Failed UNIFYNow SQL operation with error: Operation faulted: Login failed for user 'NT AUTHORITY\ANONYMOUS LOGON'.
We have a operation list which is currently failing and will only run successfully after restarting the UNIFYNow service, not sure of the exact cause of the root error but any subsequent runs fail immediately with the error:
Operation faulted: Login failed for user 'NT AUTHORITY\ANONYMOUS LOGON'. |
Operation e681ce94-daf3-4c29-b6cc-f9d03f55cce8 failed in operation list with id df1696a0-948e-41e8-939f-4b48509b5508 for the following reason. This is retry number 0: System.Data.SqlClient.SqlException (0x80131904): Login failed for user 'NT AUTHORITY\ANONYMOUS LOGON'.<br> at System.Data.SqlClient.SqlConnection.OnError(SqlException exception, Boolean breakConnection, Action`1 wrapCloseInAction)<br> at System.Data.SqlClient.TdsParser.ThrowExceptionAndWarning(TdsParserStateObject stateObj, Boolean callerHasConnectionLock, Boolean asyncClose)<br> at System.Data.SqlClient.TdsParser.TryRun(RunBehavior runBehavior, SqlCommand cmdHandler, SqlDataReader dataStream, BulkCopySimpleResultSet bulkCopyHandler, TdsParserStateObject stateObj, Boolean& dataReady)<br> at System.Data.SqlClient.SqlCommand.RunExecuteNonQueryTds(String methodName, Boolean async, Int32 timeout, Boolean asyncWrite)<br> at System.Data.SqlClient.SqlCommand.InternalExecuteNonQuery(TaskCompletionSource`1 completion, String methodName, Boolean sendToPipe, Int32 timeout, Boolean asyncWrite)<br> at System.Data.SqlClient.SqlCommand.ExecuteNonQuery()<br> at Unify.Framework.Data.AdoNetDataControl.ExecuteNonQuery(String commandText, IEnumerable`1 parameters, CommandType commandType)<br> at Unify.Product.EventBroker.OperationListExecutorBase.RunNextOperations(IEnumerator`1 operationEnumerator)<br>ClientConnectionId:a38cf1d2-62c1-4c4e-8270-29b04c876079<br>Error Number:18456,State:1,Class:14
The operation will run successfully after the restart and after some period will start failing.
It is once it starts failing that the above error is shown.
Answer
There are a few things that could cause login errors. Given that it's intermittent, it's clearly not invalid credentials. Since you say it begins to fail instantly, it's probably not a timeout either. I'm wondering if there is a limit to the number of concurrent sessions. Is there anything else that accesses the same database regularly, both other operations in UNIFYNow or other external services? Are you able to perform a SQL trace while the issue is occurring?
The stored procedure that is called flows like this:
UNIFYNow -> Reporting DB -> FIM Sync DB
The way the client has described to query I suspect it is grabbing information directly from the metaverse. :(
They have mentioned that they suspected it to be a Kerberos timeout as it has to connect to two different databases sequentially. While this may be an issue to investigate I am wondering if there is something might make it so that UNIFYNow cannot successfully run the stored procedure again until a service restart.
In the meantime I will look at getting a SQL trace next time the issue occurs.
I'm a bit confused about the multiple databases - are they both connected to from UNIFYNow via separate SQL operations, or does UNIFYNow only connect to the first, and the SP there is configured in such a way as to make the further connection to the second? If the latter, is it possible that there's connection pooling at that layer? I don't believe that UNIFYNow would be putting itself into a state that would cause these instant failures until service restart.
The Windows Event Logs might contain more information as well.
Any update Matthew? As Curtis mentioned, the Windows Event Logs are helpful here, it'll help identify what the cause is (state=1 is the hidden value, the Event Log will have the proper one).
There doesn't appear to be any evidence that UNIFYNow isn't functioning correctly. If any new information is presented please add it and I'll reopen the issue.
Customer support service by UserEcho
There doesn't appear to be any evidence that UNIFYNow isn't functioning correctly. If any new information is presented please add it and I'll reopen the issue.