Networker 8 NMO RMAN Client Configuration

Firstly, welcome to the forums, and above all, thank you for being an EMC customer.

Please consider moving this question as-is (no need to recreate) to the proper forum for maximum visibility.  Questions written to the users' own "Discussions" space don't get the same amount of attention and can go unanswered for a long time.

You can do so by selecting "Move" under ACTIONS along the upper-right.  Then search for and select: "NetWorker Support Forum".

NetWorker Support Forum

First of all, NMO is old RMAN module - new one is called NMDA.  If you are talking about Oracle, level is controlled through RMAN script which is usually done by DBAs.  I'm not sure you really wish to use incremental with Oracle, but that is DBA's decision.

I will certain look at going down the NMDA route.  Does NMDA allow me to do backups to an ASM drive or does it have to be disk on a standard file system?

ASM file system contains DB files usually and those are handled by RMAN backup and not file system backup.  NMDA will use RMAN to backup whatever you specify as backup in RMAN script.

I think I was referring to the probe based feature that NMDA does to allow backups to the FRA and NMDA will then pull them from there and will then update the RMAN catalog to reflect the new location \ device.

Well, you started with NMO and incremental backups and now you are already on different topic - are you sure about what you are asking?  I didn't test FRA option as I use RAC which is not supported combo.

Sorry I am addressing multiple issues at this time.   Basically we started using NMO and then switched to NMDA per your suggestion.  At this point the config is NW 8.1 and NMDA 1.5 on Linux RHEL6.4 on Oracle 11.2.0.4 on standy-by databases.   I am trying to achieve the best desired result in which we can implement a RMAN FULL + Incremental 1 strategy via scheduled either in NW or client based.

When I look for client backups it doesn't appear as though I can do multiple oracle instances (we are in a multi-tenant database environments, each server holds many databases) and I need the easiest and bomb-proof method to backup these databases.  some databases are 1TB+ and I don't wish to do a full daily on these databases as the maintenance window will not allow for this.

I tried using the probe based backup option using scheduled disk backups with NMDA.  Although this seemed to setup properly the probe didn't pickup the backup when DISK_BASED_BACKUP=TRUE option was set.   I haven't gone and added additional debug switches to verify the output as of yet.   The scheduled disk backups with NMDA didn't seem to allow for multiple instances.   As a result, I wouldn't know where in NW to setup to allow for this capability.

Additionally I tired using the RMAN script methodology on the server, however I saw no options to allow a schedule that would accommodate FULL + Incremental 1 during to week so that we don't have to do daily fulls.   Perhaps I am missing something but I saw the backup type only allowed one selection per client.  Additionally everytime you switch it would need to be done manually.  I saw no options to allow for it to do a FULL + Incremental 1 based on a schedule.

We have similar setup, but we do not use incr at all.  DBs are always level full where production gets backup each day, acceptance we don't backup, test and development gets DB backup once per day.  Archive logs run once per hour.  We also run Oracle hotels on Linux in RAC environment hosting some 100 databases per RAC cluster.  I'm not sure what biggest DB is, most likely DWH, but backup is rather fast - at least in our case (we use around 8 channels and use multiple nodes to send data). We never considered incremental backups as tests in the past showed that this is not so fast when it comes to restore - to my knowledge nothing has changed since.  The only thing one must pay attention if not running full DB backups on daily basis is that archive log destination might need to be a bit bigger to accommodate restores. We use shared NAS location for scripts which we use per RAC system so at the end we do not end up with so many scripts.  As RMAN script by default should be maintained by DBAs, I see no issue with that being set on client side. 

I'm not sure what you refer to wen you say multiple instances - I know that since 1.5 you can place multiple instances into CFG file and that is quite straightforward (and you call it via multiple save set lines if I remember correctly). We also do not use concept of standby databases - with RAC in use we see no use of such concept.