From: Walter N. <ne...@ma...> - 2005-01-31 03:27:55
|
I got two useful replies to this. Here is some more info that may be useful to others: It appears to be a Linux Kernel issue -- it happens with the 2.6.10 fedora kernel but not with the 2.6.9. I can fix it by adding "hdparm -B254" to my startup script. The man page for hdparm explains this parameter: "Set Advanced Power Management feature, if the drive supports it. A low value means aggressive power management and a high value means better performance. A value of 255 will disable apm on the drive." On my Hitachi drive values around -B128 already give too agressive power management. -B255 disables power management to the point that it does not kick in when I switch to battery. -B254 is a good compromise. But I cannot find any documentation of what these numbers actually mean. Does anyone know? --walter On Thu, 27 Jan 2005, Walter Neumann wrote: > > Starting a few days ago my raw count for Load_Cycle_Count has been increasing > steadily at a rate of about 2 per minute. The raw value has increased as much > in the last 10 days as it did in the previous 9 months. > > The VALUE field reported by smartctl is decrementing every two or three days > (it decrements every time the raw count increases by 10000). Current values > are > > ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED > RAW_VALUE > 193 Load_Cycle_Count 0x0012 085 085 000 Old_age Always - > 150480 > > By googling I've found that Load_Cycle_Count seems to count how often the > head was parked. Is this right? > > Can anyone suggest what might be causing this behaviour? I've checked all > software that changed around 10 days ago and can't see anything that could be > parking the disk (I'm running fedora 3 and doing automatic regular updates > with yum). > > Could this be a hardware issue? > > Any help or guesses appreciated. If the numbers are to be believed it could > cause failure in a few months. > > --walter neumann |