Found the cause of mysterious load average

[Warning: extreme geekiness ahead! Non-geeks might want to skip over this post and look at the pretty ponies instead.]

I figured out what was causing my home Linux server to think it was busy. It turns out my server has an automounted nfs share from another server, a server which happened to be down. I’d done a few df commands to check disk space before I downloaded the new Ubuntu release and these commands hung in uninterruptible sleep while they waited for the drive share that would never be found.
Continue reading

Looking busy

Just looked at my home Linux server, which from all appearances is working fine. The top tool, as well as uptime both show the load average to be 16! Yet, the CPU is listed as 96.7 % idle, as show below:

top – 13:16:58 up 61 days, 16:12, 3 users, load average: 16.00, 16.01, 16.00
Tasks: 267 total, 1 running, 264 sleeping, 2 stopped, 0 zombie
Cpu(s): 0.7% us, 1.0% sy, 0.0% ni, 96.7% id, 0.0% wa, 1.7% hi, 0.0% si
Mem: 450988k total, 447500k used, 3488k free, 121676k buffers
Swap: 987988k total, 33216k used, 954772k free, 170976k cached

I’m not sure what is going on with this server. Its actually idle but doesn’t think it is. Very strange.

cat /dev/funkywhiteboys

This may be the funniest geek humor I’ve ever seen. I was building the OSX plugin for the XMMS audio player just now when I got this ‘error’:

checking pthread.h usability… yes

checking for GTK – version >= 1.2.0… yes
checking for xmms-config… /opt/local/bin/xmms-config
checking for gawk… (cached) awk
checking for XMMS – version >= 1.2.0… yes

checking for white_boys in -lFunkyMusic.so… no
Warning: You may not have enough white boys to play your funky music.
Please obtain additional funky white boys.

configure: creating ./config.status
config.status: creating Makefile

Kudos to Bob Dean! We’ll see if it will work in spite of these limited resources. 🙂

[Update]
Must not have enough funky white boys: the plugin segfaults. 🙁