Intel Dual-CPU Server Motherboard, with 1Gig RAM, and CPUs $57.


Recommended Posts

I can happily format drives, write data, confirm data (md5), read data. What doesn't work is a lot of unMenu stuff (disk management). I thought they were a very good price, given you can pick them up elsewhere at a similar price perhaps they aren't that cheap.

For an 8 port card, the price is still pretty decent.  (For anyone with a PCI-X motherboard)

The cables are an SFF-8484 to 7 pin sata cable (see ebay 270502571488), can be called mini 32 pin sas connector.

Those are not priced too bad either.

 

Have you tried more than one of these controllers in the same server?  The Intel MB has three PCI-X slots.  If I could put more than one it would allow an eventual upgrade path to replacement of the older IDE drives in that server to SATA drives of much higher capacity.

Here are the supported and loaded drivers.

 

02:08.0 SCSI storage controller: LSI Logic / Symbios Logic SAS1068 PCI-X Fusion-MPT SAS (rev 01)

Flags: bus master, 66MHz, medium devsel, latency 72, IRQ 16

I/O ports at 9c00 [disabled]

Memory at fdffc000 (64-bit, non-prefetchable)

Memory at fdfe0000 (64-bit, non-prefetchable)

Expansion ROM at fde00000 [disabled]

Capabilities: [50] Power Management version 2

Capabilities: [98] Message Signalled Interrupts: Mask- 64bit+ Queue=0/0 Enable-

Capabilities: [68] PCI-X non-bridge device

Capabilities: [b0] MSI-X: Enable- Mask- TabSize=1

Kernel driver in use: mptsas

Kernel modules: mptsas

 

mptsas                 29204  0

mptscsih               13419  1 mptsas

mptbase                47439  2 mptsas,mptscsih

 

What version of unMENU are you using?  And what does not work on the disk-management page?

Is it the smart report stuff?  Or the spinup/spindown?  Or something else?

 

Lastly, what kind of parity check speeds are you seeing using them?

(you'll have to describe your disk configuration for that to be meaningful)

 

Joe L.

Link to comment

I have the following controller/disk config currently running.

 

Asus M2N32 WS m/b (2 x PCI-X 133, 2 x PCI-E x16, 2 x PCI-E x1 and a PCI (vga card))

6 sata (actually 10 but only 6 work in unRaid)

3 x 1TB Samsung HD103UJ inc parity (105MB/s)

 

8 sata Highpoint 1820A PCI-X 

1 x 1TB Samsung HD103UJ (104MB/s)

4 x Seagate 500GB ST3500630AS (75MB/s)

 

8 port LSI 1068 PCI-E

1 x WD 320GB WDC WD3200AAJS (101MB/s used as a test drive)

 

Parity completes at 72MB/s without the WD and 65MB/s with (WD slows down alot across its platters). PCI-X is marginally faster than PCI-E version. M/B ports are marginally faster than HBA's. 

 

The 8 port PCI-E card is an external card to add a seperate disk box as server is pretty full with only 10 drives currently possible. Any new HDDs will go in an external HDD array case. 

 

I have a second unRaid box awaiting commisioning which will be an online backup of the primary box, hence the abundance of disk controllers. I havnt tried two LSI 1068 PCI-X controllers, I only have one, the LSI card. The LSI and Marvell/highpoint PCI-X both work together. The Dell cards are 5 series cards which are the LSI 1068 PCI-X version with a PCI-X bridge chip which both work.

 

Version 1.1 of unMenu (yes I know its old....) What doesnt work? pretty much everything. The drive ID string is foobar'd which means anything the queries a disk using the ID string fails.

 

I'm probably going to order one of these cards this week for the backup box. Just waiting on shipping costs. 

Link to comment
  • 1 month later...

Put the PCI-X LSI card in today.

 

LSI SAS3080X behaves identically to the Dell 5e.

 

Below are the syslog errors recorded...

 

Jul 17 11:44:50 Tower kernel: xor: using function: pIII_sse (6498.800 MB/sec)

Jul 17 11:44:50 Tower kernel: md: unRAID driver 0.95.4 installed

Jul 17 11:44:50 Tower kernel: md: import disk0: [8,112] (sdh) SAMSUNG HD203WI S1UYJ1CZ318456 offset: 63 size: 1953514552

Jul 17 11:44:50 Tower kernel: md: import disk1: [8,48] (sdd) SAMSUNG HD103UJ S13PJDWQ302831 offset: 63 size: 976762552

Jul 17 11:44:50 Tower kernel: md: import disk2: [8,64] (sde) SAMSUNG HD103UJ S13PJDWS661831 offset: 63 size: 976762552

Jul 17 11:44:50 Tower kernel: md: import disk3: [8,80] (sdf) SAMSUNG HD103UJ S13PJDWQ302828 offset: 63 size: 976762552

Jul 17 11:44:50 Tower kernel: md: import disk4: [8,96] (sdg) ST3500630AS 9QG2CGG2 offset: 63 size: 488386552

Jul 17 11:44:50 Tower kernel: md: import disk5: [8,160] (sdk) ST3500630AS 9QG43CRM offset: 63 size: 488386552

Jul 17 11:44:50 Tower kernel: md: import disk6: [8,176] (sdl) ST3500630AS 9QG3WLYG offset: 63 size: 488386552

Jul 17 11:44:50 Tower kernel: md: import disk7: [8,128] (sdi) ST3500630AS 9QG3T4FE offset: 63 size: 488386552

Jul 17 11:44:50 Tower kernel: md: import disk8: [8,32] (sdc) SAMSUNG HD103UJ S13PJDWQ741531 offset: 63 size: 976762552

[glow=red,2,300]Jul 17 11:44:50 Tower kernel: md9: import: scsi_inquiry (std inquiry) error: -14

Jul 17 11:44:50 Tower kernel: md9: import: scsi_inquiry (vpd: unit ser no) error: -14[/glow]

 

 

Jul 17 11:46:41 Tower kernel: sdb: sdb1

[glow=red,2,300]Jul 17 11:46:41 Tower ata_id[3074]: HDIO_GET_IDENTITY failed for '/dev/block/8:16'

Jul 17 11:46:42 Tower emhttp: disk_temperature: ioctl (smart_enable): Invalid argument

Jul 17 11:46:42 Tower emhttp: clearing disk9 ...

Jul 17 11:46:42 Tower ata_id[3096]: HDIO_GET_IDENTITY failed for '/dev/block/8:16'

Jul 17 11:48:49 Tower emhttp: disk_temperature: ioctl (smart_enable): Invalid argument

Jul 17 11:48:51 Tower emhttp: disk_temperature: ioctl (smart_enable): Invalid argument[/glow]

 

[glow=red,2,300]Jul 17 12:16:21 Tower kernel: mdcmd (47): spindown 9

Jul 17 12:16:21 Tower kernel: md: disk9: ATA_OP_STANDBYNOW1 ioctl error: -22

Jul 17 12:16:34 Tower kernel: mdcmd (49): spindown 9

Jul 17 12:16:34 Tower kernel: md: disk9: ATA_OP_STANDBYNOW1 ioctl error: -22[/glow]

 

From unMenu

 

DISK_NEW new disk /dev/sdb _

 

Here is what unRAID devices shows.

 

pci-0000:02:08.0-sas-phy2:1-0x1221000002000000:2-lun0 host1 (sdb) WDC_WD20EADS-32_WD-WCAVY4026923

 

Link to comment

unRAID version: 4.5.6

unmenu.awk: Version 1.3 Revision: 97

 

Just clearing a 2TB drive currently. Seems to be progresssing nicely.

 

I have disabled spindown which has stopped it trying to spindown the drive whilst it is clearing it...

 

Leaves:

 

Jul 17 14:14:36 Tower emhttp: disk_temperature: ioctl (smart_enable): Invalid argument

Jul 17 14:14:42 Tower ata_id[5101]: HDIO_GET_IDENTITY failed for '/dev/block/8:16'

 

On startup:

Jul 17 11:44:54 Tower kernel: md9: import: scsi_inquiry (std inquiry) error: -14

Jul 17 11:44:54 Tower kernel: md9: import: scsi_inquiry (vpd: unit ser no) error: -14

 

 

Link to comment

unRAID version: 4.5.6

unmenu.awk: Version 1.3 Revision: 97

 

Just clearing a 2TB drive currently. Seems to be progresssing nicely.

 

I have disabled spindown which has stopped it trying to spindown the drive whilst it is clearing it...

 

Leaves:

 

Jul 17 14:14:36 Tower emhttp: disk_temperature: ioctl (smart_enable): Invalid argument

Jul 17 14:14:42 Tower ata_id[5101]: HDIO_GET_IDENTITY failed for '/dev/block/8:16'

 

On startup:

Jul 17 11:44:54 Tower kernel: md9: import: scsi_inquiry (std inquiry) error: -14

Jul 17 11:44:54 Tower kernel: md9: import: scsi_inquiry (vpd: unit ser no) error: -14

 

 

Are you able to use the LSI SAS3080X and assign drives connected on it on the "Devices" page to the unRAID array?

 

Apparently the driver has no idea that the disks can report temperature.  Does it appear in a SMART report?

Link to comment

Moved an existing temperature known disk from the RocketRaid to the LSI and got the same error set.

 

Jul 18 09:02:08 Tower kernel: mdcmd (79): spindown 7

Jul 18 09:02:08 Tower kernel:

Jul 18 09:02:08 Tower kernel: md: disk7: ATA_OP_STANDBYNOW1 ioctl error: -22

 

Jul 18 09:01:58 Tower ata_id[3716]: HDIO_GET_IDENTITY failed for '/dev/block/8:16'

Jul 18 08:58:42 Tower emhttp: disk_temperature: ioctl (smart_enable): Invalid argument

 

Upon doing a bit more research seems other SAS raid cards like the 3ware cards have some if not all of these issues too.

 

A smartctl -A /dev/sdb shows the temperature correctly.

 

 

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.