* AIC7xxx LUN enumeration failure and DV config failure
@ 2003-09-17 13:06 Olivier Galibert
2003-09-17 15:23 ` Patrick Mansfield
2003-09-17 16:20 ` Justin T. Gibbs
0 siblings, 2 replies; 4+ messages in thread
From: Olivier Galibert @ 2003-09-17 13:06 UTC (permalink / raw)
To: linux-kernel, linux-scsi
We have two transtec 5016 IDE disks/SCSI interface raid arrays
connected to a motherboard Adaptec AIC7902 Ultra320 SCSI adapter on
scsi1. Each is split in two LUNs, 2x1.4Tb for the first and 2x2Tb for
the second. There is also the root disk and $DEITY-knows-what on
scsi0.
With 2.4.22-pre9 aic79xx driver:
Sep 17 14:41:41 m61 kernel: SCSI subsystem driver Revision: 1.00
Sep 17 14:41:41 m61 kernel: scsi0 : Adaptec AIC79XX PCI-X SCSI HBA DRIVER, Rev 1.3.10
Sep 17 14:41:41 m61 kernel: <Adaptec AIC7902 Ultra320 SCSI adapter>
Sep 17 14:41:41 m61 kernel: aic7902: Ultra320 Wide Channel A, SCSI Id=7, PCI-X 101-133Mhz, 512 SCBs
Sep 17 14:41:41 m61 kernel:
Sep 17 14:41:42 m61 kernel: scsi1 : Adaptec AIC79XX PCI-X SCSI HBA DRIVER, Rev 1.3.10
Sep 17 14:41:42 m61 kernel: <Adaptec AIC7902 Ultra320 SCSI adapter>
Sep 17 14:41:42 m61 kernel: aic7902: Ultra320 Wide Channel B, SCSI Id=7, PCI-X 101-133Mhz, 512 SCBs
Sep 17 14:41:42 m61 kernel:
Sep 17 14:41:42 m61 kernel: blk: queue f7969618, I/O limit 524287Mb (mask 0x7fffffffff)
Sep 17 14:41:42 m61 kernel: (scsi0:A:0): 320.000MB/s transfers (160.000MHz DT|IU|QAS, 16bit)
Sep 17 14:41:42 m61 kernel: scsi1:A:0:0: DV failed to configure device. Please file a bug report against this driver.
Sep 17 14:41:42 m61 kernel: scsi1:A:1:0: DV failed to configure device. Please file a bug report against this driver.
Sep 17 14:41:42 m61 kernel: Vendor: SEAGATE Model: ST373307LC Rev: 0004
Sep 17 14:41:42 m61 kernel: Type: Direct-Access ANSI SCSI revision: 03
Sep 17 14:41:42 m61 kernel: blk: queue f7969418, I/O limit 524287Mb (mask 0x7fffffffff)
Sep 17 14:41:42 m61 kernel: Vendor: SUPER Model: GEM318 Rev: 0
Sep 17 14:41:42 m61 kernel: Type: Processor ANSI SCSI revision: 02
Sep 17 14:41:42 m61 kernel: blk: queue f78f1618, I/O limit 524287Mb (mask 0x7fffffffff)
Sep 17 14:41:42 m61 kernel: scsi0:A:0:0: Tagged Queuing enabled. Depth 32
Sep 17 14:41:42 m61 kernel: (scsi1:A:0): 160.000MB/s transfers (80.000MHz DT, 16bit)
Sep 17 14:41:42 m61 kernel: Vendor: transtec Model: Rev: 0001
Sep 17 14:41:42 m61 kernel: Type: Direct-Access ANSI SCSI revision: 03
Sep 17 14:41:42 m61 kernel: blk: queue f78ce218, I/O limit 524287Mb (mask 0x7fffffffff)
Sep 17 14:41:42 m61 kernel: Vendor: transtec Model: Rev: 0001
Sep 17 14:41:42 m61 kernel: Type: Direct-Access ANSI SCSI revision: 03
Sep 17 14:41:42 m61 kernel: blk: queue f78ce018, I/O limit 524287Mb (mask 0x7fffffffff)
Sep 17 14:41:42 m61 kernel: (scsi1:A:1): 160.000MB/s transfers (80.000MHz DT, 16bit)
Sep 17 14:41:42 m61 kernel: Vendor: transtec Model: Rev: 0001
Sep 17 14:41:42 m61 kernel: Type: Direct-Access ANSI SCSI revision: 03
Sep 17 14:41:42 m61 kernel: blk: queue f7889c18, I/O limit 524287Mb (mask 0x7fffffffff)
Sep 17 14:41:42 m61 kernel: Vendor: transtec Model: Rev: 0001
Sep 17 14:41:42 m61 kernel: Type: Direct-Access ANSI SCSI revision: 03
Sep 17 14:41:42 m61 kernel: blk: queue f7889a18, I/O limit 524287Mb (mask 0x7fffffffff)
Sep 17 14:41:42 m61 kernel: scsi1:A:0:0: Tagged Queuing enabled. Depth 32
Sep 17 14:41:42 m61 kernel: scsi1:A:0:1: Tagged Queuing enabled. Depth 32
Sep 17 14:41:42 m61 kernel: scsi1:A:1:0: Tagged Queuing enabled. Depth 32
Sep 17 14:41:42 m61 kernel: scsi1:A:1:1: Tagged Queuing enabled. Depth 32
Sep 17 14:41:42 m61 kernel: Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
Sep 17 14:41:42 m61 kernel: Attached scsi disk sdb at scsi1, channel 0, id 0, lun 0
Sep 17 14:41:42 m61 kernel: Attached scsi disk sdc at scsi1, channel 0, id 0, lun 1
Sep 17 14:41:42 m61 kernel: Attached scsi disk sdd at scsi1, channel 0, id 1, lun 0
Sep 17 14:41:42 m61 kernel: Attached scsi disk sde at scsi1, channel 0, id 1, lun 1
Sep 17 14:41:42 m61 kernel: SCSI device sda: 143374744 512-byte hdwr sectors (73408 MB)
Sep 17 14:41:42 m61 kernel: Partition check:
Sep 17 14:41:42 m61 kernel: sda: sda1 sda2 sda3 sda4
Sep 17 14:41:42 m61 kernel: SCSI device sdb: 2788016128 512-byte hdwr sectors (1427464 MB)
Sep 17 14:41:42 m61 kernel: sdb: sdb1
Sep 17 14:41:42 m61 kernel: SCSI device sdc: 2788016128 512-byte hdwr sectors (1427464 MB)
Sep 17 14:41:42 m61 kernel: sdc: sdc1
Sep 17 14:41:42 m61 kernel: SCSI device sdd: 4101521408 512-byte hdwr sectors (2099979 MB)
Sep 17 14:41:42 m61 kernel: sdd: sdd1
Sep 17 14:41:42 m61 kernel: SCSI device sde: 4101521408 512-byte hdwr sectors (2099979 MB)
Sep 17 14:41:42 m61 kernel: sde: sde1
Sep 17 14:41:42 m61 kernel: Attached scsi generic sg1 at scsi0, channel 0, id 6, lun 0, type 3
There is a DV configuration failure which limits the speed to U160,
but otherwise everything works nicely.
With 2.4.23-pre4 aix7xxx driver:
Sep 17 14:37:25 m61 kernel: scsi0 : Adaptec AIC79XX PCI-X SCSI HBA DRIVER, Rev 1.3.9
Sep 17 14:37:25 m61 kernel: <Adaptec AIC7902 Ultra320 SCSI adapter>
Sep 17 14:37:25 m61 kernel: aic7902: Ultra320 Wide Channel A, SCSI Id=7, PCI-X 101-133Mhz, 512 SCBs
Sep 17 14:37:25 m61 kernel:
Sep 17 14:37:25 m61 kernel: (scsi0:A:0): 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|QAS, 16bit)
Sep 17 14:37:25 m61 kernel: Vendor: SEAGATE Model: ST373307LC Rev: 0004
Sep 17 14:37:25 m61 kernel: Type: Direct-Access ANSI SCSI revision: 03
Sep 17 14:37:25 m61 kernel: scsi0:A:0:0: Tagged Queuing enabled. Depth 32
Sep 17 14:37:25 m61 kernel: Vendor: SUPER Model: GEM318 Rev: 0
Sep 17 14:37:25 m61 kernel: Type: Processor ANSI SCSI revision: 02
Sep 17 14:37:25 m61 kernel: scsi1 : Adaptec AIC79XX PCI-X SCSI HBA DRIVER, Rev 1.3.9
Sep 17 14:37:25 m61 kernel: <Adaptec AIC7902 Ultra320 SCSI adapter>
Sep 17 14:37:25 m61 kernel: aic7902: Ultra320 Wide Channel B, SCSI Id=7, PCI-X 101-133Mhz, 512 SCBs
Sep 17 14:37:25 m61 kernel:
Sep 17 14:37:25 m61 kernel: scsi1:A:0:0: DV failed to configure device. Please file a bug report against this driver.
Sep 17 14:37:25 m61 kernel: scsi1:A:1:0: DV failed to configure device. Please file a bug report against this driver.
Sep 17 14:37:25 m61 kernel: (scsi1:A:0): 160.000MB/s transfers (80.000MHz DT, 16bit)
Sep 17 14:37:25 m61 kernel: Vendor: transtec Model: Rev: 0001
Sep 17 14:37:25 m61 kernel: Type: Direct-Access ANSI SCSI revision: 03
Sep 17 14:37:25 m61 kernel: scsi1:A:0:0: Tagged Queuing enabled. Depth 32
Sep 17 14:37:25 m61 kernel: scsi: On host 1 channel 0 id 0 only 128 (max_scsi_report_luns) of 536870896 luns reported, try increasing max_scsi_report_luns.
Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 0 lun 0x7400616e73746563 has a LUN larger than currently supported.
Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 0 lun 0x2001202020202020 has a LUN larger than currently supported.
Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 0 lun 0x2002202020202020 has a LUN larger than currently supported.
Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 0 lun808529923 has a LUN larger than allowed by the host adapter
Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 0 lun3078 has a LUN larger than allowed by the host adapter
Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 0 lun 0x000000000f000000 has a LUN larger than currently supported.
Sep 17 14:37:25 m61 kernel: (scsi1:A:1): 160.000MB/s transfers (80.000MHz DT, 16bit)
Sep 17 14:37:25 m61 kernel: Vendor: transtec Model: Rev: 0001
Sep 17 14:37:25 m61 kernel: Type: Direct-Access ANSI SCSI revision: 03
Sep 17 14:37:25 m61 kernel: scsi1:A:1:0: Tagged Queuing enabled. Depth 32
Sep 17 14:37:25 m61 kernel: scsi: On host 1 channel 0 id 1 only 128 (max_scsi_report_luns) of 536870896 luns reported, try increasing max_scsi_report_luns.
Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 1 lun 0x7400616e73746563 has a LUN larger than currently supported.
Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 1 lun 0x2001202020202020 has a LUN larger than currently supported.
Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 1 lun 0x2002202020202020 has a LUN larger than currently supported.
Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 1 lun808529923 has a LUN larger than allowed by the host adapter
Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 1 lun3078 has a LUN larger than allowed by the host adapter
Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 1 lun 0x000000000f000000 has a LUN larger than currently supported.
Sep 17 14:37:25 m61 kernel: SCSI device sda: 143374744 512-byte hdwr sectors (73408 MB)
Sep 17 14:37:25 m61 kernel: SCSI device sda: drive cache: write back
Sep 17 14:37:25 m61 kernel: sda: sda1 sda2 sda3 sda4
Sep 17 14:37:25 m61 kernel: Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
Sep 17 14:37:25 m61 kernel: SCSI device sdb: 2788016128 512-byte hdwr sectors (1427464 MB)
Sep 17 14:37:25 m61 kernel: SCSI device sdb: drive cache: write back
Sep 17 14:37:25 m61 kernel: sdb: sdb1
Sep 17 14:37:25 m61 kernel: Attached scsi disk sdb at scsi1, channel 0, id 0, lun 0
Sep 17 14:37:25 m61 kernel: SCSI device sdc: 4101521408 512-byte hdwr sectors (2099979 MB)
Sep 17 14:37:25 m61 kernel: SCSI device sdc: drive cache: write back
Sep 17 14:37:25 m61 kernel: sdc: sdc1
Sep 17 14:37:25 m61 kernel: Attached scsi disk sdc at scsi1, channel 0, id 1, lun 0
Sep 17 14:37:25 m61 kernel: Attached scsi generic sg0 at scsi0, channel 0, id 0, lun 0, type 0
Sep 17 14:37:25 m61 kernel: Attached scsi generic sg1 at scsi0, channel 0, id 6, lun 0, type 3
Sep 17 14:37:25 m61 kernel: Attached scsi generic sg2 at scsi1, channel 0, id 0, lun 0, type 0
Sep 17 14:37:26 m61 kernel: Attached scsi generic sg3 at scsi1, channel 0, id 1, lun 0, type 0
You can see that the LUN enumeration exploded. We end up with the
first half of each RAID only. The configuration failure is still
there too. BTW, the behaviour is the same with 2.6.0-test3.
I can test and/or provide additional information as needed.
OG.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: AIC7xxx LUN enumeration failure and DV config failure
2003-09-17 13:06 AIC7xxx LUN enumeration failure and DV config failure Olivier Galibert
@ 2003-09-17 15:23 ` Patrick Mansfield
2003-09-18 11:44 ` Olivier Galibert
2003-09-17 16:20 ` Justin T. Gibbs
1 sibling, 1 reply; 4+ messages in thread
From: Patrick Mansfield @ 2003-09-17 15:23 UTC (permalink / raw)
To: Olivier Galibert; +Cc: linux-kernel, linux-scsi
On Wed, Sep 17, 2003 at 03:06:56PM +0200, Olivier Galibert wrote:
> We have two transtec 5016 IDE disks/SCSI interface raid arrays
> connected to a motherboard Adaptec AIC7902 Ultra320 SCSI adapter on
> scsi1. Each is split in two LUNs, 2x1.4Tb for the first and 2x2Tb for
> the second. There is also the root disk and $DEITY-knows-what on
> scsi0.
>
> With 2.4.22-pre9 aic79xx driver:
> Sep 17 14:41:42 m61 kernel: scsi1:A:0:0: DV failed to configure device. Please file a bug report against this driver.
> Sep 17 14:41:42 m61 kernel: scsi1:A:1:0: DV failed to configure device. Please file a bug report against this driver.
Are you sure the above is not a bug or a failure with the array (Justin)?
> Sep 17 14:41:42 m61 kernel: Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
> Sep 17 14:41:42 m61 kernel: Attached scsi disk sdb at scsi1, channel 0, id 0, lun 0
> Sep 17 14:41:42 m61 kernel: Attached scsi disk sdc at scsi1, channel 0, id 0, lun 1
> Sep 17 14:41:42 m61 kernel: Attached scsi disk sdd at scsi1, channel 0, id 1, lun 0
> Sep 17 14:41:42 m61 kernel: Attached scsi disk sde at scsi1, channel 0, id 1, lun 1
> There is a DV configuration failure which limits the speed to U160,
> but otherwise everything works nicely.
> With 2.4.23-pre4 aix7xxx driver:
The following looks like 2.6.x output, since 2.4.x does not have REPORT
LUN support.
> Sep 17 14:37:25 m61 kernel: scsi: On host 1 channel 0 id 0 only 128 (max_scsi_report_luns) of 536870896 luns reported, try increasing max_scsi_report_luns.
> Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 0 lun 0x7400616e73746563 has a LUN larger than currently supported.
> Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 0 lun 0x2001202020202020 has a LUN larger than currently supported.
> Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 0 lun 0x2002202020202020 has a LUN larger than currently supported.
> Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 0 lun808529923 has a LUN larger than allowed by the host adapter
> Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 0 lun3078 has a LUN larger than allowed by the host adapter
> Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 0 lun 0x000000000f000000 has a LUN larger than currently supported.
> Sep 17 14:37:25 m61 kernel: (scsi1:A:1): 160.000MB/s transfers (80.000MHz DT, 16bit)
> Sep 17 14:37:25 m61 kernel: Vendor: transtec Model: Rev: 0001
> Sep 17 14:37:25 m61 kernel: Type: Direct-Access ANSI SCSI revision: 03
> Sep 17 14:37:25 m61 kernel: scsi1:A:1:0: Tagged Queuing enabled. Depth 32
> Sep 17 14:37:25 m61 kernel: scsi: On host 1 channel 0 id 1 only 128 (max_scsi_report_luns) of 536870896 luns reported, try increasing max_scsi_report_luns.
> Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 1 lun 0x7400616e73746563 has a LUN larger than currently supported.
> Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 1 lun 0x2001202020202020 has a LUN larger than currently supported.
> Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 1 lun 0x2002202020202020 has a LUN larger than currently supported.
> Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 1 lun808529923 has a LUN larger than allowed by the host adapter
> Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 1 lun3078 has a LUN larger than allowed by the host adapter
> Sep 17 14:37:25 m61 kernel: scsi: host 1 channel 0 id 1 lun 0x000000000f000000 has a LUN larger than currently supported.
My guess is that something is totally screwed up with the arrays REPORT
LUN results.
Does it work OK if you turn off CONFIG_SCSI_REPORT_LUNS?
Did you check with the vendor about the problem?
You could also try modifying the arrays to report as SCSI 2, and then the
REPORT LUNS will not be sent.
If you could dump the result of sending a REPORT LUN to the device, that
would be useful. You could use the sg interface on a system (modify one of
the sg utils programs, like sg_inq.c) that found the LUNs OK, or hack the
kernel to dump the REPORT LUN output, or get a trace.
Also with CONFIG_SCSI_LOGGING enabled (hope I have this right), turn on
scsi scan logging by booting with:
scsi_mod.scsi_logging_level=0x000001c0
And post the output. The extra logging probably won't help, given that the
REPORT LUN output looks corrupted.
-- Patrick Mansfield
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: AIC7xxx LUN enumeration failure and DV config failure
2003-09-17 15:23 ` Patrick Mansfield
@ 2003-09-18 11:44 ` Olivier Galibert
0 siblings, 0 replies; 4+ messages in thread
From: Olivier Galibert @ 2003-09-18 11:44 UTC (permalink / raw)
To: Patrick Mansfield; +Cc: linux-kernel, linux-scsi
On Wed, Sep 17, 2003 at 08:23:59AM -0700, Patrick Mansfield wrote:
> The following looks like 2.6.x output, since 2.4.x does not have REPORT
> LUN support.
And it is. I had a blonde moment I guess and booted with the wrong
kernel. 2.4.23p4 works perfectly well.
> My guess is that something is totally screwed up with the arrays REPORT
> LUN results.
>
> Does it work OK if you turn off CONFIG_SCSI_REPORT_LUNS?
I'll see when I can try that.
> If you could dump the result of sending a REPORT LUN to the device, that
> would be useful. You could use the sg interface on a system (modify one of
> the sg utils programs, like sg_inq.c) that found the LUNs OK, or hack the
> kernel to dump the REPORT LUN output, or get a trace.
I'll try to do that.
OG.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: AIC7xxx LUN enumeration failure and DV config failure
2003-09-17 13:06 AIC7xxx LUN enumeration failure and DV config failure Olivier Galibert
2003-09-17 15:23 ` Patrick Mansfield
@ 2003-09-17 16:20 ` Justin T. Gibbs
1 sibling, 0 replies; 4+ messages in thread
From: Justin T. Gibbs @ 2003-09-17 16:20 UTC (permalink / raw)
To: Olivier Galibert, linux-kernel, linux-scsi
> We have two transtec 5016 IDE disks/SCSI interface raid arrays
> connected to a motherboard Adaptec AIC7902 Ultra320 SCSI adapter on
> scsi1. Each is split in two LUNs, 2x1.4Tb for the first and 2x2Tb for
> the second.
...
> Sep 17 14:41:42 m61 kernel: scsi1:A:0:0: DV failed to configure device. Please file a bug report against this driver.
> Sep 17 14:41:42 m61 kernel: scsi1:A:1:0: DV failed to configure device. Please file a bug report against this driver.
In the case of SCSI-IDE RAID arrays, many of which are of questionable
quality, this typically means that the device is not providing a valid
echo buffer size for performing domain validation. In this case, the
driver defaults to running at the maximum speed supported by the device,
but also prints out a diagnostic. I plan to remove the diagnostic for
the echo buffer issue since there seem to be lots of broken devices
out there.
Your RAID array only supports U160, so it only runs at that speed.
This has everything to do with the device, and nothing to do with the
aic79xx controller/driver.
As to the LUN issues, as already pointed out by others, it looks like
the report luns support in this device is broken. Perhaps the vendor
has a firmware update available.
--
Justin
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2003-09-18 11:44 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-09-17 13:06 AIC7xxx LUN enumeration failure and DV config failure Olivier Galibert
2003-09-17 15:23 ` Patrick Mansfield
2003-09-18 11:44 ` Olivier Galibert
2003-09-17 16:20 ` Justin T. Gibbs
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox