JosefAssadERST

I agree this is very high value. One detail which makes me a bit leery is, how do we test this? Like in the case of `make kind-up`. Should there...

Well. Not sure about optional, how about graceful failure? It's really common to have self-service account reset, and it's not our fault if someone deployed RT but didn't set SMTP...

[This](https://last9.io/blog/how-to-manage-high-cardinality-metrics-in-prometheus/) has a section titled "Find High Cardinality Metrics" which looks to me like a good place to start, i.e. figure out where we actually are. In fact you might...

Agree @LookACastle sane way to try the concept out.

Apologies for being pedantic and making the design challenge worse, but should RT core really consider the operational needs of ITPs? What I mean is, of course currently our two...

Yeah I think we discussed before and it was hard for some reason and that's entirely fair. I wanted to note the paper cut though so in the worst case...

> which can perform bad under higher load Has this actually been the case? I'm not against caching at all, but I typically prefer to try as many other things...

By the way if this really is such a big performnce problem, another way to do it without running into risk of race conditions and fiddling with TTL is for...

Bear in mind, I don't think you should do this unless the DB calls objectively are a problem. I'm proposing a cache here. That has a high price in debuggability;...

> Wrapper code shouldn't have any third-party dependency Difficult but clean. > a certain Walking Cat Why is Catwalk being developed as a RT job? > look into vendoring Pydantic...