Hi. We have a 2008 R2 Hyper-V cluster (using CSVs on iSCSI SAN). We've had a ISCSI network breakdown couple days ago. After we fixed the problem, everything came up normally, but we found out that one VM was constantly losing one virtual drive - it would appear in disk management, then disappear, then appear again. We couldn't find any problem in the VM itself, so we've turned on maintenance mode on the CSV that the problematic disk was one and ran the Repair-ClusterSharedVolume PSH cmdlet.
To my surprise, the scan reported problems, but wasn't able to fix them. So I've ran full scan (using parameters /F /V /R), but that wasn't able to fix the problem either. Here's the error output:
... ... Inserting an index entry with Id 477 into index $SDH of file 9. Inserting an index entry with Id 478 into index $SDH of file 9. Insufficient disk space to fix the security descriptors data stream. Repairing the security file record segment. 512 file SDs/SIDs processed. Security descriptor verification completed. 14 data files processed. CHKDSK is verifying file data (stage 4 of 5)... 496 files processed. File data verification completed. CHKDSK is verifying free space (stage 5 of 5)... 1073656547 free clusters processed. Free space verification is complete. The second NTFS boot sector is unwriteable.
We were out of time and there were fortunately only 2 VMs on the CSV in question, so we've moved them off and deleted and recreated the whole CSV volume.
My question is what are those two errors ("Insufficient disk space to fix the security descriptors data stream." - CSV had 3.8 out of 4TB free, and "The second NTFS boot sector is unwriteable"?