It's been a busy day so I will keep this brief. All of our plans were pretty much shot due to multiple issues that occurred while trying to migrate VPSs over to the new nodes. These issues ranged from high loads caused by busy VPSs to kernel panics caused by TUN (similar to what was experienced on FIRE0003). Unfortunately we cannot determine the cause of the kernel panics but we do know that they only occur during live migrations so from here on out will will not perform mass live migrations until the problem is resolved. That being said, we are currently migrating all VPSs over to the new nodes but they will all be experiencing short downtime (less than 1 minute). We apologize but the alternative is 10 minutes of downtime while we power cycle the node and wait for it to run a fsck and rescan each VPS to build the quota details from scratch.
As for GOLD0001's replacement. Due to time constraints and hardware failure, we have decided to order a brand new replacement server which is expected to arrive next Tuesday or Wednesday. Our current GOLD0001 will be scrapped (all 16TB of it) and we do not plan on looking back once it's gone.
Thank you for your patience in this matter and your understanding.
-The Secure Dragon Staff
Friday, November 25, 2011
