Hi Mark!
Thanks a lot for the TR! Unfortunately it does not go very deep into B/R, so I have asked Paul Hargreaves, one of the TR authors, about his opinion. Here is his answer:
From: Hargreaves, Paul
Subject: Re: Question on tr-3132
Wow, I remember that – a long time ago. We didn’t write a separate TR mainly because at the time we didn’t have enough customer interest to write a full paper and for those customers who did start using it it was trivially easy to integrate the snapshots. I know we don’t have a huge amount of knowledge internally, for example apart from the TR the last time I got involved with Progress was about 10 years ago. Fortunately Progress is a really simple environment to work with and hasn’t changed much over the years.
The basics were pretty much:
1. Follow the Progress manual to proquiet the database. That stops all write I/O from progress to the database files.
2. Use SnapDrive to take a snapshot of the volume(s) that hold the database.
3. Use proquiet to thaw the database so that clients can have disk i/o again.
I.e. 3-4 lines of script were sufficient. Progress added the proquiet ability to enable raid-1 vendors to take hot splits of the database in a more consistent manor and we just piggybacked on that technology.
What the above doesn’t cover is any roll-forward capability though. You’ll get point-in-time recovery points. Assuming the customer knows how to use after-imaging then they’ll know what to do.
Paul Hargreaves
EMEA Data ONTAP 8 SE Lead
Kind regards!
Matthias
