It seems that lemmy.one is still down.

Remember a lot of even the largest lemmy instances are hosted on a single vps. No redundancy, no scaling, and the admins barely know what they’re doing. Its volunteer work done as cheaply as possible.

My understanding (from running an instance) is that this isn't something cloudflare is going to help with, it's the database on these instances locking up. It happens to my and multiple other instances too, and i have monitoring from multiple sources so I get notified and can restart the server/docker containers. Improvements should come in the next lemmy versions.