This is a long story that seems reminiscent of the issues people had after installing Yosemite and other things associated with CoreStorage. This monster thread describes those problems: https://forums.macrumors.com/threads/caution-yosemite-may-screw-up-partitions.1741742/
I recently bought a Mac Mini Server late 2012. It originally came with Mountain Lion installed and that suited me. It had two disks, one of which had been replaced by an SSD.
So I set it up with Mountain Lion on the SSD and it was working well. Then I decided to clone it to the HDD using Carbon Copy Cloner. I also got it to make a recovery partition on the HDD. In Disk Utility, it looked fine.
I then booted it off one of the recovery disks to check. In Disk Utility, the two disks appeared in red but were otherwise normal.
http://imgur.com/a/iJcZC
Clicking on any of the red lines brought up the error message shown. Clicking Ignore just closes the window. Clicking Fix made Disk Utility start making a Logical Volume Group. After about 10 minutes, Disk Utility was a permanent spinning beachball, so I Force Quit it. (This may not have been a good idea I realised later, but it seemed to be getting nowhere.)
[Later on, I let it go and got the beachball again. I realised that it was still using about 1% CPU time, so may still be doing something.]
http://imgur.com/a/iVKZE
Disk Utility was obviously out of its depth so I did some research and found the thread above. Booted from a USB stick with 10.8, diskutil cs list showed this:
CoreStorage logical volume groups (1 found)
|
+-- Logical Volume Group 79CBA286-4F8D-46C0-B7DA-CE89D2B991DC
=========================================================
Name: Internal Drive
Status: Online
Size: 1127552614400 B (1.1 TB)
Free Space: 1119023226880 B (1.1 TB)
|
+-< Physical Volume 1DB738B0-42B2-4ED6-8272-0049C12B6E90
| ----------------------------------------------------
| Index: 0
| Disk: disk0s2
| Status: Online
| Size: 127691702272 B (127.7 GB)
|
+-< Physical Volume 4D8F8B93-1B24-4F3F-8BEF-5042CB68F7FB
----------------------------------------------------
Index: 1
Disk: disk1s2
Status: Online
Size: 999860912128 B (999.9 GB)
Following the advice in the thread, I did this for the LVG:
diskutil cs delete UUID
After a few seconds, the LVG was gone. diskutil cs list said there was no core storage.
However, the two disks were still red, and clicking on either of them brought up the error message again. So there was still something odd about them.
I put the Mini into target disk mode and connected it to a Snow Leopard iMac as I figured core storage was not around then. I erased the disks using Apple Partition Map to make sure it was all changed:
diskutil eraseDisk HFS+ xxxx APM /dev/disk0
They looked fine. Reboot the Mini off a USB stick and the disks are still red. Aaaarrrggghhh!
I would have thought that repartitioning a disk would just write everything fresh and ignore anything that was still there. Especially since the reformat used APM not GUID.
All I can think of is:
1. Despite reformatting the disks, something has survived that is causing this. Does the UUID get wiped?
2. Could the system on the USB stick be caching some information and not looking at the actual disk?
3. Some hardware problem? Strange coincidence that both disks failed at the same time.
4. Could CCC have copied the UUID to the second disk, so the system thinks they are the same disk?
I would be grateful for any ideas on how to tackle this.
I recently bought a Mac Mini Server late 2012. It originally came with Mountain Lion installed and that suited me. It had two disks, one of which had been replaced by an SSD.
So I set it up with Mountain Lion on the SSD and it was working well. Then I decided to clone it to the HDD using Carbon Copy Cloner. I also got it to make a recovery partition on the HDD. In Disk Utility, it looked fine.
I then booted it off one of the recovery disks to check. In Disk Utility, the two disks appeared in red but were otherwise normal.
http://imgur.com/a/iJcZC
Clicking on any of the red lines brought up the error message shown. Clicking Ignore just closes the window. Clicking Fix made Disk Utility start making a Logical Volume Group. After about 10 minutes, Disk Utility was a permanent spinning beachball, so I Force Quit it. (This may not have been a good idea I realised later, but it seemed to be getting nowhere.)
[Later on, I let it go and got the beachball again. I realised that it was still using about 1% CPU time, so may still be doing something.]
http://imgur.com/a/iVKZE
Disk Utility was obviously out of its depth so I did some research and found the thread above. Booted from a USB stick with 10.8, diskutil cs list showed this:
CoreStorage logical volume groups (1 found)
|
+-- Logical Volume Group 79CBA286-4F8D-46C0-B7DA-CE89D2B991DC
=========================================================
Name: Internal Drive
Status: Online
Size: 1127552614400 B (1.1 TB)
Free Space: 1119023226880 B (1.1 TB)
|
+-< Physical Volume 1DB738B0-42B2-4ED6-8272-0049C12B6E90
| ----------------------------------------------------
| Index: 0
| Disk: disk0s2
| Status: Online
| Size: 127691702272 B (127.7 GB)
|
+-< Physical Volume 4D8F8B93-1B24-4F3F-8BEF-5042CB68F7FB
----------------------------------------------------
Index: 1
Disk: disk1s2
Status: Online
Size: 999860912128 B (999.9 GB)
Following the advice in the thread, I did this for the LVG:
diskutil cs delete UUID
After a few seconds, the LVG was gone. diskutil cs list said there was no core storage.
However, the two disks were still red, and clicking on either of them brought up the error message again. So there was still something odd about them.
I put the Mini into target disk mode and connected it to a Snow Leopard iMac as I figured core storage was not around then. I erased the disks using Apple Partition Map to make sure it was all changed:
diskutil eraseDisk HFS+ xxxx APM /dev/disk0
They looked fine. Reboot the Mini off a USB stick and the disks are still red. Aaaarrrggghhh!
I would have thought that repartitioning a disk would just write everything fresh and ignore anything that was still there. Especially since the reformat used APM not GUID.
All I can think of is:
1. Despite reformatting the disks, something has survived that is causing this. Does the UUID get wiped?
2. Could the system on the USB stick be caching some information and not looking at the actual disk?
3. Some hardware problem? Strange coincidence that both disks failed at the same time.
4. Could CCC have copied the UUID to the second disk, so the system thinks they are the same disk?
I would be grateful for any ideas on how to tackle this.