'import' - 'no device' loop on starting a new unRAID server setup?

Following on from my earlier thread (linked below) testing the P5B-E motherboard, I have bought a new 500GB SATA drive, and I am now trying to setup my first working unRAID system to test.

Refer: http://lime-technology.com/forum/index.php?topic=571.0

I'm using the 4.0beta-7 with the following hardware:

Asustek P5B-E mobo with 6 x onboard ICH8R SATA (+2 JMicron SATA)

Onboard Attansic L1 GigE

Celeron D 3.06Ghz 533FSB 512M Cache

1GB DDR2 DRAM (2x 512MB G.Skill CL4)

Cheap as I could find PCI-e Video card (XFX NVidia 7100GS)

Seagate 500GB 16MB Cache SATA Drive

Everything boots up OK, and I connect to the web manager and select the deected Seagate drive as Data Disk 1, and then press 'Start'.

The web page shows that the drive is "mounting" (I initially assumed it was formatting the drive at this time).  This stays like this for maybe 10 minutes or more when I come back and refresh the page, and I can also hear disk activity.  But then if I come back a little later and refresh the page, I get no response (ie. the web server is no longer responding).

If I logon to the console as root, the server is still responding (ie. has not locked up), so I can shutdown successfully etc.

If I restart the system, I am back at square one!

I telnet'd into the server and did:

cat /var/log/syslog

Here's what I see (note: endless loop every minute):

Apr  1 08:30:19 MediaServer kernel: [ 4229.103015] md: no device

Apr  1 08:30:19 MediaServer kernel: [ 4229.103019] md0: removed

Apr  1 08:30:19 MediaServer kernel: [ 4229.103144] md: import [8,0] (sda) ST3500630AS      5QG1BYP5 offset: 63 size: 488386552

Apr  1 08:31:19 MediaServer kernel: [ 4289.102037] md: no device

Apr  1 08:31:19 MediaServer kernel: [ 4289.102041] md0: removed

Apr  1 08:31:19 MediaServer kernel: [ 4289.102161] md: import [8,0] (sda) ST3500630AS      5QG1BYP5 offset: 63 size: 488386552

Apr  1 08:32:19 MediaServer kernel: [ 4349.091064] md: no device

Apr  1 08:32:19 MediaServer kernel: [ 4349.091068] md0: removed

Apr  1 08:32:19 MediaServer kernel: [ 4349.091192] md: import [8,0] (sda) ST3500630AS      5QG1BYP5 offset: 63 size: 488386552

NB: I had renamed the server from 'Tower' to 'MediaServer' during inital setup.

I'm assuming the actual retry loop sequence is:

Apr  1 08:30:19 MediaServer kernel: [ 4229.103144] md: import [8,0] (sda) ST3500630AS      5QG1BYP5 offset: 63 size: 488386552

... 1 minute time-out?

Apr  1 08:31:19 MediaServer kernel: [ 4289.102037] md: no device

Apr  1 08:31:19 MediaServer kernel: [ 4289.102041] md0: removed

... loop - try again...

This is with the BIOS set to either 'IDE' (the default) or 'AHCI' mode for the southbridge onboard SATA (as discussed in my original post thread - linked above).

What to do now?

I'm not sure if this additional syslog information will help diagnose (which appears to be from the point of trying to 'start' the new array):

Apr  1 09:28:57 MediaServer emhttp[959]: driver cmd: start NEW_ARRAY

Apr  1 09:28:57 MediaServer kernel: [  661.898622] mdcmd (11): start

Apr  1 09:28:57 MediaServer kernel: [  661.898629] md: no device

Apr  1 09:28:57 MediaServer kernel: [  661.898764] md: import [8,0] (sda) ST3500630AS      5QG1BYP5 offset: 63 size: 488386552

Apr  1 09:28:57 MediaServer kernel: [  661.898773] md1: new disk

Apr  1 09:28:57 MediaServer kernel: [  661.898781] md: do_run: lock_rdev error:-6

Apr  1 09:28:57 MediaServer kernel: [  661.898783] md0: stopping

Apr  1 09:28:57 MediaServer kernel: [  661.898784] md1: stopping

Apr  1 09:28:57 MediaServer kernel: [  661.898940] md: no device

Apr  1 09:28:57 MediaServer kernel: [  661.898943] md0: removed

Apr  1 09:28:57 MediaServer kernel: [  661.899021] md: import [8,0] (sda) ST3500630AS      5QG1BYP5 offset: 63 size: 488386552

Apr  1 09:28:57 MediaServer emhttp[1058]: get_temperature: open: No such file or directory

Apr  1 09:28:57 MediaServer kernel: [  661.899856] md: no device

Apr  1 09:28:57 MediaServer kernel: [  661.899861] md0: removed

Apr  1 09:28:57 MediaServer kernel: [  661.899990] md: import [8,0] (sda) ST3500630AS      5QG1BYP5 offset: 63 size: 488386552

Apr  1 09:28:57 MediaServer emhttp[1059]: get_temperature: open: No such file or directory

Apr  1 09:28:57 MediaServer emhttp[959]: shcmd (19): killall -w smbd nmbd

Apr  1 09:28:58 MediaServer emhttp[959]: shcmd (20): /usr/sbin/nmbd -D

Apr  1 09:28:58 MediaServer emhttp[959]: shcmd (21): /usr/sbin/smbd -D

Apr  1 09:28:58 MediaServer kernel: [  662.946812] md: no device

Apr  1 09:28:58 MediaServer kernel: [  662.946817] md0: removed

Apr  1 09:28:58 MediaServer kernel: [  662.947128] md: import [8,0] (sda) ST3500630AS      5QG1BYP5 offset: 63 size: 488386552

Apr  1 09:28:58 MediaServer emhttp[1065]: get_temperature: open: No such file or directory

Apr  1 09:29:20 MediaServer kernel: [  684.833453] md: no device

Apr  1 09:29:20 MediaServer kernel: [  684.833458] md0: removed

Apr  1 09:29:20 MediaServer kernel: [  684.833621] md: import [8,0] (sda) ST3500630AS      5QG1BYP5 offset: 63 size: 488386552

New symptom:

I'm now getting a different symptom, that is identical to what is described in this post:

http://lime-technology.com/forum/index.php?topic=592.msg3802#msg3802

Interesting that this post is using the same Seagate 500GB drives!

ie. What I now get is that if I click refresh after a few minutes of 'starting', I end up with same 'Stopped: Configuration Valid' screen as described in the above post.

I also note that I have an '*' for temprature as well (which is probably related to the syslog errors posted above, related to temperature).

I can advise that this issue has now been fixed with the release of V4.0-beta9!

This includes using the onboard SATA configured in the BIOS as 'AHCI' mode, as indicated as the best driver to use for full SATA support under Linux kernel 2.6.20.