The maximum number of threads for checking database connections, or -1 for twice the number of sql.pool.max-connections.

Maximum time in seconds the check is allowed to take.

If the property is "true", messages concerning queries (search of content item versions) will be generated.

If the property is "true", SQL statements concerning queries will be written to the log.

If the property is "true", write messages of the connection pool to the log file.

If the property is "true", more debug messages will be generated.

Maximum number of connections to the database.

Maximum number of connections used for queries, that is, the maximum number of parallel queries.

Maximum number of retries on job that failed due to read-only DB or connection loss.

Minimum number of connections to the database.

Maximum time in seconds the opening of connections is allowed to take.

The interval in seconds in which it is checked if connections can be closed.

The time in seconds a connection must be idle before it will be closed.

Delay interval before a retry is attempted due to a read-only DB or connection loss.

If "true", will retry failed transactions due to db connection failure.

If "true", will retry failed transactions due to a read-only database.

Ratio of non-exclusive elements to pull up in scheduler queue in case of a potential deadlock (in the range ]0.0,1.0]).

The interval between two checks for potential scheduler deadlocks.

Flag to enable scheduler watchdog.

The interval in seconds in which existing connections will be checked for function.

The time in seconds a connection must be idle before it will be checked for function.

Setting the property to "true", causes the Content Server to automatically add database columns for newly defined content type properties and to fix wrong widths of String properties. This will only work if you set sql.schema.checkColumns to "true".

Setting the property sql.schema.checkColumns to "true", causes the Content Server to compare the content type definition with the existing database schema for missing columns and matching widths of String properties. If there are differences, it depends on the setting of sql.schema.alterTables if the Content Server refuses to start ("false") or if it adds and changes the columns automatically ("true"). Checking the tables consumes a considerable amount of time, so that the server starts up more slowly. If sql.schema.checkColumns is set to "false", the Content Server will not check the columns. If there are differences, you will run into Content Server exceptions later on.

Setting the property to "true", causes the Content Server to automatically create and drop indexes on content type properties, according to the Index attribute in the doctypes.xml. This flag only affects existing columns. For newly added columns and tables, an index is always created if the Index attribute is set.

The Content Server always checks if tables for the content types are missing. Setting the property to "true", causes the Content Server to create missing tables for new content types. If the setting is "false" and there are missing tables the Content Server refuses to start

The time in seconds, a blob, that has no reference to a resource, will be preserved.

The delay between two collections in seconds.

The delay in seconds before starting the first collection.

The time of the day, in seconds from 0:00h in the default time zone, when the collector should run every day. If set to -1, it does not run at a fixed time. The initial delay is always enforced as a minimal delay after server startup.

If set to true, unused blobs will not be deleted in the blob store. This can be used during backup if you have a non-transactional blob store. Alternatively, the blobcollector actuator endpoint must be used to suspend deletion of unused blobs at runtime before starting a backup of a non-transactional blob store.

Converters which are used to convert custom XML formats which have been defined for the obsolete coremedia-sgmltext.dtd.

Example: sql.store.convert.DocumentType.PropertyType=com.customer.XMLConverter

The editor in versions before SCI 4.1.38 created XML text, which was not valid according to the coremedia-richtext-1.0.dtd. If you have created content with versions before SCI 4.1.38, set the property to "true" to correct these errors (CoreMedia recommends to always use "true").

The JDBC driver used to connect to the database.

Example: oracle.jdbc.driver.OracleDriver

Maximum time to wait for an enabled folder index to become up-to-date when a content query with BELOW-clause is executed. Folder index updates may take some time, if large folders were moved. If the maximum time has been reached, the query will be executed as if the folder index was offline. Set to negative value for unlimited.

Whether the store generates blob ids on its own. Defaults to "true". If set to "false", the store relies on externally provided blob ids. This allows the use of shared media stores for MLS and RLS. This flag must not be changed after the first start of a Content Server.

Define the transaction isolation level. An empty value uses the driver default setting.

Write messages of the JDBC driver to the logs

The user name for a database login. If not set, the value of "sql.store.user" will be used to log in to the database. In some cases the login username differs from the actual user, e.g. with PostgreSQL on Azure a postfix on the user name is necessary to log in.

Example:

The password of the user at the database.

If set, denotes the maximum number of prepared statements that is kept open per database connection.

A list of names of read-only media stores. Such stores receive their binary data from another Content Server which also generates the blob ids that must then be used when creating blobs in this Content Server. Because the blob ids are provided externally, sql.store.generate-blob-ids=false is required if this property is non-empty. This makes it possible to share blobs between MLS and RLS. The blob collector will not remove any blobs from shared stores.

There are problems with zOS DB2 databases to store characters that are not in the databases character set. Such characters are converted to the character with the code "0x001A" upon read. If the property is set to "true", this character will be replaced with the character defined in sql.store.substituteCharacter, thereby avoiding the replacement character that is illegal in XML texts.

Seconds between cache statistics log entries to facility cap.server.store.sgmlcache

Total size of XML objects cached in memory. Unit defaults to bytes.

This property has been deprecated since 2404.1 and will be removed in a future version.

Use sql.store.sgml-cache-size instead.

Reason:

Aligning name with all other capacity configuration options.

The character, with which "0x001A" should be replaced.

The URL of the database to connect to.

Example: jdbc:oracle:thin@HostName:Port:CM Replace HostName and Port with the appropriate values of the database host. Don't replace HostName with "localhost", this may cause problems with some JDBC drivers.

Whether to use recursive common table expressions (aka "WITH RECURSIVE") in SQL statements for content queries with BELOW-clauses (aka descendantOf clauses) to find contents below some folder, if the folder index is not online.

The user name at the database, which must match the schema.