Seems like my relay is having some connection issues today. The relay is mostly up and reachable but some might experience connection issues here and there. Will investigate and fix it as soon as possible.
Discussion
Same here, had to tune the prune job to be more aggressive. Haven't had time to really investigate
Which prune job is that?
It's a deno typescript I wrote that expires events from the relay. Basically strfry delete with a timestamp but it selectively keeps things like profiles and NWC events
Cool. Is this script public?
My issue wasn't with disk space though, my issue seems to be nginx hitting worker connections limits.
I shared it here https://habla.news/u/blee@satstack.cloud/smol-relay-01
For worker limits I use:
```
worker_rlimit_nofile 12288;
events {
worker_connections 4096;
}
```