[parisc-linux] Dodgy SCSI in L2000
James Braid
james.braid@peace.com
Tue, 7 May 2002 19:00:40 +1200
> Which kernel version?
Ooops, forgot to include this, its:
2.4.18-pa24, and
2.4.18-pa23
> This suggests something is wrong with that device.
> Are all RAID0 disks internal? Or connected to external box?
> I see further down they seem to be working fine individually.
All disks are internal disks in the hotswap cage thing, connected to the
core I/O board (? not sure if that's what its called , it's a big board
with 3 SCSI cables on it and the ethernet jack/external SCSI). All the
disks are plugged into this, two disks per cable, and the DVD on its own
cable.
> Can you post lspci -v output from you L2000?
> I just want to verify the built-in SCSI chips are the
> same as on my A500: sym53c896.
Sure:
hypo:~# lspci -v
00:00.0 Ethernet controller: Digital Equipment Corporation DECchip
21142/43 (rev 41)
Subsystem: Hewlett-Packard Company: Unknown device 104f
Flags: bus master, medium devsel, latency 128, IRQ 128
I/O ports at 0080 [size=128]
Memory at fffffff0 (32-bit, non-prefetchable) [size=1K]
Expansion ROM at fffff800 [size=256K]
00:01.0 SCSI storage controller: LSI Logic / Symbios Logic (formerly
NCR) 53c896 (rev 05)
Flags: bus master, medium devsel, latency 0, IRQ 129
I/O ports at 0100 [size=256]
Memory at fffffff0 (64-bit, non-prefetchable) [size=1K]
Memory at fffffff0 (64-bit, non-prefetchable) [size=8K]
Capabilities: [40] Power Management version 2
00:01.1 SCSI storage controller: LSI Logic / Symbios Logic (formerly
NCR) 53c896 (rev 05)
Flags: bus master, medium devsel, latency 0, IRQ 130
I/O ports at 0200 [size=256]
Memory at fffffff0 (64-bit, non-prefetchable) [size=1K]
Memory at fffffff0 (64-bit, non-prefetchable) [size=8K]
Capabilities: [40] Power Management version 2
00:02.0 SCSI storage controller: LSI Logic / Symbios Logic (formerly
NCR) 53c875 (rev 14)
Flags: bus master, medium devsel, latency 0, IRQ 130
I/O ports at 0300 [size=256]
Memory at fffffff0 (32-bit, non-prefetchable) [size=256]
Memory at fffffff0 (32-bit, non-prefetchable) [size=4K]
00:02.1 SCSI storage controller: LSI Logic / Symbios Logic (formerly
NCR) 53c875 (rev 14)
Flags: bus master, medium devsel, latency 0, IRQ 131
I/O ports at 0400 [size=256]
Memory at fffffff0 (32-bit, non-prefetchable) [size=256]
Memory at fffffff0 (32-bit, non-prefetchable) [size=4K]
00:04.0 Serial controller: Hewlett-Packard Company: Unknown device 1048
(rev 02) (prog-if 02 [16550])
Subsystem: Hewlett-Packard Company: Unknown device 1049
Flags: bus master, medium devsel, latency 0, IRQ 132
BIST result: 00
Memory at fffffff0 (32-bit, non-prefetchable) [size=4K]
I/O ports at 0040 [size=64]
00:05.0 Serial controller: Hewlett-Packard Company: Unknown device 1048
(rev 02) (prog-if 02 [16550])
Subsystem: Hewlett-Packard Company: Unknown device 104a
Flags: medium devsel, IRQ 133
BIST result: 00
Memory at fffffff0 (32-bit, non-prefetchable) [size=4K]
I/O ports at 0500 [size=64]
08:02.0 SCSI storage controller: QLogic Corp. ISP1020 Fast-wide SCSI
(rev 05)
Subsystem: QLogic Corp.: Unknown device 0000
Flags: medium devsel, IRQ 194
I/O ports at 10000 [disabled] [size=256]
Memory at fffffff0 (32-bit, non-prefetchable) [disabled]
[size=4K]
Expansion ROM at fffff800 [size=64K]
08:03.0 SCSI storage controller: QLogic Corp. ISP1020 Fast-wide SCSI
(rev 05)
Subsystem: QLogic Corp.: Unknown device 0000
Flags: medium devsel, IRQ 195
I/O ports at 10100 [disabled] [size=256]
Memory at fffffff0 (32-bit, non-prefetchable) [disabled]
[size=4K]
Expansion ROM at fffff800 [size=64K]
20:00.0 SCSI storage controller: LSI Logic / Symbios Logic (formerly
NCR) 53c875 (rev 04)
Flags: bus master, medium devsel, latency 0, IRQ 384
I/O ports at 40000 [size=256]
Memory at fffffff0 (32-bit, non-prefetchable) [size=256]
Memory at fffffff0 (32-bit, non-prefetchable) [size=4K]
Expansion ROM at fffff800 [size=64K]
The Qlogic cards I put in as I was going to move the scsi drives onto
them, but then I found out that the Qlogic ISP driver doesn't compile
for PA-RISC (doh!). The problem happens with or without the Qlogic
boards in there.
> Can you try running two different dbench instances at the some time on
> two different disks?
On three disks (sdb/sdc/sdd):
hypo:/sdb# dbench 10
10 clients started
..........
hypo:/sdc# dbench 10
10 clients started
..............
hypo:/sdd# dbench 10
10 clients started
........................................................................
................................................
None of them finish, and I get a ton of both bus and device
RESETS/ABORTS on the GSP terminal.
On two disks (sdc/sdc):
hypo:/sdb# dbench 10
10 clients started
......................................
hypo:/sdc# dbench 10
10 clients started
...............
Pretty much the same thing, lots of RESETS/ABORTS. I can post these if
needed, but its pretty much just more of what was in my first message.
It all goes pear shaped after this with lots of I/O errors and EXT2
errors, e.g.
EXT2-fs error (device sd(8,17)): ext2_write_inode: unable to read inode
block - inode=1357217, block=2719749
I/O error: dev 08:11, sector 9961512
I/O error: dev 08:11, sector 0
I will try again with dbench on just one disk at a time as well.
Thanks, James.