New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
Why not start with a module based ovz panel? Might be a bit tough though.
All of the free panels (that I've seen) mainly focus on a single virtualization type and the entire codebase is completely built around it. In order to do something that uses plugins for each platform type, we would have to start from scratch and build a basic schema first.
@subigo why not take a look into something available like proxmox?
Proxmox is what it is, and is too dependent on hardware control, doesn't support swraid,and running KVM and ovz on the same node is just scary. OVZ inside KVM I am ok with, I don't do it, but it at least makes sense.
Why not adopt http://cvm.cryto.net/
@miTgiB
Well I said for an idea not to use it.
I was actually already working on an open-source panel (http://cvm.cryto.net/ for code, http://demo.cryto.net/ for a somewhat outdated demo, http://git.cryto.net/repo/cvm for publicly accessible Git repository), and I expect to pick up development again in 1-2 weeks (I had some IRL issues to sort out).
I'm implementing OpenVZ first, but due to the way it's designed it will be easy to add in other virtualization technologies. While it technically won't have a 'plugin system' like for example forums do, the code is sufficiently abstracted for it to allow easy integration of other virtualization platforms.
One of them had decentralized management where any node could control the others, handy if you want to take a master down, but still be able to provision.
I'd be willing to contribute some time/design if required by I can't say I'm good at coding
Adding on to those who suggested modular:
Instead of creating a core framework, create a core API.
While creating an API over a framework may be trivial, I believe it will allow the most customization.
By doing so, you're allowing those without a lot of coding skills to create their own panel, or customize the existing one. (By only reading the documentation).
Use Git and Github. Easier to sync development, track commits, and most important of all blame log (find out who changed what).
Well shit there goes my ninja ad's inserted into the code ;0)
Any naming Ideas anyone?
The statement "support is terrible" could not be more true with SolusVM. I think I've ended up figuring out/fixing every issue except for one that I've had on their system. And that was after multiple, worthless ticket replies
Less talk, more code.
By doing so, you're allowing those without a lot of coding skills to create their own panel, or customize the existing one. (By only reading the documentation).
You mean libvirt?
How can you use that as a base? The source isn't available is it.
I assumed it was, though I didn't actually take the time to download it..
Its not for download, its just a web based service that people can use to manage KVM VPS anywhere.
My bad. I could still create something from scratch, but I'm trying Cloudstack now which looks really good. I'll reply to your PM in a moment
.
Any one with a nice name?
Wouldn't Cloudstack + WHMCS integration fit your needs? Cloudstack is open source.
Yes I'm discussing this with @GetKVM_Ash - we might develop something between us.
I'd really love it if we could all come together and put something together that we can all build off of. Some of us cant code, but we can do graphics or provide hardware or cash for testing & development.
@FRCorey Agreed!
Which was basically the original idea of CVM
After spending most of my afternoon reading the Cloudstack docs, I do not see this as a good fit for the LEB provider. No OpenVZ support, not a huge issue, but running two control portals is not ideal. No IPv6 support, but that seems to be on the roadmap for the next version due soon. Template support for KVM is a huge plus, NFS for storage is a huge minus.
Those are just a few of my takes from my reading.
Here's nother name, NodeRunner?
BTW, Cloudstack is not a great solution for most of the LEB providers.
node.vm