EDIT Aug 2, 1:53 PM US Mountain Time: Maintenance has now concluded and all services are back online!
We’re preparing the server rack for an additional Proxmox node to be added to support HA failover and have identified the need to offline and move some equipment around to complete this work.
As such, we’ll have a temporary downtime for up to 1 hour from 1 PM US Mountain Time, today August 2nd, or 1 hour from this post. Though, we expect the work to update the node, move it, and bring it back online to take no longer than 30 minutes.
We’re hoping these changes will help us maintain a more stable service going forward, spread out the Sidekiq processes (the ones that receive / send data across the Fediverse), and allow us to perform maintenance as necessary without having to offline the service in the future.
We apologize for the short notice as we had hoped to perform this maintenance without any downtime.
Maintenance window: 1 PM - 2 PM US Mountain Time, August 2nd, 2023
ETA: 30 minutes to 1 hour
During the maintenance window, the following services will be unavailable:
- pawb.fun
- pawb.social
- furry.engineer
We’ll have updates throughout the maintenance window available at https://t.me/pawbsocial.
I would love a post on the pawb infrastructure. I’m super curious about what hardware, back-end software, what type of facility the server is in, and the costs of running it. Although, I think that type of post would be a super security concern.
Amusingly not really. It runs in our basement and @crashdoom posted a few pics!
Software wise: mostly just a linux container in proxmox.
Lots of hardware info in the post! Crashdoom handles the software, i am just hardware horse :P