Search Engine: Elastic

Article ID: 131633, created on Oct 18, 2017, last review on Oct 18, 2017

  • Applies to:
  • Operations Automation 6.0

Symptoms

Service 'pem' does not start after management node migration to another node (despite service 'pau' is running ok). The following error in /var/log/poa.log:

Oct 18 23:39:21 corenode.org : Caused by: org.postgresql.util.PSQLException: ERROR: column localident1_.pwd_expired does not exist
  Position: 216

Cause

Improper DB was restored.

The most probably, the similar migration has already taken place before. When migration script runs it backs up OA database to /usr/local/pem/db_backup/plesk.dump and restores later.

But the script does not overwrite the file if it is already exists. In the current case the backup created during previous migration (which might have been performed years ago) was restored thus 'pem' service could not start - it did not find the necessary fields in restores DB tables.

Resolution

Re run the migration, having removed the obsolete file /usr/local/pem/db_backup/plesk.dump before.

Feature request #POA-111103 to improve described migration script disadvantage was placed.

5356b422f65bdad1c3e9edca5d74a1ae caea8340e2d186a540518d08602aa065 e12cea1d47a3125d335d68e6d4e15e07 956c448bddc7e1f3585373687602379f 6f1456866eed87488c0f02b298a741c0

Email subscription for changes to this article
Save as PDF