Fusion MPT Ultra320 disks detected as Ultra2

DISCUSSION uprooter 3 participants 5 messages Dec 13, 2009 – Dec 14, 2009 SCSI

Share & tools

uprooter

author

2009-12-13 01:23 UTC #0 lore raw

Hello. I'm struggling a IBM x336 machine with FustionMPT LSI53C1030. This controller can't do radi0 so I disabled raid in the BIOS. Problem is that disks are detected wrong and IO is very slow hdparm -t gives only 37MB/s
dmesg shows that both disks are recognized as FAST-20 while it should be FAST-160. also cache is disabled.
I've googled around for hours, saw many problems related to the MPT being slow when compiled into the kernel in old versions (I have tried module) but I'm using using 2.6.31.
Here are some logs:

6 23:35:16 localhost kernel: Fusion MPT base driver 3.04.07
6 23:35:16 localhost kernel: Copyright (c) 1999-2008 LSI Corporation 6 23:35:16 localhost kernel: Fusion MPT SPI Host driver 3.04.07 6 23:35:16 localhost kernel: mptspi 0000:04:01.0: PCI INT A -> GSI 28 (level, low) -> IRQ 28
6 23:35:16 localhost kernel: mptbase: ioc0: Initiating bringup 6 23:35:16 localhost kernel: ata1.00: ATAPI: DV-18E, 2.0D, max UDMA/33 6 23:35:16 localhost kernel: ata1.00: configured for UDMA/33
6 23:35:16 localhost kernel: scsi 0:0:0:0: CD-ROM TEAC DV-18E
2.0D PQ: 0 ANSI: 5
6 23:35:16 localhost kernel: sr0: scsi3-mmc drive: 24x/24x cd/rw xa/form2 cdda tray
6 23:35:16 localhost kernel: Uniform CD-ROM driver Revision: 3.20 6 23:35:16 localhost kernel: sr 0:0:0:0: Attached scsi CD-ROM sr0 6 23:35:16 localhost kernel: Switched to high resolution mode on CPU 3 6 23:35:16 localhost kernel: Switched to high resolution mode on CPU 1 6 23:35:16 localhost kernel: Switched to high resolution mode on CPU 2 6 23:35:16 localhost kernel: Switched to high resolution mode on CPU 0 6 23:35:16 localhost kernel: ioc0: LSI53C1030 C0: Capabilities={Initiator} 6 23:35:16 localhost kernel: scsi2 : ioc0: LSI53C1030 C0, FwRev=01032821h, Ports=1, MaxQ=222, IRQ=28
6 23:35:16 localhost kernel: Fusion MPT FC Host driver 3.04.07 6 23:35:16 localhost kernel: Fusion MPT SAS Host driver 3.04.07 6 23:35:16 localhost kernel: Fusion MPT misc device (ioctl) driver 3.04.07 6 23:35:16 localhost kernel: mptctl: Registered with Fusion MPT base driver 6 23:35:16 localhost kernel: mptctl: /dev/mptctl @ (major,minor=10,220) 6 23:35:16 localhost kernel: Fusion MPT LAN driver 3.04.07
6 23:35:16 localhost kernel: mptlan: ioc0: PortNum=0, ProtocolFlags=08h (Itlb)
6 23:35:16 localhost kernel: mptlan: ioc0: Hmmm... LAN protocol seems to be disabled on this adapter port!
....

scsi: waiting for bus probes to complete ...

Nov 26 23:35:16 localhost kernel: scsi 2:0:0:0: Direct-Access     IBM     
IC35L146UCDY10-0 S27N PQ: 0 ANSI: 3
Nov 26 23:35:16 localhost kernel: target2:0:0: Beginning Domain Validation
Nov 26 23:35:16 localhost kernel: target2:0:0: Domain Validation skipping
write tests
Nov 26 23:35:16 localhost kernel: target2:0:0: Ending Domain Validation
Nov 26 23:35:16 localhost kernel: target2:0:0: FAST-20 WIDE SCSI 40.0 MB/s
ST (50 ns, offset 8)
Nov 26 23:35:16 localhost kernel: scsi 2:0:1:0: Direct-Access     IBM     
IC35L146UCDY10-0 S27N PQ: 0 ANSI: 3
Nov 26 23:35:16 localhost kernel: target2:0:1: Beginning Domain Validation
Nov 26 23:35:16 localhost kernel: target2:0:1: Domain Validation skipping
write tests
Nov 26 23:35:16 localhost kernel: target2:0:1: Ending Domain Validation
Nov 26 23:35:16 localhost kernel: target2:0:1: FAST-20 WIDE SCSI 40.0 MB/s
ST (50 ns, offset 8)
Nov 26 23:35:16 localhost kernel: scsi 2:0:8:0: Processor         IBM     
25P3495a S320  1 1    PQ: 0 ANSI: 2
Nov 26 23:35:16 localhost kernel: target2:0:8: Beginning Domain Validation
Nov 26 23:35:16 localhost kernel: target2:0:8: Ending Domain Validation
Nov 26 23:35:16 localhost kernel: target2:0:8: asynchronous
Nov 26 23:35:16 localhost kernel: BIOS EDD facility v0.16 2004-Jun-25, 2
devices found
Nov 26 23:35:16 localhost kernel: sd 2:0:0:0: [sda] 286749610 512-byte
hardware sectors: (146 GB/136 GiB)
Nov 26 23:35:16 localhost kernel: sd 2:0:1:0: [sdb] 286749610 512-byte
hardware sectors: (146 GB/136 GiB)
Nov 26 23:35:16 localhost kernel: sd 2:0:1:0: [sdb] Write Protect is off
Nov 26 23:35:16 localhost kernel: sd 2:0:0:0: [sda] Write Protect is off
Nov 26 23:35:16 localhost kernel: sd 2:0:0:0: [sda] Mode Sense: cb 00 00 08
Nov 26 23:35:16 localhost kernel: sd 2:0:1:0: [sdb] Mode Sense: cb 00 00 08
Nov 26 23:35:16 localhost kernel: sd 2:0:0:0: [sda] Write cache: disabled,
read cache: enabled, doesn't support DPO or FUA
Nov 26 23:35:16 localhost kernel: sd 2:0:1:0: [sdb] Write cache: disabled,

read cache: enabled, doesn't support DPO or FUA

Also tried to echo 1 > revalidate on all spi_tagets , no effect. I'll appreciate any help.
Thank you.

Signature · 2 lines

View this message in context: http://old.nabble.com/Fusion-MPT-Ultra320-disks--detected-as-Ultra2-tp26761757p26761757.html
Sent from the linux-scsi mailing list archive at Nabble.com.

James Bottomley

2009-12-13 09:29 UTC #1 lore raw

Am Sat 12 Dec 2009 23:38:00 CET schrieb uprooter <[email protected]>:

Quoted 6 lines

Hello. I'm struggling a IBM x336 machine with FustionMPT LSI53C1030. This controller can't do radi0 so I disabled raid in the BIOS. Problem is that disks are detected wrong and IO is very slow hdparm -t gives only 37MB/s
dmesg shows that both disks are recognized as FAST-20 while it should be FAST-160. also cache is disabled.

Well, the cache is easy, it isn't disabled, it's set to read through (the correct setting for enterprise disks). That's what write disabled, read enabled means.

Quoted 39 lines

I've googled around for hours, saw many problems related to the MPT being slow when compiled into the kernel in old versions (I have tried module) but I'm using using 2.6.31.
Here are some logs:
6 23:35:16 localhost kernel: Fusion MPT base driver 3.04.07
6 23:35:16 localhost kernel: Copyright (c) 1999-2008 LSI Corporation 6 23:35:16 localhost kernel: Fusion MPT SPI Host driver 3.04.07 6 23:35:16 localhost kernel: mptspi 0000:04:01.0: PCI INT A -> GSI 28 (level, low) -> IRQ 28
6 23:35:16 localhost kernel: mptbase: ioc0: Initiating bringup 6 23:35:16 localhost kernel: ata1.00: ATAPI: DV-18E, 2.0D, max UDMA/33 6 23:35:16 localhost kernel: ata1.00: configured for UDMA/33
6 23:35:16 localhost kernel: scsi 0:0:0:0: CD-ROM TEAC DV-18E
2.0D PQ: 0 ANSI: 5
6 23:35:16 localhost kernel: sr0: scsi3-mmc drive: 24x/24x cd/rw xa/form2 cdda tray
6 23:35:16 localhost kernel: Uniform CD-ROM driver Revision: 3.20 6 23:35:16 localhost kernel: sr 0:0:0:0: Attached scsi CD-ROM sr0 6 23:35:16 localhost kernel: Switched to high resolution mode on CPU 3 6 23:35:16 localhost kernel: Switched to high resolution mode on CPU 1 6 23:35:16 localhost kernel: Switched to high resolution mode on CPU 2 6 23:35:16 localhost kernel: Switched to high resolution mode on CPU 0 6 23:35:16 localhost kernel: ioc0: LSI53C1030 C0: Capabilities={Initiator} 6 23:35:16 localhost kernel: scsi2 : ioc0: LSI53C1030 C0, FwRev=01032821h, Ports=1, MaxQ=222, IRQ=28
6 23:35:16 localhost kernel: Fusion MPT FC Host driver 3.04.07 6 23:35:16 localhost kernel: Fusion MPT SAS Host driver 3.04.07 6 23:35:16 localhost kernel: Fusion MPT misc device (ioctl) driver 3.04.07 6 23:35:16 localhost kernel: mptctl: Registered with Fusion MPT base driver 6 23:35:16 localhost kernel: mptctl: /dev/mptctl @ (major,minor=10,220) 6 23:35:16 localhost kernel: Fusion MPT LAN driver 3.04.07
6 23:35:16 localhost kernel: mptlan: ioc0: PortNum=0, ProtocolFlags=08h (Itlb)
6 23:35:16 localhost kernel: mptlan: ioc0: Hmmm... LAN protocol seems to be disabled on this adapter port!
....
scsi: waiting for bus probes to complete ...

Quoted 8 lines

Nov 26 23:35:16 localhost kernel: scsi 2:0:0:0: Direct-Access     IBM
IC35L146UCDY10-0 S27N PQ: 0 ANSI: 3
Nov 26 23:35:16 localhost kernel: target2:0:0: Beginning Domain Validation
Nov 26 23:35:16 localhost kernel: target2:0:0: Domain Validation skipping
write tests
Nov 26 23:35:16 localhost kernel: target2:0:0: Ending Domain Validation
Nov 26 23:35:16 localhost kernel: target2:0:0: FAST-20 WIDE SCSI 40.0 MB/s

ST (50 ns, offset 8)

This is more curious, it seems to indicate that the disk is reporting its max transfer period as 50ns. Try raising it by echoing 25 to min_period and then 1 to revalidate to force a renegotiation.

James

Uprooter

author

2009-12-13 10:58 UTC #2 lore raw

Hello James and thanks for trying to help.

Unfortunately changing min_period didn't solve anything.

# pwd
/sys/class/spi_transport/target2:0:1

localhost target2:0:1 # echo 25 > min_period localhost target2:0:1 # echo 1 > revalidate I did it for both disks.
#dmesg | tail -10
target2:0:0: Beginning Domain Validation
target2:0:0: Domain Validation skipping write tests target2:0:0: Ending Domain Validation target2:0:0: Beginning Domain Validation target2:0:0: Domain Validation skipping write tests target2:0:0: Ending Domain Validation target2:0:1: Beginning Domain Validation target2:0:1: Domain Validation skipping write tests target2:0:1: Ending Domain Validation target2:0:1: Beginning Domain Validation target2:0:1: Domain Validation skipping write tests target2:0:1: Ending Domain Validation

# cat min_period
25
# cat period
50

hdparm still shows the same results.

# hdparm -Tt /dev/sda

/dev/sda:
Timing cached reads: 1734 MB in 2.00 seconds = 867.28 MB/sec
Timing buffered disk reads: 114 MB in 3.04 seconds = 37.55 MB/sec

# hdparm -Tt /dev/sdb

/dev/sdb:
Timing cached reads: 1732 MB in 2.00 seconds = 866.42 MB/sec
Timing buffered disk reads: 114 MB in 3.04 seconds = 37.47 MB/sec

If it helps here is what lsiutil target settings shows:

Target MB/sec | MT/sec Wide ScanID ScanLUNs Disconnect Timeout QueueTag Boot
   0     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
   1     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
   2     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
   3     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
   4     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
   5     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
   6     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
   7     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
   8     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
   9     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
  10     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
  11     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
  12     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
  13     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
  14     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes
  15     320  |   160  Yes   Yes     Yes       Yes        10     Yes    Yes

Quoted 67 lines · On Sunday 13 December 2009 11:29:25 you wrote:

Am Sat 12 Dec 2009 23:38:00 CET schrieb uprooter <[email protected]>:
Quoted 6 lines
Hello. I'm struggling a IBM x336 machine with FustionMPT LSI53C1030. This controller can't do radi0 so I disabled raid in the BIOS. Problem is that disks are detected wrong and IO is very slow hdparm -t gives only 37MB/s
dmesg shows that both disks are recognized as FAST-20 while it should be FAST-160. also cache is disabled.
Well, the cache is easy, it isn't disabled, it's set to read through (the correct setting for enterprise disks). That's what write disabled, read enabled means.
Quoted 40 lines
I've googled around for hours, saw many problems related to the MPT being slow when compiled into the kernel in old versions (I have tried module) but I'm using using 2.6.31.
Here are some logs:
6 23:35:16 localhost kernel: Fusion MPT base driver 3.04.07
6 23:35:16 localhost kernel: Copyright (c) 1999-2008 LSI Corporation 6 23:35:16 localhost kernel: Fusion MPT SPI Host driver 3.04.07 6 23:35:16 localhost kernel: mptspi 0000:04:01.0: PCI INT A -> GSI 28 (level, low) -> IRQ 28
6 23:35:16 localhost kernel: mptbase: ioc0: Initiating bringup 6 23:35:16 localhost kernel: ata1.00: ATAPI: DV-18E, 2.0D, max UDMA/33 6 23:35:16 localhost kernel: ata1.00: configured for UDMA/33
6 23:35:16 localhost kernel: scsi 0:0:0:0: CD-ROM TEAC
DV-18E 2.0D PQ: 0 ANSI: 5
6 23:35:16 localhost kernel: sr0: scsi3-mmc drive: 24x/24x cd/rw xa/form2 cdda tray
6 23:35:16 localhost kernel: Uniform CD-ROM driver Revision: 3.20 6 23:35:16 localhost kernel: sr 0:0:0:0: Attached scsi CD-ROM sr0 6 23:35:16 localhost kernel: Switched to high resolution mode on CPU 3 6 23:35:16 localhost kernel: Switched to high resolution mode on CPU 1 6 23:35:16 localhost kernel: Switched to high resolution mode on CPU 2 6 23:35:16 localhost kernel: Switched to high resolution mode on CPU 0 6 23:35:16 localhost kernel: ioc0: LSI53C1030 C0:
Capabilities={Initiator} 6 23:35:16 localhost kernel: scsi2 : ioc0:
LSI53C1030 C0, FwRev=01032821h, Ports=1, MaxQ=222, IRQ=28
6 23:35:16 localhost kernel: Fusion MPT FC Host driver 3.04.07 6 23:35:16 localhost kernel: Fusion MPT SAS Host driver 3.04.07 6 23:35:16 localhost kernel: Fusion MPT misc device (ioctl) driver 3.04.07 6 23:35:16 localhost kernel: mptctl: Registered with Fusion MPT base driver 6 23:35:16 localhost kernel: mptctl: /dev/mptctl @ (major,minor=10,220) 6 23:35:16 localhost kernel: Fusion MPT LAN driver 3.04.07
6 23:35:16 localhost kernel: mptlan: ioc0: PortNum=0, ProtocolFlags=08h (Itlb)
6 23:35:16 localhost kernel: mptlan: ioc0: Hmmm... LAN protocol seems to be disabled on this adapter port!
....
scsi: waiting for bus probes to complete ...
Quoted 8 lines
Nov 26 23:35:16 localhost kernel: scsi 2:0:0:0: Direct-Access     IBM
IC35L146UCDY10-0 S27N PQ: 0 ANSI: 3
Nov 26 23:35:16 localhost kernel: target2:0:0: Beginning Domain
Validation Nov 26 23:35:16 localhost kernel: target2:0:0: Domain
Validation skipping write tests
Nov 26 23:35:16 localhost kernel: target2:0:0: Ending Domain Validation
Nov 26 23:35:16 localhost kernel: target2:0:0: FAST-20 WIDE SCSI 40.0
MB/s ST (50 ns, offset 8)
This is more curious, it seems to indicate that the disk is reporting its max transfer period as 50ns. Try raising it by echoing 25 to min_period and then 1 to revalidate to force a renegotiation.
James

James Bottomley

2009-12-13 11:50 UTC #3 lore raw

Quoted 23 lines · Uprooter

Hello James and thanks for trying to help.
Unfortunately changing min_period didn't solve anything.
# pwd
/sys/class/spi_transport/target2:0:1
localhost target2:0:1 # echo 25 > min_period localhost target2:0:1 # echo 1 > revalidate I did it for both disks.
#dmesg | tail -10
target2:0:0: Beginning Domain Validation
target2:0:0: Domain Validation skipping write tests target2:0:0: Ending Domain Validation target2:0:0: Beginning Domain Validation target2:0:0: Domain Validation skipping write tests target2:0:0: Ending Domain Validation target2:0:1: Beginning Domain Validation target2:0:1: Domain Validation skipping write tests target2:0:1: Ending Domain Validation target2:0:1: Beginning Domain Validation target2:0:1: Domain Validation skipping write tests target2:0:1: Ending Domain Validation

Quoted 4 lines

# cat min_period
25
# cat period
50

OK, so I'd suspect a cabling problem. 50ns is the minimum period for a single ended bus. You have to move to LVD to get anything beyond. I'd say something in your SCSI cable chain is keeping the bus SE.

James

Uprooter

author

2009-12-14 19:58 UTC #4 lore raw

Quoted 29 lines · On Sunday 13 December 2009 13:50:42 James Bottomley wrote:

Quoted 23 lines · Uprooter
Hello James and thanks for trying to help.
Unfortunately changing min_period didn't solve anything.
# pwd
/sys/class/spi_transport/target2:0:1
localhost target2:0:1 # echo 25 > min_period localhost target2:0:1 # echo 1 > revalidate I did it for both disks.
#dmesg | tail -10
target2:0:0: Beginning Domain Validation
target2:0:0: Domain Validation skipping write tests target2:0:0: Ending Domain Validation target2:0:0: Beginning Domain Validation target2:0:0: Domain Validation skipping write tests target2:0:0: Ending Domain Validation target2:0:1: Beginning Domain Validation target2:0:1: Domain Validation skipping write tests target2:0:1: Ending Domain Validation target2:0:1: Beginning Domain Validation target2:0:1: Domain Validation skipping write tests target2:0:1: Ending Domain Validation
Quoted 4 lines
# cat min_period
25
# cat period
50

Quoted 6 lines

OK, so I'd suspect a cabling problem. 50ns is the minimum period for a single ended bus. You have to move to LVD to get anything beyond. I'd say something in your SCSI cable chain is keeping the bus SE.
James

Well I opened the server. nothing much to play with everything is tied up and I don't think anything except maybe disks has been changed since this server was bought.
Can It be related to disk jumper settings ? Anything else you can think of ?