Re: Dell's PERC6i

> Correct, I suppose these will be handled by mfi(4) device driver. I 
> have some Dell SC1430s that have Dell PERC 5i controllers.
> 
> One disadvantage of these is that there's no native monitoring tool 
> for mfi(4) based devices. There's only a linux utility which I
> haven't tried so far. However, it seems like porting might be easy
> and bz_at_ has idicated some interest in this work. He unfortunately
> seems to be busy these days ;-(
> 
> -- Best regards, Daniel
> mailto:danger_at_FreeBSD.org
> 
> 

I'm running 7.0-RELEASE on a Dell PE 2900II with 10x300GB 15k disks in
RAID1 + RAID5 + HS, working fine..

[angelo_at_zvr-bak-01 ~]$ cat /var/run/dmesg.boot | grep mfi
mfi0: <LSI MegaSAS 1078> port 0xec00-0xecff mem
0xfc680000-0xfc6bffff,0xfc640000-0xfc67ffff irq 16 at device 0.0 on pci1
mfi0: Megaraid SAS driver Ver 2.00
mfi0: 2664 (267900058s/0x0020/0) - Shutdown command received from host
mfi0: 2665 (4278190083s/0x0020/0) - PCI 0x041000 0x0460 0x041028
0x041f0c: Firmware initialization started (PCI ID 0060/1000/1f0c/1028)
mfi0: 2666 (4278190083s/0x0020/0) - Type 18: Firmware version 1.11.52-0396
mfi0: 2667 (4278190083s/0x0008/0) - Battery Present
mfi0: 2668 (4278190083s/0x0020/0) - Type 18: Package version 6.0.2-0002
mfi0: 2669 (4278190089s/0x0004/0) - PD 32(e0/s255) event: Enclosure
(SES) discovered on PD 20(c None/p0)
mfi0: 2670 (4278190090s/0x0002/0) - PD 32(e32/s255) event: Inserted:
Encl PD 20
mfi0: 2671 (4278190090s/0x0002/0) - Type 29: Inserted: PD 20(c None/p0)
Info: enclPd=20, scsiType=d, portMap=09,
sasAddr=5001e0f0395b0200,0000000000000000
mfi0: 2672 (4278190090s/0x0002/0) - PD 00(e32/s0) event: Inserted: PD
00(e0x20/s0)
mfi0: 2673 (4278190090s/0x0002/0) - Type 29: Inserted: PD 00(e0x20/s0)
Info: enclPd=20, scsiType=0, portMap=00,
sasAddr=1221000000000000,0000000000000000
mfi0: 2674 (4278190090s/0x0002/0) - PD 01(e32/s1) event: Inserted: PD
01(e0x20/s1)
mfi0: 2675 (4278190090s/0x0002/0) - Type 29: Inserted: PD 01(e0x20/s1)
Info: enclPd=20, scsiType=0, portMap=01,
sasAddr=1221000001000000,0000000000000000
mfi0: 2676 (4278190090s/0x0002/0) - PD 02(e32/s2) event: Inserted: PD
02(e0x20/s2)
mfi0: 2677 (4278190090s/0x0002/0) - Type 29: Inserted: PD 02(e0x20/s2)
Info: enclPd=20, scsiType=0, portMap=02,
sasAddr=1221000002000000,0000000000000000
mfi0: 2678 (4278190090s/0x0002/0) - PD 03(e32/s3) event: Inserted: PD
03(e0x20/s3)
mfi0: 2679 (4278190090s/0x0002/0) - Type 29: Inserted: PD 03(e0x20/s3)
Info: enclPd=20, scsiType=0, portMap=03,
sasAddr=1221000003000000,0000000000000000
mfi0: 2680 (4278190090s/0x0002/0) - PD 04(e32/s4) event: Inserted: PD
04(e0x20/s4)
mfi0: 2681 (4278190090s/0x0002/0) - Type 29: Inserted: PD 04(e0x20/s4)
Info: enclPd=20, scsiType=0, portMap=04,
sasAddr=5001e4f32f176804,0000000000000000
mfi0: 2682 (4278190090s/0x0002/0) - PD 05(e32/s5) event: Inserted: PD
05(e0x20/s5)
mfi0: 2683 (4278190090s/0x0002/0) - Type 29: Inserted: PD 05(e0x20/s5)
Info: enclPd=20, scsiType=0, portMap=04,
sasAddr=5001e4f32f176805,0000000000000000
mfi0: 2684 (4278190090s/0x0002/0) - PD 06(e32/s6) event: Inserted: PD
06(e0x20/s6)
mfi0: 2685 (4278190090s/0x0002/0) - Type 29: Inserted: PD 06(e0x20/s6)
Info: enclPd=20, scsiType=0, portMap=04,
sasAddr=5001e4f32f176806,0000000000000000
mfi0: 2686 (4278190090s/0x0002/0) - PD 07(e32/s7) event: Inserted: PD
07(e0x20/s7)
mfi0: 2687 (4278190090s/0x0002/0) - Type 29: Inserted: PD 07(e0x20/s7)
Info: enclPd=20, scsiType=0, portMap=04,
sasAddr=5001e4f32f176807,0000000000000000
mfi0: 2688 (4278190090s/0x0002/0) - PD 08(e32/s8) event: Inserted: PD
08(e0x20/s8)
mfi0: 2689 (4278190090s/0x0002/0) - Type 29: Inserted: PD 08(e0x20/s8)
Info: enclPd=20, scsiType=0, portMap=04,
sasAddr=5001e4f32f176808,0000000000000000
mfi0: 2690 (4278190090s/0x0002/0) - PD 09(e32/s9) event: Inserted: PD
09(e0x20/s9)
mfi0: 2691 (4278190090s/0x0002/0) - Type 29: Inserted: PD 09(e0x20/s9)
Info: enclPd=20, scsiType=0, portMap=04,
sasAddr=5001e4f32f176809,0000000000000000
mfi0: 2692 (4278190091s/0x0042/0) - Type 22: Global Hot Spare created on
PD 09(e0x20/s9) (global,rev,ea)
mfi0: 2693 (4278190091s/0x0002/0) - PD 09(e32/s9) state prior 0 new 2:
State change on PD 09(e0x20/s9) from UNCONFIGURED_GOOD(0) to HOT SPARE(2)
mfi0: 2694 (4278190104s/0x0001/0) - VD 01/1 state prior 255 new 3: State
change on VD 01/1 from UNKNOWN(ff) to OPTIMAL(3)
mfi0: 2695 (4278190104s/0x0001/0) - VD 01/1 event: VD 01/1 is now OPTIMAL
mfi0: 2696 (267900089s/0x0020/0) - Adapter ticks 267900089 elapsed 25s:
Time established as 06/27/08 16:41:29; (25 seconds since power on)
mfi0: 2697 (267900132s/0x0008/0) - Battery temperature is normal
mfi0: 2698 (267900132s/0x0008/0) - Current capacity of the battery is
above threshold
mfi0: [ITHREAD]
mfid0: <MFI Logical Disk> on mfi0
mfid0: 714880MB (1464074240 sectors) RAID volume '' is optimal
mfid1: <MFI Logical Disk> on mfi0
mfid1: 4289280MB (8784445440 sectors) RAID volume '' is optimal
Trying to mount root from ufs:/dev/mfid0s1a

I must say, this looks quite impressive:

[angelo_at_zvr-bak-01 ~]$ sudo zpool list
NAME                    SIZE    USED   AVAIL    CAP  HEALTH     ALTROOT
tank1                   668G   3.24M    668G     0%  ONLINE     -
tank2                  4.06T   1.44T   2.62T    35%  ONLINE     -


I put the ZFS pools on top of the HW RAID, because the controller does 
not support JBOD. You could export each disk as a single disk RAID0, but 
I've read that if a disk fails, ZFS won't notice, and the hotspare would 
not kick in.

Not having a native raid management tool is a very bad thing, but this 
is a non-production machine (well, at least not front-end), so if a disk 
fails the DRAC will notice me, and I can take appropriate step in the 
BIOS controller manager.


Angelo Hongens.