On Sat, Apr 13, 2019 at 5:08 PM Bryan Fields <bryan@flscg.org> wrote:
On 4/11/19 8:03 AM, Peter Canada wrote:
> Do we need hardware to run it on or is there already infrastructure in
> place to run those services?
>
> I would be interested in supporting the community by assisting with these.

On 4/11/19 8:06 AM, John Pugh wrote:
> And while i'm not as active as I'd like to be in this project overall, I
> would definitely lend a hand with grafana and the config management side
> even if it's with ansible ;)

Peter/John

I/we have a colo box with several VM's on it already, and I can spin up
another for this.  Was thinking like 3-4 cores, 16gb ram and 500g of disk
should run libreNMS, greylog and grafana nicely.  We can use it as a
staging/config management server too (ala rancid).

I think most of this can likely run in containers too which would simplify
things.  Are you guys familiar with Docker?

The hamwan.org guys have some examples of what they run, and we can steal some
of their ideas.  Really almost anything is better than what we have now :)

Hey Bryan,
I'd probably want to lay out the config to ensure those specs will work, but it *should* be ok.
Very familiar with container technologies (docker, cri-o, k8s, etc). All of the required apps (librenms, graylog, elasticsearch, mongo) have docker images available (all with different OS bases, unfortunately).
Not sure if you have config in a VCS, but I setup https://github.com/zoopster/flscg-observe for this project in my personal github account. It's empty at present, but feel free to submit a pull request. I'll start working on a terraform config based on what I know, the easiest way to share detail would be through issues in github if you're so inclined.

JP