Restrict Auto Retries In a SOA Composite (Doc ID 2445971.1)

Last updated on FEBRUARY 05, 2024

Applies to:

Goal

Need to restrict auto retries in SOA composite

Whenever we push a large/huge amount of data to an WSDL endpoint A, the WSDL endpoint A takes some time to process the data.
There is a timeout error being thrown after five (5) minutes and the problem is SOA automatically performs a retry.

We have checked from the console the following settings:
1. Deployments->soa-infra->EJB
We've set the EJB Timeout Settings to 3,600 on the following:
BPELActivityManagerBean, BPELDeliveryBean, BPELDispatcherBean, BPELEngineBean, BPELFinderBean, BPELInstanceManagerBean, BPELProcessManagerBean, BPELSensorValuesBean, BPELServerManagerBean, BPELCacheRegistryBean, BPELCacheStoreBean
(Based from https://www.oracle.com/technetwork/articles/soa/carrasco-osb-failures-3697561.html)

2. Services->JTA (Set the timeout settings to 3,600)

From the EM Console, we have set the following:
1. Set SyncMaxWaitTime to 3,600 secs
2. MaxRecoverAttempt to zero (0)
3. MinBpelWait from two(2) minutes to fifteen(15) minutes

We tried those settings above but SOA still performs a retry.

We also have removed the retry attempts from the custom code inside a bpel, and under fault-policies.xml but SOA still performs a retry

 

Solution

My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.