(no subject)

[robbat2]

So I got in to TechBC this morning, dropped off my fees, and went in to work some. Grabbed a 4.3 Gb hard drive from Dave downstairs, pulled off the controller board to fix the one pin on the IDE connector, as we think that is what the problem is. Fixed it, stuck it on sloth. No Go. It doesn't work. I go back downstairs to find Dave to tell him about the hard drive, and the power goes out.

Laptops all around bleat loudly, complaining of their interuption in the recharging cycle. Only one emergency light comes on, and there should be a lot of them. We grab a small flashlight, and head in to the server room, shut down the LCD displays to conserve UPS power. The UPS says 142 minutes at first. We head upstairs, and I grab the big flashlight we have in the fakespace. We head off through the bowels of the mall to the other network cluster. It's getting quite warm in there already, so we leave some doors open to keep it cooler, and head back to the main server room.

In the 15 minutes we were away, the temp in the server room has gone up to 37 degrees, and temp alarms are going off. We shut down a load of non-essential servers (All of the admin set, plus the disk arrays and the tape backup library). It starts to cool off now, as those are the big heat producers there. UPS runtime is up to 190 minutes now. We leave there for a bit to go and find the ET&L people to ask which of their servers to shut down. We miss them upstairs, and go back downstairs to find them. We walk into the server room, and the power comes back.

Ah! Air conditioning! In future, there needs to be at least an exhaust fan on the UPS circuit, to stop the servers overheating totally.