It's been a while, so here's an exciting e-mail to give you a peak into the future of Secure Dragon (and it's an awesome one).
First off, we apologize for the lack of communication (although if you follow
us on Twitter you've seen some nice little tweets about what's in the works this
year). Basically we've spent most of January testing out new servers, RAID
cards, and SSDs (spoiler alert: Samsung 850s out perform Crucial MX100s).
SSD Powered OpenVZ Nodes
So I'm sure you've guessed it already (or maybe you follow our Twitter and
already knew), we're slowly making the switch to 6x SSDs in RAID10 with this new
round of upgrades. Eventually all of our OpenVZ nodes will be replaced with SSD
powered nodes which will also include newer server hardware with Dual 6 Core
Intel Xeons (24 threads total), more and faster RAM, and newer RAID cards.
Our Florida OpenVZ node is the first 100% SSD node installed in our
environment boasting 6x Samsung 850s in RAID10. Clients on our current OpenVZ
node have received an e-mail regarding this migration today.
Our New Jersey OpenVZ node is the next on the list to receive the SSD upgrade
(6x Crucial MX100s in RAID10), but this was mainly due to us switching data
centers and we felt that replacing the hardware was the easiest process to keep
downtime to a minimal. Clients in our NJ location have already received an
e-mail regarding this.
We do not have a schedule in place for our other locations, but when we do
you can expect an e-mail from us when your services will be affected. We will
try to upgrade the servers in the order of node population but other factors may
be involved. We do not have an ETA on when we expect these upgrades to be
complete or how often we will be upgrading the nodes so please do not contact us
regarding this as we do not have any answers at this time.
As a result of these upgrades, migrations to and from our FL and NJ nodes
have been disabled until the migrations are complete. We have also increased our
pricing for additional disk space from $0.05/GB to $0.20/GB due to the much
higher costs of SSDs.
We recently made an announcement regarding this, but I wanted to touch on it
here so everybody is aware. As of January 22nd, 2015 all VPSs come standard with
1Gbps ports (excluding our Atlanta location, but there is a ticket open to fix
this). You can view the announcement here for the specifics (including our AUP
New Wyvern Alerting
The 22nd was a busy day for us, another announcement that was made was that
we've added some alerting to Wyvern to help clients keep closer tabs on their
VPS activity. Now Wyvern will open a ticket when a VPS is either rebooted or
throttled by our automated systems. Here's a link to the announcement with more
Wyvern Security Audit
With all of the updates we've been pushing, we wanted to make sure Wyvern was
as secure as possible for our clients and their data. On January 3rd, 2015 we
had an external security audit performed on the Wyvern code and we had it back
online and fully patched within a few hours. You can read the announcement here:
A big thanks to Vlad C. from SafeOrNot.net
for taking the time to go over all 28,000+ lines of code.
New Jersey Data Center Change
As previously mentioned, we will be migrating our clients out of COLO@'s
Weehawken, NJ location to Constant's Piscataway, NJ Tier 4 data center in the
near future. There are multiple reasons for this switch, but here's a rough
outline of why we are doing this:
1) Our NJ location has been plagued with network outages over the past few
months and our data center has been unable to provide us a reason for these,
sometimes 3+ hour, outages. Uptime is important to us and our clients.
Constant offers some features that will allow us to expand our services in the
future so making this change now will allow us to move forward with these
3) Cheaper bandwidth. Now that we offer 1Gbps ports for
all clients, we expect to use more bandwidth and in order to prevent us from
reducing bandwidth or increasing pricing we needed to find locations with
cheaper bandwidth. While cheaper usually means lesser quality, Constant offers a
much better blend of bandwidth so hopefully the network performance will remain
the same or be better than COLO@'s.
4) In addition to Constant's 100% uptime
guarantee, they also offer a 0% packet loss guarantee which is just as important
as being online all of the time.
5) Lastly, we've been trying to diversify
our infrastructure so we're not overly reliant on one specific provider.
We have received some comments and concerns regarding the quality of COLO@'s
services and we would like to ask that you do not interpret this migration away
from COLO@ to reflect negatively on them as a company. We plan to continue using
their services in other locations for the foreseeable future.
We've given Drgn.biz an update so it's easier to use and provides more
information about our infrastructure so clients have a single site they can
visit for quick stats about us. So now you can find test IPs and test files, run
network tests, view our latest Twitter update, and see our server statuses all
from one website. Visit http://drgn.biz today and
check it out.
2015 is shaping up to be a great year already and hopefully we can continue
providing the service you expect from us. As always, we're working to improve
your experience in any way that we can so please feel free to contact us if you
wish to discuss things we are doing and how we can do them better.
-The Secure Dragon Staff-
Home » dedicated server » KVM » maintenance » OpenVZ » Secure Dragon » Secure Dragon LLC: An update on 2015 and where we're going this year.