Re: shutdown no longer umounts idle USB connected drive.

On 9/5/26 12:04, Warner Losh wrote:
> 
> 
> On Sat, Sep 5, 2026, 9:04 AM Ian FREISLICH <ianfreislich@gmail.com 
> <mailto:ianfreislich@gmail.com>> wrote:
> 
> 
> 
>     On 9/5/26 09:35, Christos Margiolis wrote:
>      > On Sun Aug 23, 2026 at 7:37 PM EEST, Ian FREISLICH wrote:
>      >> On 8/23/26 10:55, Christos Margiolis wrote:
>      >>> On Sun Aug 23, 2026 at 12:05 AM EEST, Ian FREISLICH wrote:
>      >>>> Hi
>      >>>>
>      >>>> I have a USB connected hard drive which powers down when idle.
>     Recently
>      >>>> (in the last 2 weeks or so) shutdown appears to have stopped
>     waiting for
>      >>>> the drive to spin up in order to cleanly unmount. I can
>     perform a git
>      >>>> bisect but the system is relatively slow so if anyone has a
>     change in
>      >>>> mind that might be the culprit, I can start around that point.
>      >>>>
>      >>>> Ian
>      >>>
>      >>> Which driver is it using? Anything showing up in dmesg?
>      >> [router] ~ # usbconfig -d ugen0.2 show_ifdrv
>      >> ugen0.2: <External SATA Hard Drive Adapter cable PA023U3 Other World
>      >> Computing> at usbus0, cfg=0 md=HOST spd=SUPER (5.0Gbps) pwr=ON
>     (224mA)
>      >> ugen0.2.0: umass0: <ULT-Best Best USB Device, class 0/0, rev
>     3.00/41.01,
>      >> addr 1>
>      >>
>      >> I started getting these after a kernel built on July 28, but
>     only during
>      >> the cron.daily job:
>      >>
>      >> Jul 31 03:06:53 router kernel: (da0:umass-sim0:0:0:0): READ(10).
>     CDB: 28
>      >> 00 43 22 1e 28 00 00 40 00
>      >> Jul 31 03:06:53 router kernel: (da0:umass-sim0:0:0:0): CAM
>     status: SCSI
>      >> Status Error
>      >> Jul 31 03:06:53 router kernel: (da0:umass-sim0:0:0:0): SCSI status:
>      >> Check Condition
>      >> Jul 31 03:06:53 router kernel: (da0:umass-sim0:0:0:0): SCSI sense:
>      >> ABORTED COMMAND asc:47,1 (Data phase CRC error detected)
>      >> Jul 31 03:06:53 router kernel: (da0:umass-sim0:0:0:0): Retrying
>     command
>      >> (per sense data)
>      >> Jul 31 03:07:01 router kernel: (da0:umass-sim0:0:0:0): READ(10).
>     CDB: 28
>      >> 00 43 22 1e 28 00 00 40 00
>      >> Jul 31 03:07:01 router kernel: (da0:umass-sim0:0:0:0): CAM
>     status: SCSI
>      >> Status Error
>      >> Jul 31 03:07:01 router kernel: (da0:umass-sim0:0:0:0): SCSI status:
>      >> Check Condition
>      >> Jul 31 03:07:01 router kernel: (da0:umass-sim0:0:0:0): SCSI
>     sense: NOT
>      >> READY asc:4,1 (Logical unit is in process of becoming ready)
>      >> Jul 31 03:07:01 router kernel: (da0:umass-sim0:0:0:0): Polling
>     device
>      >> for readiness
>      >>
>      >> But I now can't reproduce the issue. Oh well.
>      >>
>      >> Ian
>      >
>      > Have you managed to reproduce it since this email or has the
>     issue been
>      > fixed?
> 
>     It's happened once more, I noticed it yesterday but I don't recall
>     which
>     reboot caused it. What's odd is that failure to mount doesn't stop the
>     startup and drop into single user mode and the unclean state doesn't
>     prompt a fsck.
> 
> 
> Yes. It's just a warning that the drive wasn't ready and a retry fixed it.

Well yes, after I fsck and manually mount, retries work. When I reboot 
the system, sometimes that drive is left unclean and never gets mounted 
or fscked.

Ian