and large model, an Atari version
and a Xenix version.

Henk, PA0HZP


----
From: Fred Schneider K0YUM <fls@gag.com>
Date: Sun, 16 Jun 91 21:12:17 -0600
Subject: Re: DRSI bug extingushed


f*ok
b


----
From: clark@tomcat.gsfc.nasa.gov (Tom Clark -- W3IWI)
Date: Mon, 17 Jun 91 03:56:43 UTC
Subject: NOS timer problem plus comments on DOS 5.00


I'm seeing a problem with the timer in NOS (specifically the 0531 version,
but I've seen it on other versions too). Wonder if anyone else is experiencing
the same thing. It is a pretty repeatible fault.

Here is what happens -- my tomcat host is acting as the mail exploder for
weekly AMSAT bulletins, and ends up remailing to a list of about a hundred
folks worldwide. When hit with a heavy load on weekends, the internal timer
simply stops counting. This causes two bad things -- the smtp timer stops
running (I run a 30 second timer, and it usually stops with a 29/29 reading,
although I have also seen it freeze at 0/29). I can manually force the stuff
out manually (done remotely thru the mbox sysop function) with repeated 
"smtp kick" commands. Also when the timer dies, all the old connections
at the "time wait" state never clear themselves, so the "tcp status" can
show 100 or more "time wait" entries. The uptime shown with "ps" and the
timers shown in "if" are also frozen, sometimes with bizzare values.

The 0531 NOS running on tomcat is a stripped down version to support the
internet and slip ports compiled by Gerry, N5JXS (Unfortunately I have never
been able to get NOS to compile with the most recent Turbo-C++ -- but that's
another story!). The stripped down version has a considerably larger hunk
of available memory, so I get very few yellow and zero red problems. Although
tomcat runs NOS under DV, the timer problem shows up even when running in
"nude" MessDOS also.

My guess is that some timer interrupt gets missed under high stress conditions
and never gets reinitialized.

The next thing I'm going to try is to change DOS versions. Currently tomcat
runs under MessDOS 4.01 in order to handle large disk drives as a single
volume. This weekend I picked up a new DOS 5.00 which I installed on my
host (scheat.gsfc.nasa.gov) at home which has nearly nearly identical to
tomcat and it seems very robust.

Mentioning DOS 5.00 -- locally in the DC area Egghead is running an intro-
ductory special at $40 per copy. Microsoft is calling it an upgrade, but
all you have to have is a DOS 2.x, 3.x or 4.x for the automatic installation
to work. The big ESDI drive on scheat came up fine in 5.00 with no hassles.
The easiest OS upgrade I have ever seen! QEMM (5.1) and DV (2.31) run fine --
I ignored Microsoft's EMM stuff and find that QEMM386 does the deed just
fine. Microsoft's INSTALL knows about QEMM and properly made what few
changes to CONFIG.SYS. With 5.00 having part of DOS in high memory (with
the line DOS=HIGH in CONFIG.SYS), and with all my drivers loaded into high
memory, scheat has 637k of DOS program space before loading DV, and DV windows
are in the 550-560k range. 5.00 looks like a winner to me!

Tom


----
From: clark@tomcat.gsfc.nasa.gov (Tom Clark -- W3IWI)
Date: Mon, 17 Jun 91 03:56:43 UTC
Subject: NOS timer problem plus comments on DOS 5.00


I'm seeing a problem with the timer in NOS (specifically the 0531 version,
but I've seen it on other versions too). Wonder if anyone else is experiencing
the same thing. It is a pretty repeatible fault.

Here is what happens -- my tomcat host is acting as the mail exploder for
weekly AMSAT bulletins, and ends up remailing to a list of about a hundred
folks worldwide. When hit with a heavy load on weekends, the internal timer
simply stops counting. This causes two bad things -- the smtp timer stops
running (I run a 30 second timer, and it usually stops with a 29/29 reading,
although I have also seen it freeze at 0/29). I can manually force the stuff
out manually (done remotely thru the mbox sysop function) with repeated 
"smtp kick" commands. Also when the timer dies, all the old connections
at the "time wait" state never clear themselves, so the "tcp status" can
show 100 or more "time wait" entries. The uptime shown with "ps" and the
timers shown in "if" are also frozen, sometimes with bizzare values.

The 0531 NOS running on tomcat is a stripped down version to support the
internet and slip ports compiled by Gerry, N5JXS (Unfortunately I have never
been able to get NOS to compile with the most recent Turbo-C++ -- but that's
another story!). The stripped down version has a considerably larger hunk
of available memory, so I get very few yellow and zero red problems. Although
tomcat runs NOS under DV, the timer problem shows up even when running in
"nude" MessDOS also.

My guess is that some timer interrupt gets missed under high stress conditions
and never gets reinitialized.

The next thing I'm going to try is to change DOS versions. Currently tomcat
runs under MessDOS 4.01 in order to handle large disk drives as a single
volume. This weekend I picked up a new DOS 5.00 which I installed on my
host (scheat.gsfc.nasa.gov) at home which has nearly nearly identical to
tomcat and it seems very robust.

Mentioning DOS 5.00 -- locally in the DC area Egghead is running an intro-
ductory special at $40 per copy. Microsoft is calling it an upgrade, but
all you have to have is a DOS 2.x, 3.x or 4.x for the automatic installation
to work. The big ESDI drive on scheat came up fine in 5.00 with no hassles.
The easiest OS upgrade I have ever seen! QEMM (5.1) and DV (2.31) run fine --
I ignored Microsoft's EMM stuff and find that QEMM386 does the deed just
fine. Microsoft's INSTALL knows about QEMM and properly made what few
changes to CONFIG.SYS. With 5.00 having part of DOS in high memory (with
the line DOS=HIGH in CONFIG.SYS), and with all my drivers loaded into high
memory, scheat has 637k of DOS program space before loading DV, and DV windows
are in the 550-560k range. 5.00 looks like a winner to me!

Tom


----
From: sysop@pa3eug.ampr.org (Ramon Kolb 44.137.24.31 "noli turbare meos circulos"-Archimedes)
Date: Sun, 16 Jun 91 07:23:18 GMT
Subject: WNOS 1m manual

To those who already have dowloaded WNOS1m from the thumper:
As you might have seen, due to the set-up differences compared to 
original NOS, WNOS asks for a setup file named AUTOEXEC.NOS --not .NET !!
I will send a mail containing the WNOS manual to the tcpip-area at HB9ZZ,
for those who cannot get it for the one reason or the other, please mail
me (pa3eug%pa3eug@hb9zz.ethz.ch), and I'll try to send it directly...
This manual is in GERMAN, in the next days I'll write an English trans-
lation, but I expect to have no time to do this until next weekend...
But this German manual will be understandable for most of you, Michael
DB3FL has noticed a clear syntax of every command anyway....

Lots of luck, and in case, mail me to solve your problems (and if I don't
know the answer, I'll pass them to db3fl, he surely does.....)

Best 73's, Ramon PA3EUG
BBS: PA3EUG @ PI8HRL.NLD.EU
amprnet: pa3eug%pa3eug@dl8yq (dl8yq is reachable at FlexNet at ON4UBO)
internet:pa3eug%pa3eug@hb9zz.ethz.ch


----
-- 
*****************************************************************
   ___  ____  ___    _  _ ___ ___   ___ ___ ___     ___ _  _
  /__/ / / /   /    /\ / /__   /   /__//__//   /__//__ /\ /
 / \  /   / __/_   /  / /__   /   /  //  //__ /  //__ /  /