Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!wuarchive!decwrl!ucbvax!ICAEN.UIOWA.EDU!dbfunk
From: dbfunk@ICAEN.UIOWA.EDU (David B Funk)
Newsgroups: comp.sys.apollo
Subject: Re: TCP/IP gateway troubles in 10.2
Message-ID: <9008292351.AA07851@icaen.uiowa.edu>
Date: 29 Aug 90 23:28:34 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Distribution: world
Organization: Iowa Computer Aided Engineering Network, University of Iowa
Lines: 58

In posting <9008291636.AA00462@apo.esiee.fr>, bonnetf@apo.esiee.fr (bonnet-franck) asks:

> We have a very borrowing problem with TCP/IP communications.
> Our Gateway was a DN3500 which was running 10.1, has one 380 MB disk 
> and 16 Mb of memory.                  
 [stuff deleted]
>  
> For several geographic reasons I had to change the TCP/IP gateway machine.
> So I 've installed the 802.3 controller in the new machine which is also
> a DN3500 with two 700 MB disks and 12Mb of memory.
> BUT this machine is running 10.2 Domain_os ... and has two disk instead of one.
> 
> Since this time , we've been borrowed ALL THE TIME by communications troubles 
> between our machines, TCP/IP seems to hang after a few time and there are 
> no way to restart it EXCEPT to make a TCP/IP request from ANOTHER machine 
> than one of our apollos. We have also 3 HP machines (300,500,835), how lucky
> we are ...
 [stuff deleted]
> Does anybody knows what happen to our gateway, is it a 10.2 specific BUG ?
> The release-notes says TCP/IP is more efficient in 10.2 is it a joke ?

Yes, there is a specific bug in the routing daemon '/etc/routed' that was
released with sr10.2. It would start up just fine but after some time (hours
to days) it could "time-out" a network interface and consider it down.
When this happened, it would drop that interface from its routing tables &
quit broadcasting any routing information about it. The net effect is that
the gateway would quit looking like a gateway to other machines & quit routing.
There was a patch on the January (& newer) patch tapes to fix this problem.
Quoting from the release notes:


     1.53  Patch m0108 /etc/routed

     Patch m0108 includes fixes to the /etc/routed command for nodes run-
     ning the SR10.2 version of Domain/OS. This patch is incompatible with
     all other versions of Domain/OS.

     Patch m0108 fixes the following problem (DDC72):

     The /etc/routed command was timing out active interfaces.  routed has
     been modified to prevent it from timing out and thus marking "down"
     interfaces that are configured "up". It does, however, time out inter-
     faces which have been configured "down" via /etc/ifconfig.

     Install patch m0108 on nodes running the SR10.2 version of Domain/OS
     (use the bldt command to determine the revision of the operating sys-
     tem running on your workstation).

     Patch m0108 includes the following file:

        /etc/routed        1989/11/10 20:56:21 EST (Fri)


BTW, I would strongly reccomend getting patch tape M68K_9007 (or newer) and installing
patches: m0139, m0162, & m0165. These, together with m0108, are necessary for reliable
tcp/ip service on a sr10.2 machine (fixes the infamous sr10.2 pty problem).

Dave Funk