Flashing the LSI2308 on SuperMicro X10SL7-F to IT mode. Which version to use NOW?

[Kei]

Hello,
I have purchase a Supermicro X10SL7-F-O since it's a very recommended motherboard which also has an integrated SAS controller. However, this controller need to be flashed to IT mode prior to usage, so FreeNAS can access it properly. I understand that, however I have read that you're not supposed to use the latest IT revisions otherwise FreeNAS will "complain"; instead most people seems to be using this version: PH16.0.1-IT.zip (which is IT version 16). But is this true also now in 2016?? Is there anyone who is running a much more recent version without having problems?

 

[danb35]

I understand that, however I have read that you're not supposed to use the latest IT revisions otherwise FreeNAS will "complain";

You understand incorrectly. It's not a matter of the latest version or not, it's a matter of the version that matches the driver included with FreeNAS. Since the release of 9.3.1 last August, FreeNAS has used the P20 driver, so you should use the P20 firmware.

 

[Kei]

Thank you Dan. Stupid question, does it have to be 20.0.0.0 or also 20.0.0.x will do? Also, for firther FN versions, how can I check which driver has been implemented?

 

[dlavigne]

Also note from http://doc.freenas.org/9.10/freenas_install.html#caveats:

The mps driver for 6G Avago SAS HBAs is version 20, which requires phase 20 firmware on the controller and the mpr driver for 12G Avago SAS HBAs is version 13 which requires P12 firmware. It is recommended to upgrade the firmware before installing FreeNAS® or immediately after upgrading FreeNAS®, using the instructions in Alert. Running older firmware can cause many woes, including the failure to probe all of the attached disks, which can lead to degraded or unavailable arrays. While you can mismatch your firmware version with a higher version and things will “probably still work”, there are no guarantees as that driver and firmware combination is untested.

 

[danb35]

I believe that 20.00.04.00 is the preferred version, but I'm not certain about that. For future versions, if the driver version changes, the GUI will notify you.

 

[Spearfoot]

You can pull SuperMicro's version for the X10SL7 (and other systems with the LSI 2308 chip) from their FTP server:

ftp://ftp.supermicro.com/Driver/SAS/LSI/2308/Firmware/IT/

 

[Mirfster]

Think I saw that there is an even newer version out since 04/04/2016 (20.00.07.00?); just as an FYI.

 

[Spearfoot]

Think I saw that there is an even newer version out since 04/04/2016 (20.00.07.00?); just as an FYI.

Yes, version 20.00.07.00 is available from the Avago/LSI site, and that's what I'm running on my IBM M1015.

Supermicro has the older 20.00.04.00 version on their site, and I know I installed it on my X10SL7 as soon as it became available. But I don't remember whether I installed the '07' version afterwards. I'll have to put that on my list of things to check -- but it's gonna be low priority, 'cause the system works flawlessly 'as is'...

 

[danb35]

Just run 'sas2flash -listall' from the CLI--no need to down the system to check the card firmware version.

 

[Spearfoot]

Just run 'sas2flash -listall' from the CLI--no need to down the system to check the card firmware version.

Thanks, @danb35. Still running the older '04' version:

Code:

[root@boomer ~]# sas2flash -listall
LSI Corporation SAS2 Flash Utility
Version 16.00.00.00 (2013.03.01)
Copyright (c) 2008-2013 LSI Corporation. All rights reserved

Adapter Selected is a LSI SAS: SAS2308_1(D1)

Num Ctlr FW Ver NVDATA x86-BIOS PCI Addr
----------------------------------------------------------------------------

0 SAS2308_1(D1) 20.00.04.00 14.01.30.16 07.39.00.00 00:0b:00:00

Finished Processing Commands Successfully.
Exiting SAS2Flash.
[root@boomer ~]#

 

[Kei]

Supermicro has the older 20.00.04.00 version on their site, and I know I installed it on my X10SL7 as soon as it became available. But I don't remember whether I installed the '07' version afterwards. I'll have to put that on my list of things to check -- but it's gonna be low priority, 'cause the system works flawlessly 'as is'...

Thanks for the feedback! I wonder if you'd be required to update to 20.00.07.00 with FN 9.10.
I also happen to have ordered the same motherboard and it's due next week. Could you tell me what did you use to flash the IT firmware? Mind that I'm not a windows user :P

 

[Mirfster]

Thanks for the feedback! I wonder if you'd be required to update to 20.00.07.00 with FN 9.10.

Not from what I have seen. 20.00.04.00 works fine.

 

[Ericloewe]

Thanks for the feedback! I wonder if you'd be required to update to 20.00.07.00 with FN 9.10.
I also happen to have ordered the same motherboard and it's due next week. Could you tell me what did you use to flash the IT firmware? Mind that I'm not a windows user :p

20.00.04 seems not to have any devastating bugs. It does have a few issues, most notably involving SATA 1.5Gb/s devices.

I'd apply the following criteria:

• If you're flashing a card to P20, flash it to the latest P20 firmware. If you already have other SAS2 controllers, flash them to the same version.
• If you're running P20.00.04 or newer and are not adding controllers and don't have any issues, don't bother.
• If you're running P20.00.04 and do have issues, try the latest version.

Always make sure the BIOS and UEFI extension ROMs, if installed, are the version distributed with the firmware you're installing and ensure that all installed controllers have the same BIOS/UEFI extension ROMs installed. Those things can be rather picky with mismatched cards...

 

[zoon01]

20.00.07.00 does have bug fix

Defects Fixed (1):
ID: SCGCQ01026819 (Port Of Defect SCGCQ00927500)
Headline: IOP: False Reply Parity Error Firmware Workaround
Description Of Change: When the false reply parity error occurs, clear out the interrupt and return. If it was a real parity error, then the error would either be seen upstream or downstream as well, and another parity error would be generated.
Issue Description: In the IOP a false reply parity error is being generated, which results in a 0x1500 fault.
Steps To Reproduce: Run in hostboot with specialized drivers and run IOs. Some controllers never see the issue, while others see the issue in a highly consistent manner.
Component: sctrlfw\controllerfw\iop\sched

 

[jgreco]

That seems like it's mostly an unusual problem. We've not seen problems with people on 20.00.04.00, and if that's the only significant change, you could probably run either one without too much concern.

 

[Kei]

If you're running P20.00.04 or newer and are not adding controllers and don't have any issues, don't bother.

ANd what if I wanted to bother just for the hell of it? I only need to disconnect the SATA cables of all drives and repeat the procedure, right?

 

[Ericloewe]

ANd what if I wanted to bother just for the hell of it? I only need to disconnect the SATA cables of all drives

Strictly speaking, you don't even need to do that. You can do the following, in rough order of paranoia:

• Run sas2flash in an environment in which the pool isn't being accessed (like the EFI shell)
• Disconnect all drives from the controller

and repeat the procedure, right?

Yeah.

 

[jgreco]

ANd what if I wanted to bother just for the hell of it? I only need to disconnect the SATA cables of all drives and repeat the procedure, right?

Of course. You don't strictly need to disconnect the drives, but the paranoid among us often do.

 

[Kei]

Speaking of paranoia, I run a script every twenty minutes to check the drives temperatures in case one of the fans would fail, so to autamatically power down the NAS in case of overheating. So yeah... :)

 

[Mirfster]

Speaking of paranoia, I run a script every twenty minutes to check the drives temperatures in case one of the fans would fail, so to autamatically power down the NAS in case of overheating. So yeah... :)

Seems to me that you may have had more than one instance with overheating if you had to go that far. Might want to look a bit at more at resolving the root cause... ;)

P.S. There is already alerting ability built into FreeNas for Hard Drive Temps (not sure if you were aware of this)