SUMMARY: Dual hosting A1000's SCSI errors second v480 wouldn't boot

From: Sid Wilroy (sid@instantis.com)
Date: Fri Sep 13 2002 - 17:17:43 EDT


I have 2 v480 each with a Dual-channel differential UltraSCSI host
adapter (X6541A) each channel is connect to one of the
A1000s. All SCSI IN/OUT ports on both A1000 are taken by both v480s
After installing the patches and raidmanager 6, I created the LUNs (very
easy) and mounted them, everything was working well minus a few SCSI
reset errors printing to the console.
When I tried to boot the second attached v480 it froze at the following
errors:
WARNING: glm1: SCSI bus reset recursion
WARNING: pcisch1: ino 0x1 has been blocked
WARNING: glm1: interrupt #0 has been blocked
WARNING: /pci@8,600000/scsi@1 (glm0):
unexpected SCSI interrupt while idle
WARNING: /pci@8,600000/scsi@1 (glm0):
SCSI bus reset recursion
WARNING: glm0: fault detected in device; service unavailable
WARNING: glm0: SCSI bus reset recursion
WARNING: pcisch1: ino 0x0 has been blocked
WARNING: glm0: interrupt #0 has been blocke
If I don't have the A1000 LUNs mounted, the second V480 would come up
but with some errors.

Several SUN engineers told me I had to get SUN Cluster software:
You must either have a termination point on a scsi bus, either a
terminator or a self terminating device, or
software to allow the management of the traffic on that bus.

Well this was no true after talking with one of the SUN engineers in the
HA group  Qui McDivitt
He told me to do the following on the V480 that wasnt coming up:
Bring both machines to the OK prompt (so neither v480 is using the scsi
bus or umount the LUNs)
Ok> reset-all
Ok> probe-scsi-all
Look for the whole path to each controller
/pci@6,4000/scsi@3,1 (example)
ok> nvedit
0: probe-all install-console banner
1: cd /pci@6,4000/scsi@3,1
2: 6 " scsi-initiator-id" integer-property
3: device-end
4: cd /pci@6,4000/scsi@3
5: 6 " scsi-initiator-id" integer-property
6: device-end
ctrl-c to exit nvedit
ok>nvstore
ok>nvramrc evaluate
ok>setenv use-nvramrc? true
ok>reset-all
Edit the /etc/system file (on both nodes) to set fast/wide SCSI
set scsi_options=0x3f8
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:24:56 EDT