LowEndBox - Cheap VPS, Hosting and Dedicated Server Deals

SecureDragon - $24/Year 128MB Xen VPS in Jacksonville

SecureDragon KuJoe from SecureDragon emailed me a few end-of-year specials to me yesterday. Some deals have repeated — for example $14/year 96MB OpenVZ VPS and $18/year 128MB OpenVZ VPS. The highlight, however, would be this $24/year Xen VPS offer when you get their “X128” plan with coupon code XEN4US. Direct signup link.

  • 128MB memory/256MB swap
  • 7GB storage
  • 500GB/month data transfer on 100Mbps
  • 2x vCPU cores
  • Xen/SolusVM

You can pay by either PayPal or Google Checkout. Servers with GoRack in Jacksonville FL (test IP: If you are on LowEndTalk I guess Joe needs no introduction. They have been around for more than a year now and runs a solid shop there.

Latest posts by LEA (see all)


  1. safefaek:

    yes?Test IP?

    December 28, 2011 @ 12:31 pm | Reply
  2. ipv6?

    December 28, 2011 @ 12:46 pm | Reply
    • Not at this time. We have plenty of IPv6 addresses available but the firewall we want to use does not support it yet (Untangle).

      December 28, 2011 @ 1:15 pm | Reply
      • Change the firewall? Any ideas when the firewall will support it?

        December 28, 2011 @ 1:20 pm | Reply
        • The only other firewall we’ve considered is pfSense but after 2 weeks of testing it we had nothing but issues from day one and never even put it into production. We had Untangle in production for a few days but then we found out the 1Gbps NICs we had in our server weren’t capable of exceeding 200Mbps so we had to return the server and never got around to replacing it since we found a software-based solution on the nodes themselves to mitigate the DOS attacks we were receiving at the time. We’re still evaluating other firewalls but until then we don’t want to offer IPv6 just to possibly disable it later.

          I just checked and according to Untangle’s wiki, IPv6 prcoessing is expected late 2012 so maybe we will look around for another option now.

          December 28, 2011 @ 1:49 pm | Reply
      • eric clope:

        Why don’t you use FreeBSD with PF ?

        January 5, 2012 @ 7:48 am | Reply
        • We tried pfSense for almost 3 weeks before we gave up on it. We were never able to get it to work with our network setup. It was a shame since I really liked pfSense a lot.

          January 5, 2012 @ 5:15 pm | Reply
  3. marrco:

    I still don’t get the need/urgency for IPv6. Can someone tell me the real actual usage for a IPv6 address now? Not for a webserver/mailserver nor for most (all?) of legit usage that i can think of. Some friends said that if i don’t know i don’t need it, but hey, I don’t want to to share the same node with them, i’m not that leet.

    (about pfsense i don’t know how it fit for a vps hoster, but I use it everyday with some large customer and it’s rock solid and reliable)

    December 28, 2011 @ 3:08 pm | Reply
    • Spirit:

      IPv6… nor for most (all?) of legit usage that you can think of? So IPv6 users are mostly abusers? :) Think further.. Or simply click on Dirk name (link) above to get one reason. People which use IRC (more connections, vhosts, etc..) can give you another, etc..

      December 28, 2011 @ 3:19 pm | Reply
      • Spirit is correct, there are a lot of legit reasons. To my knowledge the most prominent reason is IRC which we do not allow so luckily the demand for IPv6 hasn’t been very high for us. :)

        December 28, 2011 @ 3:30 pm | Reply
      • marrco:

        still don’t get what you can do with an IPv6 that you can’t with an IPv4, as I don’t get why you can’t chat with your friends just using your home dsl connection.

        I see it as a protocol i will have in future to manage at $dayjob. But i don’t think everybody asking for IPv6-enabled VPS just need to pass a certification test. So i’m still wondering what is it actually used for.

        But i guess this is argument for LET, so getting back on topic, i have both a xen and a openvz VPS with securedragon and really work great for me. Fast, good uptime and support. A few more info here: http://www.lowendtalk.com/discussion/889/best-startup-of-2011

        December 28, 2011 @ 4:14 pm | Reply
        • Most (if not all) you can do with ipv4 you can do with ipv6 too.
          Currently most system are talking ipv4, but more and more and starting to talk ipv6 too. More and more ISPs are deploying ipv6. So why not use the native ipv6 when it’s available?
          In my business we want be sure to give our users the possbility to use ipv4 AND ipv6 (not just one of them).

          December 28, 2011 @ 5:46 pm | Reply
        • The main reason why IPv6 is so important is that the whole world is quickly running out of IPv4 addresses. This is why many hosts here (Evorack included) take IPv6 so seriously – it is simply the future of the internet, and VPSes provide an excellent “training ground” to get to grips with it.

          Even though most good hosts here have plenty of IPv4 addresses available for the foreseeable future (Evorack included), IPv6 is still very important. I feel that by providing Evorack customers with IPv6, we are doing our little bit to help prepare and encourage the world to move forward with it :)

          Also, IPv6 is a personal passion of mine so I’m always willing to try and answer IPv6 specific questions :) So ask away!


          December 28, 2011 @ 8:39 pm | Reply
  4. Nice deal. I was looking to use it for DNS but securedragon keeps going to a 404 ;)

    Not Found

    The requested URL /apache2-default/ was not found on this server.

    Apache/2.2.9 (Debian) Server at securedragon.com Port 80

    December 28, 2011 @ 4:53 pm | Reply
    • NickM:

      That *could* be because it’s securedragon.net, not .com :)

      December 28, 2011 @ 10:58 pm | Reply
      • Correct. And we don’t run Apache so that 404 error is definitely not our site. :)

        December 29, 2011 @ 12:00 am | Reply
  5. Mike Hat:

    Any chance of the OpenVZ offers being brought back with the 500GB bandwidth limit? I see you guys offered them with 200GB in the past.

    December 29, 2011 @ 2:19 am | Reply
    • The O128 plan comes with 500GB of bandwidth already.

      December 29, 2011 @ 2:55 am | Reply
      • Mike Hat:


        Was interested in this offer but with 500GB bandwidth. I don’t expect to use anywhere near that much but it would be nice for the occasional busy month or another Reddit feature.

        Do you guys plan on bringing this deal back anytime soon?

        December 29, 2011 @ 8:26 am | Reply
        • Now I’m confused. It’s the same deal as now except with 500GB of bandwidth instead of 200GB…

          December 29, 2011 @ 8:31 am | Reply
        • Mike Hat:

          The offer I linked to was for $18/year, OpenVZ, and 200GB. I was basically just wondering if you guys were planning on bringing that offer back with a slightly higher bandwidth allocation?

          December 29, 2011 @ 8:34 am | Reply
        • Oh, I see the confusion now. Yes, the $14/year and $18/year plan is available with the specs listed on our site. This link should clear up the confusion: https://securedragon.net/announcements.php?id=103

          December 29, 2011 @ 8:40 am | Reply
        • Mike Hat:

          Thanks – it sure does.

          December 29, 2011 @ 9:36 am | Reply
  6. jlntlanyl:

    promo code expired? been getting the
    “# The promotion code entered has already been used # “

    December 30, 2011 @ 5:31 am | Reply
  7. Wow, all of the Xen promotions have been used up.

    December 30, 2011 @ 5:56 am | Reply
    • Blackstorm72:

      No wonder, I own a PV-128 of yours (not from this promotion) and I have to say, it runs our little tiny Teamspeak (non-profit 32 slot… lol) WITH a script checking our dedicated without any problems, and no packetloss (like Burst’s was).

      I will be a long time contender – love this little box.

      December 30, 2011 @ 6:30 am | Reply
      • Mike Hat:

        +1 on these guys kicking ass. I needed a new CDN node for a site that has received heavy traffic from Reddit, so I took advantage of SecureDragon’s OpenVZ offer. Setup was nearly instant even after I forgot to disable my SSH tunnel and got locked out by MaxMind for “Fraud”. Ticket responses have been lightning quick thus far.

        SecureDragon is definitely one of my “go to” providers now.

        December 30, 2011 @ 6:36 am | Reply
      • herbyscrub:

        You can get a NPL 512 slot license for TS3 in about 5 minutes on their website.

        December 31, 2011 @ 5:28 pm | Reply
        • Blackstorm72:


          Yeah I already have a NPL through Triton, this little server doesn’t need it thankfully. On the other one, it does :)

          January 1, 2012 @ 8:18 am | Reply
      • Jack:

        Burst.net’s network @ Scranton is horrible.

        Around Mid-day you get 60% Packetloss every single day.

        January 1, 2012 @ 8:22 am | Reply
        • Blackstorm72:

          Even mid-NIGHT it was causing havoc with us in July. It only took every 15-25 minutes and then… you can no longer understand the person talking and instead you hear static, and then pings go up, and everyone times out, then poof back to normal… rinse and repeat.

          And burst said it was our problem because they couldn’t find anything. Granted, this packetloss was not easy to trace down. But I knew damn well if our France Dedi (no longer in use) could hold no packetloss to our Canadian friends in the west coast then what Burst’s did, we knew it wasn’t us.

          January 1, 2012 @ 8:45 am | Reply
  8. jbarbieri:

    I am very satisfied with SecureDragon as well. Support tickets have been always responded to in a timely manner, and with resolutions.

    Patiently waiting for IPv6, I hope they get it going soon :)

    On their O256 Plan:

    dd if=/dev/zero of=test bs=16k count=16k conv=fdatasync
    16384+0 records in
    16384+0 records out
    268435456 bytes (268 MB) copied, 4.10056 s, 65.5 MB/s
    wget http://cachefly.cachefly.net/100mb.test
    --2011-12-30 16:33:39--  http://cachefly.cachefly.net/100mb.test
    Resolving cachefly.cachefly.net...
    Connecting to cachefly.cachefly.net||:80... connected.
    HTTP request sent, awaiting response... 200 OK
    Length: 104857600 (100M) [application/octet-stream]
    Saving to: `100mb.test'
    100%[=====================================>] 104,857,600 86.1M/s   in 1.2s    
    2011-12-30 16:33:41 (86.1 MB/s) - `100mb.test' saved [104857600/104857600]
    December 30, 2011 @ 4:34 pm | Reply
    • jay:

      hmm.. I thought their on 100mbps?

      January 1, 2012 @ 1:12 pm | Reply
      • It will be in the future but SolusVM disabled the throttling feature. Next week all OpenVZ/Xen VPSs will return to 100Mbps.

        January 1, 2012 @ 6:44 pm | Reply
  9. Special promotion for today only! We have 12 O96 VPS Plans for $12/Year for 2012! Use promotion code 121212 when ordering! (New clients only, 1 per client, limited to 12 sign-ups and only for January 1st, 2012.)

    January 1, 2012 @ 6:32 am | Reply
    • john:

      Why new clients only?
      Existitng royal customers whom have been supporting you don’t deserve to have cheaper price, so only new customers can have this? What is the notion for treating your royal customers like a piece of garbage?
      Or since they already paid for yearly payment, don’t give a damn about them anymore?

      January 1, 2012 @ 6:41 pm | Reply
      • This was my decision and I am sticking by it. I am sorry you feel this way but it would be unfair if we provided a $12/year promotion and somebody who already has a $12/year plan from last year grabbed up 2 since they are so limited.

        January 1, 2012 @ 6:47 pm | Reply
        • I know this will cause some outrage with current and future clients, but I feel I need to post this so everybody can see my reasoning (if it is flawed, please let me know):


          January 1, 2012 @ 7:07 pm | Reply
        • Is John and all his trolling posts u_u

          January 1, 2012 @ 7:21 pm | Reply
        • john:

          ya, I’m trolling, I already cancelled one vps with this guy.
          Happy new customers…this guy will offer $11/yr next month for new customers only…. and $10/yr following month. So funny.
          I have 4 vps with buyvm, according to you, this is not fair, right?

          January 1, 2012 @ 8:22 pm | Reply
        • John leaves out the fact that he had a $12/year VPS…

          January 1, 2012 @ 8:27 pm | Reply
        • john:

          I also had quarterly vps too, don’t leave that out.

          January 1, 2012 @ 8:43 pm | Reply
  10. john:

    wow, because I complained on LEB, you took away my two vps?
    Just take money and run?

    January 1, 2012 @ 8:33 pm | Reply
    • Your abusiveness in your support tickets over the past 6 months is no longer going to be tolerated. You bad-mouth our company every chance you get yet you have no problem purchasing our $12/year promotions AND THEN COMPLAIN ABOUT THEM. You verbally harass us in support tickets (2-3 per month for 6 months) about how horrible of a company we are and expect us to just lay down and take it? We don’t mind clients, ex-clients, or non-clients bad mouthing us in public. We appreciate all forms of criticism and feedback, we draw the line when it becomes a constant attack every time we publish a new special. Taking a look at our past offers here on LEB, you’ll see your name pop-up quite a bit with these attacks.

      I wish you and your future hosting company good luck.

      January 1, 2012 @ 8:46 pm | Reply
      • Hehe, your company and other companies… his posts are just complains and bs :|

        January 1, 2012 @ 8:48 pm | Reply
        • Heinz:

          @john take it as lesson. Don’t criticize your hosts under your real name publicly. It won’t work out. :P

          January 1, 2012 @ 10:57 pm | Reply
        • @Heinz – ironically John did use multiple names here whiling commenting. Somehow he just chose to use “john” here, if it is his real name…

          January 1, 2012 @ 11:06 pm | Reply
        • john:

          do you think Joseph Dougherty is real name?

          January 2, 2012 @ 12:39 am | Reply
        • @Heinz He was terminated due to his support tickets so it doesn’t matter what name he posted under on here.

          @john You are welcome to look up our legal information which is cross-checked by the Florida government and IRS on sunbiz.org.

          January 2, 2012 @ 1:40 am | Reply
  11. john:

    So where is my money, are you going to refund or just flat out steal it.
    Since you banned me, I can’t contact you just losing 2 vps, or stolen, tell me whether you are going to refund or just pocket it, would you? So I can move on?

    January 1, 2012 @ 8:42 pm | Reply
    • According to our Terms of Service:
      All payments made are non-refundable outside of the Money Back Guarantee. In the event you wish to cancel your service, the Secure Dragon staff may provide a full or partial refund at their discretion.


      Accounts suspended or terminated by Secure Dragon will be subject to a Reactivation Fee if eligible to be reactivated and are not able to redeem any SLAs or guarantees.

      January 1, 2012 @ 8:50 pm | Reply
      • john:

        Thanks for clarification, good luck with your biz.
        happy new year…

        January 1, 2012 @ 8:57 pm | Reply
  12. sam:

    @KuJoe: any new plan for adding more stock for the $24/year 128 XEN? i really want one for L2TP, ovz can only support pptp as in buyvm.

    January 3, 2012 @ 5:17 am | Reply
    • Nope. We’ve ordered more servers but they will be for OpenVZ so our Xen stock is too limited to run promotions at this time. :(

      January 3, 2012 @ 6:48 am | Reply
  13. User:

    I have been using SecureDragon for almost a year. Fast and direct answers to tickets. No problems at all. I see this company expanding and competing with BuyVM in the near future.

    January 4, 2012 @ 7:47 pm | Reply
    • Thanks for the confidence. We love hearing our company compared to BuyVM, Hostigations, and RAMHost (all of whom we consider to be the proven elite of the LEB market).

      January 5, 2012 @ 2:46 am | Reply
      • a:

        securedragon sucks compared to buyvm, hostigation and ramhost. Would that work kujoe? of course I was kidding lol. kujoe seems to be a really good guy

        January 5, 2012 @ 8:34 am | Reply
  14. Corey:

    Just got a server from them, rather fast, had to open a ticket due to a hiccup, most likely my fault support was fast aswell.

     100%[===================================================================>] 104,857,600 17.3M/s   in 5.9s 
     dd if=/dev/zero of=test bs=16k count=16k conv=fdatasync
    16384+0 records in
    16384+0 records out
    268435456 bytes (268 MB) copied, 3.02543 s, 88.7 MB/s 
    January 5, 2012 @ 1:33 pm | Reply
    • snape:

      Got one back on the 1st; was fine until a couple hours ago, at which point it started being extremely laggy. Not packet loss, not iowait, just slooooow. Like, Hostrail-on-a-good-day slow:

      root@durp:~# time uname -a
      Linux durp 2.6.18-238.19.1.el5.028stab092.2 #1 SMP Thu Jul 21 19:23:22 MSD 2011 i686 GNU/Linux
      real    0m0.567s
      user    0m0.001s
      sys     0m0.001s

      I’m hoping it’s just dumbasses benchmarking the node abusively…

      January 6, 2012 @ 5:12 am | Reply
      • Sometimes that happens to me o_O no idea, cpu issue maybe?

        January 6, 2012 @ 5:19 am | Reply
      • We apologize for the slowness. In the event you experience slowness please open a ticket so we can look into it immediately. We recently migrated 13 VPSs onto FIRE0002 which is causing a high load, we are migrating them back over to their temporary server while we determine which of them is causing the slowness and how.

        January 6, 2012 @ 7:14 am | Reply
      • How’s it looking now? I’m still working on moving all of the VPSs over but I believe I narrowed it down since the slowness has stopped for me at least (see tests below).


        # time uname -a | grep real
        real    0m4.489s
        user    0m0.000s
        sys     0m0.002s


        # time uname -a | grep real
        real    0m0.001s
        user    0m0.000s
        sys     0m0.001s
        January 6, 2012 @ 7:52 am | Reply
        • snape:
          root@durp:~# time uptime
           02:54:50 up 4 days, 11:08,  1 user,  load average: 0.08, 0.02, 0.01
          real    0m0.815s
          user    0m0.000s
          sys     0m0.001s


          root@durp:~# time uname -a
          Linux durp 2.6.18-238.19.1.el5.028stab092.2 #1 SMP Thu Jul 21 19:23:22 MSD 2011 i686 GNU/Linux
          real    0m2.915s
          user    0m0.000s
          sys     0m0.001s

          Something’s mauling the CPU on Fire0002, it looks like. Once a minute or so, for 15-20 seconds, the context switches/second bursts from 6-8,000 to 12-15,000, which is kinda high. You’ve also got *severe* entropy starvation going on (lots o’ VPN users?), but that’s really par for the course for an OpenVZ LEB.

          January 6, 2012 @ 8:03 am | Reply
        • @snape?
          What you mean with entropy starvation? o_O (Too much encryption-decryption???) First time I hear about that :|

          Just to share, I did this little thing

          while [ 1 ]; do
                  CUR=`grep ctxt /proc/stat | awk '{print $2}'`
                  sleep 1
                  CUR2=`grep ctxt /proc/stat | awk '{print $2}'`
                  echo $(( $CUR2 - $CUR ))

          And in my Securedragon one, I have 7-8k and little burst of 10-11k, which probably isn’t so bad :P
          IMHO isn’t a lot, I am trying this in several boxes and one of them has 20-25k ctx/s :D And is working fine

          January 7, 2012 @ 7:03 am | Reply
        • Hehe, sorry, more spam, I am excited with all this xD

          watch -n1 cat /proc/sys/kernel/random/entropy_avail

          This always give me between 130-180 in all my boxes (securedragon too)

          January 7, 2012 @ 7:17 am | Reply
        • snape:

          “This always give me between 130-180 in all my boxes (securedragon too)”

          Yep, like I said, typical of a OpenVZ LEB. It should be 2-3k, on a healthy system. Anything below about 1500 is low, and below 300 is *extremely* low. Especially when it’s happening consistently. Means most things reading from /dev/random are going to block.

          Anyway, the SecureDragon VM is being sluggish again…

          root@durp:~# time uname -a
          Linux durp 2.6.18-238.19.1.el5.028stab092.2 #1 SMP Thu Jul 21 19:23:22 MSD 2011 i686 GNU/Linux
          real    0m1.787s
          user    0m0.001s
          sys     0m0.001s

          …but, oh well. Clearly my expectations are unreasonable. :)

          January 7, 2012 @ 7:16 pm | Reply
        • I just ran cat /proc/sys/kernel/random/entropy_avail on a 1 freshly installed node with no VPSs, 0.0 load and only two active services running and I only got 145 so I’m not sure how 2000 is even obtainable. I confirmed this by building a fresh bare metal server with only CentOS 5.7 and only got 177.

          As for the sluggishness, I ask again that you open a ticket so that we are aware of the situation since the sluggishness is not displaying on any of our monitoring (load remains under 4.0, disk I/O remains above 130MB/s, network remains until 20Mbps combined).

          January 7, 2012 @ 11:00 pm | Reply
        • snape:
          snape@tiny:~$ cat /proc/sys/kernel/random/entropy_avail 
          [root@buyvmopenvz ~]# cat /proc/sys/kernel/random/entropy_avail
          [root@somepriceyopenvzvmathetzer ~]# cat /proc/sys/kernel/random/entropy_avail
          [root@quickweb ~]#cat /proc/sys/kernel/random/entropy_avail
          123sys:~# cat /proc/sys/kernel/random/entropy_avail
          [root@durp ~]#cat /proc/sys/kernel/random/entropy_avail

          But, hey, what do I know…

          January 8, 2012 @ 12:04 am | Reply
        • You’re not being very helpful at all. I’m trying to help you and you’ve done little to assist in this matter. I will continue to look into this but as far as I can tell, there are no problems with the current node.

          January 8, 2012 @ 12:26 am | Reply
        • NickM:

          snape is right about the entropy – I noticed today that it was taking forever to establish an SSH connection. Chances are, the driver for the NIC that your servers use just doesn’t call IRQF_SAMPLE_RANDOM, which adds entropy. So, you’re left with just keyboard, mouse, and IDE activity for adding entropy. You could patch your NIC driver, or use rng-tools to feed /dev/urandom back into /dev/random.

          January 8, 2012 @ 12:59 am | Reply
        • Thanks for the input @NickM, it looks like the hardware does not support rng based one everything I’ve tried. I’ll look into this more later but for now I don’t see to much of a problem with it.

          January 8, 2012 @ 4:05 am | Reply
        • Hmmm, @snape, certainly I didn’t tested the Buyvm ones, idk why it has more than 1500 :| and other boxes have the same 130-180. Even I booted some vmware stuff that I have, with GUI and the same low values :S

          I don’t have the big ssh slowness, probably 1-1.5s extra than other boxes, and also no slowness with your command (tried several times since yesterday at random times):

          $ time uname -a
          Linux sd 2.6.32-238.19.1.el5.028stab092.2 #1 SMP Thu Jul 21 19:23:22 MSD 2011 i686 GNU/Linux
          real    0m0.002s
          user    0m0.000s
          sys     0m0.002s
          January 8, 2012 @ 4:14 am | Reply
        • Because we put in extra effort to address the entropy shortage last year :P


          January 8, 2012 @ 4:21 am | Reply
        • Well I’ve definitely learned a lot about entropy in the past few hours. Just need to find a way to get more of it without buying new hardware. :(

          January 8, 2012 @ 4:25 am | Reply
        • Use this one that I posted


          Or use the rngd software (mentioned in the link btw)

          January 8, 2012 @ 4:30 am | Reply
        • install rngd (rng-utils i think on centos) and just tell it to sample from /dev/urandom and insert into /dev/random – that’s what we do.


          You can modify HRNGDEVICE=/dev/hwrng and change it to HRNGDEVICE=/dev/urandom

          While this can pose a security issue, RNG actually runs tests against the values being inserted to make sure they pass all FIPS regulations. If it doesn’t pass, it tosses the results and re-generates.


          January 8, 2012 @ 4:30 am | Reply
        • I followed that link earlier with no luck. I hadn’t tried changing the device though (tried using intel-rng which isn’t found), I’ll give that a shot. Thanks Francisco!

          January 8, 2012 @ 5:01 am | Reply
        • That worked on our test node, I’ll make the changes to our production nodes shortly. Thanks again Francisco!

          January 8, 2012 @ 5:04 am | Reply
        • The link just works if you have IPv6

          January 8, 2012 @ 5:11 am | Reply
        • Damn, doesn’t work on our production nodes. I’m guessing it’s because they are running an older kernel. :(

          January 8, 2012 @ 5:16 am | Reply
        • It works 100% fine good sir :P

          I’ll send you a message on LET in a few minutes for an init file for you to use.


          January 8, 2012 @ 5:21 am | Reply
        • PM’d.

          You can now officially say you have a bit of pony inside you.



          January 8, 2012 @ 5:30 am | Reply
        • paul:

          It’s pretty lame for ssh to be using /dev/random for ordinary connections. I’ve looked at the code but hadn’t noticed that it did that. I’d just configure it (recompiling if necessary) to use urandom. By the time you get a big node ready to run user containers (all disks spun up and mounted etc), urandom should be sufficiently stirred up for vps use.

          January 8, 2012 @ 5:34 am | Reply
        • That’s because urandom itself doesn’t have “enough good” entropy to consider it secure (That FIPS stuff previously mentioned).

          January 8, 2012 @ 5:36 am | Reply
        • paul:

          FIPS tests just check that RNG hardware isn’t broken (emitting all zeroes, etc). It makes no attempt to distinguish high from low quality randomness, and it’s intended for hardware.

          random vs urandom is a source of much misunderstanding because of the device’s poor design and documentation. urandom is almost always the right thing to use, once’s it’s seeded. Any weakness of urandom once it’s seeded could probably be turned into a weakness of ssh itself, so /dev/random won’t help. In a really high security environment, you’d want hardware crypto rather than any software rng anyway.

          http://www.pinkas.net/PAPERS/gpr06.pdf has a detailed (math and code level) explanation of random and urandom that’s much more informative than the documentation, if that’s of any interest.

          January 8, 2012 @ 5:45 am | Reply
        • I don’t think it just tests for zeroes and so


          But I can’t find the source code btw.

          And your link is interesting =) unfortunately in server environments, there are less sources to generate randomness, no keyboard or mouse. Despite that, as you said, IMHO I think urandom is virtually secure.

          January 8, 2012 @ 6:13 am | Reply
        • I had a typo on my end, but I’m using the init file you sent anyways since it doesn’t show the error messages like the other one. :)

          I’ve updated all of our servers so everything should be good from an entropy stand-point. It’s strange that our Proxmox server jumps from 3000 to 200 in less than a minute though, I’m assuming because we’re running multiple sites on strictly HTTPS though. :(

          January 8, 2012 @ 6:21 am | Reply
  15. Daniel:

    Have started with a O96mb plan at SecureDragon. Then took the promo for clients only to X128mb.
    Joe has been very kind to me and my questions, and the server is always fast, up and running. Keep up the great work.

    January 5, 2012 @ 7:14 pm | Reply
  16. Trevor:

    KuJoe, Been having issues since day 1 with the panel , keep getting “Error 324 (net::ERR_EMPTY_RESPONSE): The server closed the connection without sending any data.” when i try to reinstall to any os

    January 13, 2012 @ 4:53 pm | Reply
  17. Trevor:

    Disregard my last comment, oddly enough, it’s chromes fault, tried it with ie9 and it worked perfectly.

    January 13, 2012 @ 5:03 pm | Reply
  18. I have a plan with securedragon openvz, the technical support is always quick 10-18min, the uptime is excellent and the performance of the network is excellent, have connection with Level 3 Comunications,

    January 15, 2012 @ 2:16 pm | Reply

Leave a Reply

Some notes on commenting on LowEndBox:

  • Do not use LowEndBox for support issues. Go to your hosting provider and issue a ticket there. Coming here saying "my VPS is down, what do I do?!" will only have your comments removed.
  • Akismet is used for spam detection. Some comments may be held temporarily for manual approval.
  • Use <pre>...</pre> to quote the output from your terminal/console, or consider using a pastebin service.

Your email address will not be published. Required fields are marked *