Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
| Author |
Topic |
|
franco
Constraint Violating Yak Guru
255 Posts |
 Posted - 2005-08-04 : 10:06:25
|
| Hi all,I am aware that this is not a tipical SQL Server question, but still SQL Server is on the cluster and it suffer from the problem that I want to share with you.I would like to share with you a strange situation that we notice on a 2 node cluster, windows server 2003 - no service pack.The cluster is configured as follow:Node A hosts SQL Server 2K sp3a and file systemNode B hosts Oracle 9i and Lotus Domino Server 6.5.3.Sometimes, not always, when we move one group to the other node we end up in very strange situations, like Event id 1146 and Event id 1069.The description of event id 1146 is:The cluster resource monitor died unexpectedly, an attempt will be made to restart it.The description of event id 1069 is:Cluster resource 'Network name' in Resource Group 'Group name' failed.The problem is that when this happen, every group of the cluster are affected by this problem.I mean, for example, when we move SQL Server group to the other node and this problem appears, all the other groups, like Oracle group or Lotus group are affected by Event id 1146 and event id 1069.I know there is a fix provided by Microsoft here:http://support.microsoft.com/default.aspx?scid=kb;en-us;886652The KB article say:CAUSEThis problem occurs if the following conditions are true:• You have a network name resource with Kerberos enabled. • This name of the computer object that corresponds to the Kerberos-enabled network name resource of the cluster is located in an organizational unit that has a forward slash character in its name. So none of this conditions apply to us, but here comes my guess:When Lotus Domino was first installed on the cluster, one share was also created.I don't know if this share was created following the rules posted in this MS KB:http://support.microsoft.com/default.aspx?scid=kb;[LN];224967I know for sure that when you create a cluster share the following registry key is populated:HKLM\system\CurrentControlSet\Services\lanmanserver\SharesI also notice that when you take offline that share, also the registry key disappear.The problem is that we found the name of one Share on that key that was first created for the Lotus Domino group and then was deleted, but again I don't know how this was done.Here comes the question:Is it possible to think that this dirty registry key is the cause of the strange failover scenario described before?Sorry for this long post and for my bad english, but every suggestion is really appreciated.Cheers.Franco |
|
|
|
|
|
|
|
Event ID 1146 and 1069 - strange failover scenario