We promised we'd at least try to get some audio and video from UTOSC up as quickly as possible, so I'm sitting here at my computers waiting for my Windows computer to write a modified 23GB AVI so I can start editing the video for the keynote presentations tonight.

So, while that slow process continues, I'll write a little about how today went.

Well, for me, it began very STRESSFULLY!

I went over to Salt Lake Community College (where UTOSC is being held) on Wednesday afternoon to get things set up for shooting video starting Thursday and to help with other UTOSC-related tasks as I am a UTOSC core team member. At 4:30 p.m., a client called me to tell me their server just went down. Right after the call, the Nagios alerts came into my phone saying the same thing. Nick was with me, so I sent him out to get their server back up while I continued setting up equipment. I figured it just needed to be powered back up or something, but we weren't so lucky this time.

This particular server has been pesky and super-sensitive ever since we installed it, making it an annoyance for both us and the client. We weren't really every sure what the problem was, but I strongly suspected the motherboard was just bad even though it worked most of the time.

Nick couldn't get the server to do much. It would love the RAID controller BIOS and then reboot, or it wouldn't display anything at all on the monitor. Finally, I told him to just remove the server and bring it to me and I'd work on it later at home.

Later, when I got the server to my house, I could not get it to do anything. It would power up, but would not POST. I tried all the usual tricks: removing the power cables, disconnecting the motherboard power connector, resetting the CMOS power jumper, chanting a voodoo chant. Nothing worked, so this morning, I made replacing the motherboard my first task. I had hoped I'd be able to get it done quickly and still make it to SLCC to be of some help in the preparation for UTOSC to start at 12:30.

I made it to Universal Systems around 9:00 and they had one socket 1207 motherboard in stock, a Supermicro H8DME-2 dual-processor board. I guess I was pretty lucky they had one. I knew USI was more of an Intel shop, but I thought they'd have more than one AMD board for sale. Lucky for me, they had one. It wasn't cheap, but it was a Supermicro so that's generally a good thing.

I took the server and the new board back to the office and proceeded to install it. The Supermicro board was an EATX board which means it's about as huge as a motherboard can be. The I-Star case I was installing it in could take an EATX motherboard, but it was a tight fit. It took me about an hour or so to get the new board in, everything connected, and powered up. The box didn't have a manual in it, so I downloaded a PDF and printed off the necessary pages for jumpers and connectors.

The LSI Logic RAID controller really slowed down the boot process. I eventually just yanked it out of its PCI-X slot so I could get through BIOS and boot-up issues without waiting.

The Supermicro motherboard had a different onboard SATA chipset than the old board, so I had to install a new initial RAMdisk (initrd) for the Linux kernel. The server was running Fedora Core 6, which I didn't have any media handy for, so I downloaded a rescue disk ISO and burned it to a CD. I ripped a CD drive out of an old desktop so I could boot to the rescue disk. This, of course, all took a little time... more than I anticipated. Finally, I got the system booting by getting the new initial RAMdisk installed by way of the rescue CD. Then, I realized I had to reconfigure the networking for the server because it used a bonded ethernet configuration. All the ethernet addresses would be different, so I had to go through a tedious process of making Fedora Core forget the information it had stored about the previous ethernet ports and learn about the new ones. Finally, I had a system that was ready to go back to the client and it was about ten minutes before noon.

Things went relatively well at the client's office. I had to do a couple other little things to get things working the way they should, but I was out of there shortly after 12:30. All the hustling made me a little shakey, so I hit a local Maverik and got some hot cheesy bread. I made it back to SLCC a little before 1.

Matt Asay was well into his presentation, but Nick had both cameras rolling and I stepped in on one and took control.

Everything else throughout the day went, I thought, very smooth. We shot video for Nathan Blackham's Nagios presentation and would have shot video for Jared Smith's Asterisk presentation, but it got moved to Friday. As a result, we had a little extra time and I would have rounded up a couple people to do some on-camera interviews, but I didn't bother to shave and looked like a wild man, so we didn't do that. Instead, we loaded up the equipment and moved over to the Student Center to get set up for the evening keynote presentation.

It was good we headed over there early. It was more work than either Nick or I expected packing our equipment up, moving it, and setting it back up, so we learned a lot from that.

Dinner was pretty good. More people should have attended the dinner and the keynote presentations. A lot of people did, but I still saw empty chairs. It seems like the SLCC students didn't make it out en force to the dinner and they should have. Free food!

I got home a little after 10 p.m. and started working on this video. Now it's about 12:30 and I'm done talking about my day and this video conversion thing is still going. We'll have to see if I have the patience to get this out tonight. If nothing else, I'll get audio from the presentations to someone to make them available.