God damn it, I was just making fun of Poast for going down for maintenanceAlright here's the deal. I've been banging around on the ol' server in prep for migration. Seems like we got some filesystem corruption on the Pleroma OS partition. Bad news is that the site is down because the filesystem is in read only mode. Good news is that there's nothing all that important on that filesystem and already have backups in as part of migration prep. Bad news is that the system is on a drive without RAID that also shared by the actual hypervisor itself. The drive we're on is a ticking timebomb. I'm going to do some extra backups and then just ignore the filesystem error for now. Migration has been A C C E L E R A T E D.
ok but did you even try turning it off and then back on again?The Pleroma database has been repaired, upgraded and restored, but I forgot to prioritize the transfer of the Pleroma media files. I'll start the instance up they finish. I did start it up for a second before I realized all the files were missing, but it looks like the DB wasn't catastrophically broken after everything.
In the mean time the git and gemini services have been restored. Synapse database is being re-imported from a psql file I created just before starting an upgrade on the old server, which was happening when the Pleroma filesystem went read-only.
I severely underestimated how large our files are, Josh's proxmox configuration with ZFS has compression, I guess. Trying to pull TBs worth of data off a RAID 1 HDD pair with compressed thinly provisioned ZFS partitions is not fun.
What are the odds of you embarking on a blogger style website? I'd pay for a based blog where I can post about my niche interests without having to make threads