Method for maintaining a sequence of events function during failover in a redundant multiple layer system

• US 5,426,774 A
• Filed: 04/06/1993
• Issued: 06/20/1995
• Est. Priority Date: 04/06/1993
• Status: Expired due to Term

• Current Assignee: Honeywell Incorporated (Honeywell International Inc.)
• Original Assignee: Honeywell Incorporated (Honeywell International Inc.)
• Inventors: Banerjee, Indra, McCracken, Kevin R., McLaughlin, Paul F.
• Primary Examiner(s): Beausoliel, Jr., Robert W.
• Assistant Examiner(s): Palys, Joseph E.
• Application Number:
  US08/042,923
• Time in Patent Office: 805 Days
• Field of Search: 395/575, 371/7, 371/8.1, 371/16.5, 371/29.1, 371/11.1, 371/11.3
• US Class Current: 714/16
• CPC Class Codes: G06F 1/14 Time supervision arrangements, e.g. real time clock
• CPC Class Codes: G06F 11/2007 using redundant communication media
• CPC Class Codes: G06F 11/2023 Failover techniques
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: G06F 11/2043 where the redundant components share a common memory address space
• CPC Class Codes: G06F 2201/82 Solving problems relating to consistency

Abstract

In a process control system having a redundant multilayer hierarchical structure, each node of a layer being redundant sequence of events inputs are received from field devices by an input/output processor (IOP). The IOP is a digital input sequence of events (DISOE) IOP, the IOP being the lowest layer of the hierarchical structure. The IOP interfaces with a controller at the next layer of the hierarchy. A method for reliably maintaining a sequence of events function during a failover of any of the redundant nodes, comprising the steps of maintaining a log, a circular list, by the local DISOE. The circular list is a rolling log of all sequence of events data for a predefined time period. When a failover occurs, the new primary commands an event recovery. The event recovery process freezes the log and uses the information in the log to recreate the events data. The freeze operation inhibits background purge activity for the log thereby avoiding the deletion of information past the defined time. New events data is still entered in the log. Once the log has been processed the freeze operation is negated. The recreated data is transmitted to the controller in accordance with a predefined protocol, thereby avoiding the loss of any events data as a result of the failover.

Claims

1. In a process control system having a redundant multilayer multinode hierarchical structure, each node of a layer being redundant, and further wherein sequence of events inputs are received from field devices by a receiving input/output processor (IOP), the receiving IOP being a digital input sequence of events (DISOE) IOP, the IOP being the lowest layer of the hierarchical structure, the IOP interfacing with a controller at the next layer of the hierarchy, a method for reliability maintaining a sequence of events function during a failover of any of said redundant nodes, the method comprising the steps of:

a) receiving events data from said field devices by the receiving IOP;

b) maintaining a first buffer with events data, said first buffer being a circular list, said events data being time stamped by the receiving IOP with a relative time of receipt and a sync message ID, the relative time of receipt being a number of clock times since receipt of a time synchronizing message from the controller, the time synchronizing message further including the sync message ID, said events data in the first buffer being the events data received since a previous transmission of said events data to the controller;

c) maintaining a second buffer with events data, said second buffer being a circular list, said events data being time stamped by the receiving IOP with the relative time of receipt, said events data in the second buffer being events data received within a predetermined period such that a history of events is maintained for the predetermined period;

d) upon request from the controller, transmitting the events data and the corresponding relative time of receipt from the first buffer to the controller whereby the controller calculates a real time of occurrence of the events data, the real time of occurrence being time synchronized to the process control system; and

e) when a failover is detected, upon a request for event recovery from any of the hierarchical layers, recreating the events data from the second buffer to transmit the recreated events data to the controller in accordance with a predefined protocol thereby avoiding the loss of any events data as a result of the failover.

2. A method for reliably maintaining a sequence of events function according to claim 1, wherein the step of maintaining a second buffer further comprises the steps:

a) during a failover, freezing a purge operation of the events data in the second buffer even though the events data is older than the predetermined period of time;

b) continuing to store events data in the second buffer during the failover thereby not losing any events data; and

c) upon completing processing events data in the second buffer, unfreezing the purge operation of the second buffer to eliminate out of date events data.