Question

I have a complex context index that gets synchronized nightly. The process takes some 10 minutes, and any updates to this table that touch the index column during synchronization period result in ORA-29861: domain index is marked LOADING/FAILED/UNUSABLE exception – what can I do about it?

Answer 1

I don’t think you can do anything else while that thing is synchronizing. If the index update is unavoidable then you’ll need to find a way to queue up the requests and retry say every 15 minutes till it/they go through. I would suggest a limit of 3 tries until it goes through or fail gracefully. I make the suggestion of 3 times because if it is supposed to take 10 minutes and it doesn’t work in 45 minutes, you have bigger fish to fry I would think. Might as well make it fail gracefully than endlessly repeating on a broken system. Hopefully you don’t have so many attempted hits to the database during that period that you end up with a big queue. You could also see if you can synchronize your app with the times your infra people set up for updating indexes. So that you block these transactions at the same time. I don’t know how big your organization is or what the systems are like (if you are running Oracle you guys have enough money for something sizable). That means you might have scheduling apps that might help? In any case, unless the DBAs stop doing these updates, you’ll have to wait till they finish I would think.