Tanner's update (Week of 17 August)


(Tanner Filip) #1

I won’t be able to make it to any of the meetings this week, and probably next week either. So here’s a quick rundown of what I’ve done the last week. Sorry that I didn’t communicate this very well before/as I was doing it.

  • Crowd - Atlassian’s SSO software. This will let us have one sign-on for Jenkins, Jira, Confluence, and ideally whatever monitoring we choose to go with. It isn’t currently hooked into anything but Jenkins, I’ll probably be importing Jira users in the next week or two, once we have a security check done.

  • Jenkins - We’ll be using Jenkins for a lot of fun stuff. For one, it’ll be how we deploy MEGADISCOURSE. It’ll also be used to trigger Ansible runs, and probably some other things in the future.

  • Monitoring - I still think that we’ll be best off using Nagios for monitoring, especially because something for Megadiscourse will require very custom checks and notifications, that I’m fairly certain no hosted monitoring service can provide. I’m working on setting Nagios up, but have no ETA right now.

  • Ansible - I haven’t worked much on Ansible the last week, but all the basics are there and I personally believe it’s ready to deploy to production servers. Right now it’ll install shells (tcsh and zsh), add users for the core team, add them to groups, and update sudoers to allow our users to sudo.

That’s about it. I’ll probably be a lot less available from today on through the fall; Fridays are pretty open for me, so if you need something from me moving forward, that’ll probably be the best day to get me.


(mrz) #2

Datadog does :slight_smile: AppDynamics supports Nagios checks too!


(mrz) #3

What’s the overall strategy for Ansible?

More so around the deployment model. I’d lean towards the open source version of Tower.


(Michael Buluma) #4

+++