Ok, i think i gathered some relevant evidence. I had the drive plugged on a 16.04, and at some point i noticed it went standby (led blinking intermittently).
Last dmesg lines were [ 6217.720607] perf interrupt took too long (5053 > 5000), lowering kernel.perf_event_max_sample_rate to 25000 [ 6631.699961] sd 7:0:0:0: [sdc] tag#0 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_SENSE [ 6631.699969] sd 7:0:0:0: [sdc] tag#0 Sense Key : Hardware Error [current] [descriptor] [ 6631.699973] sd 7:0:0:0: [sdc] tag#0 Add. Sense: No additional sense information [ 6631.699978] sd 7:0:0:0: [sdc] tag#0 CDB: ATA command pass through(16) 85 06 20 00 00 00 00 00 00 00 00 00 00 00 e5 00 [ 6632.179875] sd 7:0:0:0: [sdc] tag#0 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_SENSE [ 6632.179884] sd 7:0:0:0: [sdc] tag#0 Sense Key : Hardware Error [current] [descriptor] [ 6632.179889] sd 7:0:0:0: [sdc] tag#0 Add. Sense: No additional sense information [ 6632.179895] sd 7:0:0:0: [sdc] tag#0 CDB: ATA command pass through(12)/Blank a1 06 20 da 00 00 4f c2 00 b0 00 00 i checked uptime and it was 1:55. So i took 5 min, and waited until the next DRIVE_SENSE error would appear in dmesg, monitoring the situation And guess what, the drive woke up right before the next error appeared in dmesg! Like, spin up (led blinking intermittently but much faster) and after 5-7 seconds, drive ready, and error on dmesg. 21:14:06 up 2:00, 1 user, load average: 0,11, 0,41, 1,03 [ 6032.254264] sd 7:0:0:0: [sdc] tag#0 Sense Key : Hardware Error [current] [descriptor] [ 6032.254266] sd 7:0:0:0: [sdc] tag#0 Add. Sense: No additional sense information [ 6032.254270] sd 7:0:0:0: [sdc] tag#0 CDB: ATA command pass through(12)/Blank a1 06 20 da 00 00 4f c2 00 b0 00 00 [ 6217.720607] perf interrupt took too long (5053 > 5000), lowering kernel.perf_event_max_sample_rate to 25000 [ 6631.699961] sd 7:0:0:0: [sdc] tag#0 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_SENSE [ 6631.699969] sd 7:0:0:0: [sdc] tag#0 Sense Key : Hardware Error [current] [descriptor] [ 6631.699973] sd 7:0:0:0: [sdc] tag#0 Add. Sense: No additional sense information [ 6631.699978] sd 7:0:0:0: [sdc] tag#0 CDB: ATA command pass through(16) 85 06 20 00 00 00 00 00 00 00 00 00 00 00 e5 00 [ 6632.179875] sd 7:0:0:0: [sdc] tag#0 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_SENSE [ 6632.179884] sd 7:0:0:0: [sdc] tag#0 Sense Key : Hardware Error [current] [descriptor] [ 6632.179889] sd 7:0:0:0: [sdc] tag#0 Add. Sense: No additional sense information [ 6632.179895] sd 7:0:0:0: [sdc] tag#0 CDB: ATA command pass through(12)/Blank a1 06 20 da 00 00 4f c2 00 b0 00 00 [ 7238.669972] sd 7:0:0:0: [sdc] tag#0 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_SENSE [ 7238.669981] sd 7:0:0:0: [sdc] tag#0 Sense Key : Hardware Error [current] [descriptor] [ 7238.669985] sd 7:0:0:0: [sdc] tag#0 Add. Sense: No additional sense information [ 7238.669990] sd 7:0:0:0: [sdc] tag#0 CDB: ATA command pass through(16) 85 06 20 00 00 00 00 00 00 00 00 00 00 00 e5 00 [ 7238.794856] sd 7:0:0:0: [sdc] tag#0 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_SENSE [ 7238.794863] sd 7:0:0:0: [sdc] tag#0 Sense Key : Hardware Error [current] [descriptor] [ 7238.794867] sd 7:0:0:0: [sdc] tag#0 Add. Sense: No additional sense information [ 7238.794871] sd 7:0:0:0: [sdc] tag#0 CDB: ATA command pass through(12)/Blank a1 06 20 da 00 00 4f c2 00 b0 00 00 Now, i'm not a kernel/SATA subsystem expert, but this strongly smells as the redhat bug and this problem are closely related. I'm not sure why Alan Stern refuses to look at this, and only offers patches to hide the error. There's clearly something different in this kernel (not sure how far it dates, but clearly it's not in 3.19) that doesnt let the disks sleep. And this is a SERIOUS regression. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1733068 Title: Unmounted USB drives wake up on 16.04 To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+bug/1733068/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
