Greetings,
I'm attempting to upgrade one of our servers to 6.4 (from 6.2)
Before doing so I need to stripe 3 identical drives on the
same SCSI port into one drive, the drives are on an LSI controller
with 2 ports:
May 5 02:10:15 hitme kernel: sym0: <1010-66> port 0xe400-0xe4ff mem
0xfebe0000-0xfebe03ff,0xfebd8000-0xfebd9fff irq 29 at device 6.0 on pci1
May 5 02:10:15 hitme kernel: sym0: Reserved 0x400 bytes for rid 0x14
type 3 at 0xfebe0000
May 5 02:10:15 hitme kernel: sym0: Reserved 0x2000 bytes for rid 0x1c
type 3 at 0xfebd8000
May 5 02:10:15 hitme kernel: sym0: Symbios NVRAM, ID 7, Fast-80, LVD,
parity checking
May 5 02:10:15 hitme kernel: sym0: open drain IRQ line driver, using
on-chip SRAM
May 5 02:10:15 hitme kernel: sym0: using LOAD/STORE-based firmware.
May 5 02:10:15 hitme kernel: sym0: handling phase mismatch from SCRIPTS.
May 5 02:10:15 hitme kernel: sym0: initial
SCNTL3/DMODE/DCNTL/CTEST3/4/5 = (hex) 00/4e/a0/01/00/04
May 5 02:10:15 hitme kernel: sym0: final
SCNTL3/DMODE/DCNTL/CTEST3/4/5 = (hex) 00/4e/80/01/08/04
May 5 02:10:15 hitme kernel: ioapic1: routing intpin 13 (PCI IRQ 29)
to vector 52
May 5 02:10:15 hitme kernel: sym0: [GIANT-LOCKED]
May 5 02:10:15 hitme kernel: sym0: enabling clock multiplier
May 5 02:10:15 hitme kernel: sym0: Downloading SCSI SCRIPTS.
May 5 02:10:15 hitme kernel: sym1: <1010-66> port 0xe800-0xe8ff mem
0xfebf8000-0xfebf83ff,0xfebf0000-0xfebf1fff irq 28 at device 6.1 on pci1
May 5 02:10:15 hitme kernel: sym1: Reserved 0x400 bytes for rid 0x14
type 3 at 0xfebf8000
May 5 02:10:15 hitme kernel: sym1: Reserved 0x2000 bytes for rid 0x1c
type 3 at 0xfebf0000
May 5 02:10:15 hitme kernel: sym1: Symbios NVRAM, ID 7, Fast-80, SE,
parity checking
May 5 02:10:15 hitme kernel: sym1: open drain IRQ line driver, using
on-chip SRAM
May 5 02:10:15 hitme kernel: sym1: using LOAD/STORE-based firmware.
May 5 02:10:15 hitme kernel: sym1: handling phase mismatch from SCRIPTS.
May 5 02:10:15 hitme kernel: sym1: initial
SCNTL3/DMODE/DCNTL/CTEST3/4/5 = (hex) 00/4e/a0/01/00/04
May 5 02:10:15 hitme kernel: sym1: final
SCNTL3/DMODE/DCNTL/CTEST3/4/5 = (hex) 00/4e/80/01/08/04
the drives are identical 17.5Gb U160 drives (unused)
May 5 02:10:15 hitme kernel: pass0: <IBM DPSS-318350M S80D> Fixed
Direct Access SCSI-3 device May 5 02:10:15 hitme kernel: pass0: Serial
Number ZE0E4501
May 5 02:10:15 hitme kernel: pass0: 160.000MB/s transfers (80.000MHz,
offset 62, 16bit), Tagged Queueing Enabled
May 5 02:10:15 hitme kernel: pass1 at sym0 bus 0 target 1 lun 0
May 5 02:10:15 hitme kernel: pass1: <IBM DPSS-318350M S80D> Fixed
Direct Access SCSI-3 device May 5 02:10:15 hitme kernel: pass1: Serial
Number ZE0G1630
May 5 02:10:15 hitme kernel: pass1: 160.000MB/s transfers (80.000MHz,
offset 62, 16bit), Tagged Queueing Enabled
May 5 02:10:15 hitme kernel: pass2 at sym0 bus 0 target 2 lun 0
May 5 02:10:15 hitme kernel: pass2: <IBM DPSS-318350M S80D> Fixed
Direct Access SCSI-3 device May 5 02:10:15 hitme kernel: pass2: Serial
Number ZE0B4449
May 5 02:10:15 hitme kernel: pass2: 160.000MB/s transfers (80.000MHz,
offset 62, 16bit), Tagged Queueing Enabled
May 5 02:10:15 hitme kernel: pass3 at sym1 bus 0 target 3 lun 0
The current system (6.2) is on a single drive on the other SCSI port
(IBM - identical to the others).
However, when I attempt to stripe the 3 unused drives, I recieve an
error regarding da2:
gstripe label -v st0 /dev/da0 /dev/da1 /dev/da2
GEOM_STRIPE: Device st0 created (id=2607126992).
GEOM_STRIPE: Disk da0 attached to st0.
Metadata value stored on /dev/da0.
GEOM_STRIPE: Disk da1 attached to st0.
Metadata value stored on /dev/da1.
GEOM_STRIPE: Disk da2 attached to st0.
GEOM_STRIPE: Device st0 activated.
GEOM_STRIPE: Cannot add disk da2c to st0 (error=17).
Metadata value stored on /dev/da2.
I blanked the 3 drives prior to attempting any of this, and it's
on a fresh reboot. I'm afraid I don't know how to proceed. Is it
a problem with gstripe(8)? I /really/ need to stripe these drives
so as to upgrade the system.
Thank you for all your time and consideration in this matter.
Sincerely,
Chris H