Fro automatic rotation on disk based backups, sue Advanced File System disk device, it's built in.
For the former, sue schedules to set what level when, use browse period for how long you can see individual files and retention period for how long before the backup can be overwritten (rotated).
I am sorry? where can I read about "Advanced File System disk device". Can you understand me how can I from command line set command for remove old savesets and remove phisical file in backup sets, or do it automatic (but retention policy don't remove sevesets and file - my experience from experiment.
Chapter 5 of the Administrator's Guide also explains the features of the older File Type Device and the younger Advanced File Type Device. Look for the function " Reclaiming or recovering space".
If you want to delete a save set from the command line, this is not so easy for a media in the Advanced File Type Device. On both types, each save set is stored in a separate file which you can delete. However, only for the FTD, it is easy to recognize. In general, this is what you have to do:
- find out the save set id (ssid) with the mminfo command
mminfo -q "query" -r ssid
- delete the save set from the media index with
nsrmm -d -S ssid
- delete the file "ssid.0" from the FTD directory
The other method works for both device types. It uses the fact that NW will do the cleanup itself as it is described in the book:
- find out the save set id (ssid) with the mminfo command
mminfo -q "query" -r ssid
- change the save set status to "recyclable"
nsrmm -o recyclable -S ssid
- run the automatic cleanup
nsrim
Ok. Thanks you - it's really work, but I want to know how can I run the automatic cleanup
nsrim
from Networker Administrator automatic from client, possible like sheduling every day or hour (becouse index manager located on server side) ?
You can't as nsrim has no option "-s server" which allows you to execute a program remotely. In such case you must configure a scheduler at the server.
But if I try to sheduler oracle backup
I receive error message like this
"No full backups of this save set were found in the media database performing a full backup" - I cannot undestand - why?
If I do this backup from client side through rman script - it's work very good.
And another question - how can I backup folders and files (initSID.ora,Listener.ora,...) through script and with RMAN backup from client side?
You must distinguish between file backups and database backups - they have different parameters. That's why you deal with two logical clients (two client resources for the same machine) - one for the file backup and one for the DB backup.
A 'normal' file backup is not done via an RMAN script, except for the Oracle specific files like tablespace files and control files. To do this, just start a backup for the 'filesystem' client.
Ok. Backup Oracle database and backup some files I can do. But I cannot do sheduler BACKUP ORACLE DATABASE. Can You help me what I should to do for it or where my error?
I have error - "No full backups of this save set were found in the media database performing a full backup"
script for sheduler backup in save set -
--------------------------
connect target sys/xxx@symon;
connect rcvcat rman/xxx@SONG;
run
{
allocate channel t2 type 'SBT_TAPE'
parms 'ENV=(NSR_SERVER=sector, NSR_CLIENT=ocean, TRACE=2, NSR_TRACE=2, NSR_DATA_VOLUME_POOL=Full)';
send 'NSR_ENV=(NSR_SERVER=sector, NSR_CLIENT=ocean, NSR_USER=serg, TRACE=2, NSR_TRACE=2, NSR_DATA_VOLUME_POOL=Full)';
backup full (database);
release channel t2;
}
If this is what you have done, it should work fine. If your only concern is related to this message ..
"No full backups of this save set were found in the media database performing a full backup"
.. ALTHOUGH you did successful manual backups already, this is not what you should worry about. Actually it has to to with the fact that NW uses backup levels similar to Oracle's. The issue here is that NW MANUAL backups get NO level assigned. Due to this fact, if you run an automatic backup and use the schedule Default (supposed to run incrementals throughout the week), NW will not find a FULL to refer to and overrides the actual level with the first backup. This is true for all backups of any data.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.