3TB with AXXRMS2LL080 and AXX6DRV3GEXP

[kshah10]

Hi,

Im not sure if this is the right place.

I own intel s1200btl motherboard.

I have added a axxrms2ll080 card to it and using it with a AXX6DRV3GEXP disk expander.

Recently i bought 6 3tb disk ( WDC WD30EFRX-68EUZN0 ).

When i booted the server upp the lsi card said size 0 TB.

My os did see the disks, but i did get allot of error until the disk jumped out and i needed to reboot to see the disk again.

But there was no problem connecting the disk directly to AXXRMS2LL080 and the lsi said 3 TB at boot post.

Is there a limit in AXX6DRV3GEXP ?

please note i did update AXXRMS2LL080 and after that it said at post 3TB but disk still jumped out.

S.M.A.R.T info:

Error 56742 occurred at disk power-on lifetime: 56 hours (2 days + 8 hours)

When the command that caused the error occurred, the device was active or idle.

After command completion occurred, registers were:

ER ST SC SN CL CH DH

-- -- -- -- -- -- --

04 61 0c 00 00 00 00 Device Fault; Error: ABRT

Commands leading to the command that caused the error were:

CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name

-- -- -- -- -- -- -- -- ---------------- --------------------

ef 03 0c 00 00 00 00 00 2d+08:54:29.482 SET FEATURES [Set transfer mode]

e5 00 00 00 00 00 00 00 2d+08:54:29.482 CHECK POWER MODE

ec 00 00 00 00 00 00 00 2d+08:54:29.482 IDENTIFY DEVICE

ef 03 0c 00 00 00 00 00 2d+08:54:29.232 SET FEATURES [Set transfer mode]

e5 00 00 00 00 00 00 00 2d+08:54:29.232 CHECK POWER MODE

This did not happen when i connected the drives directly to AXXRMS2LL080 .

Message was edited by: keivan shahin

[kshah10]

I also would like to know if i can upgrade AXX6DRV3GEXP from my s1200btl with IPMB since the firmware readme says:

- Intel(R) Server Board S5520HC, S5500HCV, S5500BC and Intel(R) Workstation Board

S5520SC with Fab 2 or higer, Intel(R) Hot-swap backplane AXX6DRV3GEXP, Fab 1 or

above, connected to server board or workstation board via IPMB cable

[David_A_Intel]

As noted the firmware package the update should work with those boards. I would need to investigate if the firmware is going to work with this board. The AXX6DRV3GEXP hot swap backplane was intended for 5000 series, 5500 series, and 3400 series platforms.

[David_A_Intel]

The link speed that the backplane supports is 3 Gb/s and I noticed that the RAID controller and the drives are 6 Gb/s ready.

The error does not indicate a particular issue with the link speed and the RAID controller should be able to negotiate the link speed of the drives to 3 Gb/s.

Could you share a copy of the TTY logs for the RAID controller to gather more information about this.

[kshah10]

Hi,

I seem to found the issue, as soon as there is high load on AXX6DRV3GEXP that is connected to axxrms2ll080 drives gets disconnected.

I first removed the 3tb drives from AXX6DRV3GEXP and added them direct to axxrms2ll080 and everything worked fine, as soon i started to copy information from disk in AXX6DRV3GEXP i notices that other drives that where in AXX6DRV3GEXP ( 2tb disks ) now started to lose connection as soon i was copying data over 70-100mb/sec from disk in AXX6DRV3GEXP. Some times i was lucky to see them connect again and some times i had to restart.

So i moved AXX6DRV3GEXP and connected it to my pci-e lsi 1068e, now everything works with out any disk jumping out.

Here is some info how it lookt when i was copying before i moved it to 1068e:

Nov 13 13:28:07 joker kernel: [169481.317041] sd 2:0:121:0: [sdr] Synchronizing SCSI cache

Nov 13 13:28:07 joker kernel: [169481.317335] sd 2:0:121:0: [sdr] Result: hostbyte=DID_NO_CONNECT driverbyte=DRIVER_OK

Nov 13 13:28:07 joker kernel: [169481.317508] mpt2sas0: removing handle(0x000c), sas_addr(0x5001e671499be001)

Nov 13 13:28:07 joker kernel: [169481.317990] sd 2:0:119:0: [sds] Synchronizing SCSI cache

Nov 13 13:28:07 joker kernel: [169481.318005] sd 2:0:119:0: [sds] Result: hostbyte=DID_NO_CONNECT driverbyte=DRIVER_OK

Nov 13 13:28:07 joker kernel: [169481.318126] mpt2sas0: removing handle(0x0010), sas_addr(0x5001e671499be005)

Nov 13 13:28:13 joker kernel: [169487.113794] scsi 2:0:122:0: Direct-Access ATA WDC WD20EARS-00S 0A80 PQ: 0 ANSI: 6

Nov 13 13:28:13 joker kernel: [169487.113805] scsi 2:0:122:0: STP: handle(0x0010), sas_addr(0x5001e671499be005), phy(4), device_name(0x50014ee2046d87f5)

Nov 13 13:28:13 joker kernel: [169487.113809] scsi 2:0:122:0: STP: enclosure_logical_id(0x5001e671499be000), slot(4)

Nov 13 13:28:13 joker kernel: [169487.113916] scsi 2:0:122:0: atapi(n), ncq(y), asyn_notify(n), smart(y), fua(y), sw_preserve(y)

Nov 13 13:28:13 joker kernel: [169487.113921] scsi 2:0:122:0: qdepth(32), tagged(1), simple(0), ordered(0), scsi_level(7), cmd_que(1)

Nov 13 13:28:13 joker kernel: [169487.114125] sd 2:0:122:0: Attached scsi generic sg8 type 0

Nov 13 13:28:13 joker kernel: [169487.130174] sd 2:0:122:0: [sdr] 3907029168 512-byte logical blocks: (2.00 TB/1.81 TiB)

Nov 13 13:28:13 joker kernel: [169487.162093] sd 2:0:122:0: [sdr] Write Protect is off

Nov 13 13:28:13 joker kernel: [169487.162098] sd 2:0:122:0: [sdr] Mode Sense: 7f 00 10 08

Nov 13 13:28:13 joker kernel: [169487.172765] sd 2:0:122:0: [sdr] Write cache: enabled, read cache: enabled, supports DPO and FUA

Nov 13 13:28:13 joker kernel: [169487.259893] sdr: sdr1

Nov 13 13:28:13 joker kernel: [169487.333271] sd 2:0:122:0: [sdr] Attached SCSI disk

information like that was spamming from random drives in AXX6DRV3GEXP.

Maby that could be a fix in upgrading AXX6DRV3GEXP, but as i said i dont know if i can do that since the readme file does not mention s1200btl so i hope i can do it with s1200btl over IPMB.

[kshah10]

As noted the firmware package the update should work with those boards. I would need to investigate if the firmware is going to work with this board. The AXX6DRV3GEXP hot swap backplane was intended for 5000 series, 5500 series, and 3400 series platforms

Well not sure because i asked asked some years ago what expander that worked with my old card "AXX4SASMOD" and intel support said AXX6DRV3GEXP and AXX4DRV3GEXP works with that card. i did not get any information that is was not intended for specific boards since i guessed it was the raid module that maters sadly.

[David_A_Intel]

I just compared the board model you are using and the cable to be used for the update and it is not matching as the header on the board is a 3 pin connector and the cable is a 4-wired cable.

It is possible that the issue could be related to the backplane and if it is not possible to update the firmware, I would recommend contact http://www.intel.com/p/en_US/support/contactsupport Intel Customer Support for proper warranty replacement process.

[kshah10]

Ok, thanks, but just a question what i can see my board s1200btl does have 4pin IPMB.

[David_A_Intel]

It is correct; the board has a 4-pin IPMB connector. However, the firmware is carried out through the HSBP connector instead.