Known Problems and Limitations

Opening Knowledge Trees with collections on mapped drives.


On Windows, knowledge trees with a collection on a mapped drive cannot be opened if the K2 Administration Server logs on using the Local System account. (85669)

Workaround: Configure your K2 Administration Server to log on using an account with access to the mapped drive.

a. Click Start, point to Settings, and then click Control Panel.

b. Select Administration Tools, then in the new window, select Services.

c. In the new window, double click Verity K2 Administration Server.

This displays the Verity K2 Administration Server properties (Local Computer) dialog box.

d. Tab to the Log On page.

e. Select This account: and enter the user name and password of a user with access to the mapped drive.

f. Click OK.

New documents are not indexed correctly into knowledge trees when using the ktmgr command-line tool.


Running ktmgr inserts the documents, but loses the persist setting. (89642)

Workaround: To maintain a persist setting, use the updated ktmgr command as follows.

ktmgr -kt tree -populate delta -persist 10

updates the knowledge tree every 10 seconds, but does not update the K2 Index.

ktmgr -kt tree -populate delta -persist 10 -k2

updates the knowledge tree and the K2 Index every 10 seconds.

Even with this update, the knowledge tree state with respect to the K2 Server (online or offline) must be cycled for all searches for rck2 to work. Document searches using the Root category work even if the knowledge tree is not taken offline and then placed online after being updated by ktmgr , but category searches and some document searches using subcategories of Root do not display the updates until the knowledge tree state is cycled using rcadmin or the K2 Dashboard.

Different document counts are shown between Intelligent Classifier and rck2.


The use of the <NOT> operator in a query might be invalid. Intelligent Classifier allowed an invalid placement of the operator. (74317, 74318)

Workaround: Use the latest version of the OTL file shipped with Intelligent Classifier.

Existing parametric indexes (outline or control.pi) cannot be edited with Intelligent Classifier.


Parametric indexes can be created from Intelligent Classifier, but existing parametric indexes (outline or control.pi) cannot be loaded and edited. To make changes to a parametric index in Intelligent Classifier, you must create a new one. (86398)

Case sensitivity is on by default and cannot be altered for the autocreation of a parametric index.


This might produce multiple nodes of the same name with upper and lower case differences. (101603)

Parametric index does not support the population of a bucket set that contains tree nodes if one of those nodes has a reference category whose parent category is not part of this subtree (startnode).


To populate the bucket set containing a reference category, you need to define the startnode as root. (95389)