After some routine checks, and reported issues on the systems in question, we have been made aware of some failing hard drives in node 1 and node 5 Canada.
Within the next couple of hours, we will be having the NOC technicians replace these drives. This may mean up to 30 minutes of downtime per system.
Just to note, because we use a RAID1 setup in all of our systems, no data has been lost, and we do not anticipate any data loss on the systems in question.
Thank you for your continued support and patience.
Update: Unfortunately, our datacenter in Canada was unable to comply with our request in a timely manner, which resulted in a couple of hours of downtime without the replacement. We will be rescheduling this period to have the drives replaced.
Our USA datacenter (PhoenixNAP) has just informed us of a planned network maintenance to replace some bad networking hardware on January 31st, 2013 at 2:00 AM until 5:00 AM CST.
During this time period, an outage of up to 45 minutes is to be expected. The Network Engineers will be taking as many precautions as they can to prevent any extreme downtime.
Earlier this morning we became aware of a restart on Node 23 France. We are currently diagnosing any possible hardware-related issues, and will be restarting the Multicraft daemon moments after determining that the node is safe for use.
Update as of 13:26 CST: The Multicraft daemon has been started, and Minecraft servers are now starting.
Around 17:15 CST, we became aware of a network backbone connectivity problem between our Roubaix, France datacenter and Newark, New Jersey, USA. This problem will most likely affect only players from USA who are connecting to France-based servers. We apologize for the inconvenience and will update this post when more information becomes available.
As of 20:45 CST, all issues seem to be resolved. If you still have connectivity problems, please open a support ticket.
It appears our France Multicraft is getting hit with a DDoS, large enough to mess with our virtualization software. Generally, this is a bad idea, and should’ve been replaced a long time ago, but we had not gotten to it yet due to many other projects being worked on.
We are currently working to get a new Multicraft panel online for the France location while this DDoS is ongoing.
As we receive and setup a new box for the panel, we will update this posting.
Update: We’ve received the new box, and we’re setting up the HTTP software on it.
Update 2: We’ll be adding another IP to this box in order to prevent Hamachi conflicts, however the panel will now be accessible from http://18.104.22.168/
Update 3: New IP assigned, DNS record changed. The Multicraft link for France should be working in the next 5-10 minutes.
Earlier this morning we learned of a possible hardware failure on Node 21 France. After diagnosing the issue and determining that the on-board RAM had gone bad, we put in a request to swap it out and get the server back online as soon as possible.
All went as planned and at this time all services have been restored.
Thank you for your patience!
Over the past month or so we have begun having issues with Amazon Simple Pay relating to our SSL certificate. Currently if you have been paying your invoices with this service, we suggest to switch to PayPal or Google Checkout as they seem to be much more reliable solutions, at least for the time being.
If you have recently paid your server through Amazon and have not received a payment confirmation, please submit a support ticket with your Transaction ID from the Amazon Simple Pay website.
In the meantime, we apologize for the inconvenience and we’re working to get everything back to normal; however, we can’t say for sure if we’ll continue to use Amazon in the future.
At approximately 8:30pm CST, our Arizona USA network began experiencing heavy traffic most likely originating from a DDoS attack. Our Arbor Networks firewall will activate within minutes to mitigate the attack and we expect to have full access to all affected services within an hour. All Minecraft servers on the affects systems may experience lag and/or downtime while this takes place. We apologize for the inconvenience.
Note: This notice excludes Node 30 USA.
Update @ 9:35pm CST: It appears the attack has stopped and all services should be back to normal.
Please do not submit a support ticket about this problem.
Early today we discovered some issues appearing on Node 5 USA (node5.az.us.mc.cubedhost.org) which was causing lag and some servers to go completely offline. As we investigated, we were forced to restart the node and soon realized that one of the hard drives in the system had failed. Due to the fact that we utilize RAID on every node, we did not lose any data but the drive must be replaced before we can restart the server.
Update @ 23:30 CST: Our datacenter technicians have scheduled the swap at midnight – the node should be online within the next hour.
Update @ 0:40 CST: Awaiting technicians to finish the swap.
Update @ 0:55 CST: The drive has successfully been replaced and all servers should be online again!
All customers will be credited the appropriate time on their billing period for this inconvenience.
This is the new CubedHost blog!
Here we’ll keep you updated on our latest additions, downtime, maintenance, upgrades, adventures and so much more. For support, please submit a support ticket in our billing area, and check our Twitter for real-time updates!
Have fun and check back often!