X.25 CONFIGURATION

[Thanassis Papadimitriou]

Does a Gui exist on how to configure X.25 on OpenVMS Alpha server OpenVMS 7.3-2?

How can I test if a specific configuration reaches the target?

How can I add more than 1 targets?

[Volker Halle]

Thanassis,

there is no GUI for configuring X.25 on OpenVMS.

Here is a pointer to the HP X.25 for OpenVMS documentation:

http://h71000.www7.hp.com/doc/x25.html

What physical WAN device are you using ?

Volker.

PS: It's nearly 8 years ago that I've seen X.25 being used on OpenVMS.

[JohnDite]

YahSU Thanassis,

if you don't have an application that will initiate the X.25 call then you can use
@SYS$TEST:OSIT$IVPINIT.COM
to do the same, assuming the underlying NCL configuration exists and is correct.

You talk of more than one target. What do you mean exactly? Do you mean the destination DTE (X.121 Address) or different paths to different local DTEs (client / relay configuration)?

As far as the X.25 configuration is concerned it assumes that you understand the underlying EMA (Enterprise Management Architecture) represented in Layer 3 and eventually Layer 4 in the DECnet stack.

Confused?

Well I commiserate if you are starting off from scratch.

Are we talking here of a pure X.25 application - X.25 application connectivity or is DECnet over X.25 or possibly X.25 over TCP?

More info would help.

John

[Thanassis Papadimitriou]

Hello John,

What we tried last night to achieve is the following:

x25 configuration on our production site involves 2 OpenVMS nodes. The setup was directly through local LAN and no x25 router wass used.

The same story goes on our disaster site.

The exercise was to achieve x25 connectivity between 1 production node and 1 disaster node routed from the x25 router located at production site through virtual circuits to arrive at x25 router located at disaster site with llc2 setup.

What we achieved is to setup correctly the node on production site. We had the confirmation of network team that packets are routed correctly to disaster site. By tracing dte class we observed normal behaviour.

We worked many hours to setup the disaster node unsuccessfully. The Network team told us that they see us connected on the x25 router with llc2 but no packet activity was observed.

The message that was returned when we tried to test the connectivity on the disaster node is the following. We do not understand the warning and fatal messages. By following up all the steps from X.25 OpenVMS problem guide, we reached to a point where the guide says that there is a software problem and that we have to contact the vendor of the software Digital. This sounds to me like a dead end. We gave up any further attempt to troubleshoot it.

$ set host/x29 vss-dte.1021
%PAD-W-SPEED, terminal speed not recognised - 64000 baud assumed


%PAD-F-NWERROR, unexpected error on remote terminal link
-SYSTEM-F-DEVACTIVE, device is active

PAD usage statistics:
Elapsed time: 00:00:00.04 CPU time: 00:00:00.01
Buffered I/O: 28 Direct I/O: 15
Page faults: 24

%PAD-F-END, control returned to node FIRE::

[JohnDite]

KaleeMera Thanassis,

What Versions of OpenVMS / X.25 are we talking about?

You should be able to get X.25 connectivity between 2 OpenVMS either over LLC2 or XOT.
XOT would be preferable as it's routable LLC2 is restricted to your logical LAN.


What kind of connectivity are we talking about here between your production site and disaster site? LAN / WAN ? Is DECnet involved at all?

You mention an X.25 Router, is this Alpha based?

Can you possibly send me a network diagram of the various components involved with SW Versions etc. then we can take it from there.

Efkharistó
John

[Thanassis Papadimitriou]

On the production site, the connectivity is achieved with local LAN using llc2 and not XOT. It works.

The OpenVMS versions are OpenVMS 7.3-1 & OpenVMS 7.3-2.

Between the sites the connectivity is achieved over WAN with Cisco X.25 routers. llc2 is routed over IP and not DECnet.

[JohnDite]

Thanassis,

if your local configuration works and you say that a local LAN is provided via the CISCO Routers(ie. bridging?) to your Disaster Recovery site, then either the CISCO Routers are not providing the connectivity or the receiving system are not configured in the same way as your system on your local LAN.

What does a CTF LLC2 Trace reveal on your disaster recovery site (assuming you're initiating the connection from the production site). Does the LLC2 Link get established? What does the CTF LLC2 trace on the local System reveal.

Kathika ;-)

John

[Thanassis Papadimitriou]

Here it is the output of trace:

-----------+----+-----+<--------Packet-------->+------------------------------------------------------------------------------------
Time |Evnt|Data |Chn Q Type P P |Data
hh mm ss cc| |Size | M R/S R/S |
-----------+----+-----+<---------------------->+------------------------------------------------------------------------------------
| Data 01 00 00 00|
| |
| |
02:26:07.18| Tx| 5|001 CLR C=00 D=80 |
02:26:07.18|Rx | 3|001 CLRC |
02:28:01.36|Rx | 11|001 CALL |Called DTE 1040
| Data 01 00 00 00|
| |
| |
02:28:01.36| Tx| 5|001 CLR C=00 D=80 |
02:28:01.36|Rx | 3|001 CLRC |

[JohnDite]

Thanassis,

LLC2 is a LAN connection oriented protocol equivalent eg. to HDLC/LAP-B over WAN.

On your local Production system (the one that works) you should see activity if you do the following:

$TRACE START/LIVE "LLC2 SAP LINK"

Using NCL you can also observe the state of your connection:

NCL>SHOW LLC2 SAP SAP-0 link * all

(assuming the SAP-0 has been configured).

The L3 trace is only of value if you know that your underlying link is working!