Dear TruGurus,
I have previously set up TruCluster V1.5 on Two Alpha Server 1200s (4.0D)
with sucess.
Now, over a year later I am setting up two identical (H/W) Alpha Server
1200s (4.0F) with
TruCluster V1.6., the only hardware configuration difference being that this
"new" cluster has two RA3000
controllers/two shared SCSI buses. The two shared buses are driven by two
KZPBA-CBs in each
AS1200, each on different PCI hoses. One shared bus scheme uses the Y-cable
and the other
a DWZZH-03 SCSI hub ( The Hubbed bus/ second HSZ22 was an upgrade after
delivery).
The RAID sets have been set up previously with SWCC and all sets are 0+1.
Here's the problem:
Before I installed the TruCluster software and subsequent patches and
rebooted, both servers booted
just fine and both saw all of the HSZ22 RAID sets identically. As soon as I
installed TruCluster, set it up
with two shared buses - logically numbered 16 & 17 on both members and then
immediately installed the latest
4.0F and TC16 patches, I am unable to reboot the last server coming up,
because the first server obviously
reserves (?) the HSZ22(s) and the second server just sits there in a
continual loop reported SCSI CAM errors.
Either server sill boot with the other one offline.
What did I do wrong?
What is causing this contention now that TC16 is installed and starting up
it's agent at boot time?
Is it the TC16 patches? ( This was the first time EVER I installed Tru64
patches "un-reversible" !! )
Hopefully it's a simple oversight...
Thanks in advance.
-Wayne
Received on Mon Jan 08 2001 - 19:46:23 NZDT