ZFS Commands In "D" State

[Tim Gustafson]

> procstat -kk 1425 ? (or whatever PID's that is stuck in "D" state)

I went over to my terminal to get that information for you, and found
that the "zpool scrub" command I had started when I first posted this
question had *finally* completed, and the scrub is now running, but
not making much progress:

  scan: scrub in progress since Thu Jun  8 15:15:59 2017
        1 scanned out of 103T at 1/s, (scan is slow, no estimated time)
        0 repaired, 0.00% done

So, in 50 minutes it's scanned 1 byte out of 103T.

Since the scrub command completed, I decided to try another ZFS send.
That did enter the D state as soon as it started transmitting data
again, and the output of the command you suggested is:

  PID    TID COMM             TDNAME           KSTACK
 1646 101975 zfs              -                mi_switch+0xd2
sleepq_wait+0x3a _cv_wait+0x194 zio_wait+0x5b arc_read+0x8f0
dmu_objset_open_impl+0xed dmu_objset_from_ds+0x65
zfs_ioc_snapshot_list_next+0x178 zfsdev_ioctl+0x5f5
devfs_ioctl_f+0x13f kern_ioctl+0x2d4 sys_ioctl+0x171
amd64_syscall+0x4ce Xfast_syscall+0xfb

-- 

Tim Gustafson
BSOE Computing Director
tjg at ucsc.edu
831-459-5354
Baskin Engineering, Room 313A

To request BSOE IT support, please visit https://support.soe.ucsc.edu/
or send e-mail to help at soe.ucsc.edu.