Cluster of Sandbox VMs

This topic contains 2 replies, has 2 voices, and was last updated by  Todd Wasson 1 year ago.

  • Creator
  • #47934

    Todd Wasson

    Hi everyone, I’m looking to build a small cluster of Virtualbox Sandbox VMs to test out Storm, and I’m running into some annoyances. To do this, I’ve changed the networking mode of each VM to host-only networking, with the intention of giving each one a unique fixed IP address and hostname and put all of these in each VM’s (and my host’s) /etc/hosts file. However,

    a) when I edit /etc/hosts in the VMs, those changes immediately go away upon a reboot and I can’t figure out what’s overwriting them
    b) when I set the hostname via /etc/sysconfig/network the apache server complains at boot that it can’t find the fully-qualified domain name (even though I’ve provided it)

    I suspect there might be more annoying things going on as well, but these two are immediately problematic. When I grep for sandbox in /etc/, I see there are a ton of places where the hostname is hard-coded as, so I guess my larger question from this point as well as the specific problems above is: if I’d like to have a small cluster of sandbox VMs, what’s the best way to go about it? Secondarily, if anyone can help with the above problems, I’d appreciate it.

    Thanks, and sorry for gratuitous rambling.


    PS- I did find a previous topic regarding this here but it was both outdated and not especially helpful.

Viewing 2 replies - 1 through 2 (of 2 total)

You must be to reply to this topic. | Create Account

  • Author
  • #47950

    Todd Wasson

    Hi Dave, thanks for the advice, and for saving me from going down a rabbit hole. I’ll look into going that route instead.




    Hi Todd,

    You would be better off installing your own 4 Node cluster using Ambari.
    The Sandbox wasn’t designed to be reverse engineered and forced into a cluster – the logistics of it would be much longer to do this than actually to build it yourself.



Viewing 2 replies - 1 through 2 (of 2 total)
Hortonworks Data Platform
The Hortonworks Data Platform is a 100% open source distribution of Apache Hadoop that is truly enterprise grade having been built, tested and hardened with enterprise rigor.
Get started with Sandbox
Hortonworks Sandbox is a self-contained virtual machine with Apache Hadoop pre-configured alongside a set of hands-on, step-by-step Hadoop tutorials.
Modern Data Architecture
Tackle the challenges of big data. Hadoop integrates with existing EDW, RDBMS and MPP systems to deliver lower cost, higher capacity infrastructure.