Low End Virtual

HostRail – $1.20 256MB OpenVZ VPS Exclusive Offer

Updated July 2011 — HostRail is now officially HostFail. Domain is no longer resolving. See this thread on WHT for discussions there.

HostRail Constantinos from HostRail is offering LowEndBox readers a massive recurring discount — 60% off on all VPS plans when you use promo code Lowendbox60. That applies to all their offers, including their standard fully managed OpenVZ VPS, off-shore OpenVZ VPS in Netherlands and Germany, and SHOUTcast/OpenFire VPS (purposely built, but price a bit out of range). Their “1 Power Unit VPS” will be $1.20/month after the discount (sign up link).

  • 256MB memory
  • 30GB storage
  • 200GB/month data transfer
  • 0.3GHz CPU
  • OpenVZ/SolusVM

Dallas servers I believe. 0.3GHz would mean that it is probably too slow to run anything serious, however at $1.20/month — there is no need for annual payment — it’s pretty good value, consider 30GB of storage comes with it. Whether HostRail can deliver is another matter, but I am giving them the benefit of doubt by posting their offer here 🙂

HostRail is a Cyprus registered business. Been in business since 2004 and the domain was registered since 2007. Note that this offer expires on 31 October.

600 Comments

  1. Spirit's avatar
    Spirit

    Can someone do me a favour and try to connect to some IPv6 IRC server? Does it works? I tryed 5 different IRC networks and I wasn’t able to connect to any.

    • Spirit's avatar
      Spirit

      I forgot to mention that I got allocated IPv6 addresses and ping to ipv6.google.com works – but that’s all. Traffic from saturn node to ANY IPv6 server seems still blocked.

      • Spirit's avatar
        Spirit

        Good idea!

        And no ” Unable to connect to remote host.” – it doesn’t work. I can’t reach any IPv6 site.

        • Keith's avatar
          Keith

          Try ping6 -c5 ipv6only.tk

          Works for me, it’s my BuyVM server.

  2. YW's avatar
    YW

    Ordered 8 hours ago but account has yet been activated. Order number 2580751364

  3. KLIKLI's avatar
    KLIKLI

    Nope that’s WP’s fault
    It should be in []’s but WP removed it 🙁
    In brief, you should try:
    see if this work:
    lynx http://[2001:470:0:64::2]/

    • KLIKLI's avatar
      KLIKLI

      @LEA
      Please have this post moved to the corresponding group.

  4. Spirit's avatar
    Spirit

    @KLIKLI it’s still Alert!: Unable to connect to remote host.
    @Keith I can ping it, but anything else… no

    They said that it’s not firewall issue at node but well…

    Ping and traceroute works, anything else don’t.

  5. Spirit's avatar
    Spirit

    ~# curl http://ipv6.google.com/
    curl: (7) couldn’t connect to host

    # Generated by ip6tables-save v1.4.2 on Sun Oct 31 20:39:24 2010
    *mangle
    :PREROUTING ACCEPT [89:11488]
    :INPUT ACCEPT [89:11488]
    :FORWARD ACCEPT [0:0]
    :OUTPUT ACCEPT [95:8224]
    :POSTROUTING ACCEPT [95:8224]
    COMMIT
    # Completed on Sun Oct 31 20:39:24 2010
    # Generated by ip6tables-save v1.4.2 on Sun Oct 31 20:39:24 2010
    *filter
    :INPUT ACCEPT [89:11488]
    :FORWARD ACCEPT [0:0]
    :OUTPUT ACCEPT [95:8224]
    COMMIT
    # Completed on Sun Oct 31 20:39:24 2010

  6. Spirit's avatar
    Spirit

    It’s default vps setting (i didn’t do anything yet) and they said that they don’t block nor filter any ports. They got same confirmation from their provider.

  7. rm's avatar
    rm

    I just got my IPv6 allocated and I can confirm the problem is there for me too (on the Uranus node).

    # curl ipv6.google.com
    curl: (7) couldn’t connect to host

    # nc6 -vvv6 ipv6.google.com 80
    nc6: cannot connect to yx-in-x69.1e100.net (2001:4860:800a::69) 80 [www]: Permission denied
    nc6: unable to connect to address ipv6.google.com, service 80

    The “Permission denied” message indicates it is a firewall issue somewhere on the way – either host node, or their router, or their datacenter router.

    • Constantinos's avatar
      Constantinos

      Ok, we’ve confirmed that there is some issue.

      Our Network Engineers are working to fix this on all nodes.

    • Constantinos's avatar
      Constantinos

      An update we have confirmed that curl ipv6.google.com fully works on the host nodes. We are still investigating why it doesn’t work on the VPS’s.

    • rm's avatar
      rm

      It turns out I can connect from the VPS to port 22 (SSH) on any IPv6 host which has this allowed:

      # ssh -6 http://www.freebsd.org
      The authenticity of host ‘www.freebsd.org (2001:4f8:fff6::21)’ can’t be established.
      DSA key fingerprint is 09:ae:f1:48:e0:76:c6:e1:13:ff:2b:cb:20:40:38:d0.
      Are you sure you want to continue connecting (yes/no)? yes
      Warning: Permanently added ‘www.freebsd.org,2001:4f8:fff6::21′ (DSA) to the list of known hosts.
      Password:

      I also can SSH into my VPS over IPv6. So at least port 22 is open both ways, but all/most other ports are blocked. It definitely looks like ip6tables’ work on the host node to me.

      • Constantinos's avatar
        Constantinos

        I’ve forwarded this to our Network Engineers. I will also ask for an update from them.

  8. citbun's avatar
    citbun

    just bought one.
    waiting for activation.

    • citbun's avatar
      citbun

      my VPS activated already, thanks a lot for HostRail 🙂

  9. Bernardo Andrade's avatar
    Bernardo Andrade

    I’m pretty happy with the services hostrail is giving me, for the price i paid, the VPS is pretty fast, and pretty good.

    I really like it allot.

    Their support is also responsive and this is good.

  10. Constantinos's avatar
    Constantinos

    An update that this special offer expires on the 6th of November (we have extended the offer).

    • Mike's avatar
      Mike

      Just placed an order for one of my clients (order # 8948877100). These guys are very friendly and respond to support issues very quickly. I would recommend them to anyone :)!

  11. Anand's avatar
    Anand

    Well, so far so good. Medium priority tickets are answered within 3 hours. I think thats fair considering the fact that so many orders are flowing to them. and any problem they do solve it. Had an issue with ip yesterday, they detected it and I was given a new ip very soon.

    My HyperVM is a bit slow today and yum update was very slow (download was fast but installation was slow) that I had to cancel the installation. But will wait for few hours before raising a ticket. If it gets fixed by itself, its good.

    Hope their live chat becomes live(online) after the rush calms down. But nice experience overall. Will recommend it to anyone. 🙂

    • thekreek's avatar
      thekreek

      On Debian its the same, the packages get downloaded fast, but installation takes a while, I think its due to the limited CPU 65 mhz per core, im not sure if yum and apt-get / aptitude are multicore enabled. But so far, it has been ok, I will make more tests tomorrow to see the performance.

      • Snape's avatar
        Snape

        It’s not the CPU, per se – you’re seeing crippling iowait because the hard drive(s) is/are being thrashed by one or more users/abusers. I know because I’m in the same boat – load over 2.0, sometimes as high as 4.0, on a bare system with nothing running. On a Debian VM on the ‘uranus’ server.

        • Jackk's avatar
          Jackk

          You could install sysstat which will break down the CPU usage for your VM like so.

          root@roubaix:/# mpstat
          Linux 2.6.24.th_ovz-xxxx-std-ipv4-32 (roubaix.themegaidiot.info) 11/01/10 _i686_

          09:23:14 CPU %user %nice %sys %iowait %irq %soft %steal %idle intr/s
          09:23:14 all 0.15 0.00 0.03 1.47 0.00 0.00 0.00 98.35 0.00
          root@roubaix:/#

          To install this type ‘yum install sysstat’ on a CentOS system and ‘apt-get install sysstat’ on a Debian system.

      • Jackk's avatar
        Jackk

        To test whether it is using more than the one core you could install something like htop.

        http://htop.sourceforge.net/

        apt-get install htop

        You could then run the update and see the load on the cores via htop in a separate terminal window.

    • Constantinos's avatar
      Constantinos

      Live Chat went Online for some time yesterday. We expect the Live Chat to be fully operational today afternoon/evening.

    • rm's avatar
      rm

      > Well, so far so good. Medium priority tickets are answered within 3 hours.

      I asked for TUN to be enabled, and just about 20 minutes later it was done, very nice. 🙂

  12. Jose's avatar
    Jose

    I’ve bought one, paid invoice number #HR2010-2612 but i still don’t have my vps account data, only WHM CS.

  13. tommy's avatar
    tommy

    server respond really slow:
    w
    real 0m0.142s
    user 0m0.001s
    sys 0m0.002s

    free -m
    real 0m5.666s
    user 0m0.002s
    sys 0m0.001s

  14. vRozenSch00n's avatar
    vRozenSch00n

    Hmmmmm I’ve been witnessing several good deals frenzy, and It has the same pattern:

    – Users wants to have their VPS setup immediately after payment.
    – Provider work their back to the bones because they need to setup new boxes, nodes, etc, while popping their eyes to verify that the orders are not fraud.
    – In the mean time many too enthusiastic users who got their VPS ready immediately benchmark test their new VPS. Imagine if a box is filled with 20 VPS and 16 users are testing the VPS at the same time, how do you think the performance would be?
    – Users who tested their VPS under the above condition might think that their VPS is bad or the connection is too slow and many who are new to this kind of frenzy will start screaming for a refund.

    “So, be patient grasshopper, wait and see for a fortnight after the storm go away, then decide whether or not to continue your journey in this land of penguins” – Kwai Chang Core (The NixFu Master)

  15. marrco's avatar
    marrco

    504 Gateway Time-out

    my vps is down, their control panel (hypervm enterprise) is down too. Support is offline. And i did many tests, it’s really slow compared to the 99 cents vps.

    • tommy's avatar
      tommy

      yeah, vps really slow.

      • tommy's avatar
        tommy

        update :
        take 2 minutes for finishing:
        yum search openssl

  16. Steve's avatar
    Steve

    So, just to make sure: The extensive ToS is only for shared accounts and not for VPS? So I can run Teamspeak on a VPS no problem?

    • Constantinos's avatar
      Constantinos

      Yes that is correct and our TOS will soon be updated to mention that its only for shared accounts

  17. marrco's avatar
    marrco

    my vps is offline for a few hours. I finally managed to login to the control panel (still many icons missing) and tried to rebuild the system (debian x86) .. really slow. Looks like the system is not able to rebuild my VPS.

    I the meantime i did a new test with the 99 cents vps. After a few weeks that still rocks. Looks like Costantinos has some serious problem here.

  18. marrco's avatar
    marrco

    new problems with this VPS:
    14 228 ms 170 ms 169 ms limestone-ic-135964-dls-bb1.c.telia.net [213.24874.222]
    15 169 ms 168 ms 169 ms te5-1.bdr1.core2.dllstx3.dallas-idc.com [208.115252.22]
    16 168 ms 166 ms 169 ms uranus.hostrail.com [208.115.247.130]
    17 199 ms 173 ms 168 ms 241-64-162-69.reverse.lstn.net [69.162.64.241]
    18 165 ms 169 ms 167 ms uranus.hostrail.com [208.115.247.130]
    19 168 ms 169 ms 172 ms 241-64-162-69.reverse.lstn.net [69.162.64.241]
    20 169 ms 172 ms 169 ms uranus.hostrail.com [208.115.247.130]
    21 169 ms 167 ms 169 ms 241-64-162-69.reverse.lstn.net [69.162.64.241]
    22 169 ms 170 ms 172 ms uranus.hostrail.com [208.115.247.130]
    23 167 ms 169 ms 170 ms 241-64-162-69.reverse.lstn.net [69.162.64.241]
    24 170 ms 169 ms 169 ms uranus.hostrail.com [208.115.247.130]
    25 171 ms 169 ms 171 ms 241-64-162-69.reverse.lstn.net [69.162.64.241]
    26 168 ms 169 ms 170 ms uranus.hostrail.com [208.115.247.130]
    27 168 ms 172 ms 169 ms 241-64-162-69.reverse.lstn.net [69.162.64.241]
    […]

    and when i ping my vps all i get is a ttl expired in transit answer

  19. chachiris's avatar
    chachiris

    works fine, fast.

  20. Constantinos's avatar
    Constantinos

    Regarding the slow, Uranus node is currently experiencing slow slowness, i have been informed that our Abuse Department is actively working to resolve this ASAP, they have also sent an email to all clients regarding this.

    • sleddog's avatar
      sleddog

      Good to hear someone’s working on it. I read most of this thread while waiting for an SSH login to complete. Actually I’m still waiting… 🙂

      • Constantinos's avatar
        Constantinos

        We had a technician working on it from the first moment, then we found there was some Abuse from a long term client of ours. His VPS was suspended and Uranus i confirm myself is back to normal.

  21. Constantinos's avatar
    Constantinos

    We have found the VPS abusing the server and the server has now returned back to normal.

    • sleddog's avatar
      sleddog

      Still extremely slow for me….

      [root@hrlabrad ~]# time ls -al
      total 36
      drwxr-x--- 3 root root 4096 Nov 1 08:24 .
      drwxr-xr-x 20 root root 4096 Jan 26 2010 ..
      -rw------- 1 root root 1247 Nov 1 08:40 .bash_history
      -rw-r--r-- 1 root root 24 Jan 6 2007 .bash_logout
      -rw-r--r-- 1 root root 191 Jan 6 2007 .bash_profile
      -rw-r--r-- 1 root root 176 Jan 6 2007 .bashrc
      -rw-r--r-- 1 root root 100 Jan 6 2007 .cshrc
      -rw-r--r-- 1 root root 129 Jan 6 2007 .tcshrc

      real 0m9.032s
      user 0m0.002s
      sys 0m0.002s,/code>

    • magephp's avatar
      magephp

      Still slow.
      Command “apt-get install mono-devel” is execute nearly 1 hour but still not complete

      • Constantinos's avatar
        Constantinos

        Ok Uranus is now fast, only one issue remains between communication of HyperVM and the VPS’s hosted on Uranus.

  22. Spirit's avatar
    Spirit

    Any progress regarding IPv6? 😛

    • rm's avatar
      rm

      I am waiting very eagerly too, I basically bought it because of native IPv6.

      • tommy's avatar
        tommy

        i have 20 ipv6 😛

        but i cant connect to ipv6 irc server or ipv6 website. only can ping all of them.

        i already stop ip6tables. sadly

    • Constantinos's avatar
      Constantinos

      IPv6 should fully work quite soon, i know that our Senior Network Engineer is working on that.

      • rm's avatar
        rm

        Here’s what I get on trying to connect to my VPS over IPv6:
        01:11:07.337082 IP6 xxxxxxxxxxxxxxx.55342 > 2607:ff68:100:xx::xx.www: Flags [S], seq 1530413521, win 5648, options [mss 1412,sackOK,TS val 92344317 ecr 0,nop,wscale 7], length 0
        …
        01:11:07.596211 IP6 2607:ff68:100:10:20a:cdff:fe1a:3cd8 > xxxxxxxxxxxxxxx: ICMP6, destination unreachable, unreachable prohibited 2607:ff68:100:xx::xx, length 88

        To the best of my knowledge it is ip6tables with a “-j REJECT” rule that most likely generates those “ICMP6 unreachable prohibited” messages. If I were root on the host node, the first things I’d try would be:

        # look at what's in there:
        ip6tables-save

        # remove all the rules:
        ip6tables -F

        # set all policies to ACCEPT:
        ip6tables -P INPUT ACCEPT
        ip6tables -P FORWARD ACCEPT
        ip6tables -P OUTPUT ACCEPT

        and then recheck if IPv6 started to work.

      • Spirit's avatar
        Spirit

        I hope that he doesn’t work on this issue full time whole 3 days already 😛

        • Snape's avatar
          Snape

          I like to hope that ipv6 support is secondary to concerns like the appalling i/o wait on the Uranus node, which continues sporadically even now.

          To be fair, I don’t think it’s necessarily Hostrail’s fault, per se. I think it’s abundantly clear that cheap VPS offerings like this are abuse magnets, so I suspect much of their time is spent tracking down and dealing with a plethora of spammers, crackers, torrenters, and whatnot who are making life difficult for the rest of us. If we have to wait a couple more days for ipv6 – something they didn’t even advertise as offering – while they deal with a horde of idiots, well, so be it.

        • sleddog's avatar
          sleddog

          (In reply to Snape…)

          Regards to being an “abuse magnet”, it’s interesting that my $0.99 VPS at VPS Tree never suffers this kind of abuse. It’s always fast and snappy — it’s a mystery to me how they do it, but it invites speculation….

          Comparing VPS Tree to Hostrail we have: 64MB RAM vs. 256; 5 GB storage vs 30; 20 GB transfer vs 200. Which of those differences would act as the abuse magnet for HostRail? My guess is the transfer allowance. Without the promise of a big transfer quota the VPS Tree offering is highly unattractive to the spammer/torrent/proxy etc. abuser.

          I’d be happy to be moved to the new Hostrail “nice neighbours” node, with a monthly transfer allowance of maybe 40 GB 🙂

          — sleddog

        • sleddog's avatar
          sleddog

          Sorry typo in the above. Should be “200 GB transfer vs 20”.

        • sleddog's avatar
          sleddog

          Damnit no! It was right the first time! 🙂 200 GB for Hostrail vs 20 GB for VPS Tree. Sigh.

        • Snape's avatar
          Snape

          Sleddog: I suspect the big causes of abuse are people filesharing and/or running “rapidleech” scripts, both of which you’ll want lots of disk space and bandwidth for, and both of which I suspect really thrash the disks in a user-unfriendly fashion. I mean, let’s be honest, if you’re torrenting Blu-Ray movie rips, you’re not even going to give a 5 or 10 GB VPS a second look…

          Also, to be honest, I suspect 64MB of RAM isn’t seen as attractive by script kiddies. You can do quite a bit with if if you try, as LEA has shown here, but my guess is that the vast majority of 64MB VPSes are either private VPN/proxy nodes, IRC vhosts, or backup space, which explains why the few of us who use them for “normal” server applications find them always-responsive. Offering such low-spec VMs is probably a good business plan to minimize abuse, to be honest. (Minimal resource abusage = lots of happy customers = recurring revenue, plus it’s somewhat harder to remote-exploit a VM that’s literally running SSH and OpenVPN and nothing else, than one that’s got Apache and SQL and is hosting a hundred blogs and forums and whatnot.)

        • Constantinos's avatar
          Constantinos

          Unfortunately seems we have bad news about IPv6, nothing official yet but there is a bug in the OpenVZ kernel which is not allowing full IPv6 compatibility – we have been sent a patch however we haven’t applied that to any of the production servers – we’re currently testing it at a test server to ensure that it doesn’t cause any other problems.

        • Spirit's avatar
          Spirit

          That’s strange. So far native IPv6 works at few LEB featured OpenVZ hosts (BuyVM, hitme.net.pl, etc..) without any issues.

        • sleddog's avatar
          sleddog

          Snape: I agree with what you’ve said. But the bottom line is (I think) bandwidth (or ‘transfer quota’). Without the high transfer quota the high diskspace is useless for abusers. A high diskpace/low transfer model can fit a lot of legitimate, low-resource-use users.

        • Snape's avatar
          Snape

          Sleddog: True, but I would not be at all surprised if a lot of the abusers planned to hop from provider to provider on a monthly basis, with no intention of paying for any overages. (I’m honestly not sure if VPS providers normally automatically terminate VMs that hit their transfer limits or not; I don’t think I’ve ever transferred more than about 40GB on a single VM in any given month, and all my machines have at least 70GB…)

          I guess we could pool our resources and form a short-lived VPS biz selling $3/3GB/3TB VPSes and gather some authoritative data, but… no. 🙂

        • rm's avatar
          rm

          Maybe something like this is affecting them. http://bugzilla.openvz.org/show_bug.cgi?id=1327
          However IPv6 on my BuyVM VPS does indeed work beautifully, why can’t it work here.
          (BuyVM) 2.6.18-194.8.1.el5.028stab070.2 #1 SMP Tue Jul 6 14:55:39 MSD 2010 x86_64 GNU/Linux
          (HostRail) 2.6.18-194.17.1.el5.028stab070.7 #1 SMP Fri Oct 1 14:59:55 MSD 2010 i686 GNU/Linux
          The good news here is that if a patch is already in testing, hopepully it will be rolled out soon.

      • Patrick Garman's avatar
        Patrick Garman

        Just a heads up for you guys at least (everyone else too)

        SolusVM will natively support IPv6 “by the end of the week”… this is per Phil directly.

        Expect an update in not too long! – thats all I know unfortunately 🙁

    • Constantinos's avatar
      Constantinos

      We’re rolling out the first patch. We also have a second patch to roll out however we will roll out one patch at a time.

  23. tommy's avatar
    tommy

    too much problem for the 1st day
    Alert: Could not Connect to the server uranus.hostrail.com. This is most likely due to underlying network problem. Make sure that the server is accessible from this particular node by running telnet slave-id 8889

    • Constantinos's avatar
      Constantinos

      There is a communication problem between HyperVM and Uranus. Other than this the VPS’s on Uranus are all back to normal.

      • sleddog's avatar
        sleddog

        It’s running very well for me now. The slow-slowness has disappeared and the VPS responds nicely at the command line. Well done 🙂 Just need to get the HyperVM issue sorted now.

  24. Constantinos's avatar
    Constantinos

    A note to everyone, while i try to answer all questions, there are so many questions that i would probably spend my whole day here rather than running HostRail (i’m the CEO for those of you who don’t know) so if your question hasn’t been answered don’t worry – i’ve got someone else checking and noting down any problems and checking accordingly.

    • tommy's avatar
      tommy

      my vps really slow, and I request for refund.

  25. earl's avatar
    earl

    Just ordered.. Hope setup won’t take too long.

  26. Richard's avatar
    Richard

    If you compare their TOS with Hostgator’s, you will find a lot of sentences which have been copied.

    • Constantinos's avatar
      Constantinos

      Thats interesting, our lawyers wrote our Terms of Service and they are currently re-writing them a bit as our TOS is a bit unclear on some things.

      • tommy's avatar
        tommy

        ‘Accounts that have paid monthly are not eligible for a refund.’
        you need to pre pay for 10 years then they’ll give you refund.

  27. Yomero's avatar
    Yomero

    I have one of these 1.20 VPS with you. All is axcellent. Slow as is, and cannot be better because are only 300 mhz. I think some people doesn’t know how works an AMD K6 or Pentium 2 equivalent…

    Ok, returning to the point. Someone here requested allocate the CPU resources in only one core, and Constantinos replied to put that in the “notes” field of his order. I did that, and later submitted a ticket but the support says that cannot be possible =(

    BTW.. Someone can explain me if putting the resources in one core really will show a difference? Because I have noted weird things in cpu monitoring tools, for example top, the cpu consumption doesn’t have sense. This is a normal condition in OpenVZ? Is a configuration problem? Version? Or is impossible to report accurate CPU usage? I have another provider with OpenVZ an the reports have more sense than this server xD

    Thanks in advance.

    • Snape's avatar
      Snape

      It’s slow, and you’re seeing “weird things” in top, because of really high iowait, not because it’s in any way comparable to a PII. Sumbit a ticket, and HostRail will – hopefully! – do something about the people abusing the host node.

      When things are working as they should, cpu usage works just fine, just as it does at other hosts. You’re just seeing the effects of a couple bad apples…

      • Yomero's avatar
        Yomero

        Are you sure about that? Where can I see the io wait? In top? The “wa” parameter?

        Hmmm, about the cpu consumption.. Example, I have an app that surely must use the 300-400 mhz all the time, but it shows only 4%. I am sure it must use my 100%.

        2 days ago it showed 0% ALWAYS :S

        • Snape's avatar
          Snape

          “wa” in top, or from the output of vmstat, yes.

        • rm's avatar
          rm

          > I have an app that surely must use the 300-400 mhz all the time, but it shows only 4%. I am sure it must use my 100%.
          It will never ever show more than 10% used in top, because 10% is what your account is allowed to use.
          Consider 10% to be your “100%” 🙂

      • Yomero's avatar
        Yomero

        Top shows 0% in “wa”.
        vmstat apparently rounds up the value, and shows “1”

      • Yomero's avatar
        Yomero

        But why my other provider have 8-core servers and they give me 1 core at 2.4 ghz or 4 cores at 600 mhz. And was no problem to supply me that requirement, because some applications doesn’t work well with only 600 mhz (single threaded apps)

        • rm's avatar
          rm

          > some applications doesn’t work well with only 600 mhz (single threaded apps)
          I tested this on HostRail, and there seems to be no issue with single-threaded apps.
          One thread uses 10% of the CPU, and when I launch two, they both run at 5% CPU each.
          The performance does not increase with number of threads, it divides between them evenly.
          So it is clearly not the case where 1 thread runs just at 68 Mhz and you need 4 threads to use all cores.
          Maybe your other host runs not OpenVZ but Xen (for example)?

        • eLohkCalb's avatar
          eLohkCalb

          In regards to OpenVZ, I still think it’s inappropriate to use “MHz” as the measurement unit. It’s all about time slicing if I understand it correctly, meaning your container gets assigned with a portion of the CPU time via cpuunits and cpulimit.

          Like what @rm mentioned, it is 10% of the CPU time rather than 4 x 68 MHz. OpenVZ prioritizes your program (process/thread to be exact) in a way that it only touches the CPU not more than 10% of the total time, the rest of the 90% is either idle, or given to other programs in other containers. Also, the limit of core is virtual, you still get access to all cores even if your container is “limited” to 1 core.

      • Yomero's avatar
        Yomero

        Ok, then, I must understand… If an app is not multithreaded and requieres more than the aparently 68 Mhz…Will use the 4×68 mhz like an multithreaded app?? That is what I understand of this…
        And then, the measurement of cpu usage in any tool cannot have sense in any way.
        I get 4x68mhz~300Mhz like a single core for any app that needs that 300 mhz. Is right? (Ok, maybe the cpu scheduling is more complicated…but is an idea)

        • rm's avatar
          rm

          @Yomero
          Forget about MHz, really… your account can use 10% of CPU, that’s it.
          You can use it with 1-threaded app, it will consume 10%.
          Or you can run 4 apps, or an app with 4 threads, they will get 2.5% each.
          So you don’t get ‘less’ CPU if you use 1-threaded apps.
          I have personally tested in works this way on HostRail.
          Maybe it should be clarified in the offer.

  28. Mike's avatar
    Mike

    I submitted a ticket a while ago (#853836) and have not yet recieved a response. It seems you guys might be really busy with all the new orders, which I do understand. It wasn’t marked urgent or anything but I did expect an answer before this?

    • Mike's avatar
      Mike

      I take that back. They responded as I was writing this I guess! Thanks HostRail!

  29. Nizar's avatar
    Nizar

    Not so good support, they miscalculated my upgraded server price. They gave me a new server price instead of upgraded server.

        • InsDel's avatar
          InsDel

          $25*0.4=$10. I fail to see the problem.

        • Constantinos's avatar
          Constantinos

          At the end there was a miscalculation, this was resolved and the credit was applied to the account.

      • Nizar's avatar
        Nizar

        Thanks Constantinos for the support, what a helpful support.
        Sorry, i’m late to say thanks.
        Late is better than never anyway. 🙂

  30. John's avatar
    John

    HostRail support is so helpful that they installed what I want, took long time, but they stick to the end
    and now everything works fine. Thanks Constantinos!

  31. Constantinos's avatar
    Constantinos

    We’re rolling out the first patch for IPv6. There is a second patch however we will roll out one patch at a time.

    • Spirit's avatar
      Spirit

      Any progress? 🙂

      • Constantinos's avatar
        Constantinos

        I sent an email about this. I should have IPv6 working quite quickly, i got delayed due to an issue with a VPS however i am now back to this issue – i work nearly 24 hours so i will work until this is done :). Customer Satisfaction is always my top priority and then comes everything else including sleeping 🙂

        • rm's avatar
          rm

          Come on, roll it out, and please start with Uranus 🙂
          Waiting sucks :/

  32. Taw's avatar
    Taw

    Ordered server. Discount was still valid, and VPS was delivered in 10 minutes from payment.

  33. Jackk's avatar
    Jackk

    Have just placed an order for three systems, waiting on setup now. 🙂

    • Jackk's avatar
      Jackk

      Yay! I may order some more if these go okay!

      • sleddog's avatar
        sleddog

        Don’t get too excited… the one I have is often — like now — in a state of paralysis. I think I got the “0 Power Unit” model. It’s on the Uranus node…

        [root@hr:~] time ps aux
        USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
        root 1 0.0 0.0 2160 168 ? Ss Oct31 0:00 init [3]
        root 13500 0.0 0.0 1816 256 ? Ss Nov01 0:00 syslogd -m 0
        root 13577 0.0 0.0 4496 220 ? Ds Nov01 0:00 crond
        root 13724 0.0 0.3 2272 816 pts/0 R+ 22:53 0:00 ps aux
        root 27919 0.0 0.1 2132 328 ? Ss Nov01 0:00 /usr/sbin/dropbear
        root 28537 0.0 0.3 2284 876 ? Ss 22:25 0:00 /usr/sbin/dropbear
        root 30318 0.0 0.5 2552 1344 pts/0 Ss 22:28 0:00 -bash

        real 0m53.672s
        user 0m0.001s
        sys 0m0.004s

        • Constantinos's avatar
          Constantinos

          Did you open a ticket about this issue? Uranus from what i see is currently on low load.

        • Snape's avatar
          Snape

          server:~# time uptime
          03:32:19 up 5 days, 12:19, 1 user, load average: 3.48, 2.62, 1.55

          real 0m13.558s
          user 0m0.000s
          sys 0m0.001s

          server:~# time uptime
          04:34:21 up 5 days, 13:21, 1 user, load average: 7.12, 6.48, 6.40

          real 0m19.428s
          user 0m0.000s
          sys 0m0.000s

          Uranus is pretty much unresponsive due to stalled disk i/o, Constantinos. I guess your idea of “low load” is a bit different than ours.

        • Constantinos's avatar
          Constantinos

          I just logged into SSH myself and executed a yum update on a test VPS that i setup for this purpose – the command finished within a couple of minutes – didn’t take that long. I will anyway check Uranus – if you can submit a ticket i can check your VPS and to some troubleshooting on it.

        • drmike's avatar
          drmike

          >Uranus

          >I will anyway check Uranus

          Oh, the one liners that come to mind.

          Can’t believe you actually named a server that.

          Guess I need more sleep. 😛

        • Snape's avatar
          Snape

          drmike: Uranus is overloaded. Uranus is overwhelmed. Someone is probably abusing Uranus. Uranus needs support. Yes, we get it. 🙂

          Constantinos: Ticket submitted.

    • vRozenSch00n's avatar
      vRozenSch00n

      Constantinos, you really have a good team there. Just a few minutes ago I realized that I can have a quite powerful VPS by adjusting the Addons on the shopping cart.

      So here are some settings that I tried to adjust:

      VPS – VPS 2 Power Units (hostrail200)
      » Additional CPU: 0.3 GHz CPU $1.99 USD
      $6.98 USD
      Promotional Code: Lowendbox60 – 60.00% Recurring Discount
      Subtotal: $6.98 USD
      60.00% Recurring Discount: $4.19 USD
      Total Recurring: $2.79 USD Monthly

      VPS – VPS 2 Power Units (hostrail200)
      » Additional CPU: 0.6 GHz CPU $3.99 USD
      Promotional Code: Lowendbox60 – 60.00% Recurring Discount
      Subtotal: $8.98 USD
      60.00% Recurring Discount: $5.39 USD
      Total Recurring: $3.59 USD Monthly

      VPS – VPS 2 Power Units (hostrail200)
      » Additional CPU: 0.3 GHz CPU $1.99 USD
      » Additional RAM: 256MB RAM $2.99 USD
      Promotional Code: Lowendbox60 – 60.00% Recurring Discount
      Subtotal: $9.97 USD
      60.00% Recurring Discount: $5.98 USD
      Total Recurring: $3.99 USD Monthly

      But I have already ordered a couple of days ago. So I opened a ticket and ask whether I can have the (basic price + Addon price) minus the discount instead of the (basic price – the discount) + the Addon price. So Rob Kelly the Support Department Manager immediately confirmed what I wanted and generate the calculation and invoice. Now I have VPS 2 Power Units with additional 0.3 GHz CPU and 256MB RAM for $3.99 USD.

      Great deal!

  34. I need pptp's avatar
    I need pptp

    Your VPS can’t support pptp. why not support it? Thank you

    • Constantinos's avatar
      Constantinos

      PPTP is supported, just submit a ticket and we will manually activate that feature for you.

  35. sleddog's avatar
    sleddog

    I did submit a ticket. And now the load is back to normal. I hope it can stay that way. 🙂

  36. thanks99k@gmail.com's avatar
    thanks99k@gmail.com

    Hello,

    Just need to say THANKS for Hostrail/
    My VPS setup just only 1 hour after payment !! WOW
    Also great support for my newbie in CentOS

    Anyway, hope all of you will happy with their services.

    Rgds,
    T99K

  37. vRozenSch00n's avatar
    vRozenSch00n

    Got the Power Unit 1 VPS today and paid it yearly. So far I’m satisfied with the 2 Power Unit VPS I purchased earlier, so let’s see this Power Unit 1 performance in the next 3 months.

  38. marrco's avatar
    marrco

    uranus low again:
    ok, so definitely there’s a problem somewhere with this node.
    loadimpact is sloooow
    http://loadimpact.com/result/vps.triesteuropea.it-80224793aca372069b21b1975b37d9c7

    and even a ls need a few seconds.

    vps002:~# time ls /var -la
    total 60
    drwxr-xr-x 14 root root 4096 Nov 1 18:33 .
    drwxr-xr-x 20 root root 4096 Nov 6 17:52 ..
    drwxr-xr-x 2 root root 4096 Nov 3 06:25 backups
    drwxr-xr-x 10 root root 4096 Nov 1 14:48 cache
    drwxr-xr-x 25 root root 4096 Nov 1 18:36 lib
    drwxrwsr-x 2 root staff 4096 Apr 3 2009 local
    drwxrwxrwt 2 root root 4096 Nov 6 17:52 lock
    drwxr-xr-x 11 root root 4096 Nov 6 17:52 log
    -rw-r–r– 1 root root 3139 Nov 1 18:17 logauth.log
    -rw-r–r– 1 root root 0 Nov 1 14:33 logmail.err
    drwxrwsrwt 2 root mail 4096 Nov 6 06:27 mail
    drwxr-xr-x 2 root root 4096 Apr 3 2009 opt
    drwxr-xr-x 10 root root 4096 Nov 6 17:52 run
    drwxr-xr-x 4 root root 4096 Nov 1 18:32 spool
    drwxrwxrwt 2 root root 4096 Apr 3 2009 tmp
    drwxr-xr-x 5 root root 4096 Nov 6 00:28 www

    real 0m2.274s
    user 0m0.001s
    sys 0m0.002s

    vps002:~# time uptime
    19:35:47 up 1:42, 1 user, load average: 1.57, 1.27, 0.72

    real 0m4.007s
    user 0m0.000s
    sys 0m0.001s
    vps002:~#

    it’s the third ticket i open today…

Leave a Reply

Before you comment
  • Please use your hosting provider's support system for service issues.
  • Comments may be held temporarily for spam review.
  • Use <pre>...</pre> for terminal output, or link to a paste service.

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