Chapter 6. Verifying the integrity of back-end databases

The Directory Server database integrity check can detect problems, such as corrupt metadata pages and the sorting of duplicate keys. If problems are found, you can, depending on the problems, re-index the database or restore a backup.

6.1. Performing a database integrity check

The dsctl dbverify command enables administrators to verify the integrity of back-end databases.

Procedure

  1. Optional: List the back-end databases of the instance: 

    # dsconf -D "cn=Directory Manager" ldap://server.example.com backend suffix list
dc=example,dc=com (userRoot)

  2. Stop the instance: 

    # dsctl instance_name stop

  3. Verify the database. For example, to verify the userRoot database, enter: 

    # dsctl instance_name dbverify userRoot
[04/Feb/2022:13:11:02.453624171 +0100] - INFO - ldbm_instance_config_cachememsize_set - force a minimal value 512000
[04/Feb/2022:13:11:02.465339507 +0100] - WARN - ldbm_instance_add_instance_entry_callback - ldbm instance userroot already exists
[04/Feb/2022:13:11:02.468060144 +0100] - ERR - ldbm_config_read_instance_entries - Failed to add instance entry cn=userroot,cn=ldbm database,cn=plugins,cn=config
[04/Feb/2022:13:11:02.471079045 +0100] - ERR - bdb_config_load_dse_info - failed to read instance entries
[04/Feb/2022:13:11:02.476173304 +0100] - ERR - libdb - BDB0522 Page 0: metadata page corrupted
[04/Feb/2022:13:11:02.481684604 +0100] - ERR - libdb - BDB0523 Page 0: could not check metadata page
[04/Feb/2022:13:11:02.484113053 +0100] - ERR - libdb - /var/lib/dirsrv/slapd-instance_name/db/userroot/entryrdn.db: BDB0090 DB_VERIFY_BAD: Database verification failed
[04/Feb/2022:13:11:02.486449603 +0100] - ERR - dbverify_ext - verify failed(-30970): /var/lib/dirsrv/slapd-instance_name/db/userroot/entryrdn.db
dbverify failed
  4. If the verification process reported any problems, fix them manually or restore a backup.

  5. Start the instance: 

    # dsctl instance_name start

Additional resources