Smartd won't start on T630 using H730 PERC

[NadoNate]

Hi,
I'm having an issue getting the SMART service to start on a new build.

SPECS:

Dell PowerEdge T630 - BIOS 2.7.1
CPU - Xeon E5-2680 (2)
Memory - 64GB ECC
SAS - PERC H730 HBA Mode - Firmware 4.270.00-8178
Pool - 1 Z3 9 drives
FreeNas Version - 11.1-U4

Issue:

When I try to enable the SMART service from the UI it hangs, than fails to start, accompanied by this message:

Code:

smartd[2908]: Configuration file /usr/local/etc/smartd.conf parsed but has no entries

My drives are reporting smart capabilities, they are being mounted properly:

Code:

root@ASNAS:~ # smartctl -a /dev/da1
smartctl 6.6 2017-11-05 r4594 [FreeBSD 11.1-STABLE amd64] (local build)
Copyright (C) 2002-17, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Family:	 Seagate Barracuda 3.5
Device Model:	 ST4000DM004-2CV104
Serial Number:	ZFN055SC
LU WWN Device Id: 5 000c50 0a3e7cb7e
Firmware Version: 0001
User Capacity:	4,000,787,030,016 bytes [4.00 TB]
Sector Sizes:	 512 bytes logical, 4096 bytes physical
Rotation Rate:	5425 rpm
Device is:		In smartctl database [for details use: -P show]
ATA Version is:   ACS-3 T13/2161-D revision 5
SATA Version is:  SATA 3.1, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is:	Mon Apr  9 11:29:15 2018 EDT
SMART support is: Available - device has SMART capability.
SMART support is: Enabled

Code:

root@ASNAS:~ # camcontrol devlist
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 0 lun 0 (pass0,da0)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 1 lun 0 (pass1,da1)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 2 lun 0 (pass2,da2)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 3 lun 0 (pass3,da3)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 4 lun 0 (pass4,da4)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 5 lun 0 (pass5,da5)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 6 lun 0 (pass6,da6)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 7 lun 0 (pass7,da7)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 8 lun 0 (pass8,da8)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 9 lun 0 (pass9,da9)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 10 lun 0 (pass10,da10)
<ATA TOSHIBA MG04ACA4 FK1D>		at scbus1 target 11 lun 0 (pass11,da11)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 13 lun 0 (pass12,da12)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 14 lun 0 (pass13,da13)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 15 lun 0 (pass14,da14)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 16 lun 0 (pass15,da15)
<ATA ST4000DM004-2CV1 0001>		at scbus1 target 17 lun 0 (pass16,da16)
<DP BP13G+EXP 3.35>				at scbus1 target 32 lun 0 (ses0,pass17)
<PLDS DVD+-RW DH-16AES DD1M>	   at scbus11 target 0 lun 0 (cd0,pass18)
<ADATA USB Flash Drive 1100>	   at scbus13 target 0 lun 0 (pass19,da17)
<ADATA USB Flash Drive 1100>	   at scbus14 target 0 lun 0 (pass20,da18)

Doing some research I found the H730 is more reliable using the mrsas drivers, so I've added a tunable per the recommendation here:
http://doc.freenas.org/9.10/intro.html

(Variable)

This doesn't seem to help, although I am not sure freenas is using the mrsas drivers. Is there a way to verify this?

I've read a suggestion to enable mrsas during a freebsd installation here: https://forums.freebsd.org/threads/change-storage-driver-during-install.54222/ But this may only apply to UEFI....?

 

[Ericloewe]

Is there a way to verify this?

The disks all show up as da, so it should be fine.
You can go through the logs and grep mrsas. If grep mfi returns something, that's bad.

If you confirm that mrsas is being used, you might want to file a bug report, as SMART has been iffy in recent versions.

 

[NadoNate]

If grep mfi returns something, that's bad.

Code:

"mfi" not found

Here are the results for mrsas:

Code:

Apr  9 11:06:55 ASNAS mrsas0: <AVAGO Invader SAS Controller> port 0x2000-0x20ff mem 0x91e00000-0x91e0ffff,0x91d00000-0x91$
Apr  9 11:06:55 ASNAS mrsas0: Using MSI-X with 56 number of vectors
Apr  9 11:06:55 ASNAS mrsas0: FW supports <96> MSIX vector,Online CPU 56 Current MSIX <56>
Apr  9 11:06:55 ASNAS mrsas0: FW supports: UnevenSpanSupport=1
Apr  9 11:06:55 ASNAS mrsas0: MSI-x interrupts setup success

Apr  9 11:06:55 ASNAS da11 at mrsas0 bus 1 scbus1 target 11 lun 0
Apr  9 11:06:55 ASNAS da0 at mrsas0 bus 1 scbus1 target 0 lun 0
Apr  9 11:06:55 ASNAS da2 at mrsas0 bus 1 scbus1 target 2 lun 0
Apr  9 11:06:55 ASNAS da3 at mrsas0 bus 1 scbus1 target 3 lun 0
Apr  9 11:06:55 ASNAS da11: da4 at mrsas0 bus 1 scbus1 target 4 lun 0
Apr  9 11:06:55 ASNAS da5 at mrsas0 bus 1 scbus1 target 5 lun 0
Apr  9 11:06:55 ASNAS da6 at mrsas0 bus 1 scbus1 target 6 lun 0
Apr  9 11:06:55 ASNAS da7 at mrsas0 bus 1 scbus1 target 7 lun 0
Apr  9 11:06:55 ASNAS da8 at mrsas0 bus 1 scbus1 target 8 lun 0

Apr  9 11:06:55 ASNAS da11 at mrsas0 bus 1 scbus1 target 11 lun 0
Apr  9 11:06:55 ASNAS da0 at mrsas0 bus 1 scbus1 target 0 lun 0
Apr  9 11:06:55 ASNAS da2 at mrsas0 bus 1 scbus1 target 2 lun 0
Apr  9 11:06:55 ASNAS da3 at mrsas0 bus 1 scbus1 target 3 lun 0
Apr  9 11:06:55 ASNAS da11: da4 at mrsas0 bus 1 scbus1 target 4 lun 0
Apr  9 11:06:55 ASNAS syslog-ng[1899]: Error processing log message: <ATA TOSHIBA MG04ACA4 FK1D> Fixed Direct Access SPC-$
Apr  9 11:06:55 ASNAS da5 at mrsas0 bus 1 scbus1 target 5 lun 0
Apr  9 11:06:55 ASNAS da6 at mrsas0 bus 1 scbus1 target 6 lun 0
Apr  9 11:06:55 ASNAS da7 at mrsas0 bus 1 scbus1 target 7 lun 0
Apr  9 11:06:55 ASNAS da8 at mrsas0 bus 1 scbus1 target 8 lun 0

If you confirm that mrsas is being used, you might want to file a bug report, as SMART has been iffy in recent versions.

What is the best way to submit a bug, through the UI?

 

[Ericloewe]

What is the best way to submit a bug, through the UI?

Sure, that works.

 

[NadoNate]

Ok, submitted.

You mentioned SMART has been iffy, Is there any point in switching to a nightlie or earlier stable version?

Are any Dell PERC recommended? The only dell SAS controllers recommended in the hardware guide are H200 & H310.

-Thanks for your help!

 

[Ericloewe]

Are any Dell PERC recommended?

No, but those based on SAS2008, SAS2308 or SAS3008 are occasionally cheaper than an LSI-branded one and can be crossflashed.

You mentioned SMART has been iffy, Is there any point in switching to a nightlie or earlier stable version?

Earlier version, possibly. Nightlies for production, not really.

 

[artis1sysop]

Is there a solution to this?
We have a Dell R230 with a PERC H330 controller.
Drives show up as "da0, da1, da2, da3".
SMART service won't start.

 

[Arwen]

It's helpful to use the correct terminology. Were those actual disks that showed up? Or LUNs?

Hardware RAID controllers that expose individual disks, generally do so as LUNs. And those same hardware RAID controllers generally won't pass throuh SMART information. Thus, less useful for ZFS and FreeNAS.

 

[Ericloewe]

11.2 includes a fix for SMART on mrsas controllers.