I’m working on a single-instance deploy, and I’m kinda surprised that that my only effective options are to either use the data store or redis. Redis seems overkill, as noted here Is that fine if my Redis stores only 5.48 Mb of cache? , but now I’m gonna be having needless overhead in using the data store
A comment here 🚀: Use sqlite in ":memory:" mode when selected for an in-memory store · Issue #1516 · stalwartlabs/stalwart · GitHub mentions a planned “built-in” in-memory store, which does seem appropriate, but I haven’t seen anything else here mentioning it. Is that still a planned feature?
If you guys are open to it, I could take a crack at it myself (I’m not a vibecoder, I promise. I actually have attention-to-detail and I’m not allergic to actually understanding things)