I was digging through the system logs on the MS SQL server (SQL 2000 on a Windows Server 2003 virtual machine running inside VMWare ESXi on a monster Dell box; very cool. I’ll write about it sometime…) at work just now, trying to track down a goofy slowdown that happens on occasion, and came across this System Event…
The system uptime is 1339878 seconds.
That’s one million, three hundred thirty nine thousand, eight hundred seventy eight seconds, which translates into 22,331.3 minutes, or 372.1883 hours. Which, as everybody knows, is the same as 15.5079 days (rounded to four decimal points.) Or it could be expressed as 15 days, 12 hours, 11 minutes and 18 seconds (as calculated by a quick formula I threw together in Excel.)
Filtering through the System Events, I can see an eventlog entry for every day and every time that service had been stopped or started since November, 2007. The machine has been in operation for much longer than that (probably since 2004), counting the seconds that go by one. At. A. Time… Day in and day out. It’s useful information to be sure, but why display the time in seconds? Couldn’t the geeks in Redmond be bothered to modify that to show the time in days, hours & minutes instead? It’s a computer, for crying out loud, and most computers have plenty of extra capacity.
Seeing the time in seconds — especially in numbers that big — is absolutely meaningless to me. It’s not like running a calculation to make the time count easier to read would tax the system much… A holdover from an earlier time when every processor cycle was counted as precious? Or just an item very low on the priority list? Or maybe I just have too much time on my hands today?