Job Failed to Start because of Logon Failure
Updated: May 18, 2011
Applies To: Windows HPC Server 2008, Windows HPC Server 2008 R2
A submitted job fails to start, and returns the following error message, where nodename is the name of the node on which the job failed to start:
Failed to start on node <nodename>. Error: Logon failure: the user has not been granted the requested logon type at this computer. Exception of type Microsoft.Hpc.Activation.NodeManagerException was thrown.
The following screenshot shows the error message as it appears in the Job View dialog box in HPC Job Manager:
When the HPC Job Scheduler Service starts a job on a node, it attempts to log the job owner on to the node. If the job owner does not have local logon permissions on the node, the job fails to start and displays the logon failure error message and the exception of type
Identify the users or groups that have local logon permission on one of the nodes on which the job failed to start.
Log on as an administrator to one of the nodes where the job failed to start.
Click Start, point to Administrative Tools, then click Local Security Policy.
In the Local Security Policy snap-in, expand the Local Policies folder, then select User Rights Assignment.
In the view pane, right-click Allow log on locally, then click Properties. The properties page appears.
On the Local Security Settings tab you can see the list of users and groups with local logon permissions. See if the job owner, or a group to which that user belongs, is listed.
Job owners must have local logon permissions on compute nodes in order to run jobs. For more information about local logon, see Log on locally (http://go.microsoft.com/fwlink/?LinkId=147886).
You can use one of the following methods to give a user local logon permission:
Add the job owner to a group that has local logon permissions on the compute nodes.
Add the job owner, or a group to which that user belongs, to the list of users and groups with local logon permissions on your compute nodes.
If local log on permissions are restricted by your Active Directory domain policy, ask your domain administrator for an exception to the policy for your cluster nodes. For more information about managing Group Policy, see Group Policy Management (http://go.microsoft.com/fwlink/?LinkId=147887).
To verify that the job owner has the required permissions and that the job can start, requeue the failed job, or ask the job owner to requeue the failed job.
For information about how to requeue a job, see one of the following topics: