Notes on actual use 2026

When restoring a prod db in order to make it a history db you already have a lot of data in mdrivenserversynk - you may want to trunk most of this and leaving only the tip in order to reduce backup and increase read/update speed of that table.

Some 1-1 associations (all?) get a unique constraint - and that needs to be dropped for history since they are not unique in all time. These are dropped as part of the process triggered by EnsureVersioningTablesInDB, this operation adds StartTime,StopTime columns to tables.

Main

Read here on why you may want a History slave server: Purpose of History Server

Recent changes make it possible to use SQLCompact to test the MDriven History Server, but you will most likely need more space than SQL compact can handle.

Steps:

1. On the main MDrivenServer:
   • Ensure MDrivenServerSynk
   • Switch from "Normal" to "Master"
   • Copy/backup the database from the master - restore it in the second MDrivenServer.
2. In Second MDrivenServer, change the mode to "History Slave":
   • In History Slave, ensure Version is compatible
   • In History, point out the master MDrivenServer and set the user and password for the master.

Test by making changes towards Master, within short changes show up in History slave.

Thinning, Culling, and Truncating a History Server

On classes, set the tagged value NoHistory=true to make HistoryServer skip this class for versioning. This may be desirable when having fast-moving reference data you do not own - like daily imports from other systems that you do not deem important enough to track. It is currently not supported to evolve from Versioned to not Versioned. To use NoHistory in History, make sure you have this set from the start or recreate the table to force it through.

Also, the tagged value HistoryKeepWeeks=NoOfWeeks(int) can be used on Classes. This may be good to keep shorter history on some than on others. This may lead to a false history perspective - if the old version references another object that has been culled by a shorter HistoryKeepWeeks link that will turn up as null when in fact it had a value at the time.

The history is culled via pressing a button in admin/MasterSlave/TruncHistory3Years(156 weeks) - this will change and become automatic in later releases.

In SQLServer, you can use this script to find where you have the most rows in your History server - this may be a good indicator if you need thinning:

SELECT 
    t.NAME AS TableName,
    s.Name AS SchemaName,
    p.rows,
    SUM(a.total_pages) * 8 AS TotalSpaceKB, 
    CAST(ROUND(((SUM(a.total_pages) * 8) / 1024.00), 2) AS NUMERIC(36, 2)) AS TotalSpaceMB,
    SUM(a.used_pages) * 8 AS UsedSpaceKB, 
    CAST(ROUND(((SUM(a.used_pages) * 8) / 1024.00), 2) AS NUMERIC(36, 2)) AS UsedSpaceMB, 
    (SUM(a.total_pages) - SUM(a.used_pages)) * 8 AS UnusedSpaceKB,
    CAST(ROUND(((SUM(a.total_pages) - SUM(a.used_pages)) * 8) / 1024.00, 2) AS NUMERIC(36, 2)) AS UnusedSpaceMB
FROM 
    sys.tables t
INNER JOIN      
    sys.indexes i ON t.OBJECT_ID = i.object_id
INNER JOIN 
    sys.partitions p ON i.object_id = p.OBJECT_ID AND i.index_id = p.index_id
INNER JOIN 
    sys.allocation_units a ON p.partition_id = a.container_id
LEFT OUTER JOIN 
    sys.schemas s ON t.schema_id = s.schema_id
WHERE 
    t.NAME NOT LIKE 'dt%' 
    AND t.is_ms_shipped = 0
    AND i.OBJECT_ID > 255 
GROUP BY 
    t.Name, s.Name, p.Rows
ORDER BY 
    p.rows DESC, t.Name