January 2012 Archives

FUDCon Blacksburg, day 0+1

| No Comments | No TrackBacks
FUDCon Blacksburg is getting off to a great start! I got in yesterday, and by some stroke of luck, two other FUDCon participants were on my flight to Roanoke - Justin O'Brien and Dan Walsh. Dan was at the back of the airplane, and I didn't see him until we got to the hotel, and we were both complaining about prop planes - which I think is the only thing that flies into the Roanoke airport. Being stupid, I checked my bag planeside in New York, and didn't get the bag planeside in Roanoke. They didn't seem that they were going to bring them to us, and I was at the front of the plane, so I went up to the gate and asked the gate agent where to get my bag, and she said that I needed to get it planeside, but I couldn't go back, so it would come on the luggage carousel. I figured that this would take forever, since I'm used to real airports, not those that consist of approximately 10 gates :).  Therefore, I was pleasantly surprised when my back showed up pretty much right away on the baggage belt.

Then I checked into the Inn, and met several other Fedora folks - Robyn Bergeron, Spot, Matt Domsch, Ben Boeckel, Jon Steffan, and lots of other folks were at the hotel. Stayed up til about 1AM hacking on the sysadmin study group stuff with Clint Savage trying to get a boxgrinder appliance up and running. I wound up not being able to boot the appliance for some unknown reason. I eventually figured out (just now) that you can't build a boxgrinder appliance with SELinux in enforcing mode - it won't be able to execute any %post scripts, and everything will be broken. Clint insists that he can build with SELinux in enforcing mode, but I question the veracity of this :)

Then today, I went to the Fedora Infrastructure hackfest, where we were going over the current and future state of the staging environment and trying to make it useful for things. Then I went for more boxgrinder hacking, and working with upstream in Europe to figure out why my appliances weren't working, but as mentioned above, I just figured it out now - but it took the majority of my day sadly. Once I figured out how to turn on debug logging in boxgrinder it became very obvious what the problem is.

We also had a board meeting, but there will be another post forthcoming on that.