Veritas-bu

Re: [Veritas-bu] Windows 2-node cluster backup

2008-06-16 10:39:52
Subject: Re: [Veritas-bu] Windows 2-node cluster backup
From: "Martin, Jonathan" <JMARTI05 AT intersil DOT com>
To: "Kennedy, Cameron" <Cameron.Kennedy AT bovislendlease DOT com>, <veritas-bu AT mailman.eng.auburn DOT edu>
Date: Mon, 16 Jun 2008 10:12:54 -0400
Just a shot in the dark here.  We don't do Netbackup SQL Backups.
 
Add the virtual name and all cluster IPS to the HOSTS file on the master so when the request comes in (no matter what IP the request comes from) the MASTER sees it as coming from the virtual machine.
 
We should have this problem here, but my paranoid DBAs actually want their Oracle Databases backed up via the node name.
 
-Jonathan


From: veritas-bu-bounces AT mailman.eng.auburn DOT edu [mailto:veritas-bu-bounces AT mailman.eng.auburn DOT edu] On Behalf Of Kennedy, Cameron
Sent: Monday, June 16, 2008 8:59 AM
To: veritas-bu AT mailman.eng.auburn DOT edu
Subject: [Veritas-bu] Windows 2-node cluster backup

Hi all, hoping someone has some experience with backing up a Windows SQL cluster. I am having the below issues:
Netbackup 6.5.2, AIX master + media + Windows media.
2-node Windows SQL cluster = 2 physical servers, 2 virtual servers, 2 SQL instances.
Typically we have 1 virtual server + instance on each physical server and this runs fine.
Each physical server has a dedicated backup NIC on 1GB LAN, and each has a static IP assigned.
This works for backup fine, until a failover of the cluster happens without a failback.
 
So we created virtual backup nics, that are tied to the real backup nics. We did this via Windows cluster administrator.
Now when a cluster failover happens, the virtual backup IP's follow to the active node - good.
However, if we set (in the hosts file) that the backup client is assigned the virtual NIC, when the backup starts the master server can talk to the client via the virtual backup IP, but the client responds to the master server via the real backup IP. Netbackup promptly fails the job, because of this discrepancy as seen in dbclient log.
 
If I update the hosts file to include the real backup IP of the host, then the job runs fine. This of course would mean that everytime the cluster failed, I would have to update the hosts file which would be unacceptible.
 
How are others doing this?
 
thanks
 


This email (including any attachments) is confidential. If you are not the intended recipient you must not copy, use, disclose, distribute or rely on the information contained in it. If you have received this email in error, please notify the sender immediately by reply email and delete the email from your system. Confidentiality and legal privilege attached to this communication are not waived or lost by reason of mistaken delivery to you. Lend Lease does not guarantee that this email or the attachment(s) are unaffected by computer virus, corruption or other defects. Lend Lease may monitor incoming and outgoing emails for compliance with its Email Policy. Please note that our servers may not be located in your country.

_______________________________________________
Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
<Prev in Thread] Current Thread [Next in Thread>