Hey there! 👋
It looks like you may be coming from an area of our website that is affected by a maintenance. We do apologize for any inconvenience caused by this!
Our team is currently working to upgrade the internal MySQL services powering our Prisma control panel, billing system, and so forth -- essentially, the entire core of our company. Due to the changes necessary, unfortunately, there will be more than just a few moments of interruption that would not normally require a maintenance window.
Why was I directed here?
In order to ensure the best experience possible, we've disabled certain parts of our website so that you will not experience constant errors while managing your server. Your server will remain online through this maintenance. However, scheduled tasks, restarts, etc., will be sporadically available while the maintenance is completed.
To stay up to date on the status of the maintenance, please see our status page: https://status.cubed.host/incidents/8gsqsblwmgf6
What's happening, and what's the benefit to me?
We're currently developing a solution internally to expand out our Prisma control panel PoPs ("Points of Presence"), which should overall decrease the latency to the control panel and overall improve your experience while managing your server.
Currently, our infrastructure for the control panel is available in:
- Ashburn, VA
- Dublin, Ireland
- Sydney, AU
With these changes and preparation being made, we will be working to cover all of the regions that we're currently hosting servers out of. These 3 locations work well, however, initial connection times and loading data may be less than optimal. We're fixing that.
Once the backend systems support the changes being made, we'll work to cover US west (potential location(s): California, Oregon), US central (no current potential locations), EU (potential location(s): London, Frankfurt), and APAC (potential location(s): Singapore, Tokyo).
Of course, this is currently all in the planning stage and all final locations will be listed on our status page once we've gone live with the backend changes necessary to support the database changes we're making internally.
How long is the downtime anticipated to be?
We expect no more than 30 minutes of service interruption, but it may last up to 4 hours should anything go wrong. As of 10PM CST, we will begin the maintenance that will interrupt normal management of services and will redirect our control panel to this Help Center article.
Plan for failure and be pleasantly surprised when things go smoothly.
Any technical details that can be shared?
If you're asking this, you'll likely recognize we're utilizing Amazon Web Services, or the AWS Cloud, for our web infrastructure based on some of the locations mentioned. We're primarily based out of the US East availability zones using RDS and their Multi-AZ functionality.
We've previously setup intra-VPC communication so as to prevent downtime converting our infrastructure over, but the time has come for us to correct this decision and migrate the RDS infrastructure appropriately. Once done, we'll be able to easily spin up read replicas in additional regions to decrease the overall load times for operations that solely require information to be read from the database.
If you're interested in more technical details (we'd release this as a blog post), give us a shout! While we won't go into 100% specifics, we do try our best to be as transparent as possible with our customers. Just click the button in the bottom right of the page and let us know. 😄
Thank you for your patience during this maintenance!