Update: Works with errors: LSI SAS3081E-R


erikatcuse

Recommended Posts

UPDATE 3/2/10

 

I used the card with no problem but due to unforeseen circumstances no longer have my server and will have to build a new one...I might buy the same card but I'm not sure yet.

 

UPDATE 12/17/09

 

After sending Tom some sys logs he sent me a preview of 4.5.1 and it works with a few exceptions.  SMART does not work yet (I think this has to due with the command it self since the drives are treated as SCSI and not ATA), temperatures are not reported and there is some type of error with spin down however the disks spin down.  

 

I've tested it a bit and put all 6 of my disks on the card and received a 50MB/S sync rate.

 

Here are the errors now...if others want to help troubleshoot.

 

After start up

 

Dec 16 22:00:10 Backup kernel: md5: import: scsi_inquiry (std inquiry) error: -14

Dec 16 22:00:10 Backup kernel: md5: import: scsi_inquiry (vpd: unit ser no) error: -14

 

disk_temperature: ioctl (smart_enable): Invalid argument

 

Trying to spin down

 

md: disk3: ATA_OP_STANDBYNOW1 ioctl error: -22

 

 

UPDATE 12/2/09

 

4.5beta12 sees the drive but again isn't excepted by UNRAID.

 

The error is  "Dec  2 17:25:03 Backup kernel: md: import disk4: HDIO_GET_IDENTITY ioctl error: -22"

 

I think it's still a bug with the driver but can't find any info on what the error -22 is.  I've sent Tom an email with my syslog for his review.  Anyone else have any ideas?

 

"

 

 

UPDATE 4/11/09

 

4.5beta4 sees the drives but without the kernel update I still get the HDIO_GET_IDENTITY bug.  One step closer...

 

 

UPDATE 3/24/09

 

So 4.5beta3 didn't work and I emailed Tom he was able to send me a custom emttp to try and it allowed unraid in the device manager to see the drives attached to the LSI card.  The only problem is unraid didn't recognize them.  The problem is HDIO_GET_IDENTITY doesn't work with SAS.  Tom emailed me back the following

 

Well there's good news and bad news... The good news is that this problem (sas driver correctly handling HDIO_GET_IDENTITY) has been identified and a kernel patch generated: http://article.gmane.org/gmane.linux.scsi/47723

The bad news is that this patch made it into 2.6.29 kernel and unRAID is using 2.6.28.4 at present. So... this will require updating to 2.6.29 which probably won't happen in this unRAID release cycle (4.5).  However if you have been following other discussions on forum you might know that I'm also working on unRAID 5.0.  I'll look into updating the kernel in that release.

 

As I told Tom I'll patiently wait for 5.0 beta and hope it included the 2.6.29 kernel.

 

END UPDATE

 

 

 

The LSI SAS3081E-R is an PCIE x8 card with 8 sas/sata ports on it.  It came with the cables needed for 8 drives  http://www.lsi.com/storage_home/products_home/host_bus_adapters/sas_hbas/lsisas3081er/index.html

 

I received this card today and installed it.  The first thing I did was try and update the firmware from the raid version to the target initiator version.  This took about 3 hours.  The card didn't want to flash.  I changed bios setting, moved the pci video card and tried different Dos disks.  My system doesn't have a x4 slot so I brought it over to another computer and tried it in the x16 slot and it didn't work.  So I put it into the x4 slot and about 2 minutes later it was flashed.

 

With that out of the way I added it back into my unraid server and booted up (4.4.2).  Getting into its bios is a pain.  After I figured out I needed to boot into it it was easy.  Basically when I turned the server on I had to press F11 (boot devices) and when the bios of the card appeared Ctrl-A after it initiated it would then go to the boot device menu where I could select it and it would continue to boot to its utility page.  No settings had to be changed but I could verify that it could see the disk attached to it.

 

Well once unraid was up the devices page can't see the drive.  However using unmenu.awk the main page could see the drive under: Drive Partitions - Not In Protected Array.  I then went to disk mgmt and mounted and shared the drive and could read the drive from my windows computer.

 

So the question is how do I get the drives to show up in the devices drop down menu?  I've email Tom but figured I'd post here to so if this does get resolved we'll have another option for controllers.  I got lucky and bought this one new off ebay for 169.99, so I figured it was worth a try.

 

Attached is the screen shot from unmenu.   Syslog removed due to answer from Tom.

 

Erik

Link to comment

Recieved a reply from Tom

 

Yes I see the problem.  It's the string "scsi" in front of the Model/Serial string.  unRAID OS only recognizes "ata" and "sata".  This is a bug - I will put fix in the next beta (4.5-beta2), due out any time now.

 

So now it's a waiting game for 4.5-beta2

 

WeeboTech...I had posted the file but removed it with Tom's email...thanks Erik

Link to comment

Recieved a reply from Tom

 

Yes I see the problem.  It's the string "scsi" in front of the Model/Serial string.  unRAID OS only recognizes "ata" and "sata".  This is a bug - I will put fix in the next beta (4.5-beta2), due out any time now.

 

So now it's a waiting game for 4.5-beta2

 

WeeboTech...I had posted the file but removed it with Tom's email...thanks Erik

 

I was actually reading it and writing a response, my phone rang.

I saw the issue and you can fix it via symlinks or wait for tom to fix it in emhttp.

 

what I did notice on /dev/disk/by-path is sdb is reported to be sas and not scsi.

 

lrwxrwxrwx 1 root root  9 Jan 14 07:33 pci-0000:00:1f.2-scsi-5:0:0:0 -> ../../sdg

lrwxrwxrwx 1 root root 10 Jan 14 07:33 pci-0000:00:1f.2-scsi-5:0:0:0-part1 -> ../../sdg1

lrwxrwxrwx 1 root root  9 Jan 14 07:33 pci-0000:01:00.0-sas-0xn00n0nb000bndnn0:1:0-0xnnnnnnna6b9dbbb:0 -> ../../sdb

lrwxrwxrwx 1 root root 10 Jan 14 07:33 pci-0000:01:00.0-sas-0xn00n0nn000bndnn0:1:0-0xnnnfnnna6b9dbbb:0-part1 -> ../../sdb1

 

 

So which drives are on the new controller?

 

 

I'm posting the information inline for future reference to others. Every little bit helps people understand unRAID more and more.

 

root@Backup:~# ls -lR /dev/disk
/dev/disk:
total 0
drwxr-xr-x 2 root root 520 Jan 14 07:33 by-id/
drwxr-xr-x 2 root root  60 Jan 14 07:33 by-label/
drwxr-xr-x 2 root root 320 Jan 14 07:33 by-path/
drwxr-xr-x 2 root root 180 Jan 14 07:33 by-uuid/

/dev/disk/by-id:
total 0
lrwxrwxrwx 1 root root  9 Jan 14 07:33 ata-Hitachi_HDS721010KLA330_GTAXXXXXXXXXXXX -> ../../sdf
lrwxrwxrwx 1 root root 10 Jan 14 07:33 ata-Hitachi_HDS721010KLA330_GTAXXXXXXXXXXXX-part1 -> ../../sdf1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 ata-SAMSUNG_HD103UJ_XXXXXXXXXXXXXX -> ../../sdc
lrwxrwxrwx 1 root root 10 Jan 14 07:33 ata-SAMSUNG_HD103UJ_XXXXXXXXXXXXXX-part1-> ../../sdc1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 ata-ST31000340AS_XXXXXXXX -> ../../sdg
lrwxrwxrwx 1 root root 10 Jan 14 07:33 ata-ST31000340AS_XXXXXXXX-part1 -> ../../sdg1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 ata-WDC_WD10EACS-00ZJB0_WD-WCASJXXXXXXX -> ../../sdd
lrwxrwxrwx 1 root root 10 Jan 14 07:33 ata-WDC_WD10EACS-00ZJB0_WD-WCASJXXXXXXX-part1 -> ../../sdd1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 ata-WDC_WD10EACS-00ZJB0_WD-WCASJXXXXXXX -> ../../sde
lrwxrwxrwx 1 root root 10 Jan 14 07:33 ata-WDC_WD10EACS-00ZJB0_WD-WCASJXXXXXXX-part1 -> ../../sde1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 scsi-SATA_Hitachi_HDS7210_GTAXXXXXXXXXXXX-> ../../sdf
lrwxrwxrwx 1 root root 10 Jan 14 07:33 scsi-SATA_Hitachi_HDS7210_GTAXXXXXXXXXXXXX-part1 -> ../../sdf1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 scsi-SATA_Hitachi_HDS7210_GTEXXXXXXXXXXXXL-> ../../sdb
lrwxrwxrwx 1 root root 10 Jan 14 07:33 scsi-SATA_Hitachi_HDS7210_GTEXXXXXXXXXXXXL-part1 -> ../../sdb1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 scsi-SATA_SAMSUNG_HD103UJS13PJDWQXXXXXXX -> ../../sdc
lrwxrwxrwx 1 root root 10 Jan 14 07:33 scsi-SATA_SAMSUNG_HD103UJS13PJDWQXXXXXXX-part1 -> ../../sdc1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 scsi-SATA_ST31000340AS_XXXXXXXX -> ../../sdg
lrwxrwxrwx 1 root root 10 Jan 14 07:33 scsi-SATA_ST31000340AS_XXXXXXXX-part1 ->../../sdg1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 scsi-SATA_WDC_WD10EACS-00_WD-WCASXXXXXXXX -> ../../sdd
lrwxrwxrwx 1 root root 10 Jan 14 07:33 scsi-SATA_WDC_WD10EACS-00_WD-WCASXXXXXXXX-part1 -> ../../sdd1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 scsi-SATA_WDC_WD10EACS-00_WD-WCASXXXXXXXX -> ../../sde
lrwxrwxrwx 1 root root 10 Jan 14 07:33 scsi-SATA_WDC_WD10EACS-00_WD-WCASXXXXXXXX-part1 -> ../../sde1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 usb-Kingston_DataTraveler_2.0_XXXXXXXXXXXX-0:0 -> ../../sda
lrwxrwxrwx 1 root root 10 Jan 14 07:33 usb-Kingston_DataTraveler_2.0_XXXXXXXXXXXXX-0:0-part1 -> ../../sda1

/dev/disk/by-label:
total 0
lrwxrwxrwx 1 root root 10 Jan 14 07:33 UNRAID -> ../../sda1

/dev/disk/by-path:
total 0
lrwxrwxrwx 1 root root  9 Jan 14 07:33 pci-0000:00:1a.7-usb-0:6:1.0-scsi-0:0:0:0 -> ../../sda
lrwxrwxrwx 1 root root 10 Jan 14 07:33 pci-0000:00:1a.7-usb-0:6:1.0-scsi-0:0:0:0-part1 -> ../../sda1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 pci-0000:00:1f.2-scsi-1:0:0:0 -> ../../sdc
lrwxrwxrwx 1 root root 10 Jan 14 07:33 pci-0000:00:1f.2-scsi-1:0:0:0-part1 -> ../../sdc1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 pci-0000:00:1f.2-scsi-2:0:0:0 -> ../../sdd
lrwxrwxrwx 1 root root 10 Jan 14 07:33 pci-0000:00:1f.2-scsi-2:0:0:0-part1 -> ../../sdd1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 pci-0000:00:1f.2-scsi-3:0:0:0 -> ../../sde
lrwxrwxrwx 1 root root 10 Jan 14 07:33 pci-0000:00:1f.2-scsi-3:0:0:0-part1 -> ../../sde1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 pci-0000:00:1f.2-scsi-4:0:0:0 -> ../../sdf
lrwxrwxrwx 1 root root 10 Jan 14 07:33 pci-0000:00:1f.2-scsi-4:0:0:0-part1 -> ../../sdf1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 pci-0000:00:1f.2-scsi-5:0:0:0 -> ../../sdg
lrwxrwxrwx 1 root root 10 Jan 14 07:33 pci-0000:00:1f.2-scsi-5:0:0:0-part1 -> ../../sdg1
lrwxrwxrwx 1 root root  9 Jan 14 07:33 pci-0000:01:00.0-sas-0x500605b000b5d740:1:0-0xXXXXXXXXXXXXXXXX:0 -> ../../sdb
lrwxrwxrwx 1 root root 10 Jan 14 07:33 pci-0000:01:00.0-sas-0x500605b000b5d740:1:0-0xXXXXXXXXXXXXXXXX:0-part1 -> ../../sdb1

/dev/disk/by-uuid:
total 0
lrwxrwxrwx 1 root root 10 Jan 14 07:33 1e067a1e-f6c8-4883-9dd3-0b8adb718595 -> ../../sdb1
lrwxrwxrwx 1 root root 10 Jan 14 07:33 1f3ef985-9f83-4fca-8eed-9b75a62bfe90 -> ../../sde1
lrwxrwxrwx 1 root root 10 Jan 14 07:33 2dc7e20f-9367-41f8-88d9-93d07b6abeb4 -> ../../sdd1
lrwxrwxrwx 1 root root 10 Jan 14 07:33 3448-D4A7 -> ../../sda1
lrwxrwxrwx 1 root root 10 Jan 14 07:33 36b9b201-bc2a-42f9-be6b-0c92b05877ee -> ../../sdg1
lrwxrwxrwx 1 root root 10 Jan 14 07:33 cd19b89b-f7fa-4989-bf48-94b1b2bfb766 -> ../../sdc1
lrwxrwxrwx 1 root root 10 Jan 14 07:33 d75f6b0e-b1fc-4dc1-9ac4-9b0c04d70539 -> ../../sdf1
root@Backup:~#

 

Link to comment

Only sdb is on the lsi controller

The fact that you can mount the drive outside of the array and share it indicates the controller is treating it as a stand alone disk.  That is a good thing.

 

To get it to be included in the unRAID drop-down list would required one of two things to occur.

 

Either Tom needs to allow "sas" type devices to appear in the list in future releases of unRAID

or

You can see if you can "fool" unRAID by creating an entry in /dev/disk/by-path that is linked to the correct drive, but has  "-scsi-" as part of its name.

 

I would try these two commands, but the odds are they will need to be executed before emhttp is started, so they would go above it in the config/go script.

There are two lines, Each command would need to go on its own line.

 

[pre]ln  /dev/disk/by-path/pci-0000:01:00.0-sas-0x500605b000b5d740:1:0-0x786f9937a6b9dbbb:0 /dev/disk/by-path/pci-0000:01:00.0-scsi-0x500605b000b5d740:1:0-0x786f9937a6b9dbbb:0

ln  /dev/disk/by-path/pci-0000:01:00.0-sas-0x500605b000b5d740:1:0-0x786f9937a6b9dbbb:0-part1 /dev/disk/by-path/pci-0000:01:00.0-scsi-0x500605b000b5d740:1:0-0x786f9937a6b9dbbb:0-part1

[/pre]

 

I see that Tom will add "sas" as a valid choice in the next release... nice.

 

Joe L.

 

 

 

Link to comment
  • 2 weeks later...
  • 1 month later...

updated in main post also would anyone like to help me custom compile 2.6.29?

 

From Tom

Yes you could download the source, apply patch & use the unRAID .config file to build it (also have to replace dirvers/md with unraid driver).  Replace 'bzimage' on the flash with your new kernel.  If you know how to do all that you could also download 2.6.29 and try it.  As mentioned below unRAID currently uses 2.6.28.4 - I think the .28 series goes up to .9, ie, 2.6.28.9, but I checked the source and the patch is not in any 2.6.28.x version, but does appear to be in 2.6.29.

 

A bit beyond my linux knowledge but if anyone could help I'd appreciate it and could test the SAS bug out sooner then 5.0.

 

Erik

Link to comment
  • 2 months later...

So I've just started using unRAID in the last week, and I'm running 4.5 beta 6, which appears to include kernel 2.6.29.1.  Does anybody know if this means it's safe to use with an LSI 1068E SAS controller?

 

It's safe to use- meaning that having the card won't hurt the system and the card is recognized... but I still get the HDIO_GET_IDENTITY bug.  Tom is aware of it, but I'm not sure what the status of any potential fix is.

 

Link to comment

So I've just started using unRAID in the last week, and I'm running 4.5 beta 6, which appears to include kernel 2.6.29.1.  Does anybody know if this means it's safe to use with an LSI 1068E SAS controller?

 

It's safe to use- meaning that having the card won't hurt the system and the card is recognized... but I still get the HDIO_GET_IDENTITY bug.  Tom is aware of it, but I'm not sure what the status of any potential fix is.

 

 

Are the drives on the card identified and usable in the Devices tab?

Are drives attached to this card able to be added to the array?

 

Link to comment

Are the drives on the card identified and usable in the Devices tab?

Are drives attached to this card able to be added to the array?

 

I haven't tried it since beta6 first came out, but offhand I do remember that the answer to question #1 is yes... the drive is visible and you can assign it to a slot. 

 

After that my memory is a little fuzzy... I think at that point you can start the array but you get the HDIO error in the syslog and the drive does not appear in the array, but the previous drives on other controllers carry on.

 

If I get a chance I'll try it again, though I keep thinking that by now the next release should be right around the corner... maybe everyone is out enjoying their summer.  ;)

 

Link to comment
  • 5 months later...
  • 2 weeks later...

When I run that command I get

 

mptsas 0000:01:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16

mptsas 0000:01:00.0: setting latency timer to 64

mptsas: ioc0: attaching sata device: fw_channel 0, fw_id 4, phy 0, sas_addr 0x9221b04586a4975

mptsas: ioc0: attaching sata device: fw_channel 0, fw_id 3, phy 1, sas_addr 0x4e5c394ed5a28879

mptsas: ioc0: attaching sata device: fw_channel 0, fw_id 5, phy 2, sas_addr 0x4e5c3c50d59c8978

mptsas: ioc0: attaching sata device: fw_channel 0, fw_id 0, phy 3, sas_addr 0x786e985498bef4ab

mptsas: ioc0: attaching sata device: fw_channel 0, fw_id 1, phy 4, sas_addr 0x786f9937a6b9dbbb

mptsas: ioc0: attaching sata device: fw_channel 0, fw_id 2, phy 5, sas_addr 0x4d6c35f26555414b

Link to comment
Dec 16 22:00:10 Backup kernel: md5: import: scsi_inquiry (std inquiry) error: -14

Dec 16 22:00:10 Backup kernel: md5: import: scsi_inquiry (vpd: unit ser no) error: -14

 

disk_temperature: ioctl (smart_enable): Invalid argument

 

Trying to spin down

 

md: disk3: ATA_OP_STANDBYNOW1 ioctl error: -22

 

I'm afraid that -14's and -22's sound like errno values of "Invalid argument" and "Bad address" - all of which sounds to me like unfinished coding and testing, not ready for Prime time yet.

Link to comment
  • 3 weeks later...
  • 2 months later...

UPDATE 3/2/10

 

I used the card with no problem but due to unforeseen circumstances no longer have my server and will have to build a new one...I might buy the same card but I'm not sure yet.

 

 

FYI for all...

 

I'm not sure if you were implying by this update that this card should work in 4.5.3 final, but I tried a (WD20EADS) drive on the card in my unRAID server with 4.5.3 and it did not work.  I got this for a new disk:

Mar 18 20:30:08 saraswati kernel: md6: import: scsi_inquiry (std inquiry) error: -14

Mar 18 20:30:08 saraswati kernel: md6: import: scsi_inquiry (vpd: unit ser no) error: -14

Mar 18 20:30:08 saraswati kernel: md: import disk6: [8,16] (sdb) offset: 63 size: 1953514552

Mar 18 20:30:08 saraswati kernel: md: disk6 new disk

 

and additional messages like this:

Mar 18 20:30:49 saraswati ata_id[2109]: HDIO_GET_IDENTITY failed for '/dev/block/8:16'

Mar 18 20:30:49 saraswati emhttp: disk_temperature: ioctl (smart_enable): Invalid argument

 

 

Link to comment
  • 2 months later...
  • 3 years later...

Hello,

 

I know this is old, but I have the ASrock mobo and it has the LSI SAS 3008 on it.

 

I was able to flash the BISO to IT mode, and that seems to work fine.

 

However, I can;t get unRAID to see the drives?  I just updated my unRAID to 5.05, but I didn;t install my license file yet, would that cause a problem?

 

I would think I could at least see 3 drives before I put my license file back on?  No?

 

Thanks,

Dave

 

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.