I still have not figured out why but it seemed like there was a spike in write timeouts for some period.

I’m slowly switching over everything to NFDB2.1 and a new architecture

If you want to be on the safe side, reimport your events at Nostr.land