camosoul
Well-known member
Since I'm now running my MNs on my own VM host machines, I wondered if anyone wanted to know about it.
I enjoy having a greater degree of control of the whole system and the hardware used.
As drive space requirements go up, it will no longer be so cost effective to run on VMs provided by service entities. It's unlikely that the tiny VMs we've been using will do the job, and it's cheaper to host VMs yourself on a 1U that is a generation or two old, than source that from a provider on larger footprints.
I have a good relationship with my VM host, and he just plain can't get the price down to where it is hosting on my own machine.
You can also take full advantage of IPv6, and skip all this IPv4 hogging, which is probably the biggest expense.
It's doubtful that Evolution will run on the smaller 256/384/512 nodes we're using now, and they definitely don't come with the drive space. It will no longer be cost effective to do things as we've been doing them. Why not be prepared instead of scrambling with stuff you don't understand at the last second?
It's really not that tough.
It's also a hell of a lot easier than running multiple nodes off of the same VM/instance. There's almost no config with 16.04 server. Doing everything manually, you can have a node up in a few minutes.
I enjoy having a greater degree of control of the whole system and the hardware used.
As drive space requirements go up, it will no longer be so cost effective to run on VMs provided by service entities. It's unlikely that the tiny VMs we've been using will do the job, and it's cheaper to host VMs yourself on a 1U that is a generation or two old, than source that from a provider on larger footprints.
I have a good relationship with my VM host, and he just plain can't get the price down to where it is hosting on my own machine.
You can also take full advantage of IPv6, and skip all this IPv4 hogging, which is probably the biggest expense.
It's doubtful that Evolution will run on the smaller 256/384/512 nodes we're using now, and they definitely don't come with the drive space. It will no longer be cost effective to do things as we've been doing them. Why not be prepared instead of scrambling with stuff you don't understand at the last second?
It's really not that tough.
It's also a hell of a lot easier than running multiple nodes off of the same VM/instance. There's almost no config with 16.04 server. Doing everything manually, you can have a node up in a few minutes.
Last edited:
