[Bug 1774569] Re: gsmartcontrol, hdparm, and ZFS all refuse to talk to an apparently working Seagate Backup+ Hub drive after upgrade to 18.04

Adam Novak 1774569 at bugs.launchpad.net
Tue Jun 5 05:52:28 UTC 2018


It looks like the drive is replying with an ILLEGAL REQUEST/INVALID
FIELD IN CDB error to all the interesting SCSI commands, and to pretty
much anything hdparm sends it.

I've also tried throwing sdparm at it. The only page sdparm can get out
of it is the basic identification page:

[anovak at octagon hdparm-9.54]$ sudo sdparm -i /dev/sdg
    /dev/sdg: Seagate   Backup+ Hub BK    D781
Device identification VPD page:
  Addressed logical unit:
    designator type: NAA,  code set: Binary
      0x5000000000000001

But this has convinced me that I am actually communicating with the disk
itself. Is there any way the kernel/driver could be tinkering with the
commands that hdparm used to send that worked and which now fail? Or is
there some kind of initialization that isn't being done that would put
the disk in a mode where it is willing to do more things?

-- 
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to hdparm in Ubuntu.
https://bugs.launchpad.net/bugs/1774569

Title:
  gsmartcontrol, hdparm, and ZFS all refuse to talk to an apparently
  working Seagate Backup+ Hub drive after upgrade to 18.04

Status in gsmartcontrol package in Ubuntu:
  New
Status in hdparm package in Ubuntu:
  New
Status in zfs-linux package in Ubuntu:
  New

Bug description:
  I recently upgraded from 17.10 to 18.04. After the upgrade, I noticed
  that my Seagate Backup+ Hub external drive was displaying a series of
  puzzling symptoms:

  1. gsmartcontrol can't get SMART data from the drive. I am pretty sure
  it used to report SMART data? Here's a log of it not working:

  <warn>  [hz] Warning: exit: Command line did not parse.
  <warn>  [app] execute_smartctl(): Smartctl binary did not execute cleanly.
  <warn>  [app] StorageDevice::execute_device_smartctl(): Smartctl binary did not execute cleanly.
  <warn>  [app] SmartctlParser::parse_section_info_property(): Unknown property "Physical block size"
  <warn>  [app] SmartctlParser::parse_section_info_property(): Unknown property "Logical Unit id"
  <warn>  [app] SmartctlParser::parse_section_info_property(): Unknown property "Temperature Warning"
  <warn>  [app] SmartctlParser::parse_section_data(): Unknown Data subsection encountered.
  <warn>  [hz] Warning: exit: Some SMART command to the disk failed, or there was a checksum error in a SMART data structure
  <warn>  [app] SmartctlParser::parse_section_info_property(): Unknown property "Physical block size"
  <warn>  [app] SmartctlParser::parse_section_info_property(): Unknown property "Logical Unit id"
  <warn>  [app] SmartctlParser::parse_section_info_property(): Unknown property "Temperature Warning"
  <warn>  [app] SmartctlParser::parse_section_data(): Unknown Data subsection encountered.

  2. hdparm used to be able to spin down the drive. I had it configured
  to spin it down after a few minutes of inactivity, in the hdparm
  config file. Now that no longer happens, and hdparm can't seem to talk
  to the drive meaningfully at all:

  [anovak at octagon ~]$ sudo hdparm -I /dev/sdb

  /dev/sdb:
  SG_IO: bad/missing sense data, sb[]:  70 00 05 00 00 00 00 0a 00 00 00 00 24 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

  ATA device, with non-removable media
  Standards:
  	Likely used: 1
  Configuration:
  	Logical		max	current
  	cylinders	0	0
  	heads		0	0
  	sectors/track	0	0
  	--
  	Logical/Physical Sector size:           512 bytes
  	device size with M = 1024*1024:           0 MBytes
  	device size with M = 1000*1000:           0 MBytes 
  	cache/buffer size  = unknown
  Capabilities:
  	IORDY not likely
  	Cannot perform double-word IO
  	R/W multiple sector transfer: not supported
  	DMA: not supported
  	PIO: pio0 
  [anovak at octagon ~]$ sudo hdparm -y /dev/sdb

  /dev/sdb:
   issuing standby command
  SG_IO: bad/missing sense data, sb[]:  70 00 05 00 00 00 00 0a 00 00 00 00 24 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

  I think this may be related to https://askubuntu.com/questions/1037997
  /upgraded-to-18-04-usb-harddrive-doesn-t-idle-anymore which is someone
  else having the same problem.

  3. The ZFS tools think the drive is hosed:

  [anovak at octagon ~]$ sudo zpool status hub
    pool: hub
   state: UNAVAIL
  status: One or more devices could not be used because the label is missing 
  	or invalid.  There are insufficient replicas for the pool to continue
  	functioning.
  action: Destroy and re-create the pool from
  	a backup source.
     see: http://zfsonlinux.org/msg/ZFS-8000-5E
    scan: none requested
  config:

  	NAME                               STATE     READ WRITE CKSUM
  	hub                                UNAVAIL      0     0     0  insufficient replicas
  	  ata-ST6000DM003-2CY186_ZF200PC8  UNAVAIL      0     0     0

  This may be related to the drive having adopted a new /dev/disk/by-id
  name during the upgrade? I think it was "ata-
  ST6000DM003-2CY186_ZF200PC8" when I added it to my zpool by its
  symlink under /dev/disks/by-id, but now it is "usb-
  Seagate_Backup+_Hub_BK_NA8TQC87-0:0":

  [anovak at octagon ~]$ ls -lah /dev/disk/by-id/usb-Seagate_Backup+_Hub_BK_NA8TQC87-0\:0
  lrwxrwxrwx 1 root root 9 May 31 20:52 /dev/disk/by-id/usb-Seagate_Backup+_Hub_BK_NA8TQC87-0:0 -> ../../sdb

  This *shouldn't* cause trouble; you should be able to export the zpool
  and re-import it under the new name. But zpool import shows nothing to
  import:

  [anovak at octagon ~]$ sudo zpool import
  no pools available to import

  And I also can't export or even destroy the busted zpool, because
  zpool doesn't think it exists for exporting or destroying purposes:

  [anovak at octagon ~]$ sudo zpool export hub
  cannot export 'hub': no such pool or dataset
  [anovak at octagon ~]$ sudo zpool destroy hub
  cannot destroy 'hub': no such pool or dataset

  4. The weirdest thing is that the drive itself seems to be working
  correctly. I see /dev/sdb1 and /dev/sdb9, as expected for a ZFS drive.
  I can `cat /dev/sdb1 | xxd | less` and see the data stored on the
  drive, including what I think is the ZFS label (at 0x4000, with a
  bunch of ZFS-y strings in it) that zpool is upset about not seeing. I
  see the partitions in `gparted` just fine, too; there's no indication
  that there's anything wrong with the partition table. Even the
  device's integrated USB hub seems to be working fine. This is
  definitely not a hard drive failure.

  
  If I had to speculate, I would guess that the drive is being treated as a generic USB mass storage device now, when it used to be being handled as a SATA device in a USB-to-SATA enclosure (which I think it is). That would explain the name change, and the difficulty that hdparm and gsmartcontrol have in talking to it. The ZFS weirdness with not being able to export/destroy the pool has to be another issue; it happens even when the drive is disconnected from the system entirely.

  ProblemType: Bug
  DistroRelease: Ubuntu 18.04
  Package: gsmartcontrol 1.1.3-1
  ProcVersionSignature: Ubuntu 4.15.0-22.24-generic 4.15.17
  Uname: Linux 4.15.0-22-generic x86_64
  NonfreeKernelModules: zfs zunicode zavl icp zcommon znvpair
  ApportVersion: 2.20.9-0ubuntu7
  Architecture: amd64
  CurrentDesktop: ubuntu:GNOME
  Date: Thu May 31 20:46:52 2018
  InstallationDate: Installed on 2017-08-06 (298 days ago)
  InstallationMedia: Ubuntu 17.04 "Zesty Zapus" - Release amd64 (20170412)
  ProcEnviron:
   TERM=xterm-256color
   PATH=(custom, no user)
   XDG_RUNTIME_DIR=<set>
   LANG=en_US.UTF-8
   SHELL=/bin/bash
  SourcePackage: gsmartcontrol
  UpgradeStatus: Upgraded to bionic on 2018-05-29 (3 days ago)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/gsmartcontrol/+bug/1774569/+subscriptions



More information about the foundations-bugs mailing list