Low End Virtual

BuyVM – $15/Year 128MB VPS Coming on Monday 18 April!

BuyVM “Do BuyVM have any stock?” seems to be one of the most frequently asked questions here. And behold! Pony the Almighty has spoken. Jay wrote in to point me to this announcement on FranTech:

Francisco went to bzImage and said, “Thus says The Pony, the God of Frantech, ‘Let My people go, that they may purchase VPS goodness.” But bzImage said, “Not until Monday April 18th at 12:00pm GMT-7”

New Clients: Please remember to register with accurate account details as the red sea shall not be parted for fraud.

Not too sure whether there will be 10 plagues, and the last one kills bzImage’s first born process (die, init(1) die!) but at least the date/time is clear. Monday. April 18th at lunch time. BuyVM will have stock (yes Francisco — don’t make us disappointed). Including their popular $15/year offer.

  • 128MB guaranteed/256MB burstable memory
  • 15GB storage
  • 500GB/month data transfer
  • 1x IPv4 + lots of IPv6
  • OpenVZ/SolusVM

Their servers are with Hurricane Electric in Fremont CA, and BuyVM were voted the #1 in 2011 Q1 by the LowEndTalk community. Get your world clock ready if you are not in US 🙂

461 Comments

  1. MACRO's avatar
    MACRO

    Ups, didn’t receive or went into other folder probably.
    Any ETA?

    • Vicky's avatar
      Vicky

      As far as I know hdparm doesn’t work inside a OpenVZ container/VPS.

    • dmmcintyre3's avatar
      dmmcintyre3

      hdparm does not work in openvz, but I got 100MB/s with dd when I tested about a hour ago.

  2. Shadowsurge's avatar
    Shadowsurge

    dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
    16384+0 records in
    16384+0 records out

    1073741824 bytes (1.1 GB) copied, 62.6335 seconds, 17.1 MB/s

    got this speed on node 20.. is this normal ?

    • Shadowsurge's avatar
      Shadowsurge

      Node 26
      dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
      16384+0 records in
      16384+0 records out
      1073741824 bytes (1.1 GB) copied, 122.462 seconds, 8.8 MB/s

    • Francisco Dias's avatar
      Francisco Dias

      Not at all 🙂

      We’re doing an initial check on 27 tomorrow to see if we have multiple bad disks in this set of gear. We should have a good idea by tomorrow. While adaptec has some good monitoring programs like arcconf we’ve had drives report 100% fine until a reboot happens – suddenly the drive kicks out.

      The majority of the nodes are fairly good (70MB/sec+ the majority of the time), but I’m thinking there’s a collection of bad ones in the L5420 nodes.

      We’ll know soon and roll replacements if that’s the case 🙂 We have a batch of spare drives on site so we don’t have to wait on newegg/etc.

      Thanks,

      Francisco

  3. Jacky's avatar
    Jacky

    Should really say that you need verification of id and current physical address in the first message about ordering through vpn. Had to wait for second response about it, but by then the stock of 256 was already almost out.

  4. Alex Beal's avatar
    Alex Beal

    Yay! I finally got one!

    One question, though: I was pleasantly surprised by the abundance of IPv6 addresses, but a bit puzzled by how the node doesn’t seem to have the ip6tables modules loaded. Really, if I can’t set up a firewall for the IPv6 addresses, then I’d rather have the addresses disabled. Am I missing something here?

    Also, there was a slight hic-up during setup. Apparently passwords can only be alpha-numeric, or it screws up their setup script. I had to submit a ticket to get it straightened out. So, watch out for that.

    Otherwise, I can’t wait to get started! I’ve heard nothing but good things about these guys, and it’s not a minute too late, as I’ve been having issues with my previous provider.

  5. Craig's avatar
    Craig

    i am riding one of their pony now,
    and it feels so a smooth ride 🙂

  6. macro's avatar
    macro

    server goes down every other day..

    • Francisco Dias's avatar
      Francisco Dias

      You’re on 27 then 🙁

      We scheduled some work to check into the raid. Adaptec cards have issues where they don’t actually mark drives bad at times without a reboot. We just finished checking with HE and the box will be online in just a few minutes.

      Francisco

  7. macro's avatar
    macro

    hope you let us know in advance.

    • Francisco Dias's avatar
      Francisco Dias

      We had an announcement up 🙂

      Either way! 26 is feeling much better and i’ll keep an eye on it. IOWAIT feels a load better as well. Thank you!

      Francisco

  8. c's avatar
    c

    @Francisco

    There is a problem of the ubuntu 10.04(it is LTS).

    root@u4:~# mkdir abc
    root@u4:~# touch abc
    touch: setting times of `abc’: Bad file descriptor

    And also I cannot do apt-get upgrade.

    A buyvm staff told me that I have to replace the tar with the one you buyvm provide. It is hackish.

    LTS is so important, can you fix this template?

    See ticket #617329 for detail.

    • Francisco Dias's avatar
      Francisco Dias

      Hi C,

      The issue is actually in ubuntu so i’ll check if they have a replacement version around. The replacement isn’t really hacky since the ‘proper fix’ has been in upstream debian for the better part of a year.

      Francisco

    • Francisco Dias's avatar
      Francisco Dias

      Hi C,

      We’ve:

      – Replaced your /bin/tar
      – We’ve chattr +i /bin/tar that way it isn’t replaced with the one in ubuntu (this one is from debian 6 so it’s fine)
      – I’ve run apt-get upgrade and it seems fine.

      Let me know,

      Francisco

      • c's avatar
        c

        Thx, Francisco.

        Yes, the apt-get upgrade seems has no problem now after replacing the tar binary.

        But touch a dir still shows the issue.

        root@u4:~# mkdir abc
        root@u4:~# touch abc
        touch: setting times of `abc’: Bad file descriptor

        That is one reason I don’t want to replace the tar binary, since there may be so many other bins need to be replaced.

        • Francisco Dias's avatar
          Francisco Dias

          OK how about this.

          Would 64bit work for you? 64bit has this quirk addressed and, while we don’t do it much, I have no problem rolling you out a 64bit container if you’re all backed up.

          Let the guys know, i’ve already talked to Anthony 🙂

          Francisco

        • c's avatar
          c

          Francisco,

          I don’t want a 64 bit system.

          Can’t you fix this in a few days? It hasn’t to be fixed today for me. I can wait for a few days.

        • Francisco Dias's avatar
          Francisco Dias

          We’ve already commited the error to OVZ and are awaiting a patch.

          A 64bit system would work just fine for your needs and run whatever software you need that’s 32bit as well. The 64bit option would resolve this for good.

          Let me know,

          Francisco

        • c's avatar
          c

          ok. So please roll me out a 64 bit container. Thx.

          And btw, I have no data on that box, I just bought it a few days ago, so I don’t need to back anything up.

        • c's avatar
          c

          wait wait, I just got a message from you buyvm said “We will be rebooting the nodes listed below to enable fixes for a few kernel bugs”. Is that fix this issue? If it will be fixed, I think I don’t need to move to a 64 bit container.

        • Francisco Dias's avatar
          Francisco Dias

          Unrelated 🙂 We’re still waiting word back from OVZ on your bug.

          I’ll let Anthony know.

          Francisco

  9. macro's avatar
    macro

    Hey Francisco, since there is an announcement, is it o.k to shut down every other day?
    Why don’t you just shut down with ‘announcement’?

    • Francisco Dias's avatar
      Francisco Dias

      The quirk on 27 isn’t really something that we do daily or something, we had a panic. 27 was up for well over a couple months before this and the IO quirk we were addressing.

      Francisco

  10. macro's avatar
    macro

    I didn’t sleep for 2 nights to get one of these based on your reputation, price, etc.
    This happens every other day now, and wonder what’s going to happen tomorrow.
    of course there will be an announcement of course, it justifies everything, right?
    I will just move on.
    thanks anyway…

    • Francisco Dias's avatar
      Francisco Dias

      Hi Macro,

      I apologize if you’ve had an issue with 27, but i’ve also worked hard with the guys to improve performance on the node since there was a hardware issue. Some of the reboots that happened before todays lunch one were unplanned since the node simply went south. If we were prepped for it I could have had a tech at HE help me resolve it but that wasn’t the case.

      Your spot is saved for you for now for another few days.

      I should add that this is the only node that has had any sort of crash issues. A few other nodes have some bad drives but we’re resolving those and will schedule work on those this week.

      Thanks,

      Francisco

    • cd's avatar
      cd

      Hey macro, I’m sorry to hear that you aren’t satisfied with your BuyVM. I can assure you that this isn’t their normal level of service, as many of my containers do have a higher uptime/performance rating. (See: http://servers.cedricd.org ). BuyVM is one of the best providers I’ve ever seen, providing a great price with more features then most of the other VPS hosts. I hope you can find it to stay with them a while more, as I’m confident that the issues will be resolved. Be assured that they will take care of current clients, whereas some other hosts focus more and more on the newer clients and leave the older ones in the dust. I have notified the administration that there may be an issue with the mass announcements as I have also not been getting them which is understandably annoying at times.

      ~ C

      • Yomero's avatar
        Yomero

        Hey cd, can you share the implementation of that website? n_n

        • cd's avatar
          cd

          Not at this time unfortunately – I will see into releasing it as a whole in the future though

      • Ben's avatar
        Ben

        🙂 You don’t have any “node 12” … I don’t have nice stats as you do, but I can confirm it’s working well too 🙂
        Thumbs up for Francisco and his team: I appreciate you being so open here!

        Too bad I hardly can’t access it because of the net from Thailand these days.

        • Francisco Dias's avatar
          Francisco Dias

          Maybe you should try a US IP 🙂

          We’ve had interesting results with Asia where different subnets give different latency/etc.

          Francisco

      • M.M's avatar
        M.M

        Honestly It’s more important to me -by now- of a host to stay alive as long as it can, and be serious in this regard (staying alive). Whether they have 100 clients and giving them 5 star service, or 10000 clients and giving then 0 star care.

        • Francisco Dias's avatar
          Francisco Dias

          This was a triumph.
          I’m making a note here: HUGE SUCCESS
          It’s hard to overstate my satisfaction.

          Aperture Science:
          We do what we must because we can
          For the good of all of us
          except for the ones who are dead

          But there’s no sense crying over every mistake
          You just keep on trying ’till you run out of cake
          And the science gets done and you make a neat gun
          For the people who are still alive

          I’m not even angry
          I’m being so sincere right now
          Even though you broke my heart and killed me
          And torn it pieces
          And threw every piece into a fire
          As they burned it hurt because I was so happy for you!
          Now these points of data make a beautiful line
          And we’re out of beta, we’re releasing on time
          So I’m glad I got burned
          Think of all the things we learned
          For the people that are still alive

          Go ahead and leave me
          I think I prefer to stay inside
          Maybe you’ll find someone else to help you
          Maybe Black Mesa
          That was a joke, ha ha, fat chance
          Anyway this cake is great, it’s so delicious and moist

          Look at me still talking, when there’s science to do
          When I look out there it makes me glad I’m not you
          I’ve experiments to run, there is research to be done
          On the people who are still alive

          And believe me I am still alive
          I’m doing science and I’m still alive
          I feel FANTASTIC and I’m still alive
          While you are dying I’ll be still alive
          And when you’re dead I’ll be still alive
          STILL ALIVE
          still alive

          Sorry, it was just way too tempting 😀

          Francisco

        • M.M's avatar
          M.M

          Well this really nice and lightening fast work!

          You are unique in many aspects 😉 Wish you long life :p

          Wondering do we have any new stock soon? Or as already announced early?

          Thanks 🙂

        • Craig's avatar
          Craig

          see.. ?

          they are even a good buddy when you needed it in the bar :o)

      • Antony's avatar
        Antony

        Wow, what do you do with all those buyvm boxes cd?

  11. Battletux's avatar
    Battletux

    Man…. Why do I always miss out on the best deals…. Oh well.

    • Logain's avatar
      Logain

      The Pony (BuyVM) needs a waiting list for a little ride (cheap, reliable hosting), so the ones that do not have a neat VPS with them can get a chance. Then again, I fear that even the waiting list would be flooded within hours *sighs*.

  12. c's avatar
    c

    I reinstalled with the debian6 template just now. I found that this template had the apache installed already. Why don’t you provide a pure system like the ubuntu 10.04? (I would like to ubuntu 10.04 but due to the futime issue I cannot use it until it is fixed)

    • Francisco Dias's avatar
      Francisco Dias

      We’ll be stripping our templates down more at a later time but for now we’re focused on other things 🙂

      check out 10.10? I checked the futime on there and it was fine. In fact I thought it was fine on the new 10.04 template as well.

      Francisco

      • c's avatar
        c

        I have talked to Anthony about the ubuntu 10.10. But I cannot choose 10.10 since its support will be ended next year.

        I like ubuntu, but I would not use any ubuntu other than LTS version as the server’s system since the support period of non-LTS version of ubuntu is too short.

        Finally, I chose debian 6. Because debian is like ubuntu and it has a longer support period than non-LTS versions of ubuntu.

        The reason why I want long term support is that I want the security updates there even I have been using it for years. And I don’t want to upgrade or reinstall the system, since the upgrading and reinstallation will make my service down and I may have to reinstall many things.

        With LTS, I only have to reinstall/upgrade operating system every 5 years, not every 2 years.

    • cd's avatar
      cd

      Debian 6 was very minimal for me the last time I installed it, @Francisco: did that change?

      • Ben's avatar
        Ben

        It obviously changed not long ago as I remember something about “Cedric’s bravery”… No idea who is that Cedric by the way.
        I’ll be re-using that Debian 6 template very soon again…

        • cd's avatar
          cd

          That was just a description change and not a template change – I heard they updated the Debian 6 template from the OpenVZ repos and OpenVZ thought it was a “good idea” to include apache/extra crap.

        • cd's avatar
          cd

          and it could be possible that OpenVZ thought…*

      • Francisco Dias's avatar
        Francisco Dias

        The template we had was a little bit quirky so we put the official template on for now 🙂

        When I get back from my vacation I’ll be re-rolling the majority of our templates.

        Francisco

        • cd's avatar
          cd

          How so? It was working perfectly fine for me

        • Ben's avatar
          Ben

          @Francisco: Happy vacation then 🙂

          But when you come back, don’t forget a clean Debian 6 template 🙂

        • Francisco Dias's avatar
          Francisco Dias

          There’s a lot of work going on while i’m out, namely a few extra nodes going online 😉

          We’ll have a good supply of 128’s we didn’t sell on the previous round.

          Francisco

        • drmike's avatar
          drmike

          Could have sworn that you had sold them all out. Returns and/or cancellations?

        • Francisco Dias's avatar
          Francisco Dias

          We’ve been reclaiming gear from our Frantech brand at this point to cover the last of our BuyVM sales 🙂 There was some gear that I didn’t originally mark for reclaim but after a fairly lengthly audit/reorganization I was able to reclaim another 2 – 3 128MB nodes. They won’t last long as usual but that’s how it goes 🙂

          We’ll have:

          – 2 128MB nodes based on L5320 CPU’s
          – 1 128MB node based on L5420 CPU’s
          – Either 2 1024MB/2048MB nodes based on 5520’s
          – Around 50 – 70 512MB’s left over

          The 128MB’s will have adaptec 2405 RAID cards and the 5520’s will have 5805 (8 disk) arrays.

          At this point we’ll judge what would work best for the 1024’s. For all we know we’ll take one of them for 256’s.

          Thanks!

          Francisco

  13. john's avatar
    john
    [root@ ~]# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 519.239 seconds, 2.1 MB/s
    
    • cd's avatar
      cd

      No. BuyVM and YardVPS are different companies.

      • zlotowinfo's avatar
        zlotowinfo

        both out stock, isnt strange?

        • Jamie's avatar
          Jamie

          No, not really. YardVPS has been concentrating on their Xen products for a while. They’re the budget brand of PhotonVPS.

        • LEA's avatar
          LowEndAdmin Author

          YardVPS’s OpenVZ offerings have pretty bad performance (from what I heard) and I think they have shifted their focus on Xen and back to their PhotonVPS brand.

          BuyVM just chose to slowly build up their fleet of servers.

        • Jamie's avatar
          Jamie

          I have one and I/O wait is the problem. The base plans have their I/O priority to the lowest which mainly causes it i believe.

    • Francisco Dias's avatar
      Francisco Dias

      @rm – 95% of our clients don’t read their emails for even invoices much of the time, why are they going to read for announcements?

      We post them and do our best to keep all mediums up to date but with me out of town the guys are doing their VERY best. At this point you’re just griping about every little thing and i’m more than happy to pro-rate you a refund, even the amount that LEA paid forward for the raffle last year.

      We do our best to keep things informed and minus the issues with the pony4 kernel we had fairly long periods of solid uptime on 95% of our nodes so I’m really not sure where your concerns are. If we were so terrible we’d not have the sales rushes we do.

      Our 128MB plans make up a fairly small amount of our inventory so please, either take it up in a ticket as it’s non important or just request a refund because i’m about done.

      Thanks,

      Francisco

      • cd's avatar
        cd

        Delta: Now poof and enjoy your time off, you deserve it 🙂

        • Francisco Dias's avatar
          Francisco Dias

          Haha nah I’m just handling emails and such to make sure everything is OK 🙂

          We rolled the new kernel which was, for the most part, good. Node28 really didn’t like it so Solar already logged a bug report with OVZ about it.

          Francisco

        • cd's avatar
          cd

          Ahh, glad to hear :). New kernel indeed, I hope everything works out for the best with the new one 🙂

      • rm's avatar
        rm

        I am not griping about little things, guess I am just used to you being perfect, so little things like these do really stand out and surprise in a unpleasant way. I do not want refunds, on the contrary I was leaning towards prolonging both VPSes instead of just one, but when it’s almost every week either this or that, perhaps I’ll need to think twice, for example this is past 14 hours of node 28: http://ompldr.org/vOGo4dw/2011-05-03T101333Z-buyvm.png Posted a ticked, logged in to IRC, what else can I do to have this fixed?

        • Francisco Dias's avatar
          Francisco Dias

          node28 is pretty much the only node that didn’t take kindly to the pony5 kernel. For some stupid reason it just simply stopped routing packets at one point and was spiking like mad for really no reason. At this point the guys rolled it back to pony4 and it has been stable with minimal/no CPU spikes as far as i’ve seen.

          It’s hard for me to diagnose any sort of lag since the line i’m on in Vegas is GARBAGE (a shared T1 for the entire Luxor as far as I can tell) but I’ve been keeping tabs with the guys to make sure i’m not being crazy.

          We’ve worked hard to work out any last quirks and will continue to do so. We’re disappointed that the pony5 kernel didn’t take properly on 28 but at this point we’ll just leave 28 alone until we get more word from OVZ about the potential bug.

          Fran

        • rm's avatar
          rm

          So a conversation with the BuyVM billing staff member ended in me receiving a complete refund for both of my VPSes. While it has been handled less than satisfactory in some aspects, I must say it is quite impressive that you can admit your problems and provide refund even in some circumstances where it is not per your written ToS.

          Since I am in Russia and due to PayPal restrictions can not actually receive payments, I have chosen the refund amount to be directed as a donation to the Free Software Foundation (FSF.org) instead.

  14. macro's avatar
    macro

    Don’t you know their tactic?
    If server goes down, before they tried to fix it, they write announcement first, as if it was scheduled, in order to cover their behind.
    And if you complain, they point to the announcement, what a tactic!
    Their mindset is weird, as long as there is announcement, server can go down one week.
    No problem, is there? They announced it.

  15. macro's avatar
    macro

    Edit by LowEndAdmin: @macro — watch your language.

    • Francisco Dias's avatar
      Francisco Dias

      Hi Macro,

      The only time announcements get posted after the fact is if the box is acting up and requires us to reboot it/etc.

      Any sort of planned maintenance work gets announced 24 hours before it gets done, as per our SLA/FAQ page. You’re welcome to ask one of the 120+ clients in our IRC channel for logs of the bots announcing it showing that we’re not editing dates.

      Francisco

  16. Mykeul's avatar
    Mykeul

    My VPS and manage.buyvm.net are down … no support ticket possiblity ..

  17. Adam's avatar
    Adam

    They are probably updating some network routing / DNS — They will be back within few minutes.

    • NordicVPS's avatar
      NordicVPS

      frantech.ca -> 205.185.112.61 -> Colo4 in Dallas
      buyvm.net -> 205.185.112.61 -> Colo4 in Dallas
      manage.buyvm.net -> 72.52.127.128 -> EGIHosting in California

      So most likely some delayed DNS updates ..

      • NordicVPS's avatar
        NordicVPS

        My bad – Colo4 is at our end – so basically no route to host from our terminal

  18. Hudson Nguyen's avatar
    Hudson Nguyen

    Could anyone post a notice when your VPS back online please ?

  19. Adam's avatar
    Adam

    Did anyone check their announcement site?

    • NordicVPS's avatar
      NordicVPS

      From HE.NET: A utility failure is affecting much of Fremont, CA, USA, damaging some electrical equipment. We are currently working to restore power.

    • Tom's avatar
      Tom

      I see you have time to post that Fremont has been down, but when will you get around to the crashed servers reprovisioned? The last email I have received from you is from April 25.

    • cd's avatar
      cd

      It’s not really their fault, but they are working on getting everything back from the power failure 🙂

  20. Francisco Dias's avatar
    Francisco Dias

    HE had a large scale outage in Fremont 1 again it seems :/

    We’re spinning everything quickly.

    Francisco

    • innya's avatar
      innya

      3 datacenters including HE got an outage between late last night and this morning US time.

  21. GeekInFinance's avatar
    GeekInFinance

    One of my BuyVM VPS is back, the other one is not…looks like ponies are working on it 🙂

    • Francisco Dias's avatar
      Francisco Dias

      We got an announcement up documenting each node that’s still pending some work. At this point some are just in the middle of a FSCK but a lot of them are online.

      Francisco

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 *