RE: Backup sanity check -- MIDRANGE-L

We have a similar situation.
Using BRMS, (cannot combine multiple LPARs on one volume with BRMS).
Production we do a full SWA every night.
R&D, full SWA on critical libraries, source, test config libraries, etc, incremental on rest, all test data omitted, large omit list.

20 *SAVSECDTA FFFFFFF
30 *SAVCFG FFFFFFF
40 ICOMMDL *OBJ *SYSBAS FFFFFFF
50 BRCPTDMDL *OBJ *SYSBAS FFFFFFF
60 PENSYSMDL *OBJ *SYSBAS FFFFFFF
70 PTDMDL *OBJ *SYSBAS FFFFFFF
80 UPPGMSUN *OBJ *SYSBAS FFFFFFF
90 UPICOMPGM *OBJ *SYSBAS FFFFFFF
100 UPPTM *OBJ *SYSBAS FFFFFFF
110 D1PGMFILU *OBJ *SYSBAS FFFFFFF
120 OUTQSLCCFG *OBJ *SYSBAS FFFFFFF
130 CMFILES *SYSBAS FFFFFFF
140 QUSRSYS *SYSBAS FFFFFFF
150 *ALLUSR *SYSBAS IIIIIII
160 *ALLDLO IIIIIII

Paul

-----Original Message-----
From: MIDRANGE-L [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of rob@xxxxxxxxx
Sent: Tuesday, January 19, 2016 8:56 AM
To: Midrange Systems Technical Discussion
Subject: Re: Backup sanity check

If that works, then that is perfectly acceptable.

In our world it would be a tape library, not a device.
The tape library would be hosted by VIOS.
It would then be guested out to the lpars of IBM i.
Each guested lpar wouldn't need to worry about varying it off/on.
Depending on how many drives the library had you might be able to run the backups at the same time. Otherwise, you'd just stagger the timing.

Now, if the first lpar takes too long, or can only be down from x to y and that also happens to be the only time that the second lpar can be down then you might consider creating a virtual tape and backup to that and get that moved over to the other lpar or something for the movement to physical tape. But newer tape drives are often faster than disk. Mainly because you're not doing the reading and writing to the same disk drives.

Rob Berendt
--
IBM Certified System Administrator - IBM i 6.1 Group Dekko Dept 1600 Mail to: 2505 Dekko Drive
Garrett, IN 46738
Ship to: Dock 108
6928N 400E
Kendallville, IN 46755
http://www.dekko.com

From: Richard Reeve <rjrjr64@xxxxxxxxx>
To: Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxx>
Date: 01/19/2016 08:37 AM
Subject: Backup sanity check
Sent by: "MIDRANGE-L" <midrange-l-bounces@xxxxxxxxxxxx>

All,

I recently took on a new client that is not backing up the test LPAR
at all. Really, the data isn't important, but the programs and source
files are.

The production LPAR is backed up nightly using SAVLIB to tape. My
thought was to change ENDOPT to *LEAVE for he last production library that
is saved. Then just vary off the tape drive from the production LPAR,
vary the tape drive on in the TEST LPAR, then use SAVOBJ to save the
individual source files from test to the end of the production backup.

Does this sound like a good practice or does some one have a better
idea?

As always thanks for all of you knowledge and wisdom.

Rich