Doing some more research on Microsoft and talking with the client, they re-addressed the cluster on 7/26 when they moved it into their new CISCO switch with VLAN's configured. I think they changed all the IP addresses except for the IP address of the SQL Server virtual instance. I found a couple articles on Microsoft about TCP not binding to the correct port after an IP change to the cluster. I forwarded those articles to the client, but I don't know if this is the problem either. The TDP backups worked before the IP address change, so I'm hopeful!
This box is going production in a week and the backups dont' work now! Even though it's a SQL error that is being returned, the DBA's point to TSM. So it's up to me to fix it. This is the KB article that level-2 referred to in my PMR. The DBA says that's not the problem. They are using TCP, not named-pipes. The DBA says he can back up the database via SQL Server manager, so it's not a SQL problem! Bill Boyer DSS, Inc. -----Original Message----- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] Behalf Of Del Hoobler Sent: Friday, July 30, 2004 4:08 PM To: [EMAIL PROTECTED] Subject: Re: TDP SQL errors backing up SQLSERVER Cluster on Windows2003 Bill, All I can say is that Microsoft found a problem in MDAC that they had to fix. Every time we have seen the "[DBNETLIB]ConnectionRead (WrapperRead())" error when backing up with Data Protection for SQL, it has turned out to be KB827452. I think I remember something about registry entries being invalid... or something like that. Microsoft should be able to tell you more details on what they fixed. Here is the link to the KB article: http://support.microsoft.com/default.aspx?scid=kb;EN-US;827452 Thanks, Del ---------------------------------------------------- "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> wrote on 07/30/2004 02:15:50 PM: > Windows2003 Cluster running SQL Server > TDP SQLServer 5.1.5.0 > TSM Client 5.2.2.10 > TSM Server on Windows2003 5.2.2.5 > > > We were doing TDP backups of the SQLDatabases just fine, until the cluster > was re-addressed for IP. Now when the TDP backup runs, right at the end it > fails with a DBNETLIB error. It's looking like Microsoft #827452, but it was > working. What could have changed, or not changed, when the IP addresses were > changed? The Ipaddress of both cluster nodes as well as the cluster IP > resource were changed. TSM PROMPTED scheduling works, the B/A backups of the > C: drives on each node work, but the TDP backups fail. > > I opened a PMR and they just pointed us to the Microsoft problem. > > Anyone having problems or have some suggestions??? > > Bill Boyer > "An Optimist is just a pessimist with no job experience." - Scott Adams
