We are running nw_7_3_2_jumbo.Build.11 on Solaris, and we perform
manual rman backups using the nmo module.

We have an oracle database instance ora22 running on host sol548. We
want the scheduled sol548 filesystem backups to be associated with
the client name sol548 and retained for a year, but want the manual
rman backups associated with the client name ora22 and retained for
just two weeks.

Here's the current config (extra lines removed):

              type: NSR client;
              name: sol548;
            server: solbkp01;
         client id: 188ba2b5-00000004-45f98b10-45f9c897-0021000-000000;
  scheduled backup: [Enabled]   Disabled ;
     browse policy:  2 weeks    Day    Decade   [Month]   Quarter
                     Week    Year ;
  retention policy:  2 weeks    Day    Decade    Month    Quarter
                     Week   [Year];
          save set: /, /usr, /var;
     storage nodes: sol548, nsrserverhost;


              type: NSR client;
              name: ora22;
            server: solbkp01;
         client id: 2c68df69-00000004-46251571-4628b1e1-06750000-888s4a9;
  scheduled backup:  Enabled   [Disabled];
     browse policy: [2 weeks]   Day    Decade    Month    Quarter
                     Week    Year ;
  retention policy: [2 weeks]   Day    Decade    Month    Quarter
                     Week    Year ;
          save set: All;
          priority: 500;
     remote access: solbkp01,root@sol548,oracle@sol548;
     storage nodes: sol548, nsrserverhost;

The rman backups are initiated via an oracle cron and include the
following stanza:

 send 'NSR_ENV=(NSR_SERVER=solbkp01,NSR_DATA_VOLUME_POOL=DBBACKUP, \
          NSR_CLIENT=ora22, NSR_GROUP=RMAN_2_WEEKS)';

All this seems to work fine:

 o scheduled sol548 filesystem backups are written to the normal
   pool, and savesets are associated with the client sol548.

 o manual rman backups of ora22 are initated on sol548, are written
   to the DBBACKUP pool, and savesets are associated with the client
   name ora22.

However, we have two problems:

 1) EMC support tells us this configuration is unsupported because
    there is no unique IP address (neither on sol548, DNS, nor in
    /etc/hosts) associated with the virtual Networker client name
    "ora22". We would like to believe that because we are using manual
    backups, this is irrelevant (i.e. if we had 100 oracle instances
    running on sol548, would each instance require a separate ip
    address just for backups?) We hope that we should just be able to
    do "save -c ora22 ..." from sol548, and Networker should oblige
    so long as sol548 is on the ora22 remote access list.

 2) The node sol548 is also a _dedicated_ storage node with
    san-attached tape drives, but the manual ora22 backups cannot use
    sol548 as a storage node. Even though the manual rman backups
    are initiated on sol548 (i.e. the same physical host), Networker
    complains that the virtual client ora22 cannot use sol548 as a
    storage node:

     Error: storage node contains of remote storage node sol548

    I would have hoped that networker would realize that the save of
    nsr client ora22 is in fact running on sol548, but apparently
    not. EMC has temporarily given us a "normal" storage node license,
    but we would like a more permanent solution.

Any suggestions are welcome.

Cheers,

Paul

--
Paul Robertson
Email: probertson AT unixway DOT com
Phone: 617-784-1575
GnuPG Public Key: http://unixway.com/misc/pgpkey.txt

To sign off this list, send email to listserv AT listserv.temple DOT edu and type 
"signoff networker" in the body of the email. Please write to networker-request 
AT listserv.temple DOT edu if you have any problems with this list. You can access the 
archives at http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER