EDIT: Seems to be related to the on board SATA controllers. My motherboard has two controllers, one AMD SB710 and one Marvell SE9128. The AMD SB710 does not recognize any of the drives. Here is a link to the board http://www.msi.com/product/mb/880GMA-E53.html#?div=Detail

 

Any fix for this?

 

---

///////////////////original post////////////////////////////////////

First of all I'm not very familiar with unRAID at all. I've used it as a basic media server and haven't really had to touch any thing for the past two years since initial setup so If I overlooked something simple I apologize.

 

I previously had 4.7 but wanted to upgrade and since I didn't have a lot of data that needed to be migrated over I decided to start fresh and delete everything. I ran the preclear script on all my drives (8 drives total), then formatted the usb flash drive, copied over the contents of the 5rc12a zip file and ran the makebootable script. After booting I only see one drive that is available to add to the array and when I run the preclear_disk.sh -l command it only shows the same drive. I removed the flash, reformatted it and installed 4.7 and booted from that. All my drives except one showed up in 4.7. Is there something I'm missing? I noticed there is two boot options in version 5, i see unRAID OS and unRAID OS (4gb RAM)? Should i be selecting the 4GB ram one?

 

results of preclear_disk.sh -l on 5rc12a:

root@Tower:/boot# preclear_disk.sh -l
====================================1.13
Disks not assigned to the unRAID array
  (potential candidates for clearing)
========================================
     /dev/sdb = ata-WDC_WD10EADS-00M2B0_WD-WMAV50266016
root@Tower:/boot#

 

results of preclear_disk.sh -l on 4.7:

root@Tower:/boot# preclear_disk.sh -l
====================================1.13
Disks not assigned to the unRAID array
  (potential candidates for clearing)
========================================
4     /dev/sdd = ata-Hitachi_HDS5C3020ALA632_ML0220F310HR0D
5     /dev/sde = ata-Hitachi_HDS5C3020ALA632_ML0220F310P1AD
2     /dev/sdb = ata-Hitachi_HDS5C3020ALA632_ML0220F310XPDD
3     /dev/sdc = ata-Hitachi_HDS5C3020ALA632_ML0220F31173GD
1     /dev/sda = ata-Hitachi_HDS5C3020ALA632_ML0220F311A2HD
8     /dev/sdg = ata-WDC_WD10EADS-00M2B0_WD-WMAV50266016
6     /dev/sdf = ata-WDC_WD20EARX-00PASB0_WD-WMAZA7452247

 

My server information, all HD's are in hotswap drive cages plugged directly into motherboard SATA ports.

unRAID Version:

unRAID Server Pro, Version 5.0-rc12a

 

Motherboard:

MSI - 880GMA-E53 (MS-7623)

 

Processor:

AMD Athlon II X2 250e - 3 GHz

 

Cache:

L1-Cache = 256 kB (max. 256 kB)

L2-Cache = 2048 kB (max. 2048 kB)

 

Memory:

4096 MB (max. 8 GB)

No Module Installed = BANK0, Unknown

No Module Installed = BANK1, Unknown

2048 MB = BANK2, 1333 MHz

2048 MB = BANK3, 1333 MHz

unraid_5_log.txt

It seems to just be doing this over and over and over??

 

May 18 14:31:03 Tower emhttp_event: driver_loaded
May 18 14:31:08 Tower emhttp: shcmd (40): rmmod md-mod |& logger
May 18 14:31:08 Tower emhttp: shcmd (41): modprobe md-mod super=/boot/config/super.dat slots=24 |& logger
May 18 14:31:08 Tower kernel: md: unRAID driver removed
May 18 14:31:08 Tower emhttp: shcmd (42): udevadm settle
May 18 14:31:08 Tower emhttp: Device inventory:
May 18 14:31:08 Tower emhttp: WDC_WD10EADS-00M2B0_WD-WMAV50266016 (sdb) 976762584
May 18 14:31:08 Tower kernel: md: unRAID driver 2.1.6 installed
May 18 14:31:08 Tower kernel: read_file: error 2 opening /boot/config/super.dat
May 18 14:31:08 Tower kernel: md: could not read superblock from /boot/config/super.dat
May 18 14:31:08 Tower kernel: md: initializing superblock
May 18 14:31:08 Tower kernel: mdcmd (1): import 0 0,0
May 18 14:31:08 Tower kernel: mdcmd (2): import 1 0,0
May 18 14:31:08 Tower kernel: mdcmd (3): import 2 0,0
May 18 14:31:08 Tower kernel: mdcmd (4): import 3 0,0
May 18 14:31:08 Tower kernel: mdcmd (5): import 4 0,0
May 18 14:31:08 Tower kernel: mdcmd (6): import 5 0,0
May 18 14:31:08 Tower kernel: mdcmd (7): import 6 0,0
May 18 14:31:08 Tower kernel: mdcmd (: import 7 0,0
May 18 14:31:08 Tower kernel: mdcmd (9): import 8 0,0
May 18 14:31:08 Tower kernel: mdcmd (10): import 9 0,0
May 18 14:31:08 Tower kernel: mdcmd (11): import 10 0,0
May 18 14:31:08 Tower kernel: mdcmd (12): import 11 0,0
May 18 14:31:08 Tower kernel: mdcmd (13): import 12 0,0
May 18 14:31:08 Tower kernel: mdcmd (14): import 13 0,0
May 18 14:31:08 Tower kernel: mdcmd (15): import 14 0,0
May 18 14:31:08 Tower kernel: mdcmd (16): import 15 0,0
May 18 14:31:08 Tower kernel: mdcmd (17): import 16 0,0
May 18 14:31:08 Tower kernel: mdcmd (18): import 17 0,0
May 18 14:31:08 Tower kernel: mdcmd (19): import 18 0,0
May 18 14:31:08 Tower kernel: mdcmd (20): import 19 0,0
May 18 14:31:08 Tower kernel: mdcmd (21): import 20 0,0
May 18 14:31:08 Tower kernel: mdcmd (22): import 21 0,0
May 18 14:31:08 Tower kernel: mdcmd (23): import 22 0,0
May 18 14:31:08 Tower kernel: mdcmd (24): import 23 0,0

To boot, you just want UnRAID -- the 4GB option is for folks who have one of the motherboards that won't work with > 4GB of RAM to limit their memory to 4GB.    Since you don't have more than 4GB of RAM, that's irrelevant.

 

How are your drives connected?    It sounds like you may have a SATA controller card that's supported by v4.7 but not by v5.    I don't recall exactly which ones, but I know there are a couple that require modifications to work under v5 (firmware update and I believe a BIOS setting change).

 

 

Just realized you already answered my question towards the bottom of your post ["... all HD's are in hotswap drive cages plugged directly into motherboard SATA ports."]

 

... so clearly that's not the issue.

 

Did you originally see all of the drives with the pre-clear script?  i.e. you said "... I decided to start fresh and delete everything. I ran the preclear script on all my drives (8 drives total), then ..."    ==> this implies you've already spent several days running pre-clear on all your drives ... is that correct?

 

Dang you are right! It is the SATA controllers! Ports 1-6 are a AMD® SB710 controller and ports 7-8 are Marvell® SE9128 (both on board).

 

I just moved the drives around and it checks out, with unRAID v5 I only get drives that are plugged into the Marvell SE9128 ports, anything plugged into the SB710 on ports 1-6 does not show up.

 

Update the BIOS maybe?

 

And yes, in 4.7 all the drives show up. I did a preclear, only one pass. Not sure how long it took but wasn't a couple days, under 24hrs.

 

Got it working. I updated the BIOS to the newest version and changed some of the IDE/ACHI settings. I'll have to go through and check what ended up working, spent about an hour changing things and rebooting.

Good -- glad you could get the motherboard ports to work with an updated BIOS.  Otherwise you'd have to get an add-in controller.

 

 

Got it working. I updated the BIOS to the newest version and changed some of the IDE/ACHI settings.

 

If you've had to turn off AHCI and revert to legacy IDE mode, that might not be a perfect solution, since performance can be severely restricted in IDE mode.

Yup it is IDE/Legacy mode. Is there a fix for this or am i going to need to get a new mobo? What is the actuall performance difference and whats the best way to test it? I'm not real familiar with unRAID yet since everything worked when I first set it up several years ago.

Yup it is IDE/Legacy mode. Is there a fix for this or am i going to need to get a new mobo? What is the actuall performance difference and whats the best way to test it? I'm not real familiar with unRAID yet since everything worked when I first set it up several years ago.

 

The difference isn't very significant during data transfers - it's largely a matter of more efficiency access when doing multiple accesses.  In practice, it probably isn't significant -- certainly not something I'd replace the motherboard over.

 

However ... you indicated you updated the BIOS and made several adjustments in BIOS settings.  So one thing I'd test is whether changing JUST the SATA mode (back to AHCI) causes the issue to re-appear.    You may find that this wasn't a necessary change