An app to warn about fan failure?

[DisinfectedDuck]

Recently my HSF just died. Luckily I was in the room and turned off the pc right away. I'm afraid to think what would have happened if I wasn't in the room at that time. Anyway now i'm paranoid it'll happen again Is there an app that auto-shuts down the pc when the temp is too high or the RPM's are too low? (maybe it's possible through BIOS?)

Thanks in advance.

 

[sandorski]

There is likely to be a utility as you describe on the mobos driver cd.

 

[arcas]

<anecdote>
I recall a couple years ago using my finger to stop the CPU fan in a small RS/6000 at work. The OS (AIX) detected this as a fan failure and
IMMEDIATELY powered-off the system. Didn't bother waiting a few seconds to see if the fan would restart or to see if the CPU temperatures
were rising. Didn't even bother to unmount or even sync the filesystems.
</anecdote>

Be careful what you wish for

 

[Shagga]

ASUS mobo CD's come with a utility but I can't remember what it's called.....can't find it as I'm living out of boxes at my parents until my new Pad is built....Roll on end of February.....

 

[VBboy]

A utility like that is probably included on the CD, but I wouldn't trust it. I would buy a hardware fan monitor that plugs in between the fan and the power cable, and rings an annoying signal when the fan fails. NewEgg.com sells those, I think for $7 or $14 or so. It won't alert you if the fan slows down - only if it actually stops working, though.

 

[HeinekinMan]

Check out the following link and download Motherboard Monitor:

MTM 5

I believe this utility has the feature that you're looking for although I only use it to monitor mobo/CPU temps...

 

[Syborg1211]

My motherboard has a fan monitoring thing in the bios. You can set the minimum rpm for the fan and if it gets below that it will shut off the computer. Snoop around the bios for anything of the sort, and another thing you could do is set the auto shut down temp for your cpu. If your heatsink/fan fail to work the motherboard will detect a rise in temp of the cpu and shut off once you hit that temp.

 

[DisinfectedDuck]

Syborg1211 - That's what I was looking for. I have an Epox 8K7A. Do you know if it's pssoble to do what you suggested with that mobo?

 

[MCS]

If you are running MBM5, there is an option to run a program "Shutdown Now!" should the fan stop spinning or the temp go too high (or whatever you want really). Search for and download that util and then configure MBM5 to use it. Or failing that set an alarm on your CPU fan if the RRM drops below a certain level.

 

[RalfHutter]

MBM does have a provision to shutdown your computer (through a 3rd-party app like "ShutdownNow!") if the fan speed falls below a pre-defined level. This will only work with fans that have fan speed monitoring. Some fans<cough>Panaflos<cough> don't have speed monitoring abilities. The way around this is to set MBM to shut off at a predetermined CPU heat level. I use a Panaflo on my HSF and have MBM set to run "ShutdownNow" if my CPU temp gets over 57C. This will work fine. If the fan fails, the CPU temp will start to go up and MBM will shut down my box automatically. I've tested it and it works.

 

[NesuD]

mbprobe will do what you want without the need for a 3rd pard app.Linky

 

[sharkeeper]

<< Didn't even bother to unmount or even sync the filesystems. >>



That's overreacting! If a fan fails even on a completely inadequate HSF (ANY Thermaltake product, for example!) there is always enough time to flush caches to disk, verify and close all network connections and shut down.

You can use MBM to monitor your CPU fan and set it to either shut down the system or sound an alarm when the RPM falls below a threshold you can set. The only drawback is if the computer hangs while shutting down you could still have a problem. The best protection is with hardware that will kill power when the cpu core gets to levels that endanger the chip. Usually this is around 70~75°C for sensors that are outside of the die and 90°C for an internal diode.

Cheers!

 

[arcas]

I can understand maybe not bothering to unmount NFS filesystems as they're notorious for hanging when something goes wrong. But
I completely agree that the OS overreacted by shutting down immediately. Possibly the 43P series machines has no temperature sensors.