RP/TUX 9.1 - Enhancement request: improve gateway connection retry logic (Doc ID 776978.1)

Last updated on NOVEMBER 04, 2016

Applies to:

Oracle Tuxedo / Tuxedo / 9.1
Information in this document applies to any platform

Goal

DESCRIPTION:
When a Tuxedo domain gateway connection to its peer is disconnected, it waits for retry_interval (set to 60 sec in the
customer's application) before attempting to reconnect. In many cases, even though the network has recovered quickly
and/or has an alternative path, the customer cannot do messaging for up to a minute thus impacting customer
throughput. 

The customer requests BEA to implement the following algorithm for gateway reconnect: 
1. Retry a failed gateway connection immediately. 
2. If it fails, wait for retry_increment (this would be a new parameter and probably set to 5 sec) 
3. If it fails, double it (10, 20, 40) until retry_interval value is reached (60) at which time the doubling stops and
attempts will be made every retry_interval sec 
4. When a connection is established, the "cur_retry_interval" value is reset to 0 (so as to reconnect
immediately after the next failure)

The proposed algorithm is adaptive as well as efficient enabling a quick recovery after transient network faults while
being cautious on extended network faults.

CONFIGURATION:
Tuxedo 9.1

Solution

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms