Thank you Cisco 
Attempted the upgrade from customers 6.1.4 to 9.0 this weekend and guess what..... Yep it failed, I upgraded the hardware so that it would support the upgrade to 9.0(1) but during the upgrade process it disables the RAID controller and formats the disks, this is where it failed. I followed this process: -
1/ Ran backup of customers current 6.1.4 installation o.k.
2/ Shutdown publisher and installed 4GB Memory
3/ Rebooted customers current publisher
4/ Made sure that all customer devices are registered to the subscriber
5/ Ran refresh upgrade cop file via SFTP – completed o.k.
6/ Inserted 1TB external USB disk
7/ Started upgrade process using DVD that was created from Install ISO (Image to disk)
8/ After about 2hrs starting to get concerned as the install log had not updated for some time, checked install progress from CLI using Utils System Upgrade Status – Installing
9/ Another hour later still no progress listed in the log file, no failure either, just appears to be hanging!!!
10/ Searched Internet for estimated time, anything from 2hrs to 6 hours, decided to leave and monitor it from remote.
11/ After 7hrs it was decided that this installation had failed, no reason listed in the log file for the failure
12/ Came in next morning and still no progress on the previous upgrade, rebooted publisher to still working 6.1.4 no sign of any upgrade taking place
13/ Started upgrade process again, this time from network SFTP server.
14/ log files show progress from the start and continue to update as we go, looks like the DVD was either corrupt or just wasn’t being read properly by the old DVD drive (Perhaps only reading one layer)
15/ After an hour or so the System reboots for the first time and continues install process
16/ System then says that it is stopping the RAID controller and re-booted, this caused the server to reboot and then come up with the error – “an error occurred trying to format md0, this problem is serious and the install cannot continue.”
17/ I removed the external USB and re-booted the server and got Grub loading error 17 – this is related to the master boot record being corrupt or broken
18/ re-installed CUCM 6.1.4 – this was difficult and I had to reset the RAID and format the disks before it would install
19/ Restored customer data from the backup, rebooted the cluster and all is happy back on 6.1.4
After this I checked my USB disk and saw that it had been partitioned as 2 x 36GB partitions, 1 x 19GB partition and the rest as a final partition. To me it looks like the install has tried to use my USB device as one of the HDD’s?
So after 14 hours of messing around and watching a screen do hardly anyhthing (Paint drying is more exciting) we are back to square one. Customer contact is also very concerned that we might have lost the publisher completely during this process. There must be an easier way to upgrade a Callmanager cluster that is only running 176 phones! Just give me the license, I will give you the servers when we are finished the migration...
