We have been radio silent for a while, and for good reason. We've been working on a new website for the past few weeks and we didn't want to make any new announcements until it was completed but now that it is, here we are!
New Website
We had recently hired a developer to design a new website for us that better exhibited our company where our past pre-made template designs could not. We enlisted the services of a few people, both with web hosting experience and without, to review our previous website and critique it so we knew what we needed to change. Overall reviews were that we had to much information (I know, more is better than not enough) and we didn't organize it in an easy to digest package. Another big complaint was that our website wasn't mobile/tablet friendly, something we hadn't considered until we started to research how many people utilize their phones these days. So last week, we silently rolled out our new website and waited until we had some other things in place before we made this announcement. Our new website is LIVE!
We've already received a few good reviews on it from guests and clients both so we're very happy with the new design and how streamlined it is compared to the previous one. We also made some edits to Wyvern and our Client Area to better fit the website by removing the different packages per location (i.e. FLO128, COO128, CAO128, etc... have been replaced with just O128 and your ability to select the location you want to deploy to when ordering).
So please, if you haven't checked out the new website yet, please do so and let us know what you think about it and if there's anything you miss about the old one or any information we can add that would be useful for our clients.
New Wyvern Feature
One of the most requested features for Wyvern was the ability to migrate your VPS on your own without any staff intervention, and after having the feature in testing for the good part of July, we are happy to announce that this feature is now enabled for all of our OpenVZ VPSs in Wyvern (this feature is disabled for our DDOS Protected and Backup VPSs since there are no other nodes they can migrate to). You can check out some screenshots here.
We have set a hard limit that VPSs can only be migrated once per day so please do not open a ticket to have your VPS migrated manually or it will anger the dragon. Since this process is completely automated we still ask that all clients make a backup of their data and save it off-site before migrating just as a precaution since there are no guarantees when data leaves our networks and travels hundreds or thousands of miles.
Lastly, if you are considering moving your VPS to another location, we highly recommend going to Drgn.biz and test the networks for latency, check the routing for the networks, and using the test files available to test the network speed from the locations to your home, work, or other servers.
New cPanel Resources
Since our cPanel services have received a lot of attention this year, we're taking steps to improve this service even more so clients get the best value and allow us to expand this service more in the future. This past week, all of our current cPanel services have had their disk space increased to the following:
- Entry: 2GB -> 5GB
- Standard: 5GB -> 10GB
- Advanced: 10GB -> 20GB
OpenVZ OS Templates
For those of you that read our announcements or follow us on Twitter, this won't be news for you but we figured we'd give it some room in this e-mail so everybody else will know. We've been updating our OpenVZ OS Templates regularly and with one of the recent updates we've replaced a lot of the old templates with the minimal version which uses less disk space and RAM so now our clients with less than 256MB of RAM have more options to choose from. Prior to this update, we could only recommend our custom Debian template for anything with less than 96MB of RAM, now we're glad to say that any template with the word "minimal" in it will function properly on a 32MB VPS (even though we no longer offer these plans we still have a lot of clients out there using them).
We've also remove Ubuntu 13.10 as it is now End Of Life (EOL) and added a stable template for CentOS 7 (also minimal, but only available in 64bit) for anybody that wants to play with the latest and greatest from CentOS.
Kernel Patching
Unfortunately this isn't all smiley face worthy news, we wish it was but we have to address something that seems to be a thorn in our side that can never be removed and that is kernel updates.
Since June, we've had 4 emergency reboots to all of our OpenVZ nodes due to kernel updates that address a critical exploit that can impact all of our OpenVZ clients. While kernel updates are always a good thing, emergency reboots in the middle of peak traffic times are not. We've considered those "rebootless kernel update" services, but on 2 of the 4 past occasions the release of a fix from these services has been hours after the new kernel update was available so our options would have been to either reboot anyways or go unpatched longer to wait for an update. A problem that we face that other providers might not is that before we push a kernel update to our VPS nodes, we test the kernel on 2 different test servers with our latest Wyvern code to make sure there are no adverse affects from the update which increases the time our production nodes are vulnerable. If we had to wait an additional 6 hours for the "rebootless patch" then that's an extra 6 hours on top of the time it takes to test the update, deploy the update on each node, and reboot each node individually (we will no longer been rebooting the nodes in groups like we had during a previous update).
Whenever there is a critical security exploit that requires a reboot to patch, we will always post an announcement (which will then appear on our Twitter) prior to rebooting the first server. We never send out an e-mail before or after a critical security exploit has been patched to avoid helping spread the knowledge of such an exploit (thus preventing a client from "testing" the exploit on our servers or other providers) except in certain instances where it is vital that clients are aware of the exploit and the potential impact of it. Even in our announcements, we try to be vague about the details surrounding the exploit to reduce exposure to said exploit.
We cannot apologize enough for such unscheduled reboots because we hate any form of downtime even if it's only 1-2 minutes, but the security of your data is our top priority so if the options were rebooting to secure our nodes or leaving unsecured nodes online to improve uptime then we'll always choose the option with secure nodes. We are Secure Dragon after all (although the name Uptime Dragon does have a nice ring to it).
What does the future hold?
So what does Secure Dragon have planned for the future? Well the same as always, A LOT! We're always discussing how we can make our services better and our To-Do list is getting longer each week. But that doesn't mean we have all the ideas and aren't open to suggestions. If you have any ideas that could make your Secure Dragon experience better, please let us know and at the very least we'll add it to our next meeting's agenda to discuss the idea.
The biggest benefit of being a small company is that we have the freedom to make changes without having to go through lengthy review processes or any red tape. This is why we're constantly looking outside the box for ideas because no matter how crazy it might be, we don't just dismiss it because of complexity or if it's worth it financially (originally we only released a very limited number of our 32MB VPS plans based on requests, now it's been upgraded to our 64MB plan which is easily our best selling product). So to reiterate, please feel free to drop us an e-mail if you have any ideas, suggestions, or concerns that can help us improve our services and your experience. Thank you.
-The Secure Dragon Staff
Thursday, July 24, 2014
