Preventing NW from making boostrap after each group

[denisfr]

Hello,

I've made a separate pool to save indexes and bootstrap with a cron command (savegrp -O...).

But NW (7.3.2) always makes a bootstrap when it saves saveset from itself (even for a litte file ).

Is there a way to prevent it doing this bootstrap ?

Thanks.

 

[605]

Yes and no.

NetWorker will run a bootstrap backup once a day,
- when the group starts with the NW server's client
- This means that the group must be active (enabled).

If this group is not enabled, as a precaution, NW will create a bootstrap whenever a group runs.

BTW - of course you do not need to create a separate group to filter index and bootstrap backups for a certain pool. Just do it by the save set list only and specify
index
bootstrap
and this will work for all groups.

And do not forget that savegrp -O does not backup the index for all clients but only for those which are members of this group.

 

[denisfr]

So, if I've 10 times Networker as a client belonging to an enabled group, I got 10 times a bootstrap. That does not satisfy me :-(

No way to bypass, thus ?

"And do not forget that savegrp -O does not backup the index for all clients but only for those which are members of this group."
Yes of course, but I did not mentionned that I had all my clients belonging to this group...

 

[605]

Well, each group is supposed to be started once a day anyway. What is the reason to include the same client to 10 groups?

 

[denisfr]

We have several Oracle databases on multiples machines. And we don't have enough (or we don't want to use ;-)) NMO clients. So, we use TimeFinder/FlashCopy facilities offered by EMC/IBM storage arrays. After copies are realized, we mount target devices (with vg information) on NW and let NW save data as if it was its own (it saves also bandwith, CPU and I/O usage, because the 'client' is not involved in the process)... So we've several groups representing one or multiple databases to be saved, not a the same time...

More clear ? ;-)

 

[605]

Thanks. Intersting aspect. Unfortunately you have to pay the price somehow ...

The only solution i see is not to use the NW server's but another client/snode and to backup the NW server by the standard mechanism. I do not see any other possibility.