SunFire V120 with A1000
OK, we've got a SunFire V120 server with an X6541A PCI SCSI card installed. The card is connected to an A1000 RAID Array with a good quality HVD SCSI cable. The other port on the A1000 has a HVD terminator installed. There are eight 73GB drives in the array, and there are two internal drives running on the server's built-in SCSI. Running Solaris 8 on the server.
So far, I have not been able to get the server to recognize the array at all. If I run a "probe-scsi-all" I get this:
<i>
ok probe-scsi-all
/<a href="mailto:pci@1f" target="_blank">pci@1f</a>,0/<a href="mailto:pci@1" target="_blank">pci@1</a>/<a href="mailto:scsi@5" target="_blank">scsi@5</a>,1
/<a href="mailto:pci@1f" target="_blank">pci@1f</a>,0/<a href="mailto:pci@1" target="_blank">pci@1</a>/<a href="mailto:scsi@5" target="_blank">scsi@5</a>
Fatal SCSI error at script address 10 Unexpected disconnect
/<a href="mailto:pci@1f" target="_blank">pci@1f</a>,0/<a href="mailto:pci@1" target="_blank">pci@1</a>/<a href="mailto:scsi@8" target="_blank">scsi@8</a>,1
/<a href="mailto:pci@1f" target="_blank">pci@1f</a>,0/<a href="mailto:pci@1" target="_blank">pci@1</a>/<a href="mailto:scsi@8" target="_blank">scsi@8</a>
Target 0
Unit 0DiskFUJITSU MAP3367N SUN36G 0301
Target 1
Unit 0DiskSEAGATE ST336704LSUN36G 0326</i>
As you can see, it recognizes the internal drives on SCSI address 8 and the X6541 card at SCSI address 5, but reports an error on the port attached to the A1000.
I also get the following email to root occasionally:
<i>To: root
Subject: raid Event
Content-Length: 183
An array event has been detected on Controller Unknown
Device Unknown at Host <<our domain name here>> - Time 08/23/2005 21:14:27</i>
And I get errors on boot up:
<i>Sun Fire V120 (UltraSPARC-IIe 648MHz), No Keyboard
OpenBoot 4.0, 1536 MB memory installed, Serial #53835884.
Ethernet address 0:3:ba:35:78:6c, Host ID: 8335786c.
last command: boot
Boot device: disk File and args:
SunOS Release 5.8 Version Generic_108528-17 64-bit
Copyright 1983-2001 Sun Microsystems, Inc. All rights reserved.
WARNING: /<a href="mailto:pci@1f" target="_blank">pci@1f</a>,0/<a href="mailto:pci@1" target="_blank">pci@1</a>/<a href="mailto:scsi@5" target="_blank">scsi@5</a> (glm2):
unexpected SCSI interrupt while idle
configuring IPv4 interfaces: eri0.
Hostname: <<our domain name here>>
The system is coming up. Please wait.
checking ufs filesystems
/dev/rdsk/c0t1d0s6: is stable.
/dev/rdsk/c0t0d0s7: 52762 files, 11375733 used, 10986210 free
/dev/rdsk/c0t0d0s7: (39522 frags, 1368336 blocks, 0.1% fragmentation)
/dev/rdsk/c0t0d0s4: is stable.
8/24/2005 1:14:21 GMT LOM time reference
starting rpc services: rpcbind done.
Setting netmask of eri0 to 255.255.255.240
Setting default IPv4 interface for multicast: add net 224.0/4: gateway <<our domain name here>>
syslog service starting.
Print services started.
There are no devices (controllers) in the system; nvutil terminated.
There are no devices (controllers) in the system.
fwutil failed!
Array Monitor initiated
Aug 23 21:14:27 /usr/lib/osa/bin/arraymon: No RAID devices found to check.
RDAC daemons initiated
volume management starting.
Wnn6: Key License Server started....
Nihongo Multi Client Server (Wnn6 R2.34)
Finished Reading Files
httpd starting.
Starting nrpe: Starting mysqld daemon with databases from /usr/local/mysql/data started
The system is ready.</i>
The first SCSI error you see in the boot sequence:
<i>WARNING: /<a href="mailto:pci@1f" target="_blank">pci@1f</a>,0/<a href="mailto:pci@1" target="_blank">pci@1</a>/<a href="mailto:scsi@5" target="_blank">scsi@5</a> (glm2):
unexpected SCSI interrupt while idle</i>
is sometimes replaced with:
<i>WARNING: invalid vector intr: number 0x7df, pil 0x0</i>
Any ideas at all? Is it the A1000, it's RAID controller, the X6541A, PROM settings? Anything?
Thanks!
- Matt
