Link to Original Source
Slashdot videos: Now with more Slashdot!
We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).
Link to Original Source
Actually you will just have twice the time to download from torrent seeds. "Free broadband market" will take care of this.
That's not how it generally works in the real world in my experience. Whenever an Ubuntu release happens it almost always puts our netwok links to the max. I can't imagine doubling that. A lot of people still download directly and don't bother with torrents.
I really did expect to see mention of CSOS in Lance's slides, but I digress.
That predates me by quite a bit and had never heard that story before. I'm surprised John Sechrest never told it to me when he was still living in Corvallis. Its very interesting to hear that though! I will have to remember that the next time I give this presentation. Thanks!
Also you guys are majorly dragging your feet on the Puppet thing. UO has been on it for about three years, and it is leaps and bounds better than CFEngine. It doesn't seem to help that a large portion of your staff are students with little background in system administration - not only do you have to teach them the tools, you also have to teach them the practice.
Oh I agree. Its been quite frustrating not getting very far on our puppet migration. And it has been difficult to manage considering most of our help is from students whom we have to teach/mentor at the same time. I wish we could hire more full-time sysadmins but we simply don't have the money so we have to make due with what we have.
The "legal reasons" alluded to are mostly problems with other signers on the contract for our upstream bandwidth provider. *coughDuckscough* At our bandwidth scale, tunneling is not feasible.
Indeed, tunneling IPv6 at our scale would be quite silly.
We don't run Puppet at the moment, we run CFEngine. Everybody's receiving Puppet training and there's a slow-yet-steady migration to Puppet, but these things take time. There are quite a few people depending on us to not fuck up, so we don't change our stacks without deliberation and testing.
We've been using CFengine since nearly the day we started so we have a collection of CFengine recipes that go back 5-6 years. Its going to take a while to get everything in a state considering there's only one full-timer (me) and 4-6 undergraduate students. Granted we're working on just getting a bootstrap set of modules done first (which is almost completed). Additionally we're writing our modules so that they are reusable (which takes more time) for other people and plan to post them on puppetforge eventually.
Trying to work on that plus keep up on regular maintenance, new projects, misc fires, conferences, etc all adds up. We do a pretty good job of keeping up on tasks but we do fall behind sometimes. We take pride that we run a pretty tight ship and want to continue that moving forward.
(keep in mind, they have a mirror in the midwest provided by (I think) TDS)
That is correct. We have two FTP mirrors hosted by TDS (Chicago and New York) in addition to the systems we have on campus in Crovallis.
You can see their bandwidth utilization here: http://ftpmap.osuosl.org/
Keep in mind that the Corvallis server is out of rotation currently because of some hardware issues.
From Apache to just about every Linux distro you've ever heard of, they run a mirror for it.
I max out my considerable downstream connection from them frequently. These are cool people doing a pretty cool thing.
What's funny is we sometimes get abuse emails from ISP's complaining that we are DOS'ing them when in fact its their users just using our mirrors.
Link to Original Source
The project you saw at OSCON was called RAIV (Rack And Inventory Viewer). Unfortunately it ran into a dead-end and is currently vapor-ware. Currently we're using an internal CakePHP webapp for basic inventory and customer tracking, but its very buggy and lacks many features.
We are in the midst of working on a completely new project that will cover many of the problems mentioned in this article and beyond. Think of it like an open source datacenter management webapp and backend. Its still in the planning stages, but the intent is to have a plugin based system where you can use the inventory plugin, DNS/DHCP plugin (to replace maintain), virtualization management (deployment and console access), etc all in one interface. The idea is to create an admin interface and a customer interface so that they can access and see information about their environment. We're far from having a demoed project but we hope to have something soon.