Shredder takes forever (advice please!)

RT Users
1

Hi all

We have just north of 375,000 tickets in need of a good shredding. There’s a lot of advice around here that centers mostly around building database indices, however even with that we are still only able to shred 30 tickets in a 24 hour period. Bare in mind, we made 90 new tickets in this queue yesterday alone.

I was wondering if anyone has any advice for shredding at this kind of scale? The thing that seems to take the most time is Transactions and Attachments - everything else gets shredded almost immediately.

My only thought is to start looking at ways to parallelize shredder with a list of tickets distributed to x workers, but I am hoping that is only a last resort, nuclear option.

Thanks for any advice you can give

Luna

2

You added the indexes from the docs here?

3

Yep, those are all in.

4

Are there any other options/solutions to try?

5

Im surprised it is taking that long, do these tickets have a huge amount of transactions on them?

6

Indeed… dozens. Attachments and Transactions seem to be the bulk of the delete process.

Our setup uses RT tickets to track actions on some automated pipeline work, hence the mass amount of Transactions. I suppose nobody thought to clean these up until recently.