About every 30 minutes or 45 minutes during the most active point in time of the day we are getting a series of Db connection errors. The errors themselves are part of the wierdness. They indicate the account being used to access the DB is not part of a trusted domain and cannot be used for windows authentication. On the SharePoint side of things, all indications are that the account being used is actually our farm admin account which is a domain account, local administration on all moss boxes, and DBCreator and security admin on SQL. The stsadm call to updatefarmcredentials has worked in the past to temporarily fix the issue but not permanently. When this issue occurs it comes in groups of DB connections. The first is the config DB which it fails to authenticate, after that it cycles through the SSP DB, and all content DBs. each time failing to authenticate. The authentication errors are logged on both the SQL server and MOSS servers so there is actually a connection made but for some reason it will not authenticate the user and gives us the untrusted domain issue. All servers are part of the same domain, all MOSS services accounts are also part of the same domain. on all servers I have run command to verify they use the same domain controller.The farm is 2 WFE, 1 APP/Index, clustered SQL 08. All on server 03. No firewalls between any boxes. Couple questions with this, first fo course how do get past the issue and why it would tell us the services account is part of an untrusted domain? Second would be what job is running and failing? Besides the log so far I cannot see any apparent functionality issue in the farm yet. Keeping in mind it is not being hit by production users so it is possible there are issues that we just do not see.

Sorry for the double post on other relevant detail, the farm uses NTLM

Hi, Did you install third party software on the MOSS box? If so, try to uninstall them temporary, then monitor it for a few days. Intermittent database connection issue in MOSS farm can be caused by many kinds of problems including SQL Server, network, AD, MOSS, hardware. The issue can be harmful or harmless. This usually happens when MOSS server is idle for some time. If there is a long time idle, the existing connections to SQL Server will be disconnected to same system resource. MOSS timer job may still need to connect to DB to do some scheduled work even if no one is using MOSS at night. When this happens, the first connections could fail but MOSS will retry the connection. Please check whether you enable TCP Chimney offload in your SharePoint server? Based on my experience, TCP Chimney would cause the reported error if there is any compatibility issue between the NIC driver and the OS. If it is insisted to use TCP Chimney Offload, then you should update your NIC driver to the latest version. Otherwise it would be the best to turn off TCP Chimney Offload on all servers in the SharePoint farm. For more information about TCP Chimney, please refer to the following articles: Error message when an application connects to SQL Server on a server that is running Windows Server 2003: "General Network error," "Communication link failure," or "A transport-level error" http://support.microsoft.com/kb/942861 In addition, please check whether the configuration of your cluster is correct, especially the NIC configuration. Installing a SQL Server 2008 Failover Cluster http://msdn.microsoft.com/en-us/library/ms179410.aspx If the steps above can’t resolve your issue, please post the detailed error information in the event log into the forum for further analyze. Hope this helps. Rock WangRock Wang MSFT