Health Monitoring with lm_sensors
lm_sensors creates files under /proc with sensor values in them. You can create a custom monitoring system by writing scripts that check the values from these files or parse the results of the sensors command.
The source distribution for lm_sensors includes a simple Bourne shell dæmon, healthd.sh, that sends e-mail to the root account if any sensors indicate an alarm state.
A bunch of programs are available that check lm_sensors values or display them in some cool-looking fashion, including applets for KDE and GNOME and several window managers. The resources section has some links to get you started.
I've been running the elaborate GKrellM, a GTK applet designed to make an impressive panel of monitors. The screenshot to the right shows it running on my system, with not only lm_sensors data but also with S.M.A.R.T. hard disk temperature data and the BubbleFishyMon plugin. (BubbleFishyMon represents memory load as the water level, CPU activity as bubbles and network packets as fish swimming back and forth. The rubber duck is just for fun.)
It takes a bit of work to get lm_sensors working on your system, but it's well worth it. You might catch a hardware malfunction early, before it becomes serious. Even if you don't have any problems, it's reassuring to be able to check up on the health of your system quickly.
Steve Hastings first used UNIX on actual paper teletypes. He enjoys bicycling, music, petting his cat and making his Linux computers do new things.
- The kernel doesn't really
11 hours 41 min ago
12 hours 12 min ago
12 hours 12 min ago
14 hours 17 min ago
- This should be very helpful
15 hours 31 min ago
- As much as I share your point
17 hours 50 min ago
- So girls had it better ?
21 hours 22 min ago
- Reply to comment | Linux Journal
21 hours 42 min ago
- why is GNOME 3 in the fifth position at 14.1 %?
1 day 3 hours ago
- Sublime Is Brilliant!
1 day 8 hours ago