# Interworx

**URL:** <https://forums.interworx.com/t/interworx/13048>\
**Category:** General Inquries\
**Tags:** dns, availability, round-robin, cluster, up-time\
**Created:** [November 22, 2011, 10:27am UTC](https://forums.interworx.com/t/interworx/13048 "2011-11-22T10:27:29Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![anthonycl](https://forums.interworx.com/user_avatar/forums.interworx.com/anthonycl/32/54_2.png) [@anthonycl](https://forums.interworx.com/u/anthonycl)\
**Post date:** [November 22, 2011, 10:27am UTC](https://forums.interworx.com/t/interworx/13048/1 "2011-11-22T10:27:29Z")

</div>

Hello,

I am looking for ideas and solutions to making my as reliable as possible. I am a pretty knowledgeable guy in regards to Linux but not focused for any service specifically. I am setting up my latest Interworx server on a kimsufi 24G (kimsufi.ie aka ovh.ie), however it will still be only one server. I am looking for solutions to make sure my websites stay up 24/7, or at least if the server goes down show a downtime page. I would like suggestions on my ideal solution:

Solution 1:

- Setup SolusVM or a VM solution
- Create VPS1: As the "Interworx Cluster Manager"
- VPS2: As "Cluster Slave"

Solution 2:

- One install of Interworx on server.
- Use Cloudflare for DNS, SSL, and Downtime pages.
- Turn Interworx into "Cluster Manager" if needs present themself, and purchase second server as Slave.

Thanks

---

<div class="post-metadata">

**Author:** ![IWorx-Robert](https://forums.interworx.com/user_avatar/forums.interworx.com/iworx-robert/32/15_2.png) [@IWorx-Robert](https://forums.interworx.com/u/IWorx-Robert)\
**Post date:** [January 18, 2012, 11:51am UTC](https://forums.interworx.com/t/interworx/13048/2 "2012-01-18T11:51:03Z")

</div>

Really, I’d recommend the second of your solutions. Running a VPS cluster isn’t a sensible idea - In case of hardware error, your machine would still be knocked offline.

Good news, though. Shortly we will have Cloudflare integration via a plugin.

---

<div class="post-metadata">

**Author:** ![zombie\_process](https://forums.interworx.com/letter_avatar_proxy/v4/letter/z/7ab992/32.png) [@zombie\_process](https://forums.interworx.com/u/zombie_process)\
**Post date:** [January 18, 2012, 1:29pm UTC](https://forums.interworx.com/t/interworx/13048/3 "2012-01-18T13:29:06Z")

</div>

HA and Load Balancing are two different beasts. The load balancing that interworx offers is great, but isn’t really providing HA. I played with setting up heartbeat/corosync and interworx, and eventually gave up after it became clear that there were too many knobs inside interworx I simply didn’t know about. The standard solution after that is to go with a VM solution that offers HA, get a few frontend boxes and an iscsi or FC SAN, and have the VM hypervisor manage HA for you. Such solutions aren’t cheap, though - unfortunately they’re way out of the price range I was given to work within ☹
