LoadHier Exits With Status 0, but "Cannot obtain a write lock on database" Errror Occurs and Hierdatacleaner Log Shows Exit Status 6 (Doc ID 1590407.1)

Last updated on JANUARY 09, 2017

Applies to:

Oracle Retail Predictive Application Server - Version 13.2.3 and later
Information in this document applies to any platform.

Symptoms

At the end of multi-process loadHier, the hierdatacleaner private utility is called in multi-process mode.  In the case of a large domain, one or more sub-processes may error out with lock timeouts on changearray.db or shadow.db of the master domain.  The following error occurs:


Steps to Reproduce:

This is difficult to reproduce and is based on the number of positions being purged or reclassed in a particular load.  The following steps will increase the chance of the issue happening.

  1. Set the RPAS_LOCK_TIMEOUT to a shorter interval than 60000. For example, 3000.
  2. Create either a large number of reclass loads or purge a large number of partitioned positions without purging the full domain.
  3. Run loadHier with processes > 3.
  4. LoadHier will succeed, but check the hierdatacleaner log in the output for errors.

Cause

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