A while back I wrote a post called “Why I blog” which spoke about how interaction with the community is part of the free reward you get from putting your ideas out there on the interwebs.
People make comments on what works for them, what doesn’t, how your code might be better etc etc. It’s basically free QA on your thought processes.
But there’s another reason why it is good to blog, and why, even when your content grows old and starts to the wither on the vine, that you keep it.
A couple of days ago, a customer reached out to Oracle via the standard channels. Turns out they had a Oracle 7.1.6 (!!!!) database that they would like to get onto 19c.
So what are the options here?
You will be unsurprised to learn that there is no direct upgrade path from 7.1.6 to 19c. I suspect this might have to do with the fact that the two releases are 30 years apart!
DataPump is not an option because … well … DataPump didn’t exist until version 10g, which means if you want to unload/reload your 7.1.6 database you will need to use the traditional export and import commands.
At which point I was thinking … If only there was some content out there on the internet which would guide me through maximising the throughput of traditional exp/imp commands, and lo and behold I stumbled across this post I made on my previous blogging site back in 2003 !
Some of the advice on the import side of things would no longer be relevant for a 19c database, but the approach of multiple exports, the use of direct mode, and the recordlength buffer size will all be valid strategies for this customer as they move away from version 7.1.6.
Moral of the story: Start a blog. Blog often. And don’t throw away your old blog content. You never know when it might be useful for someone out there, and that someone might be yourself.
Leave a Reply