The folks at Dreamhost, which is the provider that runs this site, have had a bad month or two. Between router problems, power problems in their datacenter, and people whining about how their baby sister could run a network better then these folks could (sorry, I doubt it). But, with their last upgrade in their datacenter which entailed a rather expensive chunk of Cisco gear, I saw something I haven’t seen in awhile on both of my shell accounts:
zagnut:~$ uptime
19:12:38 up 34 days, 14:35, 4 users, load average: 1.28, 1.46, 1.40
[laurel]$ uptime
19:14:11 up 3 days, 4:40, 4 users, load average: 1.78, 2.31, 2.34
I guess when you’re used to seeing the load averages anywhere from 100 to 300, seeing it below two is kinda scary. But it also shows just how much network issues were causing problems (if I had to guess, and I do since I don’t work there, I’d say NFS issues due to processes deadlocking on I/O).
Whatever it is/was, good job nailing it down. Now go sleep. I don’t think any of you have in at least a month.