WD Red: Firmware modifications without change in version number

Status
Not open for further replies.

MrToddsFriends

Documentation Browser
Joined
Jan 12, 2015
Messages
1,338
In my FreeNAS system I have four WD Red 3TB drives from different years. All of them show identical Firmware Version strings but the newest of them shows a different behavior regarding logs supported.

I'm on FreeNAS-11.1-U4 and I am using
smartctl 6.6 2017-11-05 r4594 [FreeBSD 11.1-STABLE amd64] (local build)
shipped with this FreeNAS version. The output of smartctl -x for all four WD Red drives is on my hand as smartctl_adaX.log with X from 2,3,4,5.

Now pointing out the differences: ada4, the newest of these WD Red drives, has support for the Device Statistics log and the Pending Defects log while the others don't.
Code:
~ # grep "Device Statistics" *ada2.log *ada3.log *ada4.log *ada5.log
smartctl_ada2.log:Device Statistics (GP/SMART Log 0x04) not supported
smartctl_ada3.log:Device Statistics (GP/SMART Log 0x04) not supported
smartctl_ada4.log:0x04	   GPL,SL  R/O	  8  Device Statistics log
smartctl_ada4.log:Device Statistics (GP Log 0x04)
smartctl_ada5.log:Device Statistics (GP/SMART Log 0x04) not supported
~ # grep "Pending Defects" *ada2.log *ada3.log *ada4.log *ada5.log
smartctl_ada2.log:Pending Defects log (GP Log 0x0c) not supported
smartctl_ada3.log:Pending Defects log (GP Log 0x0c) not supported
smartctl_ada4.log:0x0c	   GPL	 R/O   2048  Pending Defects log
smartctl_ada4.log:Pending Defects log (GP Log 0x0c) supported [please try: '-l defects']
smartctl_ada5.log:Pending Defects log (GP Log 0x0c) not supported

Year of purchase for ada4 was 08/2017, ada5 is from 06/2016 and the other two are from 12/2014.

Firmware version and device model strings for these drives, as well as serial numbers:
Code:
~ # grep "82.00A82" *ada2.log *ada3.log *ada4.log *ada5.log
smartctl_ada2.log:Firmware Version: 82.00A82
smartctl_ada3.log:Firmware Version: 82.00A82
smartctl_ada4.log:Firmware Version: 82.00A82
smartctl_ada5.log:Firmware Version: 82.00A82
~ # grep "WDC WD30EFRX-68EUZN0" *ada2.log *ada3.log *ada4.log *ada5.log
smartctl_ada2.log:Device Model:	 WDC WD30EFRX-68EUZN0
smartctl_ada3.log:Device Model:	 WDC WD30EFRX-68EUZN0
smartctl_ada4.log:Device Model:	 WDC WD30EFRX-68EUZN0
smartctl_ada5.log:Device Model:	 WDC WD30EFRX-68EUZN0
~ # grep "Serial Number" *ada2.log *ada3.log *ada4.log *ada5.log
smartctl_ada2.log:Serial Number:	WD-WMC4N0E6K1VF
smartctl_ada3.log:Serial Number:	WD-WMC4N0E18MRH
smartctl_ada4.log:Serial Number:	WD-WCC4N3SRHDH9
smartctl_ada5.log:Serial Number:	WD-WCC4N0UHPJRX

Firmware version and device model strings are identical for all four drives. First few characters of serial numbers were changed from WD-WMC to WD-WCC in the past, but with no easily detectable change related to the supported logs reported above.

This is just a little heads up for those of you interested in these details.
 

Chris Moore

Hall of Famer
Joined
May 2, 2015
Messages
10,080
It is interesting. I have noticed similar changes in the firmware of drives from other manufacturers (Seagate, HGST and Toshiba) but it usually has a different number associated to the different firmware. It is odd that the firmware is listed as the same version but displays a different behavior.
Is there a newer firmware available from WD?
 

wblock

Documentation Engineer
Joined
Nov 14, 2014
Messages
1,506
Could the difference be a default setting change rather than a firmware change? Maybe camcontrol identify would show that.
 

MrToddsFriends

Documentation Browser
Joined
Jan 12, 2015
Messages
1,338
Could the difference be a default setting change rather than a firmware change? Maybe camcontrol identify would show that.

The output of camcontrol identify applied to ada2 (w/o support for the Device Statistics log and the Pending Defects log) and ada4 (with support for these logs) differs in the following lines
Code:
~ # diff camcontrol_ada2.log camcontrol_ada4.log
1,2c1,2
< pass2: <WDC WD30EFRX-68EUZN0 82.00A82> ACS-2 ATA SATA 3.x device
< pass2: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes)
---
> pass4: <WDC WD30EFRX-68EUZN0 82.00A82> ACS-2 ATA SATA 3.x device
> pass4: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes)
7,8c7,8
< serial number		 WD-WMC4N0E6K1VF
< WWN				   50014ee65a574394
---
> serial number		 WD-WCC4N3SRHDH9
> WWN				   50014ee2b963f415

Without being familiar with the meaning of the WWN field of the camcontrol identify output I'm guessing that there's no default setting change. Or am I wrong?
 

MrToddsFriends

Documentation Browser
Joined
Jan 12, 2015
Messages
1,338
It is odd that the firmware is listed as the same version but displays a different behavior.

Yes, that's odd. In the case reported here it seems to be only a minor inconvenience not breaking compatibility. I wonder if such firmware changes without notice have something to do with the WDIDLE3 incompatibility with certain WD Blue models reported recently.

Is there a newer firmware available from WD?

Not that I'm aware of.
 

MrToddsFriends

Documentation Browser
Joined
Jan 12, 2015
Messages
1,338
Additional information: All four WD Red HDDs mentioned above are connected to the SATA 2.0 controller of a Supermicro A1SAi-2750F Avoton board, so no apparent difference here.
Code:
~ # pciconf -lv ahci1@pci0:0:23:0
ahci1@pci0:0:23:0:	  class=0x010601 card=0x72708086 chip=0x1f228086 rev=0x02 hdr=0x00
	vendor	 = 'Intel Corporation'
	device	 = 'Atom processor C2000 AHCI SATA2 Controller'
	class	  = mass storage
	subclass   = SATA
 
Status
Not open for further replies.
Top