Saturday, November 24, 2012

EDN - Same Event triggered twice in Cluster Environments

In Cluster environments, during large number of business transactions, same Business Events triggers more than once, causes duplicate transaction.

To resolve below issue, please follow below steps:

1. Log in to the 11.2 database:
   CONNECT /AS SYSDBA
    
2. Specify the following SQL command in SQL*Plus to disable the 11.2 dequeue optimizations:
   SQL> alter system set event='10852 trace name context forever, level 16384'scope=spfile;
   
3. Restart the database.

Source: Metalink 1435312.1

BPM Approval UI access URI Issue

In some cases, when you deployed BPM UI's to SOA cluster or single environment, the access URL of UI get changed to different server URL rather than SOA server URL.

This issue happens when standard IP & virtual IP configured on same machine, and Virtual IP configured to SOA server.

To resolve this issue for a while, go to soa-infra --> <Partiotion> --> Click on composite <composite name> --> Component Metrics --> Click on <HumanTask> --> Navigate to <Administraiton> --> Change the Host Name to SOA server host name --> Click Apply

But whenever the SOA server get bounced the URL get rolled back to old configuration.

In this case, try below options:
1. Check hosts.conf file(in Unix). Make sure that SOA server IP(virtual) is listed above the other server IP(standard IP)
2. Login to Console --> <Domain> --> Clusters --> Select 'SOA Cluster' --> Configuration --> HTTP --> Change 'Frontend Host' to SOA Hostname , 'Frontend HTTP Port' to SOA Port & 'Frontend HTTPS Port' to 0