Inconsistent Finish Of XA Global Transaction
Last updated on JULY 31, 2016
Applies to:Oracle Tuxedo - Version 184.108.40.206.0 and later
Information in this document applies to any platform.
On Tuxedo 220.127.116.11.0
Application committed a transaction between DB2 and MQ. You could see the xa_commit() call to one of them and then the failure occurred before xa_commit() was called on the other one. When the application came back up, Tuxedo rolled back the other RM. During this failure there was network problems that caused disks (file systems) to become unavailable. You are pretty sure you are using NFS with SAN based storage. The transaction was started in WebSphere with your JCA adapter when a call was made to Tuxedo that then enlisted the two RMs.
After machine restart, final Tuxedo client complained about the transaction. Due to this issue, users cannot complete transaction, transaction was in inconsistent state.
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