I just configure an Oracle 10 G to for Online backup to DDboost device though DFA feature in the NMDA 1.2
The backup Job took for first time 27 min for 21 GB and for second job from 13 -14 minte . NSR_DRVICE_RES=DATAdomain and yes
now I am testing recovery on the other machine
I use mminfo -s server name -c clientname -ot to browse for control file
then recover it
they ran the RMAN
}
se chan...........
restore databse
recover database
...
realease
..
)
the RMAN run succefully but I watch from server the recovery process the cleint have to bowse for each RMAN backup data file and recover it , this is very long process and it took more than 3-4 hours and really I even did not see the fininsh of it ...................
is there any reommendation to solve this issue the recover process is too long from datdomain , is it supposed to be at least same as the backup job time 27 min
The SR issue is not a question regarding how to identify where the control file is but a relocated recovery issue.
When securing Oracle Data with NetWorker the function of NetWorker is to accept the data stream produced by Oracle RMAN, secure to tape and then tell RMAN it has done this.
If you want to know where the control file is secured you should look in the RMAN catalogue and see where it is from here - all details you need should be here. You can then return to NetWorker and look for the specific saveset name to see which volume it is located on and ensure it is available for when the recovery is started - also via RMAN.
The SR is now with our Database team who should be contacting you shortly over this.
The DBA want to browse for a control file and restore it , then he want to browse for control file for he do this procedure for prevous symantec NBU and he want to make same procedure , he tell that in case the other source machine is down , I need to read where is the control file ,
the backup job is fine t will try to test that cenarion but emc recommends us to ues NMDA and DFA to send directly to ddboost devices , do you recommend to used the 1.2 RA from where I can get this software
I was just answering the question why you need to Browse the COntrol file ,
the SR is now recovery yes , I try last time and it was so slow recovery , as I mentioned in the SR the backup Job took 14 min and recovery took at least 3 hours I could not stay more to see if it ends ,
The DBA want to browse for a control file and restore it , then he want to browse for control file for he do this procedure for prevous symantec NBU and he want to make same procedure , he tell that in case the other source machine is down , I need to read where is the control file ,
the backup job is fine t will try to test that cenarion but emc recommends us to ues NMDA and DFA to send directly to ddboost devices , do you recommend to used the 1.2 RA from where I can get this software
You can restore control file from Oracle as the control file also gets backed up by Oracle. It's all about RMAN.
I can't say anything about v1.2 experience as I do not use it (I'm still working with NMO).
ble1
6 Operator
•
14354 Posts
•
56186 Points
583
0
Posted October 13th, 2011 06:00
There are two things I would test if I had this sort of problem:
a) test backup to normal disk on same host where your DD device is and compare how time does that take
b) test this with NMDA 1.1 or NMO 5
The reason I would test it against normal disk is to see:
a) if I have network issue in direction of sending data back to application host
b) if this is related to boost process
The reason I would test this with previous NMDA and NMO version is:
a) because NMDA 1.2 is RA release which did get boost integration just in this release
b) because I know NMO 5 build 347 works just fine for me so I would have valid reference (note: I do not use DD though)
As there might be several possible reasons for what you see (even worse, combination of few), I would suggest to handle this via ticket with support.