We are using NMM 8.2 to backup different MSSQL servers and 2 MSSQL Always On Clusters.
On the standalone MSSQL servers, the transaction log backups are working fine.
On the AoC, the incremental backups are backing up the entire database and the transaction logs are not flushed.
I can't find anything in the logs that is pointing towards this...
Anyone has an idea?
Solved! Go to Solution.
Apparently, there was something wrong with the group.
I created another group with exact the same settings and added the clients and the backups are working fine for a week now.
Can you say a bit more about your setup? Did you use named instances in your setup on cluster? Did you use dedicated user? Did you configure cluster correctly in NW?
We use Windows Server 2012 with MSSQL Server 2012 in an AlwaysOn cluster.
The MSSQL servers do not use named instances.
We use a dedicated user which is admin on the MSSQL servers and is specified in the Client Properties in NetWorker.
We added the 2 MSSQL servers as client in NetWorker and a client with the cluster name.
Save set is MSSQL#<AoC Availability Group name>.
Backup command: nsrsqlsv
In Remote Access all admin users of the 2 servers.
Schedule: Weekly Full, Daily Incr.
I'm doing a back with the GUI.
The client has the name of the cluster.
nsrsqlsv is the command specified in the 'Backup command' field.
In the nsrsqlsv log on the cluster members, I've found the following message:
NSR info 88 Last backup saveset and current saveset aren't same. Promoting the backup level to full.
Every time a backup is running.
Again, backup command should be what I said.
As for message, it may indicate you index is written elsewhere which should be addressed by backup command as I said.
I've changed the backup command, but results are the same.
I noticed something else.
I’ve seen that sometimes, an incremental backup was working correctly.
But this is random behavior and mostly after a full backup.
Yesterday, I’ve performed a full backup during daytime and the incremental backup at night was working fine.
Today, I started an incremental backup and it was a full again.
Do you use any other backup method to backup those DBs? Please note that if you use for example NMM VSS based backups, the cycle will be broken, and the next VDI based backup will be promoted to full.
Same would happen if you do any other type of backup, such as Windows backup or any third party backup tool.