You can configure the Identity Server instances to store access tokens in different tables according to their user store domain. This is referred to as user token partitioning and it ensures better security when there are multiple user stores configured in the system. For information on configuring user stores other than the default one, see Configuring Secondary User Stores.
To enable user token partitioning, you should change the
<AccessTokenPartitioning> elements in the
Assertions are used to embed parameters into tokens in order to generate a strong access token. You can also use these parameters later for various other processing functionality. At the moment, the Identity Server only supports UserName as an assertion.
By default, assertions are set to
false in the
You can make it true by setting the
<UserName> element to
true. You can add a user name to an access token when generating the key, and verify it by Base64-decoding the retrieved access token.
This parameter implies whether you need to store the keys in different tables or not. It can be used only if
<UserName> assertion is enabled. If it is, set the
<EnableAccessTokenPartitioning> element to true in
<IS_HOME>/repository/conf/identity/identity.xml to store the keys in different tables.
Also set the user store domain names and mappings to new table names. For example,
- if userId = foo.com/admin where 'foo.com' is the user store domain name, then a 'mapping:domain' combo can be defined as 'A:foo.com'.
- 'A' is the mapping for the table that stores tokens relevant to users coming from 'foo.com' user store.
You can provide multiple mappings separated by commas as follows. Note that the domain names need to be specified in upper case.
According to the information given above, change the
<APIKeyManager> element in the identity.xml file as shown in the following example: