Operation
Operation
The Lucene index information is stored in files arranged in a folder structure. This folder structure is replicated in every Soffid Console and every Sync Server and also is saved in the database.
In case an instance (Docker, Kubernetes, or stand-alone) detects an inconsistency, the information will be overwritten with the database data.
When you update an object, marked as the textual index, a task will be created. The soffid agent will execute this task and the Sync Server will update the database tables related to the textual index.
Folder structure
The folder structure is the following:
- ../index/<TENANT>/<SOFFID_OBJECT>
Example
1. Here you are the folder structure for the Soffid Console
2. And the folder structure for the Sync Server
DatabaseÂ
The database tables involved:
- SC_LUINPA
- SC_LUNIND
Example
1. The database structure
soffid agent
You can check the soffid agent status by visiting the Sync Server monitoring page:
Example
1. A soffid agent pending task:
Step-by-step
Example 1
1. You update one user's data and save the changes.
2. New tasks are created and executed.
3. Then Sync Server indexes the updated text and places the index file.
4. Then Sync Server and updates the database table SC_LUNIND by upgrading the LIP_TIMSTA field of the User object or by creating a new record if it did not previously exist.
5. When the following search will be performed, the very first thing to do is check the database file. If it is necessary update the file system and finally perform the search.
Example 2
1. The task engine mode is Read only
2. You update one user's data and save the changes.
3. A new task is created and executed
4. Then Sync Server indexes the updated text and places the index file.
5. Then Sync Server and updates the database table SC_LUNIND by upgrading the LIP_TIMSTA field of the User object or by creating a new record if it did not previously exist.
6. When the following search will be performed, the very first thing to do is check the database file. If it is necessary update the file system and finally perform the search.