Browse Prior Art Database

Improved Error Reporting and Field Replaceable Unit Isolation for an Optically Connected Workstation Controller and Multiple Workstations

IP.com Disclosure Number: IPCOM000108275D
Original Publication Date: 1992-May-01
Included in the Prior Art Database: 2005-Mar-22
Document File: 4 page(s) / 178K

Publishing Venue

IBM

Related People

Dancker, GA: AUTHOR

Abstract

This article describes an error detecting and reporting method that greatly improves problem determination and isolation to the failing field replaceable unit (FRU) during normal operations (as opposed to a dedicated diagnostic mode) for a complex cabling network between a workstation controller (WSC) and multiple workstations (WS). An example of this type of network would be a WSC that multiplexes 8 twinax ports over a fiber-optic cable to a demultiplexer box up to 2 miles away. The demultiplexer connects to the WSs with twinax, IBM cabling system, or telephone twisted pair.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 43% of the total text.

Improved Error Reporting and Field Replaceable Unit Isolation for an Optically Connected Workstation Controller and Multiple Workstations

       This article describes an error detecting and reporting
method that greatly improves problem determination and isolation to
the failing field replaceable unit (FRU) during normal operations (as
opposed to a dedicated diagnostic mode) for a complex cabling network
between a workstation controller (WSC) and multiple workstations
(WS).  An example of this type of network would be a WSC that
multiplexes 8 twinax ports over a fiber-optic cable to a
demultiplexer box up to 2 miles away. The demultiplexer connects to
the WSs with twinax, IBM cabling system, or telephone twisted pair.

      Fig. 1 is a diagram of a twinax protocol WSC with a fiber-optic
output.  Eight twinax ports are multiplexed over a pair of
fiber-optic cables (one for transmit, one for receive) to a
demultiplexer that can be located up to 10,000 feet away.  Up to
5,000 feet of twinax, 5,000 feet of IBM cabling system, or 1,000 feet
of telephone twisted pair (TTP) can be used to attach up to 7
displays or printers per port (maximum of 40 active at a time).

      Current other-equipment manufacturers' (OEM) twinax to fiber-
optic converters, as shown in Fig. 2, perform a similar function as
above. However, some of the OEM products do not provide much problem
isolation or error reporting: the network either works or it does
not.  In contrast, the IBM WSC, fiber-optic cable, and demultiplexor
configuration shown in Fig. 1 provides a high degree of error
detection and reporting to help with network problem determination
and failing FRU isolation during normal operations.

      Seven different error conditions are checked during normal
operations. Fig. 3 is a summary of these errors.  A more detailed
description follows.
1. Port Address Frame Parity Error
   All transmissions from the WSC to the demultiplexer are prefaced
by a port address frame.  This frame contains the address of the port
(0 to 7) that the demultiplexer will use for this operation. If bad
parity is detected in this frame, a "Port Address Parity Error" is
logged in the demultiplexer.  No data is transmitted out to the
attached WSs because, since an error was detected on the Port Address
Frame, the actual port cannot be determined. Also, this error is
logged against all eight ports of the demultiplexer for the above
reason.
   For a receive operation, the WSC will detect a "no response
timeout."  For a transmit operation, a poll is issued after this. The
response to the poll would indicate that the WS did not receive the
previous transmission.  The WSC can then check the demultiplexer
error byte to determine what error occurred.
2. Frame Parity Error from the WSC
   This error indicates that a parity error was detected while the
demultiplexer was receiving a command or data frame from the WSC. The
data will be sent out to the WS.  The WS sh...