June 2, 2009 at 12:27 am
I am trying to install SP3 on a SQL Server 2005 Active/Passive cluster.
The current version of SQL Server 2005 on the cluster is: 9.00.3282.00
I am trying to patch it to version: 9.00.4035 http://support.microsoft.com/?kbid=955706
I am running the update using an account that has admin rights on both nodes of the cluster.
The sections that is failing is: "Database Services"
Here is the error logged in: %ProgramFiles%\Microsoft SQL Server\90\Setup Bootstrap\LOG\summary.txt
----------------------------------------------------------------------------------
Product : Database Services (MSSQLSERVER)
Product Version (Previous): 3282
Product Version (Final) :
Status : Failure
Log File :
Error Number : 11009
Error Description : No passive nodes were successfully patched
----------------------------------------------------------------------------------
...
**********************************************************************************
Summary
No passive nodes were successfully patched
Exit Code Returned: 11009
The "Hotfix.log" file contains the following errors:
06/02/2009 15:58:48.292 Installing instance: MSSQLSERVER
06/02/2009 15:58:48.292 Enumerating passive cluster nodes
06/02/2009 15:58:48.307 Patching available passive node: SQL-NODE-02
06/02/2009 15:58:48.307 Patching passive node: SQL-NODE-02
06/02/2009 15:58:48.307 Waiting for first completed passive node
06/02/2009 15:58:48.386 Task Scheduler: Created remote folder for product instance target \\SQL-NODE-02
06/02/2009 15:59:23.027 Task Scheduler: Transferred payload to remote product instance target \\SQL-NODE-02
06/02/2009 15:59:23.027 Task Scheduler: Transferred payload to remote product instance target \\SQL-NODE-02
06/02/2009 15:59:23.042 Task Scheduler: Error, cannot create new scheduled task for product instance target \\SQL-NODE-02
06/02/2009 15:59:23.042 Task Scheduler: Error, cannot create scheduled task for product instance target \\SQL-NODE-02
06/02/2009 15:59:23.480 Task Scheduler: Removed remote folder for product instance target \\SQL-NODE-02
06/02/2009 15:59:23.480 Error, remote process failed for product instance target
06/02/2009 15:59:23.480 Exit code for passive node: SQL-NODE-02 = 1603
06/02/2009 15:59:28.308 The following exception occurred: No passive nodes were successfully patched Date: 06/02/2009 15:59:28.308 File: \depot\sqlvault\stable\setupmainl1\setup\sqlse\sqlsedll\instance.cpp Line: 3510
So it looks like the update is trying to create a scheduled task on the passive node to install the update.
I have configured the scheduled tasks service as a Generic Service resource on the cluster, so that scheduled tasks only run on the active node of the cluster.
Is there any way around this?
June 2, 2009 at 1:04 am
I have successfully re-run the update.
Select @@version returns:
Microsoft SQL Server 2005 - 9.00.4035.00 (Intel X86)
Nov 24 2008 13:01:59
Copyright (c) 1988-2005 Microsoft Corporation
Standard Edition on Windows NT 5.2 (Build 3790: Service Pack 2)
The Hotfix.log file now reads:
...
06/02/2009 16:44:59.375 Installing instance: MSSQLSERVER
06/02/2009 16:44:59.375 Enumerating passive cluster nodes
06/02/2009 16:44:59.375 Patching available passive node: SQL-NODE-02
06/02/2009 16:44:59.375 Patching passive node: SQL-NODE-02
06/02/2009 16:44:59.375 Waiting for first completed passive node
06/02/2009 16:44:59.439 Task Scheduler: Created remote folder for product instance target \\SQL-NODE-02
06/02/2009 16:45:35.176 Task Scheduler: Transferred payload to remote product instance target \\SQL-NODE-02
06/02/2009 16:45:35.176 Task Scheduler: Transferred payload to remote product instance target \\SQL-NODE-02
06/02/2009 16:45:35.192 Task Scheduler: Created new scheduled task for product instance target \\SQL-NODE-02
06/02/2009 16:45:35.192 Task Scheduler: SaveTaskToDisk(): Got task interface
06/02/2009 16:45:35.288 Task Scheduler: Saved new scheduled task for product instance target \\SQL-NODE-02
06/02/2009 16:45:35.288 Task Scheduler: Created scheduled task for product instance target \\SQL-NODE-02
06/02/2009 16:45:35.288 Task Scheduler: Started scheduled task for product instance target \\SQL-NODE-02
06/02/2009 16:45:35.288 Task Scheduler: Started scheduled task for product instance target \\SQL-NODE-02
06/02/2009 16:47:18.849 Task Scheduler: Scheduled task for product instance target has completed
06/02/2009 16:47:18.849 Task Scheduler: Waiting for exit code from scheduled task for product instance target \\SQL-NODE-02
06/02/2009 16:47:23.887 Task Scheduler: Received exit code 3010 from scheduled task for product instance target \\SQL-NODE-02
06/02/2009 16:47:23.887 Task Scheduler: Result code for scheduled task for product instance target has been received
06/02/2009 16:47:23.887 Task Scheduler: Removed scheduled task for product instance target \\SQL-NODE-02
06/02/2009 16:47:24.233 Task Scheduler: Removed remote folder for product instance target \\SQL-NODE-02
06/02/2009 16:47:24.233 Remote process completed for product instance target
06/02/2009 16:47:24.233 Exit code for passive node: SQL-NODE-02 = 3010
06/02/2009 16:47:26.894 First successful passive node has been completed
...
The databases are up and running Ok on the Active node.
June 2, 2009 at 1:20 am
The scheduled tasks setup to run as a Generic Service resource under the same group as SQL Server, so that the scheduled tasks only run on the Active node.
Registry replication is set for: "HKLM\Software\Microsoft\SchedulingAgent\"
The tasks are stored on the P:\ drive. which is a cluster disk resource.
This is the procedure I used to get the installation to run on the Passive node:
On the Passive Node:
Created folder: c:\ScheduledTasks
Edited registry: HKLM\Software\Microsoft\SchedulingAgent
changed: "LogPath" to: "c:\ScheduledTasks\log\SchedLgU.txt"
"TasksFolder" to: "c:\ScheduledTasks"
Started the "Task Scheduler" service.
On the Active node:
Started the SQL Server 2005 SP3 install, and waited for it to complete successfully.
On the Passive node:
halted the "Task Scheduler" service
Restarted the Active node.
Checked the Passive node, the Task Scheduler service is running again, and the registry keys have reverted back to point to the P:\ drive.
Restarted the Passive node.
The Clustered resources have all moved back to the Active node, and the databases are running Ok.
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply