Comment 140 for bug 1679898

I see this issue too. It is easy to get with a 4.4 kernel if you have an SSD hooked up to the Windows host and then you connect the SSD to the Hyper-V VM via the SCSI controller in passthrough mode. Error messages will look like this:

[ 20.476616] sd 3:0:0:0: [sdb] Synchronizing SCSI cache
[ 38.374458] scsi 3:0:0:0: Direct-Access Samsung SSD 850 PRO 256G EXM0 PQ: 0 ANSI: 0
[ 38.374703] scsi 3:0:0:1: Direct-Access PQ: 0 ANSI: 0
[ 38.374919] scsi 3:0:0:2: Direct-Access PQ: 0 ANSI: 0
[ 38.375368] scsi 3:0:0:3: Direct-Access PQ: 0 ANSI: 0
[ 38.375711] scsi 3:0:0:4: Direct-Access PQ: 0 ANSI: 0
[ 38.376526] scsi 3:0:0:5: Direct-Access PQ: 0 ANSI: 0
[ 38.376791] scsi 3:0:0:6: Direct-Access PQ: 0 ANSI: 0
[ 38.377114] scsi 3:0:0:7: Direct-Access PQ: 0 ANSI: 0
[ 38.392809] sd 3:0:0:0: Attached scsi generic sg2 type 0
[ 38.393282] sd 3:0:0:0: [sdb] 500118192 512-byte logical blocks: (256 GB/238 GiB)
[ 38.393811] sd 3:0:0:1: [sdc] Sector size 0 reported, assuming 512.
[ 38.393819] sd 3:0:0:1: [sdc] 1 512-byte logical blocks: (512 B/512 B)
[ 38.393822] sd 3:0:0:1: [sdc] 0-byte physical blocks
[ 38.393874] sd 3:0:0:1: Attached scsi generic sg3 type 0
[ 38.394262] sd 3:0:0:0: [sdb] Write Protect is off
[ 38.394267] sd 3:0:0:0: [sdb] Mode Sense: 0f 00 00 00
[ 38.394277] sd 3:0:0:1: [sdc] Write Protect is off
[ 38.394281] sd 3:0:0:1: [sdc] Mode Sense: 00 00 00 00
[ 38.395394] sd 3:0:0:2: [sdd] Sector size 0 reported, assuming 512.
[ 38.395402] sd 3:0:0:2: [sdd] 1 512-byte logical blocks: (512 B/512 B)
[ 38.395405] sd 3:0:0:2: [sdd] 0-byte physical blocks
[ 38.395553] sd 3:0:0:2: Attached scsi generic sg4 type 0
[ 38.395730] sd 3:0:0:2: [sdd] Write Protect is off
[ 38.395735] sd 3:0:0:2: [sdd] Mode Sense: 00 00 00 00
[ 38.395746] sd 3:0:0:1: [sdc] Asking for cache data failed
[ 38.395754] sd 3:0:0:1: [sdc] Assuming drive cache: write through
[ 38.400079] sd 3:0:0:2: [sdd] Asking for cache data failed
[ 38.400088] sd 3:0:0:2: [sdd] Assuming drive cache: write through
[ 38.400268] sd 3:0:0:0: [sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[ 38.400418] sd 3:0:0:3: [sde] Sector size 0 reported, assuming 512.
[ 38.400425] sd 3:0:0:3: [sde] 1 512-byte logical blocks: (512 B/512 B)
[ 38.400427] sd 3:0:0:3: [sde] 0-byte physical blocks
[ 38.400480] sd 3:0:0:3: [sde] Write Protect is off
[ 38.400483] sd 3:0:0:3: [sde] Mode Sense: 00 00 00 00
[ 38.401653] sd 3:0:0:3: [sde] Asking for cache data failed
[ 38.401660] sd 3:0:0:3: [sde] Assuming drive cache: write through
[ 38.401768] sd 3:0:0:1: [sdc] Sector size 0 reported, assuming 512.
[ 38.402068] sd 3:0:0:2: [sdd] Sector size 0 reported, assuming 512.
[ 38.402258] sd 3:0:0:3: [sde] Sector size 0 reported, assuming 512.
[...]
[ 38.415473] sdi: partition table beyond EOD, enabling native capacity
[ 38.416509] sd 3:0:0:7: [sdi] Sector size 0 reported, assuming 512.
[ 38.417072] Dev sdi: unable to read RDB block 1
[ 38.417093] sdi: unable to read partition table
[ 38.417099] sdi: partition table beyond EOD, truncated
[ 38.417516] sd 3:0:0:7: [sdi] Sector size 0 reported, assuming 512.
[ 38.417617] sd 3:0:0:7: [sdi] Attached SCSI disk
[ 38.841506] sd 3:0:0:0: [storvsc] Sense Key : Illegal Request [current]
[ 38.841510] sd 3:0:0:0: [storvsc] Add. Sense: Invalid command operation code
[ 38.841895] sd 3:0:0:0: [storvsc] Sense Key : Illegal Request [current]
[ 38.841898] sd 3:0:0:0: [storvsc] Add. Sense: Invalid command operation code

Sadly 4.4.0-67-generic #88~lp1679898Commitdf0a1b9cce531 mentioned in https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1677938/comments/6 doesn't fix the problem. It is also broken in 4.8.0-34-generic #36~16.04.1-Ubuntu but it is FIXED in 4.8.0-58-generic #63~16.04.1-Ubuntu !