LSI SAS2008 Spinning Up Attached Disks


Recommended Posts

Unraid Version: 6.8.3

LSI SAS 2008:

  • Firmware: 20.00.07.00-IT
  • Product ID: SAS9211-8i
  • BIOS: 07.11.10.00
  • NVDATA: 14.01.00.08

 

I recently purchased the above LSI card to add additional SATA drives to my server. At the moment, it has 1 HDD and 1 SSD attached.

For some reason, the HDD is being spun up, but Unraid does not know about it. Note I have 5 disks + 2 parity. The LSI attached drives are shown below:

SCSI Devices:
[9:0:0:0]disk ATA INTEL SSDSC2BF24 LH6i /dev/sdb 240GB
[9:0:1:0]disk ATA WDC WD80EDAZ-11T 0A81 /dev/sdc 8.00TB

 

I've attached the full syslog from Unraid, where the start of the log is when the new LSI card was installed. Here is a snippet from the logs showing this spinup behaviour (Disk 29 is the HDD attached to the LSI Card):

Jan  1 10:25:34 Elysia kernel: mdcmd (881): spindown 0
Jan  1 10:25:35 Elysia kernel: mdcmd (882): spindown 1
Jan  1 10:25:35 Elysia kernel: mdcmd (883): spindown 2
Jan  1 10:25:36 Elysia kernel: mdcmd (884): spindown 3
Jan  1 10:25:36 Elysia kernel: mdcmd (885): spindown 4
Jan  1 10:25:37 Elysia kernel: mdcmd (886): spindown 5
Jan  1 10:25:37 Elysia kernel: mdcmd (887): spindown 29
Jan  1 10:26:12 Elysia kernel: sd 9:0:1:0: attempting task abort! scmd(000000007fcbcbd9)
Jan  1 10:26:12 Elysia kernel: sd 9:0:1:0: [sdc] tag#435 CDB: opcode=0x85 85 06 20 00 d8 00 00 00 00 00 4f 00 c2 00 b0 00
Jan  1 10:26:12 Elysia kernel: scsi target9:0:1: handle(0x0009), sas_address(0x4433221104000000), phy(4)
Jan  1 10:26:12 Elysia kernel: scsi target9:0:1: enclosure logical id(0x5842b2b057bb1000), slot(7) 
Jan  1 10:26:12 Elysia kernel: sd 9:0:1:0: task abort: SUCCESS scmd(000000007fcbcbd9)

As you can see, about 45 seconds after the spindown command, the disk gets spun up again, but Unraid does not show the disk as Active (presumbly because it looks at the last command it issues). I noticed this behaviour as the drive was constantly reporting Temperature statistics in Grafana, and I could hear it spin up.

 

I'm entirely new to the world of SAS / HBA cards, and the above specs are those taken from the listing - I have not verified they are true (I'd have to look into how to do that). Does this look like a fault of the LSI card? Is this a Firmware issue?

 

Thanks in advance

 

elysia-syslog-20210101-1057.zip

Link to comment

The error to me looks like a slightly loose connection to the drive.

 

But, the system is not keeping the drives spun up near constantly. 

 

You've obviously got Auto Turbo mode installed as the system is bouncing back and forth between turbo mode and normal mode.  There are hours in between these 2 states.

 

Oh and Diagnostics are always preferred over syslogs

Link to comment
7 hours ago, Squid said:

The error to me looks like a slightly loose connection to the drive.

Okay, I'll double check all the connections tomorrow.

7 hours ago, Squid said:

Oh and Diagnostics are always preferred over syslogs

Apologies; Have attached them to this response if they are still necessary.

 

I did some further investigation pending my hardware checks tomorrow;

If when the drive is spun down (according to Unraid) I go to spin it up by selecting the grey circle on the Main tab, it "spins up" in 1 second. On the other hand, if I then spin it down, and then up again, each operation takes ~8 seconds.

 

I've tried to demonstrate this in the Unraid terminal;

If I issue a spindown command to a different disk that is spun down (not on the LSI card), using the time command, it takes 0.005s.

If I then issue the same command to this HDD, /dev/sdc, it takes ~0.600s.

1867847989_ArraySpindown.thumb.png.9e76c102d7033a67f2dc087328c0d950.png

 

Below is a snippet of the log again which shows my spindown command and the drive spinning itself up unbeknownst to Unraid.

Each time the block starting with "attempting task abort! scmd(0000000016cce60b)" occurs, the drive is now spun up, as the spindown command will take ~0.6s. I did this twice in succession at 22:53:27 and 22:54:05. If I issue the command again quickly (at 22:54:10), the drive has not yet spun up, and the time for the operation is 0.003s, and there is no "Power-on or device reset occurred" message.

Jan 1 22:52:38 Elysia kernel: sd 9:0:1:0: attempting task abort! scmd(0000000016cce60b)
Jan 1 22:52:38 Elysia kernel: sd 9:0:1:0: [sdc] tag#2945 CDB: opcode=0x85 85 06 20 00 d8 00 00 00 00 00 4f 00 c2 00 b0 00
Jan 1 22:52:38 Elysia kernel: scsi target9:0:1: handle(0x0009), sas_address(0x4433221104000000), phy(4)
Jan 1 22:52:38 Elysia kernel: scsi target9:0:1: enclosure logical id(0x5842b2b057bb1000), slot(7)
Jan 1 22:52:38 Elysia kernel: sd 9:0:1:0: task abort: SUCCESS scmd(0000000016cce60b)

Jan 1 22:53:27 Elysia kernel: mdcmd (969): spindown 29
Jan 1 22:53:27 Elysia kernel:
Jan 1 22:53:28 Elysia kernel: sd 9:0:1:0: Power-on or device reset occurred

Jan 1 22:53:46 Elysia kernel: sd 9:0:1:0: attempting task abort! scmd(000000006a31cdd3)
Jan 1 22:53:46 Elysia kernel: sd 9:0:1:0: [sdc] tag#1081 CDB: opcode=0x85 85 06 20 00 d8 00 00 00 00 00 4f 00 c2 00 b0 00
Jan 1 22:53:46 Elysia kernel: scsi target9:0:1: handle(0x0009), sas_address(0x4433221104000000), phy(4)
Jan 1 22:53:46 Elysia kernel: scsi target9:0:1: enclosure logical id(0x5842b2b057bb1000), slot(7)
Jan 1 22:53:46 Elysia kernel: sd 9:0:1:0: task abort: SUCCESS scmd(000000006a31cdd3)

Jan 1 22:54:05 Elysia kernel: mdcmd (970): spindown 29
Jan 1 22:54:05 Elysia kernel:
Jan 1 22:54:06 Elysia kernel: sd 9:0:1:0: Power-on or device reset occurred

Jan 1 22:54:10 Elysia kernel: mdcmd (971): spindown 29
Jan 1 22:54:10 Elysia kernel:

Also, while the GUI shows the drive is spun down, it reports an average temperature of 35 C (which is the temperature of that drive)

Temp.png.c7ca70b9bd1c35f684969dc7c621c403.png

It only changes to a * when I then issue the spindown command

1871995808_Temp2.png.a711afbde5cf7681d00004204c0d082a.png

 

elysia-diagnostics-20210101-2303.zip

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.