MIDRANGE dot COM Mailing List Archive



Home » MIDRANGE-L » April 2014

RE: comfort level in not performing TAPE backups...



fixed

We use Mimix and for years we've been building our recovery test with one from column A and two from column B--

We do an Option 22 save on each system monthly (OS, IBM libs, IBM part of IFS)

We SAVLIB the Mimixed libraries from the 2nd system daily

We SAVLIB any libraries on the primary system that Mimix doesn't save, plus the IFS.

We SAVLIB any libraries on the 2ndary system not saved elsewhere, plus the IFS

To rebuild the production system, we take the Option 22 from the live system, libraries from the live system, IFS from the live system, and Mimixed libraries from the 2ndary system. More or less. There's always tweaking, but it does make a usable recovery system. We've written procedures to check the Mimix control files so we can automatically determine which system's tape we should be using for which libraries. And then there's always WebSphere (which our programmers are quickly migrating away from-- THANK YOU!).

Going forward, we've migrated to a single system with LPARS using a V7000 for data. We are working towards Full System Flash Copy-- this will let us clone the production system and perform an option 21 full-system save anytime we want. The process works fine, but there's a lot more hand-holding required compared to letting Robot/Save wrangle tapes! The Full System Flash Copy package (from IBM) automates the making of a copy of the system. But that copy needs to be IPLd (for us it takes about an hour-- it's waking up after an abnormal termination), then we can run the Option 21 backup.

The FSFC is where we're headed-- we estimate that we can get an Option 21 completed in about 8 hours (plus 1 more for the IPL). That's right at the edge of a daily Option 21!! So far, we have a 30 page run sheet that covers firing up the copy of the system, re-assigning memory to the partition (which we may be able to skip), then running the Option 21. We can reduce the manual parts with some carefully-placed CLP, but that's still in progress.

We also hope to be able to automate some of the memory reallocation, but that's the fun of trying to talk to the HMC from an iSeries partition.

--Paul E Musselman
PaulMmn@xxxxxxxxxxxxxxxxxxxx


At 5:28 PM -0400 4/24/14, rob@xxxxxxxxx wrote:
Jim,

How do you do a recovery test of a system that does it's backup on it's
Mimix backup?
The BRMS reports of the primary don't really keep that in sync, eh?

Rob Berendt





Return to Archive home page | Return to MIDRANGE.COM home page

This mailing list archive is Copyright 1997-2014 by MIDRANGE dot COM and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available here. If you have questions about this, please contact