bubbaQ

You may not be loading the correct driver. AFAIK, without pwm_enable, you will never have success.

http://db.sanyodenki.co.jp/product_db_e/coolingfan/dcfan/dc_fan_detail.php?master_id=2606 I like Deltas for reliability.... I look for something in the 150 CFM range.

I would NOT use that fan in a server. It does not move enough air or produce enough head. Get a 2800RPM or higher fan.... minimum. Then use pwm or a fan controller to run it slower (quieter) when you don't need the maximum airflow. You size/spec your fans for the worst-case scenario... that means parity checks, with all drives hot and needing to be cooled. You can always slow down a fast fan... but that Gelid fan, you can't turn up high if you need it.

You don't have any 3-pin headers?

Here is the difference between 4-pin pwm and 2/3-pin pwm. In 2/3 pin pwm, the voltage is pulsed on pin 2. This has widely varying effects on different fans. some will stall with settings under 50. Some will double their RPM when changing pwm just a small amount, from 90 to 125. On another fam going from pwm of 90 to 150, may only increase speed a small amount. Some will have an annoying buzz. In a 4-pin pwm, the 12-volts on pin 2 is constant, and a signal is sent over pin 4 to the fan, that tells the fan a percentage of capacity to us. If the signal says 25%, then the fan is supposed to run at 25% of max. If the signal says 60%, then the fan is supposed to spin at 60% of max. Please look at figure 2 in this document: http://www.formfactors.org/developer/specs/REV1_2_Public.pdf

Geez.... I can't make it any simpler: You can only do PWM control to a 2-pin fan from a 2-pin or 3-pin fan header. You can only do PWM control to a 3-pin fan from a 2-pin or 3-pin fan header. You can only do PWM control to a 4-pin fan from a 4-pin fan header. You can stick ANY fan on ANY header and it will spin (at full speed), because pins 1 and 2 are the same for all fans and all headers. But you can only to pwm speed control in the combinations I explained above. The BIOS can not give you anything that does not meet the above.... the BIOS may REMOVE some options or disable pwm, but only within the above limits

You can't use PWM to control fan speed when you plug a 3-pin fan into a 4-wire can header. The mobo will use the 4th pin for the PWM signal, and not pulse the voltage on pin 2. You can only do PWM control to a 2-pin fan from a 2-pin or 3-pin fan header. You can only do PWM control to a 3-pin fan from a 2-pin or 3-pin fan header. You can only do PWM control to a 4-pin fan from a 4-pin fan header.

2-pin and 3-pin fans are both controlled with PWM pin 1 is ground pin 2 is v+ pin 3 is clock signal (speed sensor) (if present) pin 4 is control. (if present) http://www.formfactors.org/developer/specs/REV1_2_Public.pdf With a 4-pin wire, a signal is sent on the 4th wire to the fan, and the FAN does the pwm. With 2-pin or 3-pin fans, the mobo does the pwm. To lower the voltage from the standard 12-volt source requires something to waste (shunt to ground) the excess via a voltage regulator or a rheostat.

If the BIOS or some other setting is blocking PWM control, that may prevent pwm_enable from being available. And on some mobos, it just won't work, due to lack of support circuits on the mobo. Cheap bastards!

Did you enable PWM control with pwm_enable???

That's why you don't try to control CPU or PSU fans... they have their own self-contained control mechanisms.

I have been sleep/waking my secondary server for almost a year now, sometimes several times a day, and sometimes leaving it sleeping for weeks.... and it has worked flawlessly.

You still use s2ram in a script..... s2ram is simply a better way of doing the command "echo 3 > /proc/acpi/sleep" (better tin that it can save the video buffer and do more housekeeping, etc.)

If your mobo supports PWM fan control, you can do it, but you need to write some scripts. Or you can use a Matrix Orbital display module... the MX series displays have 3 speed-controllable fan headers and 4 ports for temperature probes.... but you'll have to write your own software to interface with it. I am working on some unRAID specific software for the Matrix Orbital, but I'm having to do it all in C and my C skills are rusty

Your mobo or BIOS settings do not support S3. S5 is powered off at the power switch.... that is not a state that you can enter manually.

Under Linux, yes.

No and you won't.... ACPI doesn't provide for it. Even for drives with power management that will power up in spin-down mode, it won't work.

I had a similar problem under 2.6.29.1 and I went to 2.6.29.6 and that fixed it.

Post the output of "uname -a"

What kernel are you using?

With the WD drives set to power up in standby, they will remain spun-down when resuming from S3 suspend state. They will even remain spun down when you access them if what you access is already cached, such as a directory listing. They will spin up when you access then for any content that is not cached.

No. s2ram has custom parameters for hundreds of mobos built into it, as well as additional command line parameters for special handling of ACPI and video quirks of a mobo.

Nooooo...... There is no package for s2ram.... what you downloaded was the source code. You have to compile it.... and for that you need a dev system, or at least a BUNCH of packages (and more than 512MB RAM if doing it under unRAID) in order to compile source.

Install and use s2ram instead of using "echo mem > /sys/power/state"

I have EXACTLY the same behavior on my test box -- a K8NE running unRAID.