[rear-users] Unexpected Netbackup Restore behavior

Gratien D'haese gratien.dhaese at it3.be
Wed Sep 3 10:43:17 CEST 2014

 Great - just create an account on GitHub and create an issue for this 
 on rear issues pages.
 If you feel lucky you could make a pull request to get it fixed...


 On Wed, 03 Sep 2014 09:42:33 +0200, Zweimueller Wolfgang 
 <w.zweimueller at salk.at> wrote:
> Hi Gratien,
> Gratien D'haese <gratien.dhaese at it3.be> writes:
>> Hi Wolfgang,
>> welcome our rear community.
> Thx!
>> Indeed seems not the expected behavior -so why don't you make an 
>> issue
>> of it.
> I will do that.
>> In the meantime you could have a look at the script
>> /usr/share/rear/restore/NBU/default/40_restore_with_nbu.sh
>> which is playing with dates to retrieve its files from NBU.
> I had a look at the script and saw the reason for my problems. It is
> the way backup images are listed and selected.
> Regards,
> Wolfgang
>> regards,
>> Gratien
>> On Tue, 02 Sep 2014 14:08:54 +0200, Zweimueller Wolfgang
>> <w.zweimueller at salk.at> wrote:
>>> Hello everyone,
>>> I am new to this list and ReaR. I am testing ReaR since about two
>>> weeks
>>> now and have to say it is very good work. Now we are in the process
>>> to
>>> set it up for production use for about 100 machines.
>>> When I do a "rear recover" I get asked for the date from which I 
>>> want
>>> to restore. There I have two choices: hit RETURN and restore the
>>> latest backup or enter a date stamp to do a Point-In-Time restore.
>>> What did I expect? When I hit the ENTER key I want a complete 
>>> restore
>>> consisting of the last full backup and all incremental backups
>>> afterwards. I think this is what most people do when they are in a
>>> situation where they need a disaster recovery. This is also the
>>> behavior of a pure "bprestore" without any time restrictions.
>>> Now my tests showed me that I really get the latest backup which is
>>> sometimes only a incremental backup! Needless to say that ReaR 
>>> cannot
>>> complete the recovery of such a system.
>>> Has anyone an explanation of how it came to this implementation? 
>>> Any
>>> plans to change that behavior?
>>> Regards,
>>> Wolfgang

More information about the rear-users mailing list