Re: Is anyone using proxmox in production?


Good feedback everyone, I appreciate it. Just to clarify, I have been a linux user since the 90s, and in addition to my service business I work as a DevOps engineer for an enterprise, we build systems at scale with kubernetes in both private and public clouds. 

My question was more centered around the reliability of proxmox in a very small business and from the standpoint of a small business service provider who doesn't want to have to fiddle with it much. Was there any reason to replace it. I am getting the feeling like there is no reason to replace it other than the ford vs chevy argument.

Again, thanks for the feedback, that was what I was looking for!
Book time with me here:
     ___ _______ 
    |   |       |
    |   |    _  |
    |   |   |_| |
 ___|   |    ___|
|       |   |    
JP (Jesse Perry)
voice/text: 907-748-2200
email: jp@...
support: helpdesk@...

On Fri, Nov 19, 2021 at 11:34 PM Arthur Corliss <acorliss@...> wrote:
On Fri, 19 Nov 2021, Jeremy Austin wrote:

> I can both agree and disagree, respectfully, with Arthur. It?s entirely
> true that one can set up HA for cluster state and roll clustered storage by
> hand. I?ve done it myself. But I disagree about free. My time is worth too
> much (well, worth too much to me), and someone else will come after me ?
> there?s something to be said for Not Maintained By Me.
> And frankly if one has less linux-fu, then arguably ESXI ? or even a ?cloud
> provider? (dirty phrase) is a better choice than Proxmox these days. The
> ?web gui? functionality is the least useful part of Proxmox for me.
> Free is only free if one ignores entire swathes of hidden costs. If it were
> truly worth rolling my own it would be worth competing against the guys who
> can  productize it; if what I want, however, is something they don?t make,
> then it is not a fair comparison and my only choice is to roll my own.
> Not to dismiss Mr Corliss? perspective; a solution is only a solution if it
> fits one?s use case.

You made the case that it had value for medium to advanced Linux users, but
I disagree, it's value proposition is precisely in the beginning Linux
users.  Advanced users won't find it that time consuming to roll out a DFS
and a more open standard infrastructure without licensing costs, moreso when
you consider any decent level of automation and/or config management system
can keep all that running, and scaling for years.  Heck, if you're going to
pay licensing costs, a good CMS is going pay a much higher ROI across the
bar than proxmox.  And that's before you consider that learning proxmox is
learning a toolset that's not used anywhere else.  You're much more likely
to run into a libvirsh-compatible infrastructure out there than proxmox.

At the end of the day, it's just math, and I don't see the business case for
it.  They've done a good job with it, I'm not dissing the quality of their
work, but it's just another apple that you're paying a premium for, not an
orange.  I know of no market-differentiating feature set that they provide.

Would I pay proxmox before I paid vmware?  Certainly.  Proxmox is a viable
commercial alternative.  But for a basic HA virtualization platform without
orchestration, I wouldn't pay for either one.  The open source alternatives
have matured enough to destroy that business case.

        --Arthur Corliss
          Live Free or Die

Join to automatically receive all group messages.